Edited by Ioan ILEANĂ Mircea RÎŞTEIU Gheorghe MARC
Smart Electronic Design
Technical support by
Descrierea CIP a Bibliotecii Naţionale a României IoanI leană, Mircea Rîşteiu, Gheorghe Marc
Smart Electronic design: Ioan Ileană, Mircea Rîşteiu, Gheorghe Marc Alba Iulia, Aeternitas, 2014 Bibliogr. ISBN 978-606-613-087-5 I.Rîşteiu, Mircea II.Gheorghe, Marc
©Authors, 2014
Publishing House AETERNITAS ISBN 978-606-613-087-5
Alba Iulia, 2014
CONTENT
Developing a Real Time Data Acquisition System with Low Power Wireless Support. A State of Arts ...... 3 An Algorithm for Multi-objective Optimization: Controlled Mutation Evolutionary Algorithm ...... 9 Design and Simulation Framework for Indoor and Outdoor LED Lighting Systems ...... 17 Implementation of Petri Nets using Analytical Theory of Markov ...... 21 Considerations Regarding the Evaluation of Documentation and Testing of Low Current Equipment Intended for Use in Explosive Atmospheres ...... 25 Method of Storing Relevant Medical Information Based on Biometric Identification ...... 29 Realization of a Performant Stand for Determining Equivalent Initiation Capability for Detonators with Modern Electronic Measuring Facilities ...... 35 The Importance of Non-Transmission Test of an Internal Ignition for Electrical Motors with Type of Protection Flameproof Enclosure ,,d” ...... 39 Developing Real-Time e-Health Parameter Monitoring Interfaces with 8- bit Microcontrollers ...... 43 Improving Security and Selectivity Requirements in Telemedicine by Combining Hardware and Software Encoders- Decoders ...... 47 PIL Development of a Multiple Channel Wind Turbine Emulator using LabVIEW FPGA ...... 51 Screw Surface Control of the Tool Made on Grinding Devices with Adjustable Step ...... 57 Power Quality Event Monitoring using Data Mining Techniques ...... 61 Integrated Channel Add-Drop Filter Based on Photonic Crystals with 2D Periodicity ...... 65 Comparison of Different Modelling Methods and Measurements for Pulse Width Modulation–Signals ...... 69 A Matrix Model of an Infinite Series of Two-Port Elementary Components ...... 73 Instrumentation to Study D.C. and A.C. Circuits ...... 77 Robots in Our Life. Support or Damage? ...... 81 A New Method to Estimate Reegineering Efficiency in Distributed Network Measurements ...... 85 Cracking the SIP Authentication with a Hybrid Password Cracking Method Specified for SIP Applications ...... 89 Emerging Areas of Photonics and Optoelectronics ...... 95 Development of Testing Methods for Specific Mechanical Parameters of Electric Mining Cables and Accessories ... 99 Author Index ...... 104
1 Technical support by 2 Technical support by Developing a Real Time Data Acquisition System with Low Power Wireless Support. A State of Arts
Risteiu Mircea, Ileana Ioan, Marc Gheorghe, Croitoru Bogdan Computer Science and Engineering Department “1 Decembrie 1918” University of Alba Iulia Alba Iulia, Romania [email protected], [email protected], [email protected], [email protected]
Barosan Ion Computer Science Department, Technical University of Eindhoven Eindhoven, The Netherlands, [email protected]
Abstract— The paper is developing a hardware/software wireless data acquisition system for monitoring a cable car GENERAL VIEW OF MEASUREMENT DESIGN system. For the research purpose, we are focusing on a high variable parameter (vibration of a cable car element) and on a Design constrains low variable (ambient temperature). The real time measurement Using MEMs sensors means low level of voltage, low up system is implemented with a wireless low data transmission rate, to ultra-low power supply requirements, voltage level and the main target is to optimize the entire process from the adapting, and protocol’s conversion (serial with many data energy consuming point of view. We have used two different low level protocol, different media converters, physical level power transceivers on ISM band, we optimized acquisition conversions). procedures, and on the processing point we have recovered the parameters variations. With a high speed current probe we They are some real examples of measurements which are measured the power consumption on both cases, and based on characterized by large amount of data, different sampling dedicated communication protocols we have estimated the periods, different data communication systems, and different consumed energy per bit for the power supply evaluation in processing requirements. Even they are not very fast, the wireless monitoring system. mentioned characteristics are contradictory from the designer point of view. This paper is focused on developing a Keywords— transportation monitoring, DAQ, wireless sensors, microcontroller- based algorithm dedicated to support sensor optimization, energy per bit, data recovery system.
INTRODUCTION Why quality function deployment Most of the science investigation is based on experiments, Data is described by many attributes. On transducers and measurements are the core role for them. In this context, electronic datasheet (TEDs) standard, few TEDS variants have the correct evaluation of measurement uncertainty leads to the been developed, and they are not covering many variations of definition of the confident interval of the measurement values. attributes yet. Even se define the largest list of attributes, it is Based on measurement systems structures, some of very hard to put data together in the same table. In the uncertainties might come from different component blocks database and measurement theory, the concept of arranging (input ports, conditioners, ADC, processing, communication). and accessing data in memory is called data alignment. Because of different parameters attributes, data have to be When some new/or emergency functions are added to the stored into memory as multiple of words sized, in order to device, like self-configuration, self-compensation, or sensor simplify memory access description. For this reason, in some recognition, the discussion goes to a more complicate cases, it might be necessary to insert between a chains of data, approach. In [1], a large analyze in shown over uncertainly some meaningless bytes. Both actions consist of data interval, acceptance criteria, or drift limits. alignment process during gathering the data. Over this memory technical design requirement, at the data user side, the process of entity assessment may vary form one moment
3 Technical support by to other moment of data evaluation. On the other hand, the clock signal, with specific pull-up resistors are needed, and as relationship between signal, features, and defined object, may many SS signals as SPI devices are connected. be changed due to time, triggering process, parameters synchronization. About core of ADC- microprocessor versus microcontroller [1] speaks about house of quality, where they introduce A very short review of the comparison could be done new designing requirement attribute- the time, where the best facing some criteria like: hardware resources, software product quality means the product which respond to the development, prototyping, and training. During the design customer momentary questions. The “voice” of customer analysis stage, some questions will raise: exceeds the technical realities. That means that the - what task system should full-fill; measurement system designer has to implement the easiest - what kind of port system requires; way to meet integration and fusion are used in two major - what is the dimension of the manipulated data; situations: integrating sensors with different parameter - how fast the system should be; attributes in the same support database, and event detection On the same time, the main constrains should be listed: record systems (different parameter attributes means different - what is the accepted implementation cost for the system; situation assessment, and different amount of stored data). - what is the powering limit of the system; - what are the limits of the physical characteristics; COMMUNICATION TECHNIQUES APPROACH - what is the system working environment A microcontroller implies 8-16-24 bus bits; clock The limits of I2C use frequency is 8-16- 32, up to 80 MHz; RAM dimension is less All bytes put on the SDA line must be 8-bits long. The than 128 Mb. On the opposite side, a microprocessor-based number of bytes that can be transmitted per transfer is system has the benefit of 32-64 bus bit, up to 1 GHz clock unrestricted. Each byte has to be followed by an acknowledge frequency, and up to 1024 MB RAM. Some time, the bit. Data is transferred with the most significant bit (MSB) microprocessor-based systems have integrated in one core first (see Fig.6). If a slave can’t receive or transmit another mathematical or graphic processor, and many communication complete byte of data until it has performed some other stacks. function, for example servicing an internal interrupt, it can On the other side, the required power in on criteria that hold the clock line SCL LOW to force the master into a wait offers a large advantage to the microcontrollers, because state. Data transfer then continues when the slave is ready for today’s technology offers ultra-low power implementations, another byte of data and releases clock line SCL. In some with sleeps facilities. cases the message can be terminated by generation of a STOP condition, even during the transmission of a byte. When we are touching the software development subject, few items should be taken into consideration: integrated In I2C transmission protocol, the acknowledge field is development environment (IDE), software components re- obligatory. The acknowledge-related clock pulse is generated usability, library or object oriented programming, debugging, by the master. The transmitter releases the SDA line (HIGH) and software loading. during the acknowledge clock pulse. The receiver will pull down the SDA line during the acknowledge clock pulse so Normally, any microcontroller has an IDE, and a that it remains stable LOW. programming environment like C/C++. The microprocessor- based systems don’t have IDE, but users can develop Running data exchange between two I2C devices means programs and libraries in Python or C/C++ [1]. On the other large amount of time spent “together”. hand, microprocessor-based systems run also using operating systems like Linux, where the development area is much The limits of SPI operation larger. The development possibility is only limited by the SPI is a synchronous protocol that allows a master device hardware/software interfacing. A simple system around to initiate communication with a slave device. Data is microcontroller accepts very easy a hardware interface that exchanged between these devices. The standard SPI operation run faster, and in embedded implementations. means that the SPI bus can operate with a single master device Prototyping is an important phase. By using only and with one or more slave devices connected with the microcontrollers, the project is done faster, with few thirty master device. If a single slave device is used, the Slave Select parts. When we are using microprocessors, if they are not (SS) pin has to be connected logic low. This operation is available system-on-chip, or package-on-package, we need required because the slaves require the falling edge (from high special tools for development, and many external resources to low state transition) of the slave select. SPI protocol is (especially for communication), by we have the opportunity to normally used when few I/O lines are available, but use external frameworks that make the project to work in communication between two or more devices must be fast and cross-platform implementations. easy to implement. Based on these considerations, [2] figure the applications When multiple slave devices are connected, an versus calculation power. independent SS signal is required from the master for each slave device. Let say, if there are 4 slave connected to the master, there has to be 4 SS pins on the master so that they can be connected. The following picture shows how a single master and 3 slaves can be connected together with all the pins shown. With a SPI transmission system, two data wire with
4 Technical support by Calibration TEDS 103 bytes, Total 565 bytes. Out of the 565 bytes, the manufacturer identification consumed 55 bytes, and product description consumed 205 bytes, for a total of 260 bytes or 46% of the TEDS.
Measurement- based on memory segmentation techniques Fast data measurements, according to Niquist theorem, ask about an amount of memory proportional with sampling rates multiplied with time of measurements. On the other hand, when an event has to be monitored continuously, all data between significant event’s parts have to be acquired. A lot of non-significant data ask to be manipulated (send, processed, stored). This situation is very often met in scope measurements, when a fast event is occurred very rarely [5]. But, them experience is useful in our data manipulation Where is the place of IEEE 1451 standard purpose (see figure). IEEE 1451 standard is dedicated to define the interfaces between transducers and microprocessors useful in industrial automation and other fields. On the next figure, there is main layer structure of the sensor hardware-software implementation. The standard defines over a physical interface a data transport serial link [2], with a triggering, interrupting and a hot swap signaling. The IEEE 1451 standard also defines a transducer electronic data sheet, TEDS, that describes the functionality of the transducer in machine- readable form. The interface supports as many as 255 independent transducer channels. These may be accessed Where t is the time between events occurrences, and it is individually or as a unit. The main target of IEEE 1451 variable, non-predictable, and can be very big, by comparing implementation is to dialog with any Network Capable with time for event occurrence. Application Processor (NCAP). In order to make an IEEE 1451 implementation operable, a specific action should be The idea we want to use, for the data manipulation, is carried on: the correction phase. Conceptually, the correction described in next figure. phase operates between a set of registers containing the raw transducer channel values and a set of registers containing application channel values expressed in the appropriate units. When new values for an actuator are provided by the application process the correction engine processes these into the appropriate raw form required by the actuators. So, basically, the sensor integration means that NCAP take the raw data, and then reconfigure it, based the correction engine, in order to store, process, and report it at the user side. As 1451 standard asks, the data manipulation is based on a multivariable polynomial function [2]. Because the specific TEDS contains information defining the data types, units, where tf1, tf2, tf3 represents time window frames, which whether a transducer is a sensor or an actuator, and other are following the signals as trailers. The time window is large information necessary in configuring the correction engine, an enough in order to cover the entire events. The main question segmentation process need to be executed. The range of each which is raised up is: “how we can know the time dimension input variable is segmented into regions effectively of the frame?” segmenting the space defined by all variables into regions. For each of these regions, the TEDS will contain a set of After the dimension of the frame is established, the all data coefficients applicable to multivariable polynomial function. is acquired and locally stored, the only significant data is taken The purpose of the segmentation is to allow a trade-off into account for manipulation. Another aspect has to be between the granularity of the segmentation and the degree of discussed- what does mean the significant data? For this the multinomial. For example, if the segmentation was fine important detail, some user interaction updates are required in enough to permit linear corrections in each cell, then the order to “recognize” the event. resulting multinomial for each cell would appear much In our purpose, it is very interesting to resume what Samir simpler. Mekid said [6] “feature recognition in order to help in the On the other hand, Texas Instruments has evaluated the identification of possible failures types and immediately possibility to develop a Smart Transducer Interface Module decide on the actions to be taken”. for running TEDs for IEEE 1451 standard [4]. This specific TEDS implementation required the following amount of memory: Meta TEDS 366 bytes, Channel TEDS 96 bytes,
5 Technical support by SYSTEM DEVELOPMENT FOR DEDICATED APPLICATIONS Reasons for embedded systems in measurements Block structure of any measuring system, based on 1. processing functions, consists in: 2. self- compensation 3. reconfigurable measurement systems
LIS331DLH: ±2G /±4G /±8G DIGITAL OUTPUT HIGH PERFORMANCE ULTRA-LOW-POWER 3-AXIS ACCELEROMETER SYSTEM DEVELOPMENT The device is a high performance ultra-low-power 3-axis linear accelerometer; it is a part of the “nano-” family of MEMS accelerometers with digital I2 C/SPI serial interface The upper figure, from hardware architecture point of view standard output, and integrates in its applications sleep and can be represented by an external bus, like in next figure. wake-up functions.
Based on its dynamically user-selectable full scales of ±2g /±4g /±8g, it is capable of measuring acceleration with output data rates from 0.5 Hz to 1 kHz. The device can be configured to generate interrupt signals as a response to inertial wake-up events, or based on the position of the device during the The left side core is used for processing and data measurements. controlling over the external bus. The right side core controls the communication protocols. When the device is powered up it automatically starts calibration procedure by downloading the calibration Without entering into details, an example of coefficients from the embedded local Flash memory to the communication implementation is shown next. internal registers. Then, in order to get data, it is necessary to select one of the operating modes through the CTRL_REG1 register [1], in order to enable at least one of the axes. In the [1] is also shown the register table with the meaning them. For instance, when the accelerometer value is changed, the register status changed the associated value as in next table: STATUS ZYX XO ZYX _REG 27h OR ZOR YOR R DA ZDA
This facility is necessary in order to optimize the reading procedure- if the value is changed, it will be red, like in next programming sequence [1]: read STATUS_REG if STATUS_REG(7) = 1 then some data have been overwritten read OUTX_L OUTX_H The microcontroller W5200, in order to communicate in read OUTY_L OUTY_H Internet network, will have loaded in the specific application. read OUTZ_L OUTZ_H The combined hardware/software architecture looks like next do something figure. According to [1] each value is represented as two bytes, so it is necessary to read both bytes for getting one axe value. According to Nyquist theorem related to sampling, this MEM device can be configured to deliver data to the output in a selectable output data rate from 50 Hz (DR1=0, DR0=0, from CTRL_REG1), to 1000 Hz (DR1=1, DR0=1, from
CTRL_REG1) [1]. Each application/ or engineering field rules under specific regulations and requirements. Sensing elements could be: On the other hand, according to [1], the reading time pressure, level, flow, temperature, vibration, acceleration, sequence, to prevent data loss, is represented in next figure. voltage, current. After condition, the ADC input will get 0- 3V, 0-5V, 0-10V linear or nonlinear.
Self- compensation: - gain error - non-linear transfer curve where T0 is reading time and is taken into account in - offset error Nyquist sampling rate formula; T1 is register reading time; T2 - hysteresis error is the new data generation- 616 µsec according to [1] . - gain variation with temperature/ time
6 Technical support by In order to develop smart applications the sleep to wake land grid array (LGA) package, and is guaranteed to operate function must be used in conjunction with the management of over an extended temperature range from -30 to +85 °C. Power mode facility. The LIS331DLH can be set in a low- power operating mode, characterized by lower date rate The main features we are taking into account are: refreshments (5 µA at 0.5 Hz [1]) or in normal mode with ■ Analog supply voltage: 2.5 V to 3.3 V higher data rate refreshments (570 µA at 1000 Hz [1]). By ■ Digital supply voltage IOs: 1.8 V using this logic, there is a possibility to continue sensing ■ Power-down mode acceleration and generating interrupt requests, even the device ■ 3 magnetic field channels and 3 acceleration channels is “sleeping”. ■ ±1.3 to ±8,1 gauss magnetic field full-scale ■ ±2g /±4g/±8g dynamically selectable full-scale The next figure shows the combined data acquisition with ■ 16-bit data out wake up threshold signal and associated new valid reading. By ■ I2C serial interface understanding this mechanism, the measured sets of data are ■ 2 independent programmable interrupt generators for optimized for continuous acceleration sensing and significant free-fall and motion detection data acquisition periods. ■ Embedded self-test ■ Accelerometer sleep-to-wakeup function The two figures offer solutions for microcontroller-based ■ 6D orientation detection data acquisition in order to reduce the associated hardware resources. The dimension of the reading time sequence is In [2] there is shown the meaning of compass coordinate settled according to Nyquist sampling formula. system. With the notations as follow- around the Zb axis clockwise at an angle (ψ) with the view from the origin to If there are enough time windows between readings, downwards; around Yb at an angle (ρ) with Xb moving multiple sensors readings could be performed. upwards; around Xb at an angle (γ) with Yb moving downwards- pitch and roll calculations are done in [2].
LSM303DLH, SENSOR MODULE: 3-AXIS ACCELEROMETER AND The rotation matrixes are: 3-AXIS MAGNETOMETER cos sin 0 , cos 0 sin 1 0 0 , The LSM303DLH is a system-in-package featuring a 3D R sin cos 0 R 0 0 0 R 0 cos sin digital linear acceleration sensor and a 3D digital magnetic 0 0 1 sin 0 cos 0 sin cos sensor. The various sensing elements are manufactured using specialized micromachining processes, while the IC interfaces For the used accelerometer raw measurements are Ax, Ay, are realized using a CMOS technology that allows the design and Az which are signed integer in terms of LSBs. In this of a dedicated circuit which is trimmed to better match the frame, the Ax1, Ay1, and Az1 are the normalized values after sensing element characteristics [1]. The LSM303DLH has a applying accelerometer initial settings with the meaning the linear acceleration full-scale of ±2g / ±4g / ±8g and a magnetic floating point values less than 1 in terms of g (earth gravity), field full-scale of ±1.3 / ±1.9 / ±2.5 / ±4.0 / ±4.7 / ±5,6 / ±8.1 and the root sum of their squared values should be equal to 1 gauss , both fully selectable by the user. The LSM303DLH when the accelerometer is still. includes an I2C serial bus interface that supports standard According to [1], [2], the matrix values of the measured mode (100 kHz) and fast mode (400 kHz). The internal self- accelerations is: test capability allows the user to check the functioning of the whole module in the final application. The system can be Ax1 cos cos cos sin sin 0 A cos sin sin cos sin cos cos sin sin sin cos sin .0 configured to generate an interrupt signal by inertial y1 wakeup/free-fall events, as well as by the position of the Az1 cos sin cos sin sin sin cos sin cos sin cos sin 1 device itself. Thresholds and timing of interrupt generators are then the pitch/roll values are: programmable on the fly by the end user. Magnetic and accelerometer parts can be enabled or put in power-down Pitch arcsin(Ax1 ) , Roll arcsin(Ay1 / cos ) mode separately. The LSM303DLH is available in a plastic
7 Technical support by CONCLUSIONS For measuring the data with this sensor, powering up In our setup, by using routing facilities, we consider a data conditions must be accepted, and then the controller has to transmission session, the entire stage of from data sensing to wait for the measurement to complete. This takes a maximum PC data storage. In between this to end points, they are of 20/80/320 ms for a 8/12/14bit measurement [8]. The time integrated some devices. varies with the speed of the internal microcontroller’s oscillator and can rich the level up to 30%. The controller Our proposed architecture started from two different must wait for the Data Ready signal before restarting SCK to approaches. First of all there are a lot of device which are readout the data. Measurement data is stored until readout, “connected” to the web. For instance, publishing room therefore the controller can continue with other tasks and temperature on the web is very easy today. On the other hand, readout at its in our lab we have tested a web-based application for green energy measurement. The core device around our convenience. This detail is used in our measurement implementation is a simple wireless card built with 802.11b/g implementation for forcing “parallel” tasks. Two bytes of protocol. measurement data and one byte of CRC checksum (optional) will then be transmitted. The microcontroller must From wiring and signal consideration points of view, only acknowledge each byte by pulling the DATA line low. All two ideas must be underlined here. First idea is related to the values are sent with MSB first. wire length when we are carrying the SCK and DATA signal parallel and in close proximity (e.g. in wires) for more than 10cm may result in cross talk and loss of communication, REFERENCES according to [8]. As for all digital devices, this restriction may be resolved by routing VDD and/or GND between the two data signals and/or using shielded cables. Since the interface [1] Jamie Coble, Pradeep Ramuhalli, Ryan Meyer, and Hash Hashemian, Online sensor calibration assessment in nuclear power stations, IEEE consists of fully static logic there is no minimum SCK Instrumentation & Measurement Magazine, Vol.16.No.3, June 2013, frequency. ISSN 1094-6969 Second related idea is related to data pin. The tri-state [2] John Eidson, IEEE1451.2 Correction Engine Applications Intergrated DATA pin is used to transfer data in and out of the sensor. As Solution, HP, HPL-98-143 August, 1998 [3] http://www.webopedia.com/quick_ref/OSI_Layers.asp, January, 21st in any I2C interface, for sending a command to the sensor, 2013 DATA is valid on the rising edge of the serial clock (SCK) [4] http://ieee1451.nist.gov/senoct11-2col.pdf, February 3rd, 2013 and must remain stable while SCK is high. After the falling [5] http://cp.literature.agilent.com/litweb/pdf/5989-7833EN.pdf, February edge of SCK the DATA value may be changed. For safe 2nd , 2013 communication DATA valid shall be extended in time before [6] Samir Mekid, Further Structural Intelligence for Sensors Cluster the rising and after the falling edge of SCK, respectively. For Technology in Manufacturing, Sensors 2006, 6, 557-577, Special Issue reading data from the sensor, DATA is valid TV after SCK "Intelligent Sensors" has gone low and remains valid until the next falling edge of SCK.
8 Technical support by An Algorithm for Multi-objective Optimization: Controlled Mutation Evolutionary Algorithm
Rotar Corina Departament of Computer Science “1 Decembrie 1918” University Alba Iulia, Romania
Abstract — The paper presents a novel multi-objective their ability in quickly finding the approximations of the optimization evolutionary algorithm (MOEA). The main idea is multiple solutions (convergence). These two issues are linked to design an intuitive evolutionary technique which substitutes since by emphasizing of one would disfavors the other and, complicated procedures for keeping population diversity and therefore, each MOEA needs a proper way to balance them. speeding convergence to the Pareto front. Controlled Mutation Evolutionary Algorithms (CMEA) is enhanced with a procedure Fitness assignment is another critical subject in EMOO. In that estimates the closeness to the target and the crowdedness in one-criterion optimization problems, it is relative simply to the vicinity of each potential descendant by using several decide which solution is better than the others as it is not reference parents. In this manner, CMEA does not need extra problematic to establish an order relation based on the values parameter for estimating the crowding and encourages surviving of the objective functions. Contrarily, when more criteria are those individuals which are closer to the Pareto optimal solutions. involved it is necessary to find a relevant definition of the Our algorithm does not use the concept of Pareto dominance to optimum. Pareto definition of the non-dominance provides a determine the performance of a candidate; instead, Preference proper basis in the fitness assignment as proven by many state notion is defined. The proposed algorithm is validated using of the art MOEAs (NSGA [12], NSGAII [8], SPEA2[20]). several standard test problems and popular performance indicators. The experiments show that the proposed approach is The rest of the paper is organized as follows. Section 2 competitive and can be considered a viable alternative to solve presents a brief overview of some of the most popular Pareto- multi-objective optimization problems. based multiobjective evolutionary techniques and their features. Section 3 describes the CMEA algorithm and Keywords— multi-objective optimization, preference condition, introduces the definition for the Preference notion in the controlled mutation multiobjective context. Section 4 presents the results of the experiments and a comparative analysis of CMEA technique I. INTRODUCTION with a well-known algorithm for multiobjective optimization: NSGAII. Finally, the conclusions of the paper are given in Because of the generality and the flexibility of the section 5. evolutionary algorithms, their popularity expands continuously. Increasingly more, these algorithms became a suitable alternative to the traditional approaches in solving LITERATURE OVERVIEW hard optimization problems. One of the major areas of the The most popular concept of optimality for multicriteria Evolutionary Computation is the evolutionary multiobjective optimization problems, namely Pareto optimality, classifies optimization (EMOO), where the research activity expands the evolutionary multicriteria techniques into two categories. and provides numerous suitable techniques. Although the The Pareto techniques are evolutionary techniques for literature reveals many efficient evolutionary techniques for multicriteria optimization, which explicitly make use of Pareto multiobjective optimization, there is still a continuous interest optimality definition. Non-Pareto techniques are tools that do for developing new algorithms or improving the existing ones. not use the concept of Pareto optimality in the assessment of Evolutionary algorithms prove to be preferable techniques individuals. for multiobjective optimization problems (MOPs). As long as, Well-known methods as NSGA2[8] and SPEA2 [20] make because of the intrinsic parallelism, EAs are able to discover, use of Pareto dominance by explicitly calculating individual in a single run, not one but many solutions, this feature makes performance. CMEA, in turn, uses a new relaxed concept for them appropriate for MOPs. Also, the standard EA is more a evaluating the solutions (preference). template than a strictly-designed algorithm by giving the opportunity to develop, to improve and to be creative in Multicriteria optimization problem represents a situation finding new techniques of that kind. As a proof of the where premature convergence of the evolutionary algorithm to importance of this subject, the EMOO repository [15] offers a a unique solution prevents the detection of Pareto optimal set. comprehensive list of the bibliographic references on EMOO. Correcting this drawback consists in enriching evolutionary algorithm with an additional mechanism that allows obtaining, Two main issues of the multiobjective optimization finally, a better spread of the solutions. This is closely related evolutionary algorithms (MOEAs) are: their ability in to the degree of diversity of the population. Regarding the maintaining the diversity among the possible solutions and diversity preservation there were proposed many mechanisms
9 Technical support by including clustering [10], [16], considering the diversity as a Ph ,h Ph ,h (3) new objective [14], crowding entropy-based diversity measure j k k j strategy [17], adaptative mutation [13]. Overall preference of h against the other solutions from the Diversity mechanism described in this paper uses a i controlled mutation operator and a survival mechanism that current generation Population(t) is computed as follows. mutually maintain the required diversity degree. This approach has the advantage of simultaneously assessing card(Population) (4) diversity and solution quality in a single fitness value and also Phi ,Populationt Phi ,hk k 1 has the advantage of being parameter free insofar as the number of objectives in the problem, is used to dictate the Each solution obtained by using a variation operator (for number of reference parents required. instance: mutation) can be evaluated according to the diversity preservation and the convergence requirements. Therefore for ALGORITHM DESCRIPTION a bi-objective optimization problem, two parents - p1 and p2 are selected as reference points in the objective space. For Controlled Mutation Evolutionary Algorithm (CMEA) these two reference points, the middle point and the target works with a constant size population of individuals which point (ideal vector in objective space) are determined as encodes the possible solutions from the search space. The follows: search process is conducted through a variation operator and a selective replacement procedure which will be described (5) p1. f1 p2. f1 p1. fm p2. fm further in this paper. An alternative solution to the non- Middle ( ,..., ) dominance concept is described through the preference notion. 2 2 The preference value counts the objectives for which an (6) individual is better than others from the same population. Target (minp1. f1, p2. f1,...,minp1. fm, p2. fm)
According to the desideratum of establishing the balance Two measures are computed: an indicator of diversity and between exploration and exploitation of the search space, the an indicator of convergence. First one represents the distance modified fitness function for qualifying a possible descendent to the middle point of the line segment between two reference takes into account two computed values: one to quantify the parents: diversity of the population and the second to measure the (7) closeness to the Pareto front. Selection mechanism promotes Idiversityc Distance(c,Middle) the better individuals to become reference points for further improvement of the local search zone. The principal units of Second measure represents the inverse of the distance to CMEA will be described next. the target point corresponding to ideal vector:
A. Fitness assignment 1 (8) Iconvergenc ec An alternative technique for computing the performance of 1 Distance(c,Target) possible solutions is described next. The following formulas Fitness of an individual is computed by using the present the manner in which each solution is qualified for a bi- following formula: objective optimization problem. Let us consider a minimization problem involving f1, f2,...,fm objectives. Let Fitness(c) Pc,PopulationtI c I c (9) diversity convergence be h j and hk - two possible solutions of the same generation. The idea is to prefer from among the new created Definition1. Preference of h j against hk is computed as descendants the ones which are closer to the target point and also far from the middle point. The principle is intuitive: follows: closeness to target point is related to the requirement of fast (1) convergence; deviation from the middle point induces the Ph ,h card{i f h f h ,i 1...m} j k i j i k diversity preservation.
Ph j ,hk - represents the number of objectives in which h j is strictly better than hk . Also,
Ph ,h m Ph ,h (2) k j j k
Definition2. We state that solution h j is preferred against hk if and only if: Fig. 1. Graphical representation for the reference parents, middle and target points in objective space.
10 Technical support by Centroid point (middle point for bidimensional space) procedure rises accordingly. For this reason, the proposed represents the ideal descendent according to the similarity approach maintains the number of successors to a smaller with both parents as it is the unique descendent which inherits value (for instance k=m). the same quantity of genetic material from the parents. Also, Controlled_Mutation by generating the centroid points for all mating parents the diversity of the population decreases dramatically. The (Input:PT(i),Output: Parents(PT(i),New) similarity with both parents in the same proportion makes the For k=1 to NoObj centroid point a weak candidate for the next generation in Parents(PT(i),k)= Select(PNew) order to preserve the diversity of the population. Otherwise, Next k similar solutions will proliferate during evolution. Compute Centroid(Parents(PT(i))); The target point represents the ideal descendent according Compute Idiversity to the environment. It inherits best features of the parents, and Compute Target(Parents(PT(i))); it is qualified to survive into the next generation. Compute Iconvergence Unfortunately, none of the ideal or centroid points could Offspring = {} certainly be created or excluded by using probabilistic/random For k=1 to NoObj mutation or crossover. Therefore, among the anticipated Offspring = Offspring Mutation (PT(i)); offspring, those which are closer to the ideal and far from the centroid are favored for survival. Next k Evaluation(Offspring); For an m-objective minimization problem, where m>2, the New=Best(Offspring) descendant is created by taking into account m reference parents (pi), i=1..m, and choosing the mutated individual that End is the closest to the target point and the farthest from the The fittest among the newly-generated individuals centroid in the vector space. The generalized fitness survives, and compete with the parents. Two variants of the assignment formula is computed accordingly. parents’ replacement are described next. The first strategy is based on the non-domination condition. Instead the second Considering: one uses only the preference value in comparing two candidates. In both situations the new individual might survive, and replace its weaker parents. pi. f pi. f Cetroid ( 1 ,..., m ) First variant of the survival strategy assures a better m m convergence to the Pareto front when there are no local Pareto fronts which could attract the population. The second variant Target (minpi. f1,...,minpi. fm) (11) of the survival strategy appears to be more suitable for the many local Pareto fronts multi-objective problems, being less elitist regarding the preservation of the non-dominated The fitness of the descendent is computed by next formula, solutions. where alpha - represents a constant (e.g. alpha=1) used to The fitness of the descendants is computed according to prevent division by 0 when individual c overlaps the target: the formula (12).
Pc,Populationt Distance(c,Centroid) Survival_variant_I(PT(i), Parents(PT(i)), New) Fitness(c) alpha Distance(c,Target) (12) For each PNew(k) from Parents(PT(i))
If Fitness(New)>Fitness( PNew(k)) AND
B. Controlled Mutation operator Nondominated( PNew(k))=false Controlled Mutation operator represents the main variation PNew(k)=New ////replace weaker parent operator of the CMEA algorithm. The principle of the endIf mutation procedure relies on obtaining the improved offspring next k at each stage of the search process. The operator involves m reference parents that are selected according to preference End values by using a binary tournament selection mechanism. The Survival_variant_II(PT(i), Parents(PT(i)), New) parents produce one descendent that will further replace a //survival mechanism exclusively based on Preference weaker parent. A standard mutation mechanism is applied for For each PNew(k) from Parents(PT(i)) each selected parent. The result consists in several possible If Preferred(New, P (k))=true descendants which further compete for survival. In order to New induce a higher selection pressure in the decision process of PNew(k)=New //replace parent the outcome of the controlled mutation procedure, the number endIf of the mutated descendants of each parent could be increased. Next k Therefore, the basin of possible competitive individuals End expands, but in this case the computational cost of the
11 Technical support by C. Replacement mechanism Pareto front. NSGA II algorithm represents one of the a state The mutation endowed with the replacement procedure is of art techniques for multiobjective optimization. an elitist mechanism, by not allowing the generation of a In the first scenario, both algorithms CMEA and NSGAII weaker individual and also a regression in the search process. runs 10 times; each time the termination condition is given by The result of the mutation procedure can replace an individual attaining the maximum number of 100 generations. The of the current generation. In this phase, the quality of Pareto original NSGAII algorithm uses Simulated Binary Crossover non-dominance is taken into consideration. Therefore, a (SBX) and Polynomial crossover. A population of 100 restriction is made: a Pareto non-dominated individual from individuals is used. The parameter settings for NSGAII are: the last generation will not be replaced by any newly crossover probability pc=0.9 and mutation probability is generated individual. This practice assures that none of the pm=1/n, where n is the number of decision variables. The proven Pareto non-dominated solutions of the current distribution indices for crossover is 15 and distribution index population will not be replaced unless its statute of non- for mutation operator is 20 [7]. dominance shifts during the evolution process. Obviously, the non-dominance feature of a solution could be changed in the The main variation operator of CMEA is uniform mutation next generation, but, through the restrictive replacement with probability pm=1/n, where n represents the number of mechanism, the replacement of advantageous solutions is variables; the mutation operator is applied only if the parent is delayed in a favorable way and the loss of the non-dominated a Pareto non-dominated solution in the current generation. solutions is prevented. Consequently, the stable, promising Also, the mutation operator is applied for generating the zones of the search space survive from one generation to possible candidates of the selected parents, and it is embedded another. into a control mechanism for a wise replacement of the candidates. The number of reference parents for controlled The CMEA algorithm is outlined next. mutation operator is equal to the number of objectives. The CMEA Algorithm size of the population is set to 100 individuals.
Population initialization P0; T=0; In order to evaluate the performance of the proposed While (T Survival_variant_I(PT(i),Parents(PT(i)),New) non-dominated vectors found. Convergence metric evaluate Next i how far from the true Pareto front the solutions obtained in the final population are. For i=1 to sizePop If Nondominated(PT(i))=false then For a close comparison of two non-dominated sets resulted by using different algorithms, Coverage [19] metric is used. P (i)= P (i) T+1 New Coverage computes the fraction of solutions in one set of non- Else dominated solutions (found by one algorithm) that are PT+1(i) = Mutation(PT(i))with probability 1/n dominated by those obtained by the other algorithm. EndIf The following table summarizes the average results of Next i CMEA and NSGA II in 10 runs with 100 generations, T=T+1; regarding Spacing, Convergence and Hypervolume metric. In order to illustrate the performance of the proposed CMEA EndWhile[11 algorithm, we used several well-known test problems ZDT1, End Algorithm ZDT2, ZDT3 [18] as the two objectives test functions, and DTLZ1, DTLZ2 and DTLZ3 [9] problems as the three RESULTS objectives test functions. NSGA-II [8] represents a revised version of the NSGA [12] which is based on a procedure of sorting the individuals TABLE I. CONVERGENCE METRIC COMPUTED FOR THE ZDT AND DTLZ of a given population according to the level of non- TEST PROBLEMS, 100000 FITNESS EVALUATION domination. Deb, Agrawal, Pratap, Meyarivan, overcome the CONVERGENCE metric Average StdDev inconvenients of the first version by designing a fast, elitistic algorithm. The authors proposed a mechanism for estimating ZDT1 CMEA 6.45E-03 1.93E-03 the crowding degree of the solutions eliminating the need of NSGA2 3.48E-02 2.35E-02 defining a user defined parameter for diversity preservation ZDT2 CMEA 6.50E-02 8.78E-03 among the population. NSGA2 1.41E-01 1.22E-01 NSGA-II based on nondominating sorting approach has ZDT3 CMEA 1.33E-03 5.36E-04 proven to be one of the most efficient algorithms for multi- objective optimization due to simplicity, excellent diversity NSGA2 6.05E-03 6.04E-03 preserving mechanism and fast convergence towards the true 12 Technical support by DTLZ1 CMEA 6.58E-02 4.80E-02 In the second scenario, we run each algorithm for 10 times, NSGA2 1.14E+00 8.61E-01 each time a number of 200 generations were produced. For the ZDT test problems, CMEA perform better than NSGA DTLZ2 CMEA 8.59E-03 6.59E-03 regarding Convergence and worse regarding diversity ( , ). NSGA2 6.87E-04 1.87E-04 Average Spacing values are similar as recorded for the first DTLZ3 CMEA 7.19E-02 4.83E-02 scenario which involved 100 generations per run. For 200 generations per run, average Convergence values decrease NSGA2 2.27E+00 1.29E+00 accordingly for both algorithms. As a conclusion, increasing the number of generations for the bi-objective test problems TABLE II. SPACING METRIC COMPUTED FOR THE ZDT AND DTLZ taken into consideration, nothing remarkable happens: CMEA TEST PROBLEMS, 100000 FITNESS EVALUATION preserves its advantage regarding convergence and NSGAII SPACING metric Average StdDev offer a more diverse approximation Pareto front. ZDT1 CMEA 1.43E-02 2.41E-03 For a higher number of generations, we notice that NSGA2 NSGA2 7.18E-03 1.18E-03 overcomes CMEA for DTLZ1 test problem, both in respect ZDT2 CMEA 1.64E-02 5.01E-03 with convergence and spacing. An explanation could be the fact that keeping the non-dominated solutions during the NSGA2 2.64E-03 4.00E-03 transition from one generation to the next one, makes CMEA ZDT3 CMEA 2.37E-02 7.08E-03 unable to converge more and the population gets stuck on a NSGA2 8.31E-03 1.25E-03 local Pareto front. CMEA seems to be unable to quit soon enough the many local Pareto fronts which slow down the DTLZ1 CMEA 3.71E+00 2.64E+00 convergence. Apparently, by using Pareto non-dominating NSGA2 1.72E+00 1.90E+00 sorting in order to qualify the solutions and not the Preference DTLZ2 CMEA 1.24E-01 1.29E-02 values as CMEA does, it could be a better choice for solving multiobjective optimization problems which involves many NSGA2 5.88E-02 4.23E-03 local Pareto fronts. Nevertheless, for the second multimodal DTLZ3 CMEA 5.99E+00 3.12E+00 problem- DTLZ3, CMEA preserves its advantage against NSGA2 1.20E+01 7.19E+00 NSGAII as in the previous scenario. Even the generations’ number increases, NSGAII do not succeed to overcome CMEA results regarding the closeness to the true Pareto front. TABLE III. AVERAGE HV FOR 10 RUNS. RESULTS OBTAINED IN 100 GENERATIONS FOR A POPULATION OF 100 INDIVIDUALS Otherwise, the population diversity after 100 and, respectively after 200 generations, is similar for both algorithms as average Hypervolume Metric CMEA NSGA2 spacing metric reveals. Both NSGA II and CMEA show no Average Stddev Average Stddev improvement regarding Spacing metric when the number of generations increases. ZDT1 8.28E-01 6.85E-03 7.75E-01 2.02E-02 ZDT2 7.00E-01 2.31E-02 5.22E-01 7.74E-02 ZDT3 6.42E-01 2.90E-03 5.80E-01 2.40E-02 DTLZ1 9.70E-01 1.04E-02 7.47E-01 1.17E-01 DTLZ2 8.67E-01 6.65E-03 8.90E-01 3.69E-03 DTLZ3 3.71E-01 6.84E-02 1.04E-01 4.09E-02 TABLE IV. FIRST SCENARIO – NSGAII VERSUS CMEA AFTER 100 GENERATIONS Fig. 2. ZDT1: results in 200 generation CMEA, NSGA2 Algorithm ZDT1 ZDT2 ZDT3 DTLZ1 DTLZ2 DTLZ3 Average Convergence NSGA II worse worse worse Worse better worse CMEA better better better Better worse better Average SPACING NSGA II better better better Better better worse CMEA worse worse worse Worse worse better Average HV Fig. 3. ZDT2: results in 200 generation CMEA, NSGA2 NSGA II worse worse worse Worse better worse CMEA better better better Better worse better 13 Technical support by ZDT1 1.00E+00 2.29E-06 8.24E-01 2.43E-02 ZDT2 1.00E+00 4.83E-07 6.37E-01 1.12E-01 ZDT3 1.00E+00 3.24E-09 6.19E-01 1.44E-02 DTLZ1 6.82E-01 5.57E-02 8.29E-01 1.11E-01 DTLZ2 7.00E-01 2.21E-02 8.90E-01 3.91E-03 DTLZ3 6.80E-01 2.90E-02 1.54E-01 6.25E-02 Fig. 4. Figure 1 ZDT3: results in 200 generation CMEA, NSGA2 TABLE VIII. SECOND SCENARIO – NSGAII VERSUS CMEA AFTER 200 GENERATIONS The following tables summarize the comparisons of CMEA and NSGA2 results obtained in 200 generations. Algorithm ZDT1 ZDT2 ZDT3 DTLZ1 DTLZ2 DTLZ3 Average Convergence TABLE V. CONVERGENCE METRIC COMPUTED FOR THE ZDT AND DTLZ NSGA II worse Worse worse Better better worse TEST PROBLEMS, 200000 FITNESS EVALUATION CMEA better Better better Worse worse better Convergence metric Average Std Average SPACING ZDT1 CMEA 5.70E-04 2.84E-04 NSGA II better Better better Better better worse NSGA2 1.01E-02 1.40E-02 CMEA worse Worse worse Worse worse better ZDT2 CMEA 1.47E-03 7.27E-04 Average HV NSGA2 3.39E-02 2.67E-02 NSGA II worse worse worse better better worse ZDT3 CMEA 1.00E-03 4.59E-04 CMEA better better better worse worse better NSGA2 1.08E-02 9.30E-03 DTLZ1 CMEA 6.58E-02 4.80E-02 It appears that the shape of the Pareto front (multimodal, NSGA2 1.14E+00 8.61E-01 linear, spherical) imprints different behavior of the compared DTLZ2 CMEA 8.59E-03 6.59E-03 algorithms: NSGAII converge better toward linear Pareto NSGA2 6.87E-04 1.87E-04 front, and CMEA performs better for spherical Pareto front, when dealing with multimodality. Instead, for the DTLZ2 test DTLZ3 CMEA 7.19E-02 4.83E-02 problem, even the shape of the Pareto front seems to NSGA2 2.27E+00 1.29E+00 advantage CMEA, NSGAII results are better than CMEA results in respect with convergence and diversity, too. TABLE VI. SPACING METRIC COMPUTED FOR THE ZDT AND DTLZ TEST PROBLEMS, 200000 FITNESS EVALUATION GENERATIONS Spacing metric Average Std ZDT1 CMEA 1.45E-02 5.06E-03 NSGA2 7.49E-03 6.31E-04 ZDT2 CMEA 1.52E-02 3.22E-03 NSGA2 3.58E-03 3.78E-03 ZDT3 CMEA 2.50E-02 8.08E-03 NSGA2 8.32E-03 8.95E-04 Fig. 5. DTLZ1: results in 200 generation CMEA, NSGA2 DTLZ1 CMEA 1.51E+00 2.52E+00 NSGA2 1.75E-01 1.30E-01 DTLZ2 CMEA 6.46E-02 7.77E-03 NSGA2 5.58E-02 5.31E-03 DTLZ3 CMEA 1.03E+00 3.43E-01 NSGA2 1.40E+00 7.20E-01 TABLE VII. AVERAGE HV FOR 10 RUNS. RESULTS OBTAINED IN 200 GENERATIONS FOR A POPULATION OF 100 INDIVIDUALS Fig. 6. DTLZ2: results in 200 generation CMEA, NSGA2 Hypervolume CMEA NSGA2 Metric Average Stddev Average Stddev 14 Technical support by Average 2.80e-01 1.72e+00 9.74e-01 1.20e+01 Best 3.65e-02 3.52e-01 2.21e-01 2.55e+00 Worst 2.34e+00 5.62e+00 2.59e+00 2.83e+01 Median 4.85e-02 7.31e-01 7.55e-01 1.07e+01 Std 7.24e-01 1.90e+00 6.72e-01 7.19e+00 Fig. 7. DTLZ3: results in 200 generation CMEA, NSGA2 Third scenario involves only the multiobjective problems with many local Pareto fronts (DTLZ1 and DTLZ3). The replacement procedure of CMEA is based on preference, not on Pareto non-dominance concept, as it is described in the second variant of the survival procedure. Therefore, the best Fig. 8. DTLZ1: results obtained in 100 generations by using preference in candidate according to the altered fitness value will replace its CMEA parents only if the descendent is better in respect with the preference value. In this scenario, the results of CMEA outcome the results of NSGA II regarding closeness to the Pareto front. TABLE IX. presents the average Coverage metric computed for each pair of non-dominated sets obtained by CMEA and NSGAII in 10 runs. Also, a better population diversity than NSGAII is obtained with CMEA as shown in TABLE X. . TABLE IX. CMEA VERSUS NSGA2 REGARDING COVERAGE METRIC FOR DTLZ1 AND DTLZ3 TEST PROBLEMS DTLZ1 Fig. 9. DTLZ3: results obtained in 100 generations by using preference in CMEA CMEA with Non-dominance CMEA with Preference Metric C(A,B) C(B,A) C(A,B) C(B,A) CONCLUSIONS Average 1.00e-04 9.90e-01 9.70e-01 4.40e-01 In this paper, a simple evolutionary algorithm for Best 1.00e-02 1.00e+00 1.00e+00 1.00e+00 multiobjective optimization is proposed. The principle of the algorithm is to encourage the convergence and maintain the Worst 0.00e+00 8.60e-01 4.70e-01 0.00e+00 diversity of the population by estimating the closeness to the Median 0.00e+00 1.00e+00 1.00e+00 4.00e-01 target and the crowdedness in the vicinity of each potential descendant. The core of the algorithm is given by the Std 1.00e-03 1.90e-02 1.00e-01 3.60e-01 controlled mutation operator that uses several reference DTLZ3 parents from the current generation in order to decide the best candidate for replacement. Each candidate is evaluated CMEA with Non-dominance CMEA with Preference according to the preference value, the distance to the target Metric C(A,B) C(B,A) C(A,B) C(B,A) point and the distance to the centroid of the parents’ set. Among the created candidates, the best one replaces the Average 9.20e-02 6.40e-01 6.70e-01 3.80e-01 weaker parents. In this phase, the non-dominance attribute of Best 3.70e-01 1.00e+00 9.70e-01 9.70e-01 the parents represents a significant factor: the parents that are weaker according to the fitness value and also non-dominated Worst 0.00e+00 1.00e-01 9.00e-02 0.00e+00 will be preserved further into the next generation. By using Median 7.00e-02 6.60e-01 7.90e-01 3.10e-01 this mechanism, the promising solutions from a previous generation are kept longer during evolution. Therefore, the Std 9.70e-02 2.60e-01 2.50e-01 2.70e-01 diversity of the population is maintained even the convergence speed of the algorithm is slower. TABLE X. SPACING METRIC COMPUTED FOR CMEA WITH PREFERENCE IN 10 RUNS, 100 GENERATIONS One advantage of the CMEA is given by the parameter- SPACING DTLZ1 DTLZ3 free mechanism of estimating the crowding of the populations. Metric Therefore, by using the replacement mechanism based on the CMEA NSGA2 CMEA NSGA2 altered fitness value (equation 12) there is no need of 15 Technical support by estimating the diversity of the current population. In fact, at [8] Deb, K., Pratap, A., Agarwal, S., and Meyarivan, T., “A fast and elitist each generation, it is anticipated that no crowded multi-objective genetic algorithm: NSGA-II”, IEEE Transactions on subpopulation appears in the next generation. Evolutionary Computation, 6 (2), 182-197, 2002. [9] Deb, K., Thiele, L., Laumanns, M. , Zitzler, E., Scalable Multi-Objective Fitness of an individual is computed by using the Optimization Test Problems. CEC 2002, 825 - 830, IEEE Press, 2002. preference definition. Fitness assignment based on the Pareto [10] Gzara M., Essabri A., Balanced Explore-Exploit clustering based non-domination without a good mechanism of diversity Distributed Evolutionary Algorithm for Multi-objective Optimisation, Studies in Informatics and Control, ISSN 1220-1766, vol. 20 (2), 97- preservation proves to be inefficient. The preference relation 106, 2011. introduces a more relaxed optimality definition than Pareto [11] Schott, J. R. , Fault tolerant design using single and multicriteria genetic non-dominance, in multiobjective space. The fitness algorithm optimization, M.S. thesis, Dept. Aeronautics and Astronautics, estimation mechanism based on preference is accompanied by Massachusetts Inst. Technol., Cambridge, MA, May 1995. the replacement decision based on closeness to the target point [12] Srinivas, N., Deb, K., Multiobjective function optimization using and deviance from the centroid of the reference parents. nondominated sorting genetic algorithms, Evolutionary Computation, vol.2, no.3, pp.221–248, 1995. We analyzed the influence of Preference quality in [13] Tan K.C., Goh C.K., Yang Y.J., Lee T.H., Evolving better population replacement procedure. For MOP with multiple local Pareto distribution and exploration in evolutionary multi-objective front, the replacement based on Preference appears to produce optimization, European Journal of Operational Research 171 (2) 463– better solutions regarding the closeness to the true Pareto 495, 2006. front. [14] Toffolo A., Benini E., 2003. Genetic diversity as an objective in multi- objective evolutionary algorithms. Evol. Comput. 11 (2), 151-167, 2003 On the whole, the proposed algorithm can be considered as [15] The evolutionary multi-objective optimization (EMOO) repository, an alternative to the popular NSGAII as it outperforms the http://www.lania.mx/~ccoello/EMOO/, Cited 1 February 2012. mentioned algorithm for several test problems. Also, it is [16] Wang Y., Dang C, Li H, Han L., Wei J.. A clustering multi-objective clear that CMEA converges better when preference feature is evolutionary algorithm based on orthogonal and uniform design. In taken into consideration for survival procedure. The proposed Proceedings of the Eleventh conference on Congress on Evolutionary Computation (CEC'09). IEEE Press, Piscataway, NJ, USA, 2927-2933, CMEA optimization algorithm can be considered as a viable 2009. method for multiobjective optimization problems that involve [17] Wang Y., Wu L.-H., Yuan X., Multi-objective self-adaptive differential many local Pareto fronts. evolution with elitist archive and crowding entropy-based diversity measure, Soft Computing 14 (3), 193–209, 2010. [18] Zitzler, E., Deb, K., Thiele, L., Comparison of Multiobjective Evolutionary Algorithms: Empirical Results. Evolutionary Computation, 8(2):173-195, 2000. REFERENCES [19] Zitzler, E., Thiele, L., Multiobjective optimization using evolutionary algorithms a comparative case study. In A. E. Eiben, T. Back, M. Schoenauer and H. P. Schwefel (Eds.), Fifth International Conference on Parallel Problem Solving from Nature (PPSN-V), Berlin, Germany, pp. 292 – 301, 1998. [7] Deb, K., Jain, S., “Running Performance Metrics For Evolutionary [20] Zitzler E., Laumanns M., and Thiele L., SPEA2: Improving the Strength Multi-Objective Optimization”, Indian Institute of Technology, Kanpur, Pareto Evolutionary Algorithm. Technical Report 103, Eidgenössische India, Tech. Rep. KanGAL Report Number 2002004, 2002. Technische Hochschule Zürich (ETH), Gloriastrasse 35, CH-8092 Zurich, Switzerland, 2001. 16 Technical support by Design and Simulation Framework for Indoor and Outdoor LED Lighting Systems Ileana Ioan, Risteiu Mircea, Marc Gheorghe, Sofalca Ionut Department of Engineering and Computer Science, “1 Decembrie 1918” University of Alba Iulia, Alba Iulia, Romania [email protected], [email protected], [email protected], [email protected] Abstract — The paper approaches the design of lighting efficiency and color rendering index - CRI) and different systems using power LED’s. Starting with existing solid-state spatial positioning of source. lighting - SSL in the paper we propose a design and simulation technique in order to find the distribution geometry of light In our approach we used a Matlab script that allow the sources (sources number, spatial coordinates and fixing angles) specification of optical and geometrical parameters of sources that simultaneously satisfy the lighting regulation requests and and compute the flux distribution (illuminance) on interest energetic efficiency. A Matlab script was created that allow area, as in fig. 1. Even in the figure the sources have all the simulating and analyzing different spatial distributions of LED same height, the program allow different vertical positioning lighting sources concerning the illuminance of interest area. Our of source. There is no limitation on sources number. main contribution is the modeling and simulation of lighting scenarios involving a large number of LED device types and LIGHTING MODELING AND SIMULATIONS USING MATLAB different spatial lighting configurations, in order to achieve a uniform lighting of the area of interest that stays in conformity We developed a Matlab script that, starting with the type with actual standards. and geometric distribution of the LED lamps compute the illuminance distribution in AOI and also determine the Keywords— LED lighting, spatial sources distribution, isolux average illuminance. The illuminance produce by a lighting curves. source (in our case a LED source) in a point is: I I. INTRODUCTION E cos (1) d 2 In the last years a great effort was dedicated by the producers, local communities and researchers to implementing where I is light intensity, d is the distance between source efficient indoor and outdoor lighting systems based on and incidence point and α is the incidence angle. The semiconductor devices namely power LEDs [2], [3], [6]. coordinate system used is shown in fig. 1. Power LED lighting is more suited for automation, distance managing, and even smart systems compared to classical indoor and outdoor lighting systems. Smart control systems S1(x1,y1,z1) S2(x2,y2,z2) have to monitor the surroundings conditions (traffic, temperature, humidity, natural lighting etc.) and adapt the lighting to them on one side; on the other side they have to z deliver the proper electrical commands for standard compliance with maximum energetic efficiency. For example efficient street lighting systems were presented in [3], [4], [6]. Such systems are of uttermost practical importance, since they y can improve the traffic safety while in the same time maximizing energy savings. Concerning the indoor LED lighting systems there are also many advantages in W comparison with traditional systems: a light more similar to the natural light, energetic efficiency, automation and Area of interest (AOI) intelligent adaptive control. In each situation, an outdoor or indoor lighting system O x must provide the specified (sometimes stated in official L regulations) illuminance of the interest area. This appropriate light flux may be generated using different LED sources Fig. 1. The spatial geometry of our framework (power LED’s with individual directivity diagram, lighting 17 Technical support by For more than one lighting source the overall illuminance Illumination) standard in particular case of a pedestrian will be found by summing the individual illuminances. crosswalk without traffic lights. The illuminance or, in some Usually, the producers provide the light flux Ф of the LED cases the luminance for different category of roads are and no intensity. One can compute the intensity using equation specified by the CIE standard 180:2007 [1] as in the table 1, in 2: which we highlighted the road category for this paper. I (2) TABLE XI. CIE STANDARD FOR ROAD ILLUMINANCE [1] where Ω is the solid angle of emission ant is provided in Category Average level U0 U1 TI Residential areas, pedestrians and 4-5 lux 0,2 n/a n/a LED datasheet many non-motorized vehicles. The LED sources used in our approach have directivity Largely residential, but some 4-5 lux 0,2 n/a n/a properties provided in their datasheets. This directivity motorized vehicles Major access roads, distributors and 0,5 cd/m2 (~8 0,4 0,5 n/a diagrams show the dependency of relative intensity Ir on the minor main roads lux) radiation angle as depicted in fig. 2. Important rural and urban traffic 1,0 cd/m2 (~15 0,4 0,6 20% routes lux) Using this relative intensity, the illumination in a point of High-speed roads, dual 1,5 cd/m2 (~25 0,4 0,7 15% coordinate P(x,y), having an incidence angle α and an angle θ carriageways lux) between the normal to LED surface and the vector SP (where S is the source point) is: The meanings of symbols in table 1 are: . U0: Overall uniformity of luminance, or illuminance, I I () both across and along the roadway, defined as the E r cos (3) d 2 minimum divided by the average. . U1: Uniformity of luminance, or illuminance, along the axis of the road, usually an axis which coincides with a typical driver’s eye position, defined as the ratio of the minimum to the maximum. . TI (threshold increment) defines the glare. As glare has the effect of reducing contrast, a luminaries' "glare performance", or optical control, can be expressed in terms of the increase in background luminance necessary to compensate. The lower the figure the better. Our goal is to find an optimal lighting system having such Fig. 2. Directive characteristic of CLL042-1818A5503M1 LED, used in an architecture that provide energy efficiency but also obey simulations the lighting standard for circulation safety. The geometrical parameters of the simulations are: the length of the interest Before achieving the spatial positioning simulations we area (AOI, fig. 3) L, the width of AOI, W, the height of the evaluated a great number of power LED types in order to find lamp, H, the fixing angle of the lamp β and also the positions, the best concerning the color temperature, CRI and electrical in xOy coordinate system shown in figure 1, of the lighting command values (tension and current) taking into account the pole(s). The Oz axis is vertical. driving system. A good compromise for these requirements is β provided by the power LED's CITIZEN CLL042-1818A5- β 503M1A2 and CITIZEN CLL042-1218A5-303M1A2 that have CRI>80, a color temperature of 5000K, respective 3000K, a light flux of 7170lm, respective 4755lm and a junction current of 1.08A. Area of interest H H (AOI) In order to compute the overall illuminance in AOI we discredited the space (L and W) in100 steps resulting 104 points in which we computed the illuminance, than we computed the average illuminance and the uniformity factor. d SIMULATION AND ANALYSIS RESULTS EXAMPLES d An example of using our design and simulation framework W is the lighting of a pedestrian crosswalk [4]. The geometry of simulated case is illustrated in figure 3. We intend to find the spatial positioning of light sources so L that light flux ɸ must obey CIE (International Commission on Fig. 3. The geometric parameters of simulated system 18 Technical support by The objective of the simulation is to find the optimal We performed a great number of simulations maintaining spatial distribution of lamps and their number in order to obey constant values for L (19m), W (10), H (10) and varying the the standard specifications from table 1. Although the software number of lighting poles, the number of lamps on every pole developed allow arbitrary values, for case take into account in and the angles of lamps. Several results are shown in table 2 our work we used: L=19m, W=10m, H=10m. and figure 4. After establishing, by simulations, the most convenient Although the practical implementation using only one LED devices (power: 60W, 7170 lm, T= 7000K, 10m high, lighting pole is more attractive, the simulations show that the inclination of the axe LED= 300, for a surface framed into standard is not respected even with three LED lamps on the 19m x 10 m), we have imagined some different scenarios of pole. As opposite, the using of two lighting poles, on each side lighting system placement on the street. of the road allows the respect of standard requirements. TABLE XII. SEVERAL SIMULATION RESULTS Number of Number of Lamp's angle Average Maximum Minimum Case Uniformity U lighting poles lamps on pole (degree) illuminance (lx) illuminance (lx) illuminance (lx) 0 1. 2 1 30 15.42 18.84 10.29 0.66 2. 2 2 10, 30 25.32 31.5 18.21 0.72 Simulation results Isolux curves 10 12 12 13 13 12 13 14 14 13 15 14 15 14 8 16 16 15 15 16 18 17 17 16 18 17 6 17 18 17 18 16 17 16 18 18 16 16 18 4 16 Width Width [m] 14 15 15 14 17 17 12 2 16 16 15 14 15 13 Illuminance level [lx] 13 14 14 12 13 13 15 0 12 12 10 0 2 4 6 8 10 12 14 16 18 10 Length [m] 5 5 Length [m] Width [m] 0 0 Fig. 4. The illuminance distribution and the isolux curves on the illuminated crosswalk, with two LED lamps The best results from the illumination uniformity point of an indoor or outdoor LED lighting system. It allows the view are when two LED lamps are located on both sides of the specification of light sources characteristics and of street (both LED-s have T=7000K). The configuration with geometrical distribution of sources and provide the two lighting poles having coordinates (1;5) and (18;5) in illuminance of AOI and also his uniformity. figure 2 each having one LED CITIZEN CLL042- 1818A5503M1 lamp places at 300 from horizontal obey the Varying the lighting optical and geometrical parameters a requirements of standard. great number of lighting scenarios may be investigated theoretically in very short time and then a decision of the best Moreover, a more complicated setup with two lighting configuration may be taken. Moreover, the framework allows poles, each having two lamps (LED CITIZEN CLL042- its using in very different geometrical configurations. 218A5303M1 and CITIZEN CLL042-1818A5503M1) and placed at 100, respective 300 assure a medium illuminance Actually this work is part of a more complex project level over 25 lx and a uniformity factor of 0.72. aiming the design and implementation of a smart lighting system. One of consequences of these simulations is the using of an intelligent management system of LED lighting sources CONCLUSIONS that will allow the optimization of light flux in real traffic The design and simulation framework presented in this conditions of cars and pedestrians. This means the monitoring paper is a useful tool for establishing the optimal structure of of traffic level and surrounding conditions (temperature, 19 Technical support by humidity, natural lighting etc.) in order to efficient use the REFERENCES existing resources. As consequence the whole system contains (fig. 5): . Light sources, power LED's; [1] ***The International Commission on Illumination (CIE), [CIE 180:2007 . Power driver that provide the requested current to Technical report, ROAD TRANSPORT LIGHTING FOR DEVELOPING COUNTRIES], http://files.cie.co.at/180.pdf, (2014). LED's, for needed light flux; [2] *** US Department of Energy, [Final Report prepared in support of the . Sensor's block that sense the light level and the traffic DOE Solid-State Lighting Technology Demonstration GATEWAY conditions; Program, Pedestrian Friendly Outdoor Lighting], http://apps1.eere.energy.gov (2014). . Microcontroller that uses the acquired data and an [3] *** European Commission, [Lighting the Cities - Accelerating the original algorithm in order to properly command the Deployment of Innovative Lighting in European Cities], 2013 drivers. [4] Ileana I.,Risteiu M., Marc G., Sofalca I., “Analysis and simulation of an automated LED lighting system for pedestrian crosswalk”, paper presented at the 7th edition of the International Conference "Advanced The smart character of the system concerns the light flux Topics in Optoelectronics, Microelectronics and Nanotechnologies", control having an operating regime for reduced traffic and 2014. another operating regime when pedestrian traffic is detected. [5] Krishnadas A., Rasheed R., "Simulation of a Two Stage High Power Factor Driver for White Power Led", International Journal of Modern Engineering Research (IJMER), Vol. 4, Iss. 3, Mar. 2014, 38, www.ijmer.com. LED Driver Power LED [6] Nguyen1 H.Q., Choi1 J.-H., Kang M., Ghassemlooy Z., Kim D. H., Lim Sensors Microcontroller S.-K., Kang T.-G., Lee1 C. G., "A MATLAB-based simulation program block for indoor visible light communication system", http://www.researchgate.net/profile/Professor_Zabih_Ghassemlooy/publi LED Driver Power LED cation/224176034_A_MATLAB-based_simulation_program_for_indoor_ visible_light_communication_system/(2014). [7] Rajput K.Y., Khatav Gargeyee, Pujari Monica, Yadav Priyanka, Figure 5. The smart lighting system "Intelligent Street Lighting System Using Gsm", International Journal of Engineering Science Invention, Volume 2, Issue 3, March. 2013, pp.60- 69, www.ijesi.org [8] Santha Sheela K., S.Padmadevi, "Survey on Street Lighting System Based On Vehicle Movements", International Journal of Innovative Research in Science, Engineering and Technology, Vol. 3, Issue 2, February 2014, pp 9220- 9225, www.ijirset.com (2014). [9] Xuan-Hao Lee, Ivan Moreno, and Ching-Cherng Sun, "High-performance LED street lighting using microlens arrays", OPTICS EXPRESS Vol. 21, No. 9, 10612 -10621, 2013. 20 Technical support by Implementation of Petri Nets using Analytical Theory of Markov Ciortea Elisabeta Mihaela University “1 Decembrie 1918” of Alba Iulia, Department of Science and Engineering, Blvd. N. Iorga, No. 11-13, Alba Iulia, Romania, [email protected] Aldea Mihaela University “1 Decembrie 1918” of Alba Iulia, Department of Science and Engineering, Blvd. N. Iorga, No. 11-13, Alba Iulia, Romania, [email protected] Abstract— This paper presents a model of network events complexity of real systems, the dimensions are very large Petri that continue to apply to the general case of system Markov net models. networks using Petri. Performance measures for system behaviour can be introduced with markov models. The problem is that it is not clear what kind of measures should be generalized and calculated, because each system has different system THE ANALYSIS OF NETWORKS requirements. The basic system requirements can be generalized into four classes: system availability, system reability, system Structure general linear programming model is established performance and task completion. Common effects of resource A , A ,..., A primarily by the set of activities { 1 2 n } composing sharing and queuing are naturally represented by means of queuing systems, while synchronization can be easily specified the analyzed system, the set of resources used R , R ,..., R with petri net. In this paper, a method is presented to obtain { 1 2 m } as well as the technical relationships performance parameters with Generalized Stochastic Petri Net in between them. The link between activities and resources is order to be able to analyze the stochastic behaviour of the system. The embedded Time Markov Chain is derived from the determined by the manufacturing technology for each activity GSPN and the Markov Chain theory is used to obtain the A j performance parameters. It was design a case of tool to obtain (j=1,...,n) and may be characterized by the column vector some performances that were discussed to a case study. a ,a ,...,a a number a(j) component ( 1 j 2 j mj ). Elements { ij , i Keywords— Petri, stochastic Petri Net, manufacturing = 1,...,m; j = 1,...,n} are called technical coefficients or specific consumption coefficients and show how much of the R P I. INTRODUCTION resource i consumed to produce a unit of product j . All Petri networks have a number of properties arising from "technology" column vectors defined manufacturing a(j) can the behavior of the network or the network marking the first or be organized in a matrix A with m rows and n columns; Each the structure of the network or the network topology. Ri Modeling real systems using Petri nets model is Petri line refers to a resource (i = 1,...,m) and each column checkpoints which qualitative properties required, such as A j vividness, reversibility, limitation and consistency. relates to an activity (j = 1,...,n). The functioning of a number of automated systems can be The restrictions described above makes the link modeled using the concept of discrete event system. For between resources and activities through the m linear construction specifications and the behavior patterns that constraints. reflect real processes, the correct identification of state-space The linear programming problem 0 contains restrictions structure and the multitude of events that coordinate and criteria of "performance" to assess the effectiveness of transitions in state space, discrete event systems play a each activity. Depending on the purpose, we choose the fundamental role. criterion of efficiency indicator that measures the effort, one Properties characterize the dynamic behavior of the system that measures the outcome or indicator expressed as the ratio built. Vivacity ensures that a deadlock state will not interfere between output and effort. in the process, limiting the number of process components Maximum efficiency means minimizing effort and guarantees an upper limit. reversibility allows the system to maximizing outcome and optimal concept is defined as a return to its original state regardless of which touch at a time program that minimizes or maximizes an objective function and consistency is considered a structural property which is a and at the same time meet all technical restrictions. necessary condition for reversibility. Because of the 21 Technical support by Assuming that each component of the vector line Aij Sbj S c (c ,c ,...,c ) when t , is the interaction matrix with bi 1 2 n measured efficiency of a unit of the result B influencing state of the entire system, pq - A j of , then we can introduce the linear u k x k matrix interaction between q and p that f (x) c1 x1 c2 x2 ... cn xn function: (1) f pq uq kx p k evaluating the performance of any program x. influencing state the f x k,u k Summarizing, we obtain the following linear programming entire system, (5) software: x t optimf (x) - internal interaction matrix between states their x (2) overall system which affects the state at time t , a Relations (2) constitute part of the general pattern of whole system next time t 1 . a linear programming problem, each having a specific role: Equations of state are subject to conditions and n restrictions: f x c j x j - relations (2), where j1 efficiency is - for S continue t called the objective function of the problem, evaluate the N performance of each program options x; e t k ,te T n - for discrete e1 aij x j bi j1 t ct. T N t - the relationship of the type type is if e that e e , restricted resources; x x x 0 n - 0 - known initial state S, and 0 for a x b kj j k , - restriction on the type j1 relates to the technical limitations of qualitative; R x k ,u k 0 - , R expressing a set of x 0 restrictions on the matrix, - relationship j j = 1,...,n, called the non-negativity condition variables, ensures a feasible solution from the point v k q x k - , and of view of economic logic. v 0 q x 0 v When determining optimal production assortment 0 for . structure, known quantities available of each raw material bi, a x technological coefficients ij , maximum quantities j and x j PETRI NET ANALYSIS minimum that can be produced from each type of unit A Petri net [8] is composed of a particular type of directed p profits in the period under review and j , of each type of graph, an initial state called the initial marking. product. Formal definition of Petri network type is of the form, General dynamic model of manufacturing system is given P,T, F,W by the matrix vector equation of the form: (6) were: P {p1, p2 ,..., pm} is the set of positions, x k 1 A x k B u k f x k,u k a ij a pq a T {t1,t2 ,...,tn}is the set of transitions, (3) F PTT Pis a binary relationship, called the 1,n,i, j n n,p,q n m,,i, j, p,q relationship flow network, W : F N is the function of (4) the network share W f is called the share element x k 1 f . where is the matrix of system states when Either type a network location / translation, marking the t 1,x k u k and are arrays of states and inputs network is called any function M : P N with 22 Technical support by property Mp Kp, for any p P , were dependent on the marked, but the user can select an operation independent of the goal. K : P N is the function of network capacity . By using generalized stochastic Petri nets to obtain Network P/T, which is obtained manufacturing system model analyzed. Execution times of N P M M : P N p PMp Kp transitions are distributed exponential law which generates a graph of the Petri net marking, which is a homogeneous (7) Markov process. If the network has infinite capacity only when N P coincides with the set of the P to N. The model presented in this paper may be useful for managers to analyze changes, system designers can include A network P/T is a pair ,M 0 , where is a tracking and control cells. support mesh network , and M 0 marking the grid is The current trend analysis of transport systems is to original . develop large structures, the development of new algorithms for transport and analysis of transport networks in order to For network transitions are considered functions, increase productivity, degree of automation and workflow t ,t : P N and t : P Z defined by: tracking. t p Wp,t, t p Wt, p, If stochastic networks only exponential distributions can be used to assign time ranges of transitions. For transitions in tp t p t p, for any p P . conflict, the shortest transition that allows the choice to be The dynamic evolution of a network known three types of made without using priorities or probabilities. Do not allow semantics [8]: multiple execution of the same transitions, even if permitted by marking content input positions; this means that the 1. sequential evolution of the network is given by the transition to be executed once, and after the allotted time is rule of transition, which consists of: running out, can be executed again if the current marking is - (RA –applicability rule), transition t it is possible to appropriate. [5] [6]. mark M in and note Mt if there is the relation: Modeling distributed systems using Petri nets is performed Wp,t Mp and at state level: determine which actions occur in system that precede these actions and conditions in the state will go after M p W t, p K p ,p P the action system. Conditions simulating the Petri nets model (8) describing the system behavior is obtained. [3]. ' - (RC – calculation rule), marking M is produced by the In many areas of research studying the behavior of real appearance transition t marking M and note MtM ' , if system not directly, but indirectly by means of the model. The - meets the characteristic properties for the object or system Mt and under consideration. Studying the given system model can M ' p M tp, p P or deduce new information without the costly expense [3]. M ' M Ot It Failures with respect to time of the equipment or components will vary according to the law of distribution of (9) reliability. Failure occurs when the level of damage exceeds 2. concurrent developments relative to a lot is the easiest the limit. Maintenance is defined as a strategy to maintain the way to capture the look of concurrent application of a availability or operational conditions of a facility, using all transition. This method consists in generalizing the concept of possible means and methods to ensure optimal working "step" through the transition from a transition to a lot of conditions. Maintenance helps to improve the safety of the transitions that can be applied competitor of labels. equipment system. Maintenance program, with appropriate 3. type evolution process [8], type subset does not capture intervals predetermined plan must be adapted to the situation aspects of the concurrent application of competition that and lead to achieving real manufacturing system. [3]. transitions herself. This process introduces the concept of a The basic concept is to provide opportunities to Petri nets network using network appearances lined. modeling a network as a real network. As a method of In many areas of research, studying the behavior of real describing graphics Petri nets for adding resources to represent system not directly, but indirectly, by using the model. The locations and check if there is any conflict or error in the model is characterized by the fact that meet desirable system. When transitions are triggered, changing the state and properties of the object or the system under consideration. the dynamic behavior of Petri nets are obvious. ANALYSIS SYSTEM Graphics can simulate and analyze the behavior of Sequence of transitions executed is controlled exclusively discrete systems, efficient events. This instrument is widely by exponential intervals to ensure equivalence of Markov applied in information protocols, software engineering and chains. By default, the rate of execution of transitions is flexible manufacturing systems to help reduce processing time and improve the process of determining the errors and system reliability management. It is a tool for detecting and counting 23 Technical support by failures to replace components after successive failures, CONCLUSIONS system state description and automatic shutdown to reset the A model was developed transportation system that set monitoring and preventive maintenance. [3]. sizing, internal structure and the corresponding couplings. In The bow which is shown in the following figure: addition to the workstations in the system included control station subsystem and the corresponding logistics. In the analyzed system was considered a location where loading and unloading is part of the process and a variant are shown separately transport system, processing system and ancillary elements. - The simulated using actual production parts made of HIPS pack. The graph is done using: - Petri Nets, where the instruments available to provide information related to the mathematical model (the incidence matrix, graphic design, text and presentation model node and Fig. 1. Classical Petri net transport and deposit containing two processing transition invariants). centers Fig. 2. Petri net control system maintenance in the two processing centers REFERENCES [4] Mihăilă N. Introducere în programare liniară, Editura didactică şi pedagogică, Bucureşti, 1964 [5] Păstrăvanu O., Matcovschi M., Mahulea C. Aplicaţii ale reţelelor Petri în studierea Sistemelor cu evenimente discrete, Editura Gh. Asachi, 2002, ISBN 973-8292-86-7 [1] Ciortea E. M., L. Morar L. Analysis of the influence of the [6] Păstrăvanu O. Sisteme cu evenimente discrete. Tehnici calitative bazate transportation system in increasing the performances of the production pe formalismul reţelelor Petri, Editura MatrixRom, Bucureşti, 1997, system, ACTA UNIVERSITATIS CIBINIENSIS, Vol. LII, Technical ISBN 973-9254-61-6 Series, Sibiu, 2005 [7] Gladcov P., Neagu C., Gladcov C., Pregătirea fabricaţiei, Editura [2] Ciortea E. M. Analysis of manufacturing systems modeling by petri nets, MatrixRom, Bucureşti, 2004, p. 106 ACTA UNIVERSITATIS APULENSIS, Mathematics-Informatics, No.11/2005, ISSN 1582-5329 [8] Camerzan I., Proprietăţi structurale ale reţelelor Petri temporizate, Thesis in Computer Science, State University of Tiraspol, Chişinău, 2007-12- [3] Ciortea E. M. Analytical interpretation to optimize transport systems, 15. ModTech 2014, 13-16 Iulie Gliwice Poland 24 Technical support by Considerations Regarding the Evaluation of Documentation and Testing of Low Current Equipment Intended for Use in Explosive Atmospheres Colda Cosmin, Burian Sorin, Ionescu Jeana, Darie Marius, Csaszar Tiberiu D.S.I.E.A. I.N.C.D.-INSEMEX Petrosani, Romania [email protected] Abstract— Low current electrical equipment can be used in in accordance with Directive 94/9/EC, annex 3. The all hazardous areas, the intrinsic safety type of protection level ia documentation submitted to the notified body must include the being the type of protection that can be used in zone 0 where the following: danger is permanent. Evaluation of documentation and testing of low current electrical equipment is made in accordance with a general type-description; Directive 94/9 / EC and specific standards in force. design and manufacturing drawings and layouts of Keywords— testing type, intrinsic safety, type of protection components, subassemblies, circuits, etc.; descriptions and explanations necessary for the I. INTRODUCTION understanding of said drawings and layouts and the In order to evaluate the test for low voltage electrical operation of the product; equipment in accordance with Directive 94/9 / EC are used a list of the standards referred to in Article 5, applied in European harmonized standards that require performance type full or in part, and descriptions of the solutions adopted testing to verify the explosion protection characteristics. The to meet the essential requirements of the Directive Directive 94/9/EC applies to equipment and protective where the standards referred to in Article 5 have not systems intended for use in potentially explosive atmospheres. been applied; This Directive is also cover to the Safety devices, controlling devices and regulating devices intended for use outside results of design calculations made, examinations potentially explosive atmospheres but required for or carried out, etc.; contributing to the safe functioning of equipment and tests reports. protective systems with respect to the risks of explosion. The notified body shall: The SR EN 60079 part 11 specifies the construction and testing of low current apparatus intended for use in an [1] examine the technical documentation, verify that the explosive atmosphere and for associated apparatus, which is type has been manufactured in conformity with the intended for connection to intrinsically safe circuits which technical documentation and identify the elements enter such atmospheres. This type of protection, intrinsic which have been designed in accordance with the safety "i", is applicable to electrical equipment in which the relevant provisions of the standards referred to in electrical circuits themselves are incapable of causing an Article 5, as well as the components which have been explosion in the surrounding explosive atmospheres. The designed without applying the relevant provisions of intrinsic safety “i” is based on the restriction of electrical those standards; energy within equipment and of interconnecting wiring [2] perform or have performed the appropriate exposed to the explosive atmosphere to a level below that examinations and necessary tests to check whether the which can cause ignition by either sparking or heating effects. solutions adopted by the manufacturer meet the essential requirements of the Directive where the II. ASSESSMENT OF TECHNICAL DOCUMENTATION FOR LOW standards referred to in Article 5 have not been applied; VOLTAGE EQUIPMENT [3] perform or have performed the appropriate examinations and necessary tests to check whether The conformity assessment of technical documentation for these have actually been applied, where the equipment used in potentially explosive atmospheres is made 25 Technical support by manufacturer has chosen to apply the relevant The most important tests that the notified body must standards; perform for low voltage equipment are: [4] agree with the applicant the location where the Spark ignition test - all circuits requiring spark ignition examinations and necessary tests shall be carried out. test shall be tested to show that they are incapable of causing ignition under specified conditions ; Temperature tests - All temperature data shall be III. SPECIFIC TESTS FOR LOW VOLTAGE EQUIPMENT USED IN referred to a reference ambient temperature of 40 °C or HAZARDOUS AREAS the maximum ambient temperature marked on the apparatus. Tests to be based on a reference ambient temperature shall be conducted at any ambient Low voltage equipment testing shall be made in temperature between 20 °C and the reference ambient accordance with European standard SR EN 60079 part 11. The temperature. The difference between the ambient low voltage equipment and associated apparatus which has a temperature at which the test was conducted and the type of protection listed in SR EN 60079-0 for use in the reference ambient temperature shall then be added to appropriate explosive atmosphere, shall be grouped in the temperature measured unless the thermal accordance with equipment grouping requirements of SR characteristics of the component are non-linear, for EN 60079-0 and shall have a maximum surface temperature or example batteries. temperature class assigned in accordance with the temperature Dielectric strength tests - The supply shall have requirements of SR EN 60079-0. sufficient volt-ampere capacity to maintain the test The low voltage equipment and intrinsically safe parts of voltage, taking into account any leakage current which associated apparatus are divided into levels of protection "ia", may occur. The voltage shall be increased steadily to "ib" or “ic”. the specified value in a period of not less than 10 s and then maintained for at least 60 s. The applied voltage I. Level of protection "ia"- with Um and Ui applied, the shall remain constant during the test. The current intrinsically safe circuits in electrical apparatus of level of flowing during the test shall not exceed 5 mA r.m.s. at protection "ia" shall not be capable of causing ignition in each any time. of the following circumstances: Tests for intrinsically safe apparatus containing a) in normal operation and with the application of those piezoelectric devices - measure both the capacitance of non-countable faults which give the most onerous condition; the device and also the voltage appearing across it when b) in normal operation and with the application of one any part of the intrinsically safe apparatus which is countable fault plus those non-countable faults which give the accessible in service is impact tested in accordance with most onerous condition; the “high” column of Tests for resistance to impact c) in normal operation and with the application of two Type tests for diode safety barriers and safety shunts - countable faults plus those non-countable faults which give the tests are used to demonstrate that the safety barrier the most onerous condition. or safety shunt can withstand the effects of transients. II.Level of protection "ib" - with Um and Ui applied, the Transformer tests - for safe electrical isolation is intrinsically safe circuits in electrical apparatus of level of satisfied if the transformer passes the routine test, the protection "ib" shall not be capable of causing ignition in each type test described and subsequently withstands a test of the following circumstances: voltage of 2 U + 1 000 V or 1 500 V, whichever is the greater, between any winding(s) used to supply a) in normal operation and with the application of those intrinsically safe circuits and all other windings, U non-countable faults which give the most onerous condition; being the highest rated voltage of any winding under b) in normal operation and with the application of one test. countable fault plus the application of those non-countable faults which give the most onerous condition. Cable pull test - The cable pull test shall be carried out as follows: III. Level of protection "ic"- With Um and Ui applied, the intrinsically safe circuits in electrical apparatus of level of - apply a tensile force of minimum value 30 N on the protection "ic" shall not be capable of causing ignition in cable in the direction of the cable entrance into the normal operation and under the conditions specified in this apparatus for the duration of at least 1 h; standard. - although the cable sheath may be displaced, no visible displacement of the cable terminations shall Can be made a link between the level of protection and the be observed; area where the equipment is used such: the level of protection "ia" can be used in zone 0, the level of protection "ib" can be - this test shall not be applied to individual used in zone 1 and the level of protection "ic" can be used in conductors which are permanently connected and do not form part of a cable. zone 2 in accordance with European standards. 26 Technical support by Optical isolators tests - the tests shall be performed if An example of a transfer curve obtained with the energy optical isolators are used to provide isolation between limited circuit on the developed test rig is shown in Figure 2. intrinsically safe circuits and non-intrinsically safe circuits and are not adequately protected against overload by external protection components Dielectric and short-circuit test - optical isolators shall be subjected to a dielectric strength test, followed by a short-circuit current test and if applicable to the current limited short-circuit current test described below, followed by a dielectric strength test. Routine tests for diode safety barriers - a routine test shall be carried out on each completed barrier to check correct operation of each barrier component and the resistance of any fuse. The use of removable links to allow this test shall be acceptable provided that intrinsic safety is maintained with the links removed Figure 2: The change of the current curve obtained using the acquisition card After performing these tests certification body is able to certify the equipment under test, if it passes all tests at which it was submitted. The output energy in the assessed circuit is calculated IV. RESULTS OBTAINED ON SAMPLES OF LOW VOLTAGE using the relationship 1 and 2 but also taking into account the EQUIPMENT tolerance value of the load resistor in the worst case condition To illustrate the importance of testing was chosen a power (in this case the positive tolerance value): supply whit a safety barrier used in hazardous environments, n to which was made the determination of transitory energy. WWi i1 The assessment of the safety barrier with intrinsic safety protection, which is intended for use in potentially explosive 2 2 WR1tolIimin(Iadm,Ii)Ii1min(Iadm,Ii)titi1 environments, consists in determining the maximum transfer energy. This can be done by viewing with the help of achieved stand Figure 1, the peak current for a given value of load, In order to achieve the rated energy in the assessed circuit, respectively the determination by integrating the curve use may be made of the values obtained directly on the obtained, of the maximum output power to the circuit. measuring current channel (CH1) or on the voltage measuring channel (CH2). In case we have available the instantaneous voltage values Ui, the value of the current Ii in that moment will be determined according to equation 3, also taking into account the tolerance value in the worst case condition (in this case the negative tolerance value). Ui Ii R(1tol) In a sample that was not encapsulated, was shunted the switching power supply, remaining in the power supply circuit only the safety barrier, rectifier bridge and filter capacitors. It was supplied all the circuit to an alternative voltage of 1,1 x 127 V a.c. and was measured using the oscilloscope the transient voltage characteristic at the output of the safety barrier for more load resistances Figure 3 . Figure 1: Connecting the energy limited circuit on the test stand 27 Technical support by 60 50 40 30 20 Energia disipata [microJ] 10 0 0 0,5 1 1,5 2 2,5 3 3,5 4 4,5 5 Rezistenta de sarcina [ohmi] Figure 3: The energy dissipated on load resistance For the type of protection intrinsic safety, not all the low voltage equipment undergoing testing cycle presented in standard. It is up to the certifier to require testing laboratory only certain tests, in the test report will be mention the spelling omitted test. REFERENCES [1] Bottril G., Chueyne D., Vijayaraghavan G., (2005), Practical electrical equipment and installations in hazardous areas, Elsevier Publishers, 149- 150. [2] Burian S., Ionescu J., Darie M., Csaszar T., Andriş A., (2007), Requirements for Installations in Areas with Explosive Atmosphere, Other than Mines, (in Romanian), INSEMEX – Petrosani Publishing House, Romania. [3] Csaszar T., Modernisation of experimental methods for assessing energy supply circuits of electrical equipment used in potentially explosive atmospheres, 2011 University of Petroşani, Romania. [4] EC Directive, (1994), Directive 94/9/EC of The European Parliament and The Council of 23 March 1994 on the approximation of the laws of the Member States concerning equipment and protective systems intended for use in potentially explosive atmospheres, Official Journal of the European Communities no. L100, 19.04.1994, Brussels.K. Elissa, “Title of paper if known,” unpublished. [5] Standard SR EN 60079-11: 2012 - Explosive atmospheres -- Part 11: Equipment protection by intrinsic safety "i". 28 Technical support by Method of Storing Relevant Medical Information Based on Biometric Identification Dobra Remus 1, Leba Monica 1, Ionica Andreea 2 1University of Petrosani, Department of Computer and Electrical Engineering Str. Universităţii nr. 20, 332006 Petrosani, Romania Tel.: + 40-254-549011, Fax: +40-254.546238, Email: [email protected], [email protected] 2University of Petrosani, Department of Management Str.Universitatii nr. 20, 332006 Petrosani, Romania Email: [email protected] Abstract— A method of storing relevant medical information ones. If the matching templates is found then the patient is based on biometric identification is proposed and for this reason recognized and counted as known by the system. we have developed an optimal system that allows the person identification based on fingerprint, the storage/access to MEDICAL INFORMATION STORING USING BIOMETRIC information in a centralized database and the deliverance of reports containing relevant personal and medical data. The FINGERPRINT SENSOR developed biometric system provides a method for storing Biometric technologies based on fingerprint sensor are relevant primary health information based on biometric basically pattern recognition systems that use data acquisition identification that lead to a prompt, easy and secure devices such as dedicated scanners in order to gather biometric determination of the identity and relevant medical information of characteristics which are distinctive between users. When the people who require medical emergency intervention. It creates digital system identifies a proper fit the fingerprints are the possibility of taking the right decisions and immediate actions extracted and encoded into a biometric template that is a by authorized medical staff due to the access to personal mathematical representation of a person biometric unique information (name, ID, address, phone number, picture, contact characteristic. person) and relevant medical information (blood type, RH, allergies, chronic diseases, organ donor option, resuscitation option) stored in a central database. A. Purpose of Study The medical record is a useful tool that allows the Keywords— medical information; fingerprint scanner; databse; emergency medical personal to track the patient medical biometric identification; fingerprint template history and identify problems or patterns that may help determine the course of health care. Current procedures to I. INTRODUCTION identify individuals and medical information storage are based Continued development of biometric technology has on two principles: allowed its usage in different biometric applications with great Classical one consisting on storage of information in advantages. An important application refers to the paper files; identification methods of those who need emergency medical services. The goal of emergency medical services is to provide Computerized one consisting on storage of information treatment to those patients in need of urgent medical care, or in databases on dedicated servers. preparing for fast movement of the patient to the next medical These procedures do not allow a prompt, easy and secure facility, but in order to do make the right decisions in this identify of people who need emergency medical services. amount of time they need to know with whom they are dealing Current procedures based on biometric information are used to with. Using a biometric system that has integral networking identify people based on fingerprint in order to increase the functionality, with a wireless protocol, the medical personnel security of banking transactions, access to mobile devices can read the patient medical information stored in a central (phone, tablet, laptop) and fixed devices (secured access database on a server. interfaces). Biometric technology used in emergency medicine requires a collection of data representation using a fingerprint B. Method advantages sensor, of physiological characteristics unique for every The development of this innovative product follows the individual person. This digital representation of biometric data well-known lifecycle, consisting in requirements capture, is transformed using a dedicated algorithm in order to produce system design, development and implementation of the system a unique template usually stored in smart card, in a central (hardware and software), testing. The research methodology database on a server, or directly on the sensing device. These was based on the collaboration between specialists from stored templates can be accessed when the finger is presented different fields, like computer engineering, electronics, quality to the biometric sensor interface and the identification is management and medicine, and led to the objective achieved by comparing the finger template with the stored 29 Technical support by achievement, the biometric identification system for protocol the microcontroller system will connect with a emergency medical situations. mobile device (laptop or tablet). After the fingerprint identification on the tablet will be displayed in the shortest This developed product is based on a method that time a medical report regarding the victim. eliminates the classical methods insufficiencies by optimizing the time response, facilitating access and by ensuring an If the victim shows up at the emergency hospital building increased security regarding the primary medical information the procedure will be similar with that described above, for emergency situations. A new approach is developed for because a biometric system will be used in this case also and storage the primary healthcare information based on biometric the person will be quickly identified both in terms of its identification by means of fingerprint sensor. identity as well as her health status. C. Method description FLOWCHAT OF THE BIOMETRICAL STORING METHOD Primary care is crucial for the health status being the most Fig. 2 describes the flowchart for relevant medical accessible and less expensive and it is the responsibility of information storage based on biometric identification family doctors. The primary healthcare provider must use regarding the basic fingerprint reader. Check the current status modern techniques based on medical decision support systems of biometric system in terms of hardware integrity and if the and this is a must because the communication between family hardware self-test was successful when the biometric system doctors and emergency services doctors is deficient, because with microcontroller will signaling the current status; if after there is no real coordination between preventive services and the hardware self-test the biometric system has any defects or emergency services. does not match in terms of authenticity, the system will The medical information storing, begins in the family generate an error ID and will signal this current state doctors cabinets which gather information for operational appropriately, the system may be restarted only after the administration of personal data contained in a medical record defect remedy compare the current supply voltage of the specific to each person (Fig. 1). Data loading is possible by system with reliable running threshold imposed by the means of biometric fingerprint sensor that is based on an standards. optical or capacitive transmitting image captured by the If the current value of the supply voltage is less than 20% microcontroller, every patient signature will be loaded in to a of the rated value; if the supply voltage is above the threshold data base and a medical record will be created also. This of 20% reliable function, the system will generate a unique medical record will contain relevant information about the identifier numeric specific to the equipment. patient like, the name, NPC, address, blood type, allergies, chronic diseases, organ donor or resuscitation option, and so on. Figure 2. Basic fingerprint reader flowchart Fig. 2 describes the unique identification number generated by the biometric system with microcontroller for family doctors ID. If it is a valid ID and corresponds to the Figure 1. Method of accessing patients medical information based on their cabinets of family doctors, the biometric sensor belonging to fingerprints the microcontroller system detects the fingerprint of the All data loaded in the family doctor cabinets will be stored person whose personal data is to be stored; if the biometric in a central database which will contain the digital patient’s sensor does not detect the fingerprint or the finger was not signatures together with their relevant medical data. Using properly placed, the device allows successive attempts until password or PIN to control access to programs and computer achieved fingerprint detection. Is checked the fingerprint and files is a pretty good method, but can be quite easily hijacked the information about it is transmitted to the central database; and for that reason we have used a biometric method able to if the person fingerprint is recognized and it is already make permanent identify of the user. This system should not included in database, the system allows that authorized allow access to data until the time the operator is identified as personal from family doctor cabinet can add new personal or an authorized user. medical information in the patient medical record; If the fingerprint is not recognized and therefore is not in database, When an emergency comes along the victim is taken by then the system initiates the procedure for adding a new the emergency ambulances which will be equipped with fingerprint through its validation; after this stage the biometric identifications systems. The victim finger is placed biometrical system allows the user to create patient medical on the biometric sensor and via USB cable or a bluetooth 30 Technical support by record and if it is desired to load everything it into database, If it is valid ID and corresponds to emergency services or the system initiates an appropriate procedure. legal medicine, then the biometric sensor corresponding to the numerical system with microcontroller detects the person fingerprint for data reading; if the biometric sensor does not detect the fingerprint or the finger was not properly placed, the device allows successive attempts until fingerprint detection will be achieved; Is checked the fingerprint and the information about it is transmitted to the central database; if the person fingerprint is recognized and it is already included in database, the system displays in the shortest time a patient summary medical report containing its relevant personal information; If the fingerprint is not recognized and therefore is not in database, system indicates that in database is no template related to the concerned patient (victim). If during this time there is a biometric system shutdown command or a fault occurs, the entire system will reset giving the user the chance to start all over again. FLOWCHAT OF THE BIOMETRICAL STORING METHOD The processing flow for biometric template electronic submission contains basic fingerprint readers that submit the person fingerprint to a central database via smart device. These submissions will originate from fingerprint sensors at family doctors or emergency services levels. Local Figure 3. Flowchart for ID that corresponds to the cabinets of family doctors submissions may be processed by a family doctor and transmitted via internet to a central database for storing. If a template of the fingerprints in database is not wanted, the system initiates a reset command that will allow a resuming of the entire procedure in order to store biometric D. Hardware processing flow data. If during this time there is a biometric system shutdown Fig. 5 describes the block diagram basic fingerprint readers command or a fault occurs, the entire system will reset giving that ensure the fingerprint scanning of persons in need of the user the chance to start all over again. emergency services. Fig. 2 describes the unique ID generated by the biometric The identification is made using a biometric digital system for emergency services or legal medicine. interface, which takes over information regarding the fingerprint using a fingerprint sensor and transmit them to the microprocessor, via a serial protocol. Biometric system identification is powered from an accumulator and is switched on/off using button, and its current status is signaling by LED block. Figure 5. Basic fingerprint readers Fig. 5 describes the block diagram of the storing medical information using biometrical fingerprint sensor. Loading persons data is made in family doctors offices and the reading part take place in the emergency services, using a classical PC or laptop (2), connected to the digital biometric Figure 4. Flowchart for ID that corresponds to the cabinets of emergency system (1), via a USB cable, either using wireless or Bluetooth services protocol. Data transmission to the central system, for storing 31 Technical support by data in the central database, located on the server (3), is done Fig. 5 describes a simplified version of the developed through the internet protocol. database. Through the graphical interface the user can view the current status of basic fingerprint reader, can choose between enroll new person and indentify new person types. Database provides information regarding the patient picture, name and surname, birth date, contact person, blood type, ID, organ donor. The basic fingerprint reader hardware contains a fingerprint sensor connected to the microprocessor, via a serial protocol. This biometric assembly work together with the PC (developed database) trough USB cable (Fig.8). Figure 6. Architecture of medical storing data Storing templates in a central database on a server mans that users authenticating from multiple locations and the access to the encryption keys will have only doctors from emergency services or legal medicine based on patients fingerprints. E. Storing medical information database The biometric system is used for verification, enroll and identification of persons in need of emergency services. In verification process, a patient that has a certain identity will put his finger on the fingerprint sensor, and the biometric system performs a comparison between the offered biometric and the biometric reference information stored in the central database. Figure 8. Basic fingerprint reader hardware Person fingerprint enrollment is a process that is responsible for registering individuals in the central database. For this project was used optical fingerprint readers During the enrollment process, the biometric characteristic of because are the most common at present. They are based on a subject is first captured by a fingerprint scanner to produce a reflection changes at the spots where the finger lines touch the sample. A quality check is often performed to ensure that the readers surface. All the optical fingerprint readers comprise of acquired sample can be reliably processed by successive the source of light, the light sensor and a special reflection stages. A feature extraction module is then used to produce a surface that changes the reflection according to the pressure. feature set. The template creation module uses the feature set Some of the readers are fitted out with the processing and to produce an enrollment template. memory chips as well. In order to identify, a comparison is performed between The fingerprint technology can be used for identification the offered biometric template and all available reference even within large databases. Fig. 9 presents database graphical information stored in database to reveal the identity of a interface, in which we tried to identify a new person which patient. does not have the fingerprint template stored in database. Figure 9. Database interface – identify person routine Figure 7. Developed database interface 32 Technical support by Fig. 5 describes enrolling situation, which means assigning ID to the corresponding fingerprint template. Figure 10. Basic fingerprint reader hardware – fingerprint processing Fingerprint device used do all the processing by the hardware. Connect to any microcontroller or system with TTL serial, and send packets of data to take photos, detect prints, hash and search. This basic fingerprint reader also enrolls new fingers directly - up to 162 finger prints can be stored in the onboard FLASH memory. There's a red LED in the lens that lights up during a photo when working. In this case, when the person fingerprint was not stored in database, right after the sensor detect the finger, database Figure 12. Developed database interface – enroll new person displays “Unidentified Fingerprint” (Fig.11). So, if the fingerprints is not enroll, that means no assigning ID's to each Fig. 13 show the enroll routine when the developed print they can’t query them later. Using optical fingerprint database ask person to place the same finger again. After this sensor and the developed database will make adding step the fingerprint template is stored and some fields can be fingerprint detection and verification simple. filled, also a person photo can be uploaded. Figure 11. Developed database interface Figure 13. Developed database interface – fingerprint stored 33 Technical support by The research results, through the biometric identification system for emergency medical situations, allow the response time optimization, easy access and enhanced security ensuring the primary emergency medical information. This paper presents a new approach to primary healthcare information storage based on biometric identification, using a fingerprint sensor. More specifically, in case of an emergency medical situation, the system allows the identification of the patient by the medical personnel based on fingerprint which represents the key to access the relevant medical information, previously stored in database. REFERENCES Figure 14. Developed database interface – identify person Fig. 14, 15 describe routine when the developed database query all stored fingerprints in order to find the one [1] J. Anil, L. Hong, S. Pankanti, "Biometric identification." corresponding to the queried person. As shown accessing data Communications of the ACM 43.2, 2000, pp. 90-98. from database displays a medical repot about the person and [2] Sanchez-Reillo Raul, Carmen Sanchez-Avila, Ana Gonzalez-Marcos. also his corresponding fingerprint ID. "Biometric identification through hand geometry measurements." Pattern Analysis and Machine Intelligence, IEEE Transactions on 22.10, 2000, pp. 1168-1171. [3] Ribaric Slobodan, Ivan Fratric. "A biometric identification system based on eigenpalm and eigenfinger features." Pattern Analysis and Machine Intelligence, IEEE Transactions on 27.11, 2005, pp. 1698-1709. [4] Im Sang-Kyun, et al. "An biometric identification system by extracting hand vein patterns." JOURNAL-KOREAN PHYSICAL SOCIETY 38.3, 2000, pp. 268-272. [5] De Luis-Garc a Rodrigo, et al. "Biometric identification systems." Signal Processing 83.12, 2003, pp. 2539-2557. [6] Kwapisz Jennifer R., Gary M. Weiss, Samuel A. Moore. "Cell phone- based biometric identification."Biometrics: Theory Applications and Systems (BTAS), 2010 Fourth IEEE International Conference on. IEEE, 2010, pp. 1-7. [7] M. Leba, R. Dobra, A. Ionica, Method of Storing Relevant Medical Information Based on Biometric Identification, O.S.I.M. Romanian Patent Application, 2014 [8] Van der Putte, Ton, Jeroen Keuning. "Biometrical fingerprint Figure 15. Developed database interface – accessing data recognition: don’t get your fingers burned." Smart Card Research and The identification system recognizes an individual by Advanced Applications. Springer US, 2000, pp. 289-303. [9] Mong Gary M., C. E. Petersen, T. R. W. Clauss. Advanced fingerprint searching the entire enrollment template database for a match. analysis project fingerprint constituents. No. PNNL-13019. Pacific It conducts to comparisons to establish if the individual is Northwest National Lab., Richland, WA (US), 1999. present in database and if so, returns the identifier of the enrollment reference that matched. In an identification system, the system establishes a subject’s identity without the subject having to claim an identity. The template may be stored in internal storage devices like the basic fingerprint reader internal memory or external internal storage like the developed database. V. CONCLUSIONS The procedures for individuals' identification and medical information storage must ensure prompt, easy and safe identification of those who need emergency medical services. This requires a means to identify people based on a cheap technology, using easy matching analysis that does not require complex electronic devices and this is possible by means of fingerprint scanning. 34 Technical support by Realization of a Performant Stand for Determining Equivalent Initiation Capability for Detonators with Modern Electronic Measuring Facilities Rus Daniela Carmen 1, Kovacs Attila 2, Ghicioi Emilian 3 1,2,3 Department SEAP, National Institute for Research and Development in Mine Safety and Protection to Explosion – INCD- INSEMEX Petrosani, Hunedoara / Romania, [email protected], [email protected], [email protected] Abstract -This paper concerns the development The group of means of initiation falls detonators and infrastructure of the testing and evaluation of provision the detonating fuse containing explosive substances essential safety requirements for determining equivalent characterized by high sensitivity, to initiate explosives and initiation capability, to initiate explosive detonators in detonate very easily when subjected to a heat, shock or according to the essential safety requirements of Annex 1 of dynamic impulse. In this way, dynamic or initiating impulse Directive 93/15/EEC evaluation implemented in Romanian legal regulations. wave forms, the energy producing the blasting explosives detonation. (Dumitru F, „Ingineria împuşcărilor materiale Series of standards SR EN 13763, harmonized with şi tehnici de lucru”, Volumul 1) [2]. Directive has CEN developed, and adopted in the Romanian For detonators and relays are in force harmonized Standardization Committee "Explosives for civil uses" standards taken as Romanian standards, SR EN 13763 series regarding essential safety requirements with describing the technical requirements to be applied, to perform laboratory of standards. and functioning (blasting) tests, in order to assess the essential DESCRIPTION OF THE TECHNICAL SOLUTIONS safety requirements in accordance with the Directive. ADOPTED TO ACHIEVE WITH EQUIPMENT Keywords— detonators, initiation capacity, explosion IMPLEMENTED IN INSEMEX GROUP OF pressure, aquatic environment, pressure sensors, measurement LABORATORIES (GLI), TEST PROCEDURE PI-85 system. INCD INSEMEX had accredited and applied the procedure PI ETI 05.1 "Evaluation/ Determination of initiation I. INTRODUCTION (capacity) of electric detonators (including serial Essential safety requirements on explosives for functioning), of pyrotechnic detonators, non-electric civil uses are listed in the Annex I of the Directive initiation systems and relays retarders." Change and 93/15/EEC, HG 207/2005 and these requirements describe adjustment of procedure required after applying new measures to evaluate any risk of detonators use, in all techniques to test and integrate it in quality system within situations and all operations related thereto [3]. INSEMEX GLI. These requirements stipulated in Anexe 1 are structured as The procedure describes how to check the operation of follows: electric detonators in series and the ability to initiate - General requirements for all explosives; (capacity) for all types of detonators in order to ensure full - Special requirements for all different types: satisfaction of the requirements of Directive 93/15/EEC • For all types (groups) of explosives for civil use; Annex 1 and in the technical file of the product [2]. • For different (groups) such as: The main modules of the testing equipment and their ▪Blasting explosives; functional roles are: ▪Detonating cords, safety fuses, other fuses and - a 1000 l capacity tank in metal grille to ensure hydraulic environment of plastic sono absorbent, with appropriate size shock tubes; requirements; ▪Detonators (including delay detonators) and relays; - a sensor positioning system that is trying the detonators in ▪Propellants and rocket propellants. the tail; - thermo - Hygro - Barometer PCE type - THB 40; Assessment of requirements primarily involves - blasting system (blasting machine and blasting cable); identifying and selecting specific requirements that are - data acquisition system consists of piezoelectric pressure applicable to a particular product, evaluating and recording transducer KISTLER type 603B (Pmax = 200 bar, response their witch did not apply to some other requirements set out time <2μs), KISTLER type 5018 signal amplifier (analog in Directive / HG 207/2005. 35 Technical support by output controlled by PC) and with an appropriate sensitivity generates a shock wave and a spherical volume of gas that amplification factor type oscilloscope OX 6152-C METRIX expands and then break as the bubbles rise on water surface. which takes the data measured with the possibilities of The shock wave and gas volumes are a finite relation to the receiving and signal processing, data storage and provide it energy released. to the IT facilities; Thus are made, by measuring: - computing platform from test results consisting in laptop • Pressure shock wave; computer on that have been run processing programs and • Time interval between the peak pressure of the data post processing from oscilloscope (SX-METRO shock wave and the first burst of bubbles of gas and Version 4.4./04.dec.2006 and Microsoft Excel). calculating the proportional parameters: • Equivalent energy of the shock wave; • Equivalent energy of the gas bubble; The developed energy of detonators can be compared with the energy raised from reference detonators to whom the producer declares equivalence. To test ability to initiate underwater described equipment was used mounted on the new stand on following figures. Explosion tank (pool) (capacity 1000 l) with metal grille, to ensure hydraulic environment of sono-absorbent plastic, with appropriate size (Figure 2) [1]. Fig.1. The water tank for carrying out the test, measurement system and data processing. Pilot tests were conducted applying a new procedure with code GLI PI-85, which transposes the requirements of EN 13763-15, completing provisions PI- ETI-05.1 accredited procedure „Determination of equivalent initiating capability for detonators and relays” [4]. The integrated system has been mounted inside the ground test facilities of INSEMEX and pilot tests were conducted using electric blasting detonators type Fig.2. Tank with capacity of 1000 l ERGODET no. RO/01/2013 Polish manufactures and MMSED no.68/E/2012 Bosnia-Herzegovina manufactured Positioning system for pressure sensor and detonator. The of various batches with different numbers of delay steps. distance between the sensor and the position of the The data acquisition and processing proved to be detonator should be (400 ± 5) mm. The bottom of detonator compatibility with the objectives of the test, introducing and the sensor must be at (400 ± 5) mm below the surface of database record results. the water. The distance from the every wall from detonator must be at least 200 mm. (Figure 3) [1]. FULFILLING THE REQUIREMENTS OF THE STANDARD SR EN 13763-15 „EXPLOSIVES FOR CIVIL USES. DETONATORS AND RELAYS. PART 15: DETERMINATION OF EQUIVALENT INITIATING CAPABILITY” WITH THE NEW INFRASTRUCTURE The project aims to achieve a advanced equipment to conduct the test to determine initiation capacity regarding the requirements in Annex I of Directive 93/15/EEC to ensure a higher technical level. The initiation of the explosive charge of detonators should provide an explosion with an sufficient energy to initiate the explosive or detonating wire or shock tube with Detonator. which it is intended to use [4]. Pressure sensor with a response time of <2 m Test on "ability to initiate underwater" - is based on Fig.3. Positioning system the principle that an explosive charge detonated under water 36 Technical support by Amplifier with proper amplification and facilities for - is the time in seconds at which the sensor power down to connecting sensors and oscilloscope. P max / E, where P max is the maximum measured pressure (Figure 4). and e is the base of natural logarithms. Memory oscilloscope with a sampling frequency of min.10 MHz. (Figure 5). It will calculate the individual values, the mean and standard deviation for detonators tested and the reference detonators. Calculate it will the individual values, the mean and standard deviation for detonators test and the detonators staples. B. Energy equivalent of the bubble Energy of the bubble in s3 can be calculated using the following formula based on the elapsed time between the peak pressure of the shock wave and the first burst of gas bubbles produced by the reference detonation: E = t 3 [s3] (2) b b Fig.4. KISTLER signal amplifier type 501 where: tb - the bubble period, in seconds, from the tip of the shock wave pressure and breaking the first gas bubbles produced by the detonation gases. Have calculated individual values, mean and standard deviation of test detonators and reference detonators. SAMPLE RESULTS FROM UNDERWATER TESTING Typical curves for peak pressure and bubble period according to SR EN 13763-15 determined with the new equipment [4] Fig.5. METRIX type memory oscilloscope OX 6152-C No. Individual value Medium Value Thermo hydro barometer to measure atmospheric Equivalent Energy Equivalent Energy pressure type PCE-THB 40. energy of equivalent of energy of shock equival shock wave the bubble wave ent of Product 2 3 2 CALCULATION OF RESULTS [Pa . s] [s ] [Pa . s] the bubble 3 A. Equivalent energy of shock wave [s ] 1943301,27 4,3527 x10-13 By means of the output voltage of the pressure 3953794,03 1,728 x10-15 -14 sensor, the computer software based on the curve calculated 15033094,30 1,75 1. X 34116318,6 3,2768 x10 full pressure / time square, from which one can obtain -13 34226318,6 3,2868 x10-14 x10 equivalent impact energy Es, Pa2 s, with the general PRODUCT -13 925738,78 3,7325 x10 formula: 16791836,9 4,096 x10-15 2376799,23 3,2768 x10-14 t -14 5958433,31 6,1 x10- 2 2 2. Y 2422788,9 3,0899 x10 14 E P dt [Pa . s] (1) -14 s 4967254,48 6,4 x10 t0 PRODUCT 3233487,03 1,7562 x10-13 where: P - is the pressure measured in Pascal; 37 Technical support by non-electric) with various explosive charges as type of detonators and quantity [1]. Tests carried out on the equipment are designed for on assessments certification of products, technical expertise as a result of incidents, at the request of the authorities, namely the use of new research projects aimed detonators. REFERENCES [1] Daniela Carmen Rus, and others, PN 07 45 03 09 – „Development of laboratory logistics by creating a test equipment for detonators to t determine the initiation capacity”. National Research Program, 2013. Legend: 1 Pressure (kPa); [2] Dumitru Fodor, „Engineering materials and techniques of blasting 2 Time (ms) work", Volume 1, Publisher NAMASTE Timişoara, and Publisher CORVIN Deva, 2007. Fig.6. Example of pressure curve of commercial detonators tested, [3] Council Directive 93/15/EEC of 5 April 1993 on the harmonization of detonated underwater in accordance with the method described in this the provisions relating to the placing on the market and supervision of document (PRODUCT X). explosives for civil use. [4] SR EN 13763-15 – „Explosives for civil uses. Detonators and relays. P Part 15: Determination of equivalent initiating capability” ASRO, November 2005. t Legend: 1 – Pressure (kPa); 2 – Time (ms); 3 - Bubble period. Fig.7. Registered data of bubble, base period of elapsed time according to EN 13763-15 (PRODUCT Y). The first part of the diagram (0 ms to 5 ms) is showed the compressed version of the reading. Loop period is the time between the peak and the peak shock of the explosion wave of the first burst of the bubble. Conclusions: The test for determining the equivalent initiating capability consist in tests measurement according to SR EN 13763-15 ensure explosion pressure measurement in a hydrostatic environment, accurate records for two indicators, namely equivalent shock wave energy expressed in Pa2x s respectively in equivalent energy of the gas bubble that expressed in s3. The development of equipment allows for testing and evaluation of different types of detonators (electric, 38 Technical support by The Importance of Non-Transmission Test of an Internal Ignition for Electrical Motors with Type of Protection Flameproof Enclosure ,,d” Fotău Dragoș, Burian Sorin, Friedmann Martin, Magyari Mihai, Moldovan Lucian, Rad Daniel, Department D.S.I.E.A National Institute for Research and Development in Mine Safety and Protection to Explosion – INCD – INSEMEX Petrosani, Hunedoara / Romania [email protected], [email protected], [email protected], [email protected], [email protected] Abstract—The type of protection flameproof enclosure applies generally to electrical equipment which in normal operation THE CONSTRUCTION OF FLAMEPROOF ENCLOSURE PROTECTION produces electrical arcs and sparks, and consists in placing the EQUIPMENTS parts that could ignite an explosive atmosphere inside of an The explosive environment which may appear in relation enclosure that can withstand the pressure developed during an internal explosion of an explosive mixture and which prevents the to the application of explosive hazardous materials within the explosion transmission to the explosive atmosphere surrounding technologic process is able to reach the interior of the the enclosure. The paper presents why is important the non- enclosure, where it may catch fire and therefore explode. transmission test for the electrical motors, usen in hazardous But due to the fact that all the exterior connections of the areas. enclosure are made within a special slot, calibrated joints, the Keywords—Flameproof enclosure, test, non-transmission, type explosion is not transferred from the enclosure to the of protection. environment. Therefore, the joints of the enclosure are made using I. INTRODUCTION geometrical interstitions and precise dimensions. The property The use of electrical equipments in potentially explosive of these interstitions is to cool down the flame and the debris atmospheres requires special attention, namely from the point resulted from an interior explosion, due to the absorption of of view of their construction and maintenance. Together with heat by the material of the flanges and of the adiabatic the technological developments, these objectives can easily be expansion (Figure 1). Therefore, cooled down, the burnt gases met, objectives which are defined by European standards and the resulted metal particles inflamed by the explosion, concerning the construction, use and maintenance of explosive they are not capable of igniting the explosive mixture of the proof equipments. environment upon their exit. The protection of the equipment with flameproof The flameproof enclosure protection method is based on enclosures is one of the most spread means of ensuring the the mechanical resistance of an enclosure to cope with an protection from explosion of electric machineries, devices and internal explosion. It should not allow for the internal equipments, assuming therefore the enclosure of their explosion to propagate within the inflammable environment electricity conducting parts in solid, resistant cases, capable to surrounding the enclosure. retain the pressure of an explosion without damaging or permanently deforming it and which do not allow the explosion to be transferred to the explosive atmosphere. Generally, the flameproof enclosed electric devices are used in: - chemical and petrochemical industry; - oil and gas extraction; - pumping stations for explosive environments; - car industry and light industry dyeing houses, etc.; - inflammable materials warehouses, etc.; The flameproof enclosed electric equipment is generally Fig 1. Flameproof enclosure example used in any place where gases, vapors or mists may appear, which mixed with air may give birth to explosive mixtures. Each enclosure shall allow to be opened in order to allow the installation of the interior equipments. It results therefore, 39 Technical support by as a natural consequence, that when the enclosure is closed, TYPE TESTS FOR EQUIPMENT WITH PROTECTION TYPE there may be an interstition between the surfaces of the lid and FLAMEPROOF ENCLOSURE those of the enclosure. The joint surfaces need to meet a series Acording to ATEX Directive 94/9/EC there are some of dimensional requirements, because in the case of an internal conditions wich have to be fulfilled before putting the ignition, the resulted debris cannot exit through the void products used in potentially explosive atmospheres on the between the joints. The length and the interstition of the market. flameproof joint need to be well defined. The enclosure itself needs to be verified considering the operation together with The flameproof enclosure equipments are build according the equipments mentioned which can be found inside it. with the requirements of SR EN 60079-0 and SR EN 60079-1. After the equipment are build it must be testesed. Inside the The dimensions of the enclosure are determined in order specific standard for flamepeproof enclosure equipments (SR for the enclosure to bear the pressure created during the EN 60079-1) there is specified that there are 3 specific tests explosion of the mixture inside without suffering any damages for those enclosures or deformations capable of modify the dimensions of the flameproof joints. The different compartments of the -determination of the explosion pressure (reference flameproof enclosures are isolated by partition walls. The pressure); assembling elements such as bolts and tenons should not cross -overpressure test; the wall of the enclosure because if such an element is lost -test for non-transmission of an internal ignition. then the flameproof character is lost by creating a direct Before introducing into the test chamber, the equipment connection of the interior and exterior of the enclosure. have to be prepared, like in figure 2. Normally, the bolts and tenons are made of OL37. Cylindrical head bolts and hexagonal socket, or hexagonal head bolts are used. DETERMINATION OF EXPLOSION PRESSURE The reference pressure is the highest value of the Explosionproof electrical apparatus, with the type of maximum smoothed pressure, relative to atmospheric protection flameproof enclosure „d”, is included in one (or pressure, observed during these tests. possible both) of the following two groups: Each test consists of igniting an explosive mixture inside Group I: electrical apparatus for mines susceptible to the enclosure and measuring the pressure developed by the firedamp; explosion. Group II: electrical apparatus for places with an The number of tests to be made and the explosive mixture explosive gas atmosphere other than mines susceptible to to be used, in volumetric ratio with air and at atmospheric firedamp. pressure, are as follows: For the type of protection flameproof enclosure, group II – electrical apparatus of Group I: three tests with (9,8 electrical apparatus is divided in 3 subgroups: IIA, IIB and 0,5) % methane; IIC. – electrical apparatus of Group IIA: three tests with This dividing of group II in 3 subgroups (for the type of (4,6 ± 0,3) % propane; protection flameproof enclosure) was made in the basis of the – electrical apparatus of Group IIB: three tests with (8 maximum experimental security gap (MESG), which is ± 0,5) % ethylene; decreasing in value from group IIA to Group IIC, as follows: – electrical apparatus of Group IIC: three tests with (14 ± 1) % acetylene and three tests with (31 ± 1) % hydrogen. -in IIA subgroup are included gases with MESG > 0,9 mm; -in IIB subgroup are included gases with 0,5 mm < MESG < 0,9 mm; OVERPRESSURE TEST -in IIC subgroup are included gases with MESG < 0,5 mm. This test shall be made using either of the following The specific elements that ensure the protection to methods, which are considered as equivalent. explosion for a flameproof “d” equipment are represented by a) Overpressure test – First method (static) flameproof joints. A flameproof joint represents the place The relative pressure applied shall be where the corresponding surfaces of two parts of an enclosure, or the conjunction of enclosures, come together, and which -1,5 times the reference pressure, with a minimum of prevents the transmission of an internal explosion to the 3,5 bar, or explosive atmosphere surrounding the enclosure. -4 times the reference pressure for enclosures not subject There are three types of flameproof joints: non-treaded, to routine overpressure testing, or threaded and special joints. -at the following pressures, when reference pressure determination has been impracticable. 40 Technical support by Volume Pressure Group cm3 bar Group iE / iC Mixture 10 I,IIA, IIB, IIC 10 Group I 0,75 55 % H2 ± 0,5 >10 10 0,6 50 % H2 ± 0,5 >10 I,IIA, IIB,IIC 15 Group IIA 0,75 50 % H2 ± 0,5 >10 20 0,6 45 % H2 ± 0,5 Group IIB 0,75 28 % H2 ± 1 The period of application of the pressure shall be at least 0,6 28 % H2 ± 1 10 s and the test is made once. at 1,4 bar b) Overpressure test – Second method (dynamic) –precompression of the normal test mixtures according to The dynamic tests shall be carried out in such a way that the following formula: the maximum pressure to which the enclosure is subjected is 1,5 times the reference pressure, but with a minimum of iC 3,5 bar. Pk = 0,9 iE When the test is carried out with mixtures specified for the determination of the reference pressure, these may be where Pk is the precompression factor. precompressed to produce an explosion pressure of 1,5 times the reference pressure. b) If enclosures of Groups IIA and IIB could be destroyed The test shall be made once only except for electrical or damaged by the test for non-transmission of an internal apparatus of Group IIC for which each test shall be made ignition, it is permitted that the test be made by increasing the three times with each gas. gaps above the maximum values specified by the manufacturer. The enlargement factor of the gap is 1,42 for Group IIA electrical apparatus and 1,85 for Group IIB TEST FOR NON-TRANSMISSION OF AN INTERNAL electrical apparatus. The explosive mixtures to be used in the IGNITION enclosure and in the test chamber, in volumetric ratio with air Gaskets shall be removed and the enclosure is placed in a and at atmospheric pressure, are as follows: test chamber. The same explosive mixture is introduced into the enclosure and the test chamber, at atmospheric pressure. -electrical apparatus of Group IIA: The flamepath lengths (engagement) of threaded joints of (4,2 ± 0,1) % propane; the test specimen(s) shall be reduced. -electrical apparatus of Group IIB: Flanged gaps of spigot joints, where the width of the joint (6,5 ± 0,5) % ethylene. L consists only of a cylindrical part shall be enlarged to values of 1 mm for Groups I and IIA, 0,5 mm for Group IIB and The test for non-transmission of an internal ignition shall 0,3 mm for Group IIC. be made five times. The test result is considered satisfactory if the ignition is not transmitted to the test chamber. ELECTRICAL APPARATUS OF GROUPS I, IIA AND IIB a) The gaps iE of the enclosure shall be at least equal to 90 % of the maximum constructional gaps iC as specified in the manufacturer's drawings (0,9 iC iE iC). The explosive mixtures to be used, in volumetric ratio with air and at atmospheric pressure, are as follows: –electrical apparatus of Group I: (12,5 ± 0,5) % methane- hydrogen [(58 ± 1) % methane and (42 ± 1) % hydrogen] (MESG = 0,8 mm); –electrical apparatus of Group IIA: (55 ± 0,5) % hydrogen (MESG = 0,65 mm); –electrical apparatus of Group IIB: (37 ± 0,5) % hydrogen (MESG = 0,35 mm). If the gaps of a test specimen do not fulfill the above condition, one of the following methods may be used for the type test for non-transmission of an internal ignition: Fig.2 Example of a flameproof equipment test sample preparation for testing in explosive atmosphere – a gas/air mixture with a smaller MESG value: 41 Technical support by REFERENCES CONCLUSIONS This paper revealed the importance of non-transmission test [1] L. Moldovan M. Friedmann M. Magyari and D. Fotau “Considerations of an internal ignition and he influence of the gap over the Regarding The Tests In Explosive Mixtures Made Upon The Electrical precompression factor of the explosive mixture for the specific Apparatus With The Type Of Protection Flameproof Enclosure “d” tests. [2] SR EN 60079-1:2008, Explosive atmospheres - Part 1: Equipment protection by flameproof enclosures "d", ASRO, 2008 This type test is made with the purpose of verifying the [3] *** SR EN 60079-0:2010, Explosive atmospheres - Part 0: Equipment – capacity of not-transmitssion the the internal ignition of an General requirements, ASRO, 2010 explosion mixture outside the equioment, ant that is the final [4] Directive 94/9/EC of the European Parliament and the Council of 23 March 1994 on the approximation of the laws of the Member States and the most importan test for the equioments with tipe of concerning equipment and protective systems intended for use in protection flameproof enclosure,,d”. potentially explosive atmospheres. Official Journal of the European Communities No. L 100, 1994-04-19, Brussels If the equipment passes this test, is ready and safe to be used in hazardous areas. 42 Technical support by Developing Real-Time e-Health Parameter Monitoring Interfaces with 8- bit Microcontrollers Marc Gheorghe, Boca Maria Loredana University “1 Decembrie 1918” of Alba Iulia, Department of Science and Engineering, Blvd. N. Iorga, No. 11-13, Alba Iulia, Romania [email protected], [email protected] Ömer Meşedilci University “Gelişim” of Istanbul, Department of Science, Dolum Tesisleri yolu, Sht. Pyd. Er. Murat Şengöz Sk. No:8 Avc lar/ İstanbul, Turkey, [email protected] Abstract— In present telemedicine is an increasing areain two 802.15.3/4 (Howitt 2003, FCC) technology, UWB/WiMedia aspects: in the diagnosis of some disease and in the treatment and (Soomro 2004) systems, but in the same time local wireless intervention in critical situations. networks (WLANs), for example IEEE 802.11a/b/g/e/h/I Often to establish one exact diagnosis, the doctor needs to (Cordeiro 2003). At this we can add follow the patient's condition over several days and his reactions, the patient's vital functions during different times of the daily wireless wide area network (WWAN), like 2G, 2.5G şi 3G, activities and in different conditions of exercise or rest. and Internet which can be used for improve health system On the other hand, in the treatment of some critical diseases care for a wide range of users. we need a short time of intervention. In this paper we treated the monitoring of physiological To achieve greater mobility and global care connectivity, it parameters using microcontrollers and remote data transmission requires the integration of heterogeneous wireless via the Internet and creating a database. technologies. 3G cellular network has many restrictions and is not able to support medical applications that require real-time work for many patients. Keywords— telemedicine, monitoring, real-time, e.-Health As result, WBAN, WPAN and WLAN technologies, which typically can support high data communications, but I. INTRODUCTION smaller coverage area than cellular systems, can be integrated and used to achieve connectivity in non- medical and medical Telemedicine is an area of increasing current sick care, environment in hospitals, clinics and home or residential care. especially those chronically ill and enables remote receiving data related to patients or real-time diagnosis of patients who In the past, medical services have allocated particular are critically ill in remote locations [1]. frequency bands. Now, it is good that this technologies WBAN , WPAN and WLAN used in medical applications has Telemedicine offers the possibility of simultaneously to use same infrastructure and same frequency bands as well remote monitoring of a large number of patients and allows as non - medical applications: administration, entertainment, fast intervention in critical cases. voice communication. This is necessary to: 1 Frequency bands allocated to medical applications - such as WMTS (Wireless Medical Telemetry Service) are too small to handle large flows of data required to achieve quality care. II. TELEMONITORING- CURENT CASE Radio systems which works in bands without license - Recent advances researches in wireless technology have such as WBAN , WPAN and WLAN are becoming popular created new opportunities that could revolutionize the delivery and used in hospitals and homes. of health services ( Soomro 2007). Wireless connectivity and Shield eHealth V2.0 allows Arduino and Raspberry Pi support infrastructure provides everywhere mobile real-time users, to perform biometric and medical applications, when is monitoring of the patient (Istepanian 2004) as well as location necessary to use 10 different sensors for monitoring human tracking system in case of alarm ( emergency ) ( Liszka 2004). body: pulse, blood oxygen ( SpO2 ), airflow ( breathing), body These are just some example applications on medical wireless temperature , electrocardiogram (EKG ) , meters , galvanic technologies, but the opportunities still increasing. For skin response ( GSR - sweat) , blood pressure ( blood pressure example, personal wireless networks develop (Wireless ) , body position sensor ( accelerometer ) and muscle ( EMG ) Personal Area Networks – WPANs), wireless networks . WBAN (Wireless Body Area Networks - WBANs), 802.15.1/Bluetooth technology (Lorincz 2004), IEEE 43 Technical support by This information can be used to monitor in real-time a patient or obtain sensitive data for further analysis in medical diagnosis. Biometric information can be sent using wireless communication such as: Wi - Fi , 3G , GPRS , Bluetooth , 802.15.4 and ZigBee depending on the application . If is needed to make a real-time diagnostic image, a camera can be attached to the 3G module, used to send photos and videos of the patient to a medical diagnostic center . To make a permanent storage or viewing real-time data, this data are sent directly to a laptop or smartphone. To support this were created applications for iPhone and Android in order to easily see the patient information. Fig. 11. The e-Health shield over Arduino [2] Fig. 10. Equipments for health monitoring Fig 3. The e-Health shield over Arduino [2] 2.1. The electrical characteristics of the platform 2.3 The e-Health sensor platform The e -Health Shield can be powered from the PC or from The e-Health sensor platform use C++ classes that allows external power source. Some of the USB ports on computers to easily read all sensors and send information using any are not able to provide all the power needed to operate the available wireless interfaces. This class provides a simple module and if the module has problems when is running, can open source system to used . be use an external power supply (12V - 2A ) of the Arduino / RasberryPi . In order to ensure the same code on both platforms ( Arduino and Raspberry Pi ) is good to use ArduPi libraries 1.2. Shield e-Health that allow developers to use the same code. The images below shows the e -Health shield over 2.4 Characteristics of the E -Health Platform Sensor Arduino: Breathing sensor characteristics Abnormal respiratory frequency and respiratory rate changes are an important indicator of physiological instability, and in many cases respiratory rate is an important indicator of this instability. It is therefore very important to monitor respiratory rate as an indicator of the human condition. Airflow sensor can provide early warning of hypoxemia and apnea. Breathing sensor is a device used to monitor a patient's respiratory that needs help to breathe [6]. 44 Technical support by It consists of a flexible wire that sits behind the ear and the two nasal devices that are placed in the nostrils. It is specifically designed to allow the thermocouple sensor to be placed in optimum position to observe exactly the changes in air flow and nasal air temperature. Breathing sensor has the advantage that is comfortable, adjustable and easy to install. Fig 5. USB output, using the terminal serial port of Arduino IDE [3]. IV. TEMPERATURE SENSOR CHARACTERISTICS Body temperature depends on where measurement is made in the human body, the time of day and the level of activity of a person. Different body parts have different temperatures. Fig 4. Breathing sensor One adult people has a breathing rate of 15-30 times per one minut. III. THE FUNCTIONS OF BREATHING SENSOR CLASSES This sensor is connected at Arduino via an analogue input Fig. 6. Temperature sensor and return values between 0-1024. With following functions can get directly this value and print a waveform on monitor Accepted average temperature taken on human body is series. 37.0 ° C ( 98.6 ° F ). In healthy adults , the body temperature fluctuates by about 0.5 ° C (0.9 ° F ) throughout the day , with lower temperatures and higher temperatures in the morning, Example: afternoon and evening, depending on the needs and activities body. { Measuring body temperature has a great medical int airFlow = eHealth.getAirFlow(); importance because many diseases has like symptom changes eHealth.airFlowWave(air); in body temperature. Also during some diseases can be monitoring the body temperature, and such way the doctor can } check the effectiveness of treatment initiated [7]. Once the code is loaded, it is follow the result on the serial Body temperature sensor accuracy is enough for most monitor. Here is the USB output, using the terminal serial port applications, but though the calibration process can be of Arduino IDE . improved the accuracy. When using the temperature sensor actually measuring a voltage that is related to the temperature 45 Technical support by of the sensor. If the measurement voltage errors can be Fig. 7 Block scheme of a temperature sensor [6] avoided and can be show the relationship between voltage and temperature, it can be obtained better readings of temperature. REFERENCES Calibration is the process of measuring the actual values of voltage and resistance. [1] Electronic integrated services, S. Puscoci, R. Dragomir, Designing Ambient Interactions for Older Users, European Conference on Ambient Intelligence – AmI09, 18-21 Nov. 2009, Salzburg, Austria ONCLUSION AND FUTURE WORK V. C [2] Home homecare teleassistance system, Sorin Pușcoci, Dorin Berian, Florin Șerbănescu, RO-MEDINF 2010 – 31st National Conference On Medical Informatics, 18-20 Nov 2010, ARAD, Romania The practicalanalysis and measurements during testing [3] Integrated teleassistance platform with enhanced accessibility to revealed safe operation monitoring device and data information – TELEASIS, S. Pușcoci, L. Stoicu-Tivadar, V. Stoicu-Tivadar, transmission without loss of information. D. Berian, F. Șerbănescu, S. Ioniță, F. Băjan 6th IEEE International Symposium On Applied Computational Intelligence And Informatics (SACI It also can be transmitted in addition to the monitoring data 2011), May 19-21, 2011 In Timisora, Romania various warning signals to overcome some parameters that can [4] Romanian multimedia platform for medical teleservices be set in the microcontroller software. implementation, S. Puşcoci, F. Șerbănescu, F. Băjan, B. Dionisie, Healthcare IT Management vol.3, Issue 3, Autumn 2007 Future concerns will be focused first on improving the quality of the sensors and the use of components with low [5] Tele-assistance integrated services, Sorin Puşcoci, Revista : energy consumption, mainly following the models of Telecomunications nr.2/2010 monitoring devices, increasing the power life of the battery [6] TELEASIS –, Sorin Pușcoci, Lăcramioara Stoicu-Tivadar, Florin using low power devices. Șerbănescu, Flavius Băjan, Ion Bogdan, Silviu Ioniță, Cristian Ciortan, Revista : Telecomunicaţii, nr.1 2010 [7] Webservice-based solution for an intelligent telecare system, Vasile Stoicu-Tivadar, Lăcrămioara Stoicu-Tivadar, Sorin Puşcoci, Dorin Berian,Vasile Topac, Individual Book Chapter in Studies in Computational Intelligence, Volume 378/2012, 383-408, DOI: 10.1007/978-3-642-23229- 9_18 46 Technical support by Improving Security and Selectivity Requirements in Telemedicine by Combining Hardware and Software Encoders-Decoders Marc Gheorghe 1, Dobra Remus 2 1University of “1 Decembie 1918” Alba Iulia, Department of Exact Sciences and Engineering Alba Iulia, Str. Gabriel Bethlen Nr.5, Alba, Romania Tel.: +40 258 806 130, Email: [email protected] 2University of Petrosani, Department of Computer and Electrical Engineering Str. Universităţii nr. 20, 332006 Petrosani, Romania Tel.: + 40-254-549011, Fax: +40-254.546238, Email: [email protected] Abstract— Voice recognition in computer engineering can be Speech coding is an application of data compression of described as a way of entering data into the computer by digital audio signals containing speech. Speech coding uses identifying spoken words. The developed hardware and software speech-specific parameter estimation using audio signal coders-encoders system provides a method for improving processing techniques to model the speech signal, combined security and selectivity requirements in telemedicine and one of with generic data compression algorithms to represent the the main uses of the system can be voice commands to operate resulting modeled parameters in a compact bitstream [1]. various appliances lighting for people with limited mobility or immobilized. Currently voice command is often used in practice to different commands for driving robots of smart appliances, or HARDWARE SYSTEM FOR SPEECH RECOGNITION even securing some military strategic areas, all this, however, The paper deals with the possibility of transmitting requires specialized software and ultimately leading to high costs. commands to various actuators drive using LabVIEW In the present work we have achieved a method for control development environment and facilities for processing and devices without modifying them. transmission of signals provided by microcontrollers. Keywords— telemedicine; encoders; decoders; speech Fig. 1 describe the block diagram of speech recognition recognition; system where voice command is transmitted from a wireless microphone is received by the computer, after that the signal is I. INTRODUCTION processed using LabVIEW environment and transmitted to the microcontroller which encodes the digital transmission by one In the medicine sector, speech recognition can be of the execution devices. implemented in front-end or back-end of the medical documentation process. Front-end speech recognition is where the provider dictates into a speech-recognition engine, the recognized words are displayed as they are spoken, and the dictator is responsible for editing and signing off on the document. Back-end or deferred speech recognition is where the provider dictates into a digital dictation system, the voice is routed through a speech-recognition machine and the recognized draft document is routed along with the original voice file to the editor, where the draft is edited and report finalized. Deferred speech recognition is widely used in the industry currently. Most voice codecs utilize a source-filter model to encode the speech signal. The signal is modeled as the convolution of excitation signal generated by the vocal chords with a time- varying all-pole filter that represents the shape of the vocal tract. The transform-domain signal is then directly quantized and encoded in some manner. Because speech recognition features are derived from the Fourier transform of the input Figure 1. The operation principle of the speech recognition system signal, it is more efficient to compute them from transform- Voice recognition is the process of converting a speech domain frequency representation. signal to form a sequence of digital data words by means of an 47 Technical support by algorithm implemented as a computer program. Recognized command and the wireless transmission of encoded command words can be the final result, for applications such as by the identified device that must be controlled. command and control, data entry and document preparation. Fig. 2 shows the main source LabVIEW program that has a Recognition systems can be designed in many ways to structure of type state machine that seeks emergence of a new achieve a specific goal or performance criterion: event (a voice command) and when a new voice command is given is detected the corresponding case structure command. System dependent / independently of speaker, for the systems dependent of speaker the user must use Within these case structure subsystems is given the words or predefined phrases like we did in this switching off or switching on of LED corresponding to the application. These acoustic signals form the training mentioned command voice. It should be noted that the use of data that are used for voice recognition. Since these speech recognition function of the Windows operating system systems are used only for predefined speaker, their requires transmission controls only in English language. performance is higher compared to the speaker independent systems. Recognition of isolated words, it is also called discrete recognition system. In this system, there must be pauses between speaking words. Therefore, the system must take care to find the voice pauses between words. Continuous speech recognition, these systems are the ultimate goal of a process of recognition, no matter how or when a word is spoken, they are recognized in real time and the appropriate action is performed. Changes in the rate of speech, sloppy pronunciations, end of the word detection are the main problem with this way of recognition. The size of the vocabulary, the vocabulary size as a recognition system is higher than the performance is Figure 2. Basic fingerprint reader flowchart recognized. Specific tasks can use smaller Fig. 3 describes the front panel of the program written in vocabularies. However a natural system should LabVIEW and as one can see is a big advantage possibility to recognize the speaker independently and continuous track on the display the command and confirmation of the to a large vocabulary, which is more difficult. correctness of its execution. The codified transmission is Keyword recognition, these systems are used to performed using integrated circuits manufactured by Holtek detect a word in continuous speech. For this reason, Semiconductor, HT 12E and HT 12D types. they can be isolated as good recognition, these having the ability continuous administration. Coding and decoding the control signal is performed by means of specialized integrated circuits HT 12E – HT 12D, controls coding possibilities being 256, but in this paper the number of controls was limited to one execution device, with the mention that they can be easily extended to multiple of 4. From the encoder the command is transmitted by radio control actuators and the encoded signal is selected by one of the devices running. For of each execution device is assigned a digital code, so that he will recognize the radio that has the same code, thus only one of them can be activated by a command. One of the main uses of the system can be voice commands to operate various appliances lighting for people with limited mobility or immobilized. SOFTWARE SYSTEM FOR SPEECH RECOGNITION Figure 3. HT 12E Application circuit Transmission of voice commands is made through a For coding the HT 12E integrated circuit the chip has 8 wireless lavalier microphone via the speech recognition of 8 Windows and the signal is processed in LabVIEW pins, and the number of addresses is 2 which is the maximum programming environment, and transmitted to an electronic number of devices which can be controlled using this system, system with microcontroller which processes, selecting the therefore a sufficiently large number of it in order to meet any 48 Technical support by application requirements. HT12E has a transmission enable HT12D is a CMOS LSI IC and is capable of operating in a pin which is active low. When a trigger signal is received on wide voltage range from 2.4V to 12V. Its power consumption TE pin, the programmed addresses/data are transmitted is low and has high immunity against noise. The received data together with the header bits via an RF or an infrared is checked 3 times for more accuracy. It has built in oscillator; transmission medium. we need to connect only a small external resistor. As HT12E, it is available in 18 pin DIP (Dual Inline Package) and 20 pin HT12E begins a 4-word transmission cycle upon receipt of SOP (Small Outline Package) as given below. a transmission enable. This cycle is repeated as long as TE is kept low. As soon as TE returns to high, the encoder output Fig. 4 schematically show the coding and decoding completes its final cycle and then stops. circuits, as well as the radio transmission system between the two devices, was conducted with the experimental model type transceiver TWS-DS-6 and receiver RWS-371-6. Figure 4. Flowchart for ID that corresponds to the cabinets of family doctors Decoding is performed by the control signal integrated Figure 6. Encoding and decoding circuits diagram circuit HT 12D that can execute up to four independent devices or transmit four different commands to the same The two switches S1 and S2 are actually coding system device. HT12D is a 212 series decoder I2C (Integrated Circuit) and assignment of ID for each device to be controlled. for remote control applications manufactured by Holtek. It is Wireless radio transmitter module power high enough to cover commonly used for radio frequency (RF) wireless a transmission site and has the following main characteristics: applications. By using the paired HT12E encoder and HT12D Range is 433.92 MHz frequency, modulation mode decoder we can transmit 12 bits of parallel data serially. type is ASK (amplitude shift keying). HT12D simply converts serial data to its input (may be received through RF receiver) to 12 bit parallel data. These 12 Form of the circuit is SAW (surface acoustic wave) bit parallel data is divided in to 8 address bits and 4 data bits. data transmission rate is 8Kbps supply voltage is Using 8 address bits we can provide 8 bit security code for 4 1.5V to 12V bit data and can be used to address multiple receivers by using the same transmitter. Electrical current consumption varies depending on the supply voltage, so we 23mA 5V to 9V and 12V we have 40mA 53mA Power is 10dBm at 5V, 14,5dBm at 9V and 20dBm at 12V, temperature is from -20 to 85 ° C, sealing temperature of 230 ° C (10 seconds), plus high sensitivity. The receiver used in this application is RWS-371-6 and it has the following important features: Frequency range of up to 433.92 MHz; Module type is ASK modulation with LC circuit form; Spacing between channels is 500KHz supply voltage is 5V, the design makes it a very high sensitivity passive and is easily integrated into applications; Data transfer rate is 4800bps, the selection is made at -10dBm. Figure 5. HT 12D Application circuit 49 Technical support by The objectives pursued in making voice control device [4] J.C. Moreira, P.G. Farrell, ”Essentials of error-control coding”Editura were: Jphn Willey & Sons Ltd, The Atrium, anul 2006, Anglia. [5] Jan Axelson, ”Programming and circuits for RS-232 and RS-485 Links Create a simple voice command device, easy to use, and Networks”, Editura Lakeview Research, anul 1998, USA. and high reliability; [6] S. Barret, D. Pack, ”Atmel AVR Microcontroller Primer: Programming and Interfacing”, Editura Morgan&Claypool, anul 2008. Using some simple commands consisting of a single [7] S. Barret, ”Embeded Systems Design with the AVR Microcontroller”, word for the best recognition of the given command Editura Morgan&Claypool, anul 2010. thereby increasing the operational safety. [8] M. Mazidi, S. Naimi, ”The AVR microcontroller and embeded system using assembly and C”, Editura Prentice Hall, 2011. Basically we developed an electronic device driven by [9] Radu Lucian Lupşa, ”Noţiuni de teoria informaţiei”, Editura Casa Cârtii microcontroller with wireless transmission of voice commands de Ştiinţa, 2008. so that almost any device or electrical system be able to [10] Shannon, Robert V., et al. "Speech recognition with primarily temporal receive commands the only condition being that the device to cues." Science 270.5234, 1995, pp. 303-304. be supplied from the mains supply by means of a small static [11] Mohri Mehryar, Fernando Pereira, Michael Riley. "Weighted finite-state contactor, which in addition to switching function can receive transducers in speech recognition." Computer Speech & Language 16.1 an ID for identification and control selecting. (2002), pp. 69-88. [12] Lippmann, Richard P. "Speech recognition by machines and humans." Speech communication 22.1 (1997): 1-15. V. CONCLUSIONS [13] Pallett, D.S., Fiscus, J.G., et al., 1995. 1994 benchmark tests for the ARPA Spoken Language Program. Proc. Spoken Language Systems Making experimental module has led to several conclusions Technology Workshop. Morgan Kaufmann, Austin, TX, pp. 5–36. like: experimental device works securely up to a distance of [14] Paul, D., Baker, J., 1992. The design for the Wall Street Journalbased 100 m what is sufficient for the applications for which it was CSR corpus. Proc. DARPA Speech and Natural Language Workshop. made; in execution of commands no interferences from other Morgan Kaufmann, Austin, TX, pp. 357– 360. devices was founded or existing devices within range of the [15] Peskin, B., Connolly, S., Gillick, L., Lowe, S., McAllaster, D., Nagesha, V., Van Mulbregt, P., Wegmann, S., 1996. Improvements in wireless transmission system; in case of taking over false Switchboard recognition and topic identification. Proc. IEEE Internat. commands, control correction is very easy especially since Conf. Acoust. Speech Signal Process., pp. 303–306. there is the possibility of monitoring and confirmation of [16] Rabiner, Lawrence R., and Biing-Hwang Juang. Fundamentals of speech execution; this device is accessible and easy to use that recognition. Vol. 14. Englewood Cliffs: PTR Prentice Hall, 1993. ensures the user complete independence by making all [17] Allopenna, Paul D., James S. Magnuson, and Michael K. Tanenhaus. wireless controls, makes it especially useful for people with "Tracking the time course of spoken word recognition using eye movements: Evidence for continuous mapping models." Journal of certain mobility disabilities or persons temporarily memory and language 38.4 (1998): 419-439. immobilized. [18] Cooke, Martin, et al. "Robust automatic speech recognition with missing and unreliable acoustic data." Speech communication 34.3 (2001): 267- 285. REFERENCES [19] Pearce, David, and Hans-Gnter Hirsch. "The aurora experimental framework for the performance evaluation of speech recognition systems under noisy conditions." INTERSPEECH. 2000. [20] Friesen, Lendra M., et al. "Speech recognition in noise as a function of [1] M. Arjona Ramírez, M. Minami, "Low bit rate speech coding," in Wiley the number of spectral channels: comparison of acoustic hearing and Encyclopedia of Telecommunications, J. G. Proakis, Ed., New York: cochlear implants." The Journal of the Acoustical Society of America Wiley, 2003, vol. 3, pp. 1299-1308. 110.2 (2001): 1150-1163. [2] A. Spătaru, ” Teoria Transmisiunii Informaţiei”, Editura Didactica si [21] Lee, C. H., et al. "Acoustic modeling for large vocabulary speech Pedagogica, anul 1983, Bucureşti. recognition." Computer Speech & Language 4.2 (1990), pp. 127-165. [3] I. Angheloiu, ”Teoria Codurilor”, Editura Militara, anul 2002, Bucureşti. 50 Technical support by PIL Development of a Multiple Channel Wind Turbine Emulator using LabVIEW FPGA Topor Marcel Sorandaru Ciprian, Muntean Nicolae Politehnica University of Timisoara Politehnica University of Timisoara Department of Electrical Engineering and Industrial Department of Electrical Engineering Informatics, Hunedoara, Romania Timisoara, Romania marcel.topor@upt.ro [email protected], [email protected] Abstract—The purpose of this paper is to present the platform launched in 2008, which is based on the Zynq-7010 advantages of using LabVIEW FPGA tools for PIL development All Programmable SoC.(system on a chip ) The NI cRIO of a multiple channel wind turbine emulator. features 8 RIO (reconfigurable I/O) and the platform is based on a mix of discrete processors, FPGAs, and pluggable I/O Keywords—wind turbine, emulator, processor in the loop, modules. The addition of an FPGA into the NI RIO platform LabVIEW FPGA permits large performance improvements, on the order of 10x or more. This provides a significantly better performance in I. INTRODUCTION process-control loops for control applications based on the Processor in the loop or PIL is relatively new strategy for cRIO platform. testing software for control of real applications. PIL is a development process which is centered on a system model, from requirements capture and design to implementation and test. During the development of the control system the controller is developed using a generation of system design processes on the of graphical models developed in specialized software to design, analyze, and implement the software that models process performance and behavior. One of the steps of Model Based Design is Processor-in-the-Loop. During the processor-in-the-loop (PIL) phase the control algorithm is compiled and downloaded into an real embedded target processor and communicates directly with the plant model via standard communications such as Ethernet, Modbus Etc. The plant model however is is modeled using a high degree of fidelity using either MATLAB - Simulink, C++, or any other modeling language. In this case, no I/O devices are used for the communication and the controller operates in purely Fig. 1. PIL concept for emulation virtual plant. The advantage of this method is that it can provide a real world controller to be tested in a many scenarios configurations without requiring the presence of the real plant. In this paper we present the development of a emulation for wind turbines emulation using the national Instruments CRIO platform. There a many real-time hardware platforms to implement a prototype HIL. In this paper we consider the use of National instruments CompatRIO platform. CompactRIO is ideally suited due to its packaging, ruggedness, and flexibility. The CompactRIO embedded system includes a real-time processor that can execute control algorithms deterministically, perform data logging, serve up Web pages, Fig. 2. NI RIO 9068 Real time FPGA control etc. CompactRIO also has integrated in the chassis an FPGA, REAL TIME EMULATION OF WIND TURBINES which provides the flexibility and performance for high-speed In the renewable energy systems very often is required to signal acquisition and generation. The NI cRIO 9068 is face some design issues. The effective design of wind turbines relatively new platform hardware/software development 51 Technical support by relies on detailed knowledge of several distinct subject areas 1 TRUC 32 (3) which are briefly summarized below 2 T First the conceptual design problem is very since it is Another very important factor is the aerodynamic important to have a means of expressing the efficiency of a construction of the blades determines the torque generated by wind turbine. Fundamentally, this is ratio of the actual power turbine, however the mathematical models used here are production of the machine to the total energy available, and complex and require expert knowledge. In practice, the torque gives a dimensionless power coefficient, CP, whose basic developed by a rotor is typically calculated on the basis of formula is given by Eq. 1. empirical measurements. P C out (1) The characteristic shape of the CP-λ curve for a turbine with a p 1 UA3 fixed blade pitch (see Figure 3) indicates that maximum 2 efficiency is only achieved at a particular tip speed ratio. where ρ is the density of air (1.225 kg/m3) By adjusting the pitch angle, however, a turbine can be U is the wind velocity (m/s) made to maintain its rated power output at wind speeds higher A is the circular area swept by the turbine blades (m2) than rated Active pitch regulation requires that additional R mechanical features to allow each blade to rotate about its own (2) U longitudinal axis. This necessarily includes a shaft and bearings for each blade, but the selection of individual or In the formula for CP, the maximum possible value is 0.593, known as the Betz limit. Because the Betz limit is central actuator is a function of detailed turbine design derived from theoretical principles rather than with reference Finally the dynamic behavior of the turbine can be represented to any particular turbine design, it provides a fixed upper by a simple two mass model (Figure 4) bound for the efficiency of any turbine. The actual value of the power coefficient is affected by the geometry of the rotor blades and also varies with wind speed. The interaction of these quantities can be captured with reference to another proportional characteristic known as the tip speed ratio, λ. Effect of blade number on performance 5 C one blade 4.5 p C three blades p C two blades 4 p C four blades p C five blades 3.5 p Fig. 4. Two mass model of the wind turbine drive train 3 where T is mechanical torque p mech 2.5 C n is the gear ratio 2 Jrotor is the rotor inertia 1.5 Jgenerator is the generator inertia 1 ωgenerator is the rotational speed of the generator 0.5 Tgenerator is the generator torque 0 0 2.5 5 7.5 10 12.5 15 TJmech rotor dgenerator lambda (4) 2 JTgenerator generator Fig. 3. Power curves characteristic n n dt The power coefficient can be plotted against tip speed ratio the type of generator and the requirement for other electrical for a given turbine design to give a characteristic performance components depends on the detailed turbine design. curve which can be used in real time emulation of the turbine. Using the previous equations a complete model of wind Figure 3 shows this curve for turbines with different numbers turbines can be assembled. of blades. One may observe that fewer blades produce a broad, The complete wind turbine model is presented in Fig 7 flat curve where CP remains roughly constant over a wide range of λ, but the maximum value of CP is low. More blades give a higher maximum value for CP, but the curve has a narrow peak making the design sensitive to changes in λ. A design with three blades produces the highest maximum CP, but a two bladed design gives more consistent performance over a wider range of λ. An very important performance indicator is the torque coefficient, CT. the torque coefficient can also be calculated by dividing the power coefficient by the tip speed ratio for a given wind speed This can be useful in determining instantaneous torque Q using Eq. 3 52 Technical support by incorporates two dual core ARM processors and a FPGA. In our application the control task is performed by the FPGA and the communication and data transfer is performed by the RT OS based on Linux running on the ARM cortex processor. Using the mathematical equations of the turbine the model was implemented in the real time NI RIO 9068 controller at the FPGA logic level. We have considered that if we use the model based on FPGA logic we can take advantage from the computational power of the Zynq core. Since the FPGA has a naturally parallel data processing we can implement in this way a a micro wind farm emulator by using several emulation loops in parallel. In Figure 6 is presented main FPGA Main VI. The VI contains a control loop running in parallel with an RT watchdog loop. - The RT Watchdog Loop checks for a periodic Fig. 5. Proposed wind turbine emulator communication from the RT Main VI to ensure IMPLENTATION OF THE WIND TURBINE MODEL IN LABVIEW communications between RT and FPGA are maintained The NI 9068 employs an Xillinx Artyx 667 MHz dual-core ARM Cortex-A9 processor, with 1 GB nonvolatile storage, 512 MB DDR3 memory. The architecture is hybrid since it Fig. 6. FPGA code for emulation 53 Technical support by - The Control Loop performs control calculations on four the measured torque and speed from the AC drive. The torque input/emulation channels in parallel, and outputs the results to and speed can be measured from a dedicated torque or speed four analog output channels. transducers or it can be obtained from the estimated The images above show the front panel and block diagrams of torque/speed from the drive converter. The characteristic the example emulation algorithm. There are four analog curve of the wind turbine was implemented in VI using a inputs to the controller: reference wind speed and speed are parameterized model. Fig. 7. Wind turbine model with a parameterized CP characteristic Fig. 8. Power coefficient computation in LabVIEW using characteristic coefficients AC induction motor drive with vector field control is USER AND DATA PROCESSING INTERFACE simulated using Multisim software. In this way the emulation algorithm is validated. The plant simulation code is running on In figure 9 the RT VI communication with the FPGA is the PC and the measured signals are transferred to the real presented. The communication with the RT host is realized time controller. using TCP/IP communication. This way the system can be The Multisim inverter setup is presented in fig. 11.It consists operated through internet remotely. in a back to back inverter used for the control of AC induction motor. The motor reference speed is provided by the wind turbine emulator and the simulated speed and torque are used CODE VALIDATION as input for the wind turbine emulation. The evaluation results are presented in fig 12. The emulator is tested in PIL using a very simple setup. The real time controller is running the emulation software and the 54 Technical support by Fig. 9. Real time code used for communication with the FPGA target Fig. 10. Vector of the induction motor used for simulation The development of the complex wind turbine emulator CONCLUSIONS for multiple wind turbines power plant testing is presented. The turbine emulator can perform satisfactorily under The developed simulator was implemented by a high- steady state wind profile and turbulence. The system could performance FPGA system controller developed by LabVIEW provide all necessary parameters of the wind turbine system language. Wind speed can be easily programmed based on the such as wind speed, output torque, torque coefficient, output wind power spectrum, or from recorded wind speed data or power, power coefficient, and tip speed ratio. FPGA based from manual set-up. The advantages of the simulators are that simulation is complex but the NI LabVIEW graphical various wind profiles and wind turbines can be incorporated as desired in the control software and it includes the data programming tools can provide a very fast solution. The acquisition to verify the control algorithms and display the ability to test/debug/validate LabVIEW FPGA code on the parameters. desktop using co-simulation software is a very useful feature. This can provide a confirmation that all the controller works correctly by comparing fixed-point FPGA simulation results to the floating point simulation results (which can be from any simulation tool). 55 Technical support by 3-Phase Back-to-Back Inverter VDC_Link_A ConnectDCLinks VDC_Link_B - V(p-p): + V(rms): sConnectDC V: I(rms): I: Probe2 Probe5 S22 L1 I(p-p): I(rms): Probe3 1mH C2 Vgrid uH_A vH_A wH_A C4 uH_B vH_B wH_B XCP2 3300µF V1 VDC XCP1 Vu_A 3300µF Vu_B 1 mV/mA Iu_A S18 S10 24V 1 mV/mA Iu_B S7 s4_A S13 s6_A S15 S17 S9 S1 s4_B S2 s6_B S3 Vu_A s2_A s2_B Vu_B C1 XCP5 Vgrid 120Vrms C3 XCP3 Vv_A 3300µF GND Vv_B 1 mV/mA Iv_A 60Hz 3300µF 1 mV/mAIv_B Vgrid 0° Vv_A Vv_B sPC_A sPC_B XCP6 XCP4 Vw_A T2 T1 Vw_B 1 mV/mA Iw_A S21 1 mV/mA PC_A 4.5:1 4.5:1 Iw_B + PC_B U7 Vw_A + Vw_B - NTC1 - NTC2 10Ω Vgrid_v_A 10Ω uL_A vL_A wL_A Vgrid_v_B S19 S20 S11 S12 uL_B vL_B wL_B Vgrid_u_A Vgrid_u_B s1_A S8 s3_A S14 s5_A S16 s1_B S4 s3_B S5 s5_B S6 3-Leg Inverter A Pre-Charge Contactor Full Bridge Rectifier Full Bridge Rectifier Pre-Charge Contactor 3-Leg Inverter B XSC1 GND Ext Trig + _ s2_B A B + _ + _ V: Vu_B Vw_B s3_B I(rms): DIRECTION w Probe6 PWM_FULL_BRIDGE Vu_A V(rms): I(rms): V(rms): + PWM_HALF_BRIDGE Probe7 I(rms): s4_B CONNECT s6_B Vv_A Probe4 V: V: I: I: s1_B Vw_A Te Probe1 Probe8 R_BRAKE Key = Space s5_B θ 27Ω Key = C - θ λsd λsq 1V λrd BRUSHED_DC_MOTOR Key = M 0V λrq Vv_B Key = B DC_MOTOR_DUTY_CYCLE BRAKE_CHOPPER_DUTY_CYCLE GND INDUCTION_MOTOR PWM_3PHASE s2_A P1 s1_A 3 Precharge_Contactor_Control P1 s4_A P2 s3_A sPC_B P2 s6_A 10ms 10s P3 s5_A VCC P3 sPC_A 3.3V PWM_SINUSOIDAL_3PHASE Fig. 11. Induction machine Multisim simulation used for PIL testing of the NI cRIO 9068 real-time controller 5 Wind speed [m/s] 4 3 2 2501 Turbine200 torque [Nm] 15 0 100 50 0 01:01 14:55 Time Fig. 12. PIL testing of the real time simulator [2] https://decibel.ni.com/content/groups/power-electronics-development- REFERENCES center/blog/2012/02/25/new-250-page-platform-guide-ni-power- electronics-control-platform-introduction [3] http://sine.ni.com/nips/cds/view/p/lang/ro/nid/211620 [1] https://decibel.ni.com/content/groups/power-electronics-development- center/blog/2011/11/23/complete-reference-design-3-phase-inverter- [4] Diana Paula Petrila, Energy Conversion and Storage Control for Small rapid-control-prototyping-and-hardware-in-the-loop-simulation-using- Wind Turbine Systems, Ph.D. Thesis, Politehnica University of labview-fpga Timisoara, 2013 56 Technical support by Screw Surface Control of the Tool Made on Grinding Devices with Adjustable Step Micaciu Alexandru Cătălin Technical University of Cluj – Napoca, Faculty of Mechanical Engineering Muncii Blvd., No. 103 - 105, Cluj – Napoca, Cluj County, Romania [email protected] Vușcan Gheorghe Ioan Technical University of Cluj – Napoca, Faculty of Mechanical Engineering Muncii Blvd., No. 103 - 105, Cluj – Napoca, Cluj County, Romania [email protected] Abstract: This paper presents a device for grinding with the movement, moving the guiding shoe too (3), set at the possibility of continuous tuning step of the corrected surface on angle . tools with active spiral surfaces. Based on the mathematical w determination of the helical line on the active spiral surface tool there are offered the data for controlling the 3D machines of the precision tools cutting edges of this kind used on CNC tool- machines too. Key words: helicoidally surface I. INTRODUCTION Most machine parts used in the construction of Figure 1: End mill with rotary helical surface clearance machines are processed by cutting because until now it is the corrected according to current requirements. only way to ensure precise shape and size. The shoe, (16) integrated in the transversal slide, due Due to the friction between the cutting tool and the to the longitudinal displacement of the guiding shoe (3), will processing piece there is a gradual removal from the tool move the transversal slide (11) and the worm screw (17) also, active surfaces of a material quantity, thus causing wear of the being engaged with the worm screw wheel (18). Thus the cutting tool. worm screw (17) will rotate the worm screw wheel (18) and Sharpening is the grinding of the tool cutting face in with it and the tool also, resulting combining movements of order to obtain an appropriate geometrical shape and smooth displacement and rotation, named the helical motion. With the surfaces to provide optimum cutting conditions. Sharpening is help of the tableland (14) the division can be performed in done on sharpening machines or grinding machines with special devices. order to sharpen all the tool teeth. Through the hand wheel (4) it can be performed the cutting depth adjustment when the tool Sharpening tool must comply with the following is being sharpened. conditions: - the geometric elements of the cutting tool; The helical surface step which will be generated, H f is - the appropriate smoothness of the active surface and cutting achieved by adjusting the inclination angle, of the guiding edges. w u − is displacement speed of the point M, on the helicoidally shoe (3). The worm screw (17) has m 1 mm axial mode motion (figure 1). and the worm wheel with which it is in engagement, has the An actual exploitation of cutting tools requires for sharpening number of teeth of z 30 . Therefore, the diameter of the to comply the helical pitch area clearance. Obviously it is worm wheel driving is. required the best possible rectified surface (Fig. 1). Severe (1) conditions are imposed to keep the numerical value of helical dw m z 130 30 mm pitch area clearance, marked on the tool in a position with The relationship between helical surface step and the 77.72 mm. inclination angle of the guiding shoe (3) has the expression: In order to achieve these technical conditions it was designed H f a device for grinding helical surfaces, executed by S.C. ctg w (2) SCULĂRIA SRL Cugir. According the cinematic scheme d w [Fig.2 (b)], the tool (5) which is to be sharpened is mounted The swivel angle of the guiding shoe (3) [Fig.3 (b)] is between the edges (2). By the crank (1), the longitudinal shoe adjusted by the principle sinus ruler, taking into account the (15) of the mechanical system it is operated in a translation 57 Technical support by construction of the mechanical system [Fig.3 (c)], resulting in the following formula for the package gauge blocks Ew : Ew 55,3 65,6sin w (3) By the adjusting formula (9), the calculated gauge blocks package Ew , ensures the implementation step H f of the helical surface to be generated by the mechanical system shown. The mechanical system described can be adapted to the grinding machines and its construction allows grinding surfaces clearance of cutting tools with helical surfaces. The following describes a case study for the realization of sharpening tool from (Fig.1) with which is found the correct setting to the sharp device of cutting tools helical surfaces. The helical surface step of the cutting tool has the value: H f 77,72 mm. For the calculation of the control angle of the guiding shoe (3) it is used the formula (2) in the next mode: 77,72 0 w arcctg 50 2923,4 30 The guiding shoe (3) will lean on the principle of 0 sinus ruler to the angle w 50 2923,4 . For this purpose it will be used the control formula (3) as it follows: 0 Fig. 2. (c) Ew 55,36 65,6sin50 2923,4 105,9711678 mm. Figure 2 - Mechanical system for generating helical movement by: (a) overview; (b) cinematic scheme; (c) helical pitch adjustment on the principle of sinus ruler: 1-crank, 2-edges, 3- guiding shoe, 4-hand wheel with graduated cylinder, 5- sharpening subject tool, 6-stroke limiter, 7-tool mount, 8- package gauge blocks, 9- guiding shoe lock lever, 10-frame, 11-transversal slide, 12-graduated hand wheel drum, 13- worm locking screw, 14- division plate system, 15-transversal slide, 16-shoe, 17- worm screw, 18- worm screw wheel. The calculated value of the package gauge block and tool setting on the device will achieve the cutting tool Fig.2. (a) sharpening shown in Figure 3 and 3 b. Fig.2. (b) (a) 58 Technical support by (b) Fig.3. Making rectification surface clearance to an end mill with a adaptable device to a pan grinder The sharpening control was achieved in SC Star Transmission Cugir SRL through tool assimilation with slanted teeth gear. The measurement protocol Fig. 4 from control of the Klingelnberg measuring machine sharpening P26 Fig.5 and the angle of the tool propeller Fig.6 measured on the Surfcom conturograf Fig.7 have allowed the following conclusions: Fig.6. Measuring protocol for the inclination angle of the cutting tool propeller Fig.5. Klingelnberg P26 measuring machine Fig.6. Surfcom Conturgraf 59 Technical support by REFERENCES 1. Murgulescu, E., Flexi, S., Kreindler, O.,Sacter, O., Târnoveanu, M., (1962), Analitic and differential geometry, Didactical and Pedagogical House Publising, Bucharest; 2. Şteţiu, G., Oprean, C., Lǎzǎrescu, I.-D., Şteţiu, M. (1994), The theory and the practice of the cutting tools, Vol. 1, 2 and 3, House Publishing of Sibiu University, Sibiu; 3. Vuşcan, I., Vuşcan, M., (1998), Helicoidally couple with free cylindrical rolls, Letter patent OSIM, No 3182, Bucharest; 4. Vuşcan, I., (1998), Helicoidally couple with bearings, Letter patent OSIM, No 113084, Bucharest; Fig.7. III CONCLUSIONS The designed device generates the helical motion; from measurements results that the device ensures the required accuracy to the clearance surfaces correction of cutting tools. ACKNOWLEDGEMENT This paper is supported by the Sectorial Operational Programme Human Resources Development project POSDRU/159/1.5/S/137516 financed from the European Social Fund and by the Romanian Government. 60 Technical support by Power Quality Event Monitoring using Data Mining Techniques Marc Gheorghe, Muntean Maria, Risteiu Mircea, Ileana Ioan Exact Sciences and Engineering Department “1 Decembrie 1918” University of Alba Iulia Alba Iulia, Romania [email protected], [email protected], [email protected], [email protected] Abstract—This paper presents an attempt of using data disturbances in power systems. In order to reduce the mining techniques for analyzing data acquired in power quality computational cost of the proposed method, Binary Decision monitoring in an industrial environment. The huge data quantity Tree is created and a Support Vector Machine classifier is registered may contain behavioral patterns of monitored system trained for every node of the tree. which can be discovered by data mining models. Moreover an automated notification system can be developed and used in real In this paper we applied unsupervised and supervised time for the optimization of the monitored system. learning strategies of data mining in power quality data analysis in order to identify useful patterns within the dataset Keywords—power quality; working condition; data mining; for the four possible states of the system (events): the normal cluster analyzis. running, the notification regime, heavy duty running of the system, the idle-running of the system. Significant results I. INTRODUCTION obtained from Farthest First cluster analysis and Decision Trees classification technique are finally used to take Recent research in powering the compressed air production decisions about system repairs and management utility in plants focuses on identification and diagnose of power quality order to avoid unexpected system running interruption. disturbance problems, prediction of system abnormalities or failure, and alarming of critical system situations. In such systems, the unscheduled stopping of the machines lead up to EXPERIMENTAL RESULTS laborious operations to restart the process flow, and to In this paper we propose two data mining techniques for significant loss of production and storing a large number of outlier detection and prediction, using Farthest First clustering, rejects. a k-means clustering and Decision Trees classification The huge amount of power quality monitoring data cannot technique. To implement the experiments we chose the Weka be processed anymore through traditional techniques, because environment [3] as clustering tool. this process would be a time consuming one, without real time diagnoses. Data mining tools are an obvious candidate for A. Dataset description assisting in such analysis of large scale power quality A large amount of information, obtained by the data monitoring data. Essential in applying data mining tools to collection equipment, was recorded and accumulated in the power quality data is the ability to identify the various database used for the experiments. The dataset obtained is underlying classes associated with power quality disturbances presented in Table 1. of interest, and events understanding and definition. TABLE XIII. THE DATASET USED IN THE EXPERIMENTS RELATED WORK Dataset Number of instances Number of attributes In [1] the authors underline classes in harmonic data from Set1 1714 11 medium and low voltage power distribution systems using clustering techniques and link analysis tools to merge the The original dataset was preprocessed, so, the quality of obtained clusters into super-groups. Finally, the characteristics the data was improved and the data mining tasks became of these super-groups are discovered using various algorithms efficient. for classification techniques and a priori algorithm of association rules is used to find the correlation between the Before the experiments, the dataset was prepared by harmonic currents and voltages at different sites (substation, performing the following steps: residential, commercial and industrial) for the interconnected • The missing values were deleted or replaced; super-groups. • The data was transformed in forms appropriate for the The paper [2] proposes the use of Support Vector mining process; Machines in order to identify and classify voltage and current 61 Technical support by • Some additional attributes were included; reassigning objects to clusters to improve the partitioning is • The data was exported to a required .arff or .csv format referred to as iterative relocation. Eventually, no redistribution file in order to be built as input file for the clustering of the objects in any cluster occurs, and so the process component. terminates. The k-means algorithm returns the best results when the clusters are well separated [5], [6], [7]. The original and modified formats of dataset are shown in Figure 1 and Figure 2. Fig. 12. Original Data Fig. 14. Variation of average current in time Fig. 13. Preprocessed data Data visualization is also a very useful technique because it helps to determine the difficulty of the learning problem. There were visualized pairs of attributes (2-d) using Weka data mining software [3]. Figures 3 - 7 show the variation of each Fig. 15. Variation of average voltage in time parameter in time (legend: cluster 0: idle_running, cluster 1: heavy_duty_running, cluster 2: notification_regime, cluster 3: normal_running). In the performed experiments, a large volume of data was processed and samples of data corresponding to the four situations that may arise were chosen. B. Cluster Analysis The k-means exploratory data analysis technique provided two categories of data which helped us in labeling data. According to the k-means algorithm for partitioning [4], where each cluster’s center is represented by the mean value of the objects in the cluster, we arbitrarily chose four objects as the four initial cluster centers. Each object was distributed to a cluster based on the cluster center to which it is the nearest. Next, the cluster centers were updated. That was, the mean value of each cluster was recalculated based on the Fig. 16. Variation of average active power in time current objects in the cluster. Using the new cluster centers, the objects were redistributed to the clusters based on which cluster center is the nearest. The process of iteratively 62 Technical support by Notification regime (cluster 2) is due to the influences of external factors that can be attrition of the equipment systems, and improper operation of electric power networks and require intervention in determining and eliminating the cause of its appearance. If no action is taken to surgery and remove the cause of increasing current values, the heavy duty running will be reached (cluster 1). So, it will be registered a malfunction of the system, the system will enter in another class of operation (emergency) which can cause its stop running by activating protections. Although the idle-running (cluster 0) is a normal operation, the running efficiency of the system is low, and the system should be limited. This regime results from negative values for reactive power (-3450.0) and low current values (139.36, 116.05, 129.68). Fig. 17. Variation of average reactive power in time Figure 9 presents the distribution of instances in the four clusters found by FarthestFirst algorithm. Cluster 0 idle_running Cluster 1 heavy_duty_running Cluster 2 notification_regime Cluster 3 normal_running Fig. 20. The dataset cluster distribution Fig. 18. Variation of average power factor in time C. Decision Trees classification We used the Weka FarthestFirst implementation of the above clustering technique and we discovered that the four After performing several experiments, we concluded that categories of objects discovered corresponded to the four J48 decision tree is the most suitable classifier for our dataset. possible situations from our system: the normal running, the The J48 classification accuracy was equal to 95.79%, and notification regime, heavy duty running of the system, the idle- the confusion matrix had the following values: running of the system. The resulting clusters returned by the clustering process are presented in Figure 8. === Confusion Matrix === a b c d <-- classified as 578 0 6 11 | a = idle_running 0 51 6 0 | b = heavy_duty_running 2 4 754 15 | c = notification_regime 7 0 21 259 | d = normal_running The Confusion Matrix shows that the classifier recognized the normal_running class for 259 instances, assigning the notification_regime class only for 21 instances and idle_running class for 7 instances. It can be noticed that no normal_running was assigned to heavy_duty_running class. The J48 discovered tree is presented in Figure 10 and Figure 11. Fig. 19. The discovered clusters 63 Technical support by Fig. 22. J48 right discovered tree CONCLUSIONS In this paper we addressed the anomaly detection of power supply of compressed air production plants with data mining techniques. Two methods were approached: the first one was based on a k-means clustering technique and accurately discovered the anomaly values clusters in our dataset, and the other was a classification technique based on Decision Trees. The last one also classified with a high accuracy the anomaly values form the dataset and could be used in order to predict the class of a new registered instance in order to prevent the unscheduled stopping of the machines. Monitoring power quality parameters will be recorder in real time, and after their classification, some decisions will be taken about repairs, planning and management utility to avoid unexpected system running interruption. Fig. 21. J48 left discovered tree REFERENCES [1] Asheibi, A., Stirling, D. and Robinson, D., Identification of Load Power Quality Characteristics using Data Mining, Canadian Conference on Electrical and Computer Engineering, Ottawa, Canada, 7-10 May 2006, 157-162. [2] Milchevski, A., Kostadinov, D., Taskovski, D., Classification of Power Quality Disturbances using Wavelets and Support Vector Machine, Elektronika IR Elektrotechnika, ISSN 1392-1215, Vol. 19, No. 2, 2013. [3] http://www.cs.waikato.ac.nz/ml/weka/ [4] Han, J., and Kamber, M., Data Mining: Concepts and Techniques, Second Edition, Morgan Kaufmann Press, Elsevier Inc, San Francisco, 2006. [5] Nazeer, K., A., A., Kumar, S., D., M., Sebastian, M., P., Enhancing the k-means clustering algorithm by using a O(n logn) heuristic method for finding better initial centroids, 2011 Second International Conference on Emerging Applications of Information Technology, Kolkata, West Bengal, India, February, 2011, pp. 261-264. [6] Esteves, R., M., Pais, R., Rong, C., K-means clustering in the cloud - a Mahout test, 2011 Workshops of International Conference on Advanced Information Networking and Applications, March, 2011, Biopolis, Singapore, pp. 514-519. [7] Yuepeng, S., Min, L., Cheng, W., A Modified k-means Algorithm for Clustering Problem with Balancing Constraints, 2011 Third International Conference on Measuring Technology and Mechatronics Automation, January, 2011, Shanghai, China, pp. 127-130. 64 Technical support by Integrated Channel Add-Drop Filter Based on Photonic Crystals with 2D Periodicity Mițǎ Oana, Şchiopu Paul Bostan Cazimir-Gabriel Optoelectronics Research Center (CCO), Optoelectronics Research Center (CCO), Politehnica University of Bucharest, Bd. IuliuManiu 1-3, Politehnica University of Bucharest, Bd. IuliuManiu 1-3, Bucharest, ROMANIA Bucharest, ROMANIA [email protected] Abstract—A photonic crystal with two-dimensional (2D) based on self-collimation effect in photonic crystals with 2D periodicity integrated channel add-drop filter is presented in this periodicity - paper. A combination of waveguide – photonic crystal ring resonator – waveguide based on photonic crystal with 2D Fig.1(b)) (if the modes are excited with frequencies that are periodicity is used in order to obtain an integrated, high outside the photonic bandgap, light can propagate within the performance channel add-drop filter. Through the Finite- photonic crystal with almost no diffraction, without having to Difference Time Domain simulation of light propagation, we introduce defects – this is self-collimation effect) [4]. demonstrate a good performance regarding the band and area of the filter that can be used in photonic integrated circuits. Keywords—Photonic crystal, Finite-Difference Time Domain, supercell, waveguide, ring resonator, filter I. INTRODUCTION Due to the continuous and aggressive scaling of CMOS technologies, electrical interconnect faces many difficult challenges - larger power consumption, an increase in the parasitic elements, pins count and routing complexity. In addition, the need for fast data propagation across chips is (a) becoming too big for electrical interconnect to satisfy the requirements regarding the speed and power [1], [2], [3], [4]. An alternative and a promising approach to the electrical interconnect for on-chip applications is an optical interconnect layer, which allows for the propagation of light through passive and active optical devices across a chip ( Fig.1 (a)) [1]. An integrated optical layer will enable a significant bandwidth increase, immunity to electromagnetic noise, a decrease in pins count and electrical power consumption, synchronous operation within the circuit and with other circuits and will reduce dependence on the thermal gradients. There are also some constraints regarding the developing of this layer: (i) all fabrication steps have to be (b) compatible with integrated circuits technology; (ii) the additional cost should be as small as possible [4]. Fig.1: (a) A proposed architecture using photonic and electronic systems on The compatibility between the optical interconnect layer same chip [1]; (b) SEM picture of Si membrane with hexagonal lattice and state-of-the-art integrated circuit requires a thorough photonic crystal with 2D periodicity [3], [4]. design of the former. In order to "siliconize" photonics [5] and Photonic crystals (Error! Reference source not to integrate on the same chip the photonic and the electronic found.(b)) are engineered periodic optical nanostructures that circuit, it is necessary to reconsider the optical routing because affect the propagation of photons in a similar way that the the conventional waveguide technology is a limiting factor of periodicity of a semiconductor crystal affects the motion of the integration of photonic integrated circuits (large waveguide electrons (they are often call optical semiconductors). Because bends are needed). One solution is the optical interconnect of the high refractive index contrast, light will be scattered very strongly throughout the structure and the scattered waves 65 Technical support by from each period can either add up or cancel out, depending The substrate we use to pattern photonic crystal with 2D on the wavelength of the light. periodicity devices is silicon (refractive index of silicon is For a well-chosen geometry and a unit cell with 3.45 at 1.55μm wavelength). Silicon is mostly known as sufficiently high refractive index contrast, the scattering from material intensively used for microelectronic integrated each cell can interfere destructively, so that all light inside the circuits. But it can play an important role in photonics too. crystal within a certain wavelength range is cancelled out, Although silicon is often considered an impractical material propagation is forbidden in this case and controlling the flow for photonic application because it is an inefficient light of light in a structure becomes possible [2]. There are two emitter (because of the indirect bandgap structure), several possible working regimes - photonic spectral bandgap (the optical functions can be accommodated on silicon as a wavelength range is called photonic bandgap (PBG) – and substrate: light can be split, distributed, amplified, switched allowed frequencies where exhibit a wide variety of and detected. Crystalline silicon of high purity is completely anomalous refractive effects (like superprism effect,self- transparent at the infrared wavelength (third generation fiber collimation effect, negative refraction). Due to their ability to optic system operates at 1.55μm), so being suitable for control light propagation in a small area, photonic crystals telecommunication applications [6]. have many application in photonic integrated circuits. One of the important subsystems of photonic integrated circuits is The photonic crystals with 2D periodicity were chosen due represented by filter because it can act as demultiplexer: it to their high versatility to vary the material and geometric selects a specific wavelength in a wavelength multiplexed parameters and because mature fabrication processes used for system (as it is Dense Wavelength Division Multiplexed CMOS technology can be adapted for this application. 2D communication system) [6]. periodicity can be interpreted in this way: it is assumed that In this paper, we designed the channel add-drop filter the third dimension is either zero case or infinite and uniform. using the properties of photonic crystal in photonic spectral This case is an ideal one, so real photonic crystal of finite bandgap regime. The filter is composed of two identical thickness with 2D periodicity should be modeled as a waveguides and a photonic crystal ring resonator. complete 3D structure. Because it is possible to separate the second order effects, the modeling and simulation in this paper CHANNEL ADD-DROP FILTER DESIGN are performed 2D. But for many photonic crystal based devices with 2D periodicity, the difference between 2D and By putting aring resonator next to two waveguides, we can 3D simulation is a small shift in frequency [6]. obtain a channel add-drop filter (Fig. 2). A signal traveling through input waveguide is coupled to a resonance of ring B. Waveguide resonator and transferred to the dropping (output) waveguide (within the resonance bandwidth of the ring resonator). We consider the same structure as in [6]: in a photonic crystal with 2D periodicity with rectangular lattice, dielectric (silicon) pillars in air, pillar’s diameter - 0.4∙a, a pillars line (row) has been removed. Introducing this linear defect (in this paper, along Γ–X) by removing a row of pillars (so deacreasing the amount of high index material), a single mode waveguide is obtained , waveguide which has the property that there is at most one guided mode at a given frequency (if multiple rows are removed, multi mode waguides are obtained). In order to determine the defect mode, a band calculation is realized using a (1x11) supercell (so, the new Brillouin zone is 11 timessmaller) [6]. Comparing the two diagrams presented in Fig. 3, we can Fig. 2: Schematic representation ofring resonator based add-drop filter [7]. observe the guided mode that appears in the projected gap So this type of structure can change the transmission and region [6]. dispersion of propagating modes of the waveguide: the optical input signal is transmitted to the dropping (output) waveguide C. Photonic crystal ring resonator for frequency in a narrow band centered on resonance frequency of the photonic crystal ring resonator and it is Generally speaking, a ring resonator is a very interesting transmitted through bus (input) waveguide for slightly and useful device that acts as a high quality factor filter for different frequencies. The main components (two waveguides wavelength filtering applications. This device is characterized and a photonic crystal ring resonator) are analyzed and the by many resonances, so it can filter out many different simulation results for the entire system (filter) are presented. wavelengths. In this paper, we implement a square ring in a rectangular lattice photonic crystal with 2D periodicity (Fig. A. Substrate 4(a)). The same photonic crystal as in waveguide design is used. The advantages of photonic crystal ring resonator over 66 Technical support by point/line defect photonic crystal resonator are the scalability SIMULATION RESULTS in size and the flexilibility in mode design due to their multi- mode behavior [8]. Filter schematic is presented in Fig.5. We implemented a silicon channel add-drop filter for 1.55μm. It can be observed the input waveguide, photonic crystal ring resonator and the output waveguide (Fig. 5). (a) (a) (b) Fig. 3: (a) Photonic band structure for a photonic crystal with 2D periodicity (b) with rectangular lattice, dielectric (silicon) pillars in air, pillar’s diameter is 0.4∙a, TE polarization; (b) Projected band diagram along Γ–X for a photonic crystal with 2D periodicity with rectangular lattice, dielectric (silicon) pillars Fig. 4: (a) Ring resonator formed by missing 20 pillars in a square lattice of in air, pillar’s diameter is 0.4∙a, a pillar line has been removed, TE radius 0.4∙a, silicon pillars in air; (a) photonic band structure for a (9, 1, 9) polarization [6]. supercell, TE polarization: the band structure is highly folded, a few flat bands appear in the spectral range of photonic bandgap indicating defect modes, so a resonances. In order to determine the defect states – the defect mode of a point defect in a photonic crystal with 2D periodicity, the Single mode photonic crystal waveguides are the case of supercell method is used (the supercell method is an extension interest, and then the main requirement in the design of the of Plane Wave Expansion Method). It is obvious that the ring resonator for efficiently coupling it to the waveguide is to domain must contain the unit cell that contains the defect. have its resonance frequency (Fig. 4(b)) at the single mode Also, the supercell has to contain some part of the lattice operation frequency range of the waveguides (Fig. 3(b)): this because the defect represents a defect with respect to the condition it is fulfilled. lattice (a bigger supercell has benefits for the computation accuracy, but it is time consuming) [6]. The calculated band structure is highly folded, a few flat bands appear in the spectral range of photonic bandgap indicating defect modes, so resonances. We choose a resonance having normalized frequency a/λ = 0.339289, frequency that can be found inside TE photonic bandgap and also it represents a propagating mode for photonic crystal waveguide. 67 Technical support by presented in Fig.6 (b). Through Finite Difference Time Domain simulation, it is confirmed that for the resonance frequency, the light reaches the drop port of the filter, while Fig. 5:Channel add-drop filter schematic representation. for a slightly shifted frequency the reaches the through port. To confirm the filtering functionality predicted by previous numerical investigation, two Finite Difference Time Domain CONCLUSION simulations are done. For the two cases, we consider as input In this paper, we have designed a channel add-drop filter based on photonic crystal structure with 2D periodicity patterned with a rectangular lattice having silicon circular pillar. We analyze the components of this kind of filter: ring resonator and waveguides. This design represents a compact solution for filtering/demultiplexing and routing optical signals with good transmission in a small area (10×10μm2, so less area than their “classical” counterparts). Therefore the photonic crystals provide a new alternative to fabricate very compact photonic components using the advances in the technologies available in the microelectronic industry. The transmittance and frequency selectivity values are good, though they can be improved through optimization (the optimization can be done including in photonic crystal ring (a) resonator “scatterers” and using multiple ring resonators). REFERENCES [1] O. Miţă, C. G. Bostan, P. Şchiopu, “On chip optical signal routing based on self collimation effect in two-dimensional photonic crystals”, International Semiconductor Conference (CAS), October 2008 [2] I. O’Connor, “Optical Solutions for System Level Interconnect”, Laboratory of Electronics, Optoelectronics and Microsystems, Ecole Centrale de Lyon, 2004 [3] C. G. Bostan, “Design and fabrication of quasi – 2D photonic crystal components based on silicon - on – insulator technology”, PhD thesis, University of Twente, Holland, 2005 [4] O. Miţă, C. G. Bostan, P. Şchiopu, “Structureless interconnects for photonic integrated circuits”, UPB SCIENTIFIC BULLETIN Series A (b) Applied Mathematics and Physics ISSN (print) : 1223-7027 / (online) 2286-3672, no.1/2014. Fig. 6:Electric field profile for two frequencies for a channel add-drop [5] www.intel.com filters: (a) resonance frequency; (b) a frequency slightly different. [6] O. Miţă, C. G. Bostan, P. Şchiopu,“Integrated Filter based on Photonic crystals with 2D periodicity”, International Conference “Advanced signals two Gaussian sources having the spectral width a, a – Topics in Optoelectronics, Microelectronics and Nanotechnologies”, lattice period. The first simulation is done for the resonance August 2012 normalized frequency: [7] S. Robinson and R. Nakkeeran, "Advances in Photonic Crystals", book edited by Vittorio M.N. Passaro, ISBN 978-953-51-0954-9, Published: a/λ = 0.3392949 (1) February 13, 2013 under CC BY 3.0 license. [8] J. Romero-Vivas, D. N. Chigrin, A. V. Lavrinenko, and C. M. If the working wavelength is 1.55μm, the lattice period is: Sotomayor Torres, "Resonant add-drop filter based on a photonic a = 0.3392949∙λ = 526nm (2) quasicrystal," Opt. Express 13, 826-835 (2005) The result is presented in Fig.6 (a); the second one is done for a slightly different frequency (a/λ= 0.33) and the result is 68 Technical support by Comparison of Different Modelling Methods and Measurements for Pulse Width Modulation–Signals Tulbure Adrian, Pocan Ioan, Cioflica Dumitru University “1 Decembrie 1918” of Alba Iulia, Department of Science and Engineering, Blvd. N. Iorga, No. 11-13, Alba Iulia, Romania [email protected], [email protected], [email protected]. Abstract— This contribution has as goal one procedure to asynchronous or synchronous device of the SAB80C167. The generate a standard PWM signal (Pulse Width Modulation). It serial ports of the PC can be set in the configuration file. explains the internal structure of hardware that implements software to generate PWM signal. To overview whole, they enumerate the types of called routines: initial, intermediate and final, used during the development of industrial applications. The proposed article ends with some relevant oscilloscope captures. Keywords—Modelling, Measurament, signal, PWM I. INTRODUCTION 1.1 Software presentation Fig. 1. Development flow of the EDE A classical Program Development Environment (DE) is used to develop applications for systems with only one target and In order to communicate with the PC the evaluation board host system. Thereby is possible to run a application directly EVA167 uses the program which contains the following files: from the Integrated DE. - MON16X.EXE (monitor user interface) - MON16X.OVL (overlay file) In an Embedded Development Environment (EDE) this is no - BOOT.167 (bootloader file) longer true. To run an application for a embedded system, a - MON.167 (loadable monitor) simulator or target hardware is required. The EDE can both - MON16X.CFG (configuration file) compile error free a module and offers a number of simulators - MON16X.SCR (start recorder file) and target hardware. - LD167.BAT (loader for microcontroller programs) - TERM.OUT (demo to the TERM command) The used Embedded Development Environment EDE can be - TEST.REC (demo recorder file) divided into three main parts: The monitor program MON16X uses a memory region, which EDIT, BUILD and DEBUG, like in the fig.1 In the Edit part will be downloaded into the board memory. MON16X have one can select the toolchain, create and maintain a project, available several command groups, like as: memory, register, edit some sources to be used in a project and specify the execute, monitor and other commands. development options. In the Build part a makefile (created by The attached batch file LD167.BAT is able to load files the Edit part) is used to invoke the needed toolchain directly. It recognizes automatically the following formats: components, resulting in an absolute object file. In the Debug Tasking-format (*.out), OMF166-format (*.66), and HEX- part one can use this absolute file to debug proposed project. format (*.H66). By starting the monitor, this searches first for the configration file „MON167.CFG“. This file contains The Evaluation Board offers two possibilities for connection parameters, which the monitor users to configure the to a PC: through serial ports COM1 or COM2 and the experimental board. /* TERM.C Program demonstrativ pentru MON167 The following file „TERM C“ shows a classical program, */ that wants to use the computer as a terminal. /* */ /************************************************* /************************************************* **************************/ **************************/ 69 Technical support by #include #define MAIN_C /* Parametrii biţilor */ #define BIT15 0x8000 #define BIT14 0x4000 ...... /* Definirea perioadelor PWM (în unităţi timer i.e. 400ns / bit) */ #define PROC_FREQ 20000000L /* Frecvenţă CPU (în Hz) */ #define PWM_FREQ 1000L /* Frecvenţă Timebase (în Hz) */ Fig. 2. Diagram of the development board #define GPT1_FREQ (PROC_FREQ / 8L) /* Frecvenţă de clock GPT1 */ ... Port P2 (fig.1. SAB80C167 links) is a 16-bit bidirectional I/O ... port. It is also bit-wise programmable for input or output via #define PWM_PERIOD (GPT1_FREQ / PWM_FREQ) /* direction bits. Port 2 outputs can be configured as push/ pull or Perioada PWM (unităţi GPT1 ) */ open drain drivers. The input threshold of Port 2 is selectable #define DUTY_CYCLE 90L /* % PWM Duty (TTL or special). Cycle */ Port P3 is a 15-bit (P3.14 is missing) bidirectional I/O port #define HIGH_PERIOD ((PWM_PERIOD * also bit-wise programmable. The input threshold of Port 3 is DUTY_CYCLE) / 100L) selectable (TTL or special). The Port 3 pins can serve for #define LOW_PERIOD (PWM_PERIOD - alternate functions like Timers, Transmisions Units, Ext. HIGH_PERIOD memory signals and system clock. Port P4 is an 8-bit bidirectional I/O port. It is bit-wise programmable. In case of an external bus configuration, Port 4 can be used to output the segment address lines. II. HARDWARE DESCRIPTIONHHARDWARE Port P5 is a 16-bit input-only port with Schmitt-Trigger PRESENTATION Properties. The pins of Port 5 also serve as the (up to 16) analog input channels for the A/D converter, where P5.x equals ANx (Analog input channel x), or they serve as timer inputs: The used controller is a derivative of the C16x Family of full Port P6 is an 8-bit bidirectional I/O port. Port 6 outputs can be featured single-chip CMOS microcontrollers /3/, /6/. It configured as push/ pull or open drain drivers. The Port 6 pins combines high CPU performance with peripheral functionality also serve for alternate functions like: chip select, master hold, and enhanced IO-capabilities, like in fig.2. bus request. A lot of features contributes to his performance: short Port P7 is an 8-bit bidirectional I/O port. It is bit-wisePort 7 execution time per instruction acceptance of many data types outputs can be configured as push/ pull or open drain drivers. like bit, byte and word, internal on-chip integrated RAM- The input threshold of Port 7 is selectable (TTL or special). Memory, interrupt vectors with more Priority Levels, The following Port 7 pins also serve for alternate functions Peripheral Event Controller and Intelligent Peripheral like: pwm- and CapCom-Output. Components. Port P8 is an 8-bit bidirectional I/O port. The input threshold Port P0 (fig.1. SAB80C167 right) consists of the two 8-bit of Port 8 is selectable (TTL or special). Many Port 8 pins also bidirectional I/O ports. Both are bit-wise programmable via serve for alternate functions like CapCom Outputs. direction bits as input or output. For a input configuration, the The following data describes the capabilities of C167: output driver is put into high-impedance state. -2 kB On-Chip Internal RAM (IRAM) and 2 kB On-Chip Port P1 consists also of the two 8-bit bidirectional I/O ports. Extension RAM (XRAM) This port is used as the 16-bit address bus (A) in -100 ns Instruction Cycle Time at 20 MHz CPU Clock demultiplexed bus modes and also after switching from a -500 ns Multiplication (16 ´ 16 bit), 1 ms Division (32 / 16 bit) demultiplexed bus mode to a multiplexed bus mode. -up to 16 MBytes Linear Address Space for Code and Data -1024 Bytes On-Chip Special Function Register Area -8-Channel Interrupt-Driven Data Transfer Facilities via Peripheral Event Controller 70 Technical support by -16-Priority-Level Interrupt System with 56 Sources, Sample- Rate down to 50 ns -16-Channel 10-bit A/D Converter with 9.7 ms Conversion (Counter 8-bit 10-bit 12-bit 14-bit Time -Up to 111 General Purpose I/O Lines, with Selectable Input resolution) PWM PWM PWM PWM Threshold / Hysteresis. f (50 ns) 78.13 19.53 4.88 1.22 CPU Mode 0 KHz KH KHz KHz III. THE PWM-GENERATION fCPU (50ns) 39.1 9.77 2.44 610 Mode 1 KHz KHz KHz Hz The on board integrated Pulse Width Modulation Unit consists of PWM channels, which allows the generation of up to 4 The Port 7 posses 8 Bit. By use of the Register DP7 (Direction independent patterns (Fig.3). Each PWM-channel possess the Port 7) the main data direction will be established (In- or following elements: a period register PPx, up/down counter Output). PTx, pulse width register PWx, 16-bit shadow latch, two The Mode of the Output (push/pull or open drain) will be comparators, and the corresponding control logic /3/, /5/, /6/. adjusted in ODP7 Register (Open Drain Control Register Port 7). For center aligned signals the frequency range is 4.8 Hz up to 5 MHz. The minimum values depend on the width and the resolution (CLK) of the PWM timers. The maximum values assume that the PWM output signal changes with every cycle of the respective timer. The on chip PWM unit can operate in four differentmodes /3/, /4/: I) PWM standard mode 0 (EdgeAligned PWM); II) PWM symmetric mode 1 III) Burst mode and IV) Single shot mode. The operation modes are set depending on the states of the registers PTx , PWx and PPx [x =1..4]. Generally the output signal is set on “logic 1” when the counting value is greater or equal than register from PWx. IV. MEASUREMENTS AND CONCLUSIONS Fig 3. Design of the Pulse-Width-Modulation Unit TABEL 1 CONNECTIONS OF PORT 7 P P7.0 POUT0, PWM model ch 0 out O P7.1 POUT1, PWM model ch 1 out R P7.2 POUT2, PWM model ch 2 out T P7.3 POUT3, PWM model ch3 out 7 P7.4 CC28IO, In /compare out ch 28 P7.5 CC29IO, In /compare out ch 29 P7.6 CC30IO, In /compare out ch 30 P7.7 CC31IO, In /compare out ch 31 Figure 4. The development board. TABEL 2 FREQUENCIES OF PWM SIGNALS 71 Technical support by 1-serial connector 2 –external memory 3 -jumper, 4 - I/O connector, 5 - microcontroller chip The kit EVA167 contain the software tools and the hardware device for real-time evaluation of the C167 microcontroller. The board is delivered with two static RAMs which con be extended until 2 Mbytes. This functionality can be tested by starting the monitor program. With the help of open drain circuit on Port7 one can achieve the combination of more PWM outputs through different pull up configuration/connections. This provides the short circuit for any PWM channel. Fig. 5 and 6 shows the measurement results by scope at connector 4 from fig.4 for different sets of the control registers. The measured signals are used to control a step-down chopper for experimental tests in the field of electrical drives. This control task needs quickly signals and high resolution signals. Fig. 6. PWM-signal from microcontroller. Measurement by 1 kHz and duty The achieved results demonstrated the flexibility of the cycle 50%. programmable logic compared with the wiring logic. By setting some program-variables the parameters of the The article presents and illustrates the required steps for generated signals can be adjusted without modifying the practical implementation of a program for generating a PWM hardware part. waveform on embedded structure. The experiments confirm the functionality and flexibility of developed applications. Dedicated routines were implemented on a board with a famous controller, which dominate the market in recent years. REFERENCES [1] A.Tulbure, M.Erdmann 16 Bit Embedded System to Control the High Power Supply Units. SIMPRO 05. Petrosani 2005 [2] A..Tulbure, A..Egri, R..Joldes Cages Switching at an induction machine using the 16-bit microcontroller. MicroCAD, University of Miskolc 2007. [3] Ertec GmbH Evaluation Board EVA167, Erlangen 1993 [4] Mattheis/Storadt MC-Tools 17, Otmar Feger, Hard+SoftwareVerlag, Traunstein 1999 Fig. 5. PWM-signal from microcontroller. Measurement by 2 kHz and duty [5] Siemens C167 Derivatives. 16 Bit Microcon-trollers. cycle 90%. User Manual 1996 72 Technical support by A Matrix Model of an Infinite Series of Two-Port Elementary Components Olteanu Emil “1 Decembrie 1918” University of Alba Iulia Alba Iulia, Romania [email protected] Abstract — The long lines are very widely used in in cascade. The first diagram is a series circuit and the second communications equipment, IT and most modern electronic diagram is a parallel circuit. devices. For their study a common method is the matrix, applied The equations system from the first diagram series particularly to the two-port elementary components study. The circuit: transmission line model, and are based on Maxwell's Equations. It is current the transmission line model figured of a transmission line as an infinite series of two-port elementary components, each U1 U2 ZI2 representing an infinitesimally short segment of the transmission (1) line. We simulate this aspects by using an Matlab model. I1 I2 Keywords— electrical circuit, distributed parameters, where: transmission line, transmission line, simulate. Z ω Z Z Z jωl (2) I. WHAT IS OUR INTENTION? r l r In this paper we propose to analyse some aspect The system (1) may be written in matrix format: concerning matrix model of an infinite series of two-port elementary components. U2 U1 A11 A12 U1 1 Z U1 An electric line is described by four distributed A parameters: resistance r; inductance l; capacitance c; I2 I1 A21 A22 I1 0 1 I1 conductance g. (3) The model consists of an infinite series of the elements shown in the figure, and that the values of the components are The name of A is caracteristical matrix of seriel specified per unit length so the Figure 1, of the component can circuits. be misleading. The parameters r, l, c, and g may also be The equations system from the second diagram functions of frequency. These quantities can also be known as parallel circuit: the primary line constants to distinguish from the secondary line constants derived from them, these being the propagation U1 U2 constant, attenuation constant and phase constant. (4) I1 Y U2 I2 The next figure symbolized a four terminal model electrical Γ circuit format. where: The elements: U1, U2, I1, I2, Z and Y are complex C space vectors. Yω Yg Yc Yg jωc (5) The system (4) may be written in matrix format: U U A A U 1 0 U 2 A 1 11 12 1 1 I2 I1 A21 A22 I1 Y 1 I1 (6) The name of A is caracteristical matrix of parallel circuits. The four terminal model electrical Γ circuit format There are two type of load parallel circuits: can be break down in two simplify diagram who are connected in empty (blocked impedance) 73 Technical support by The two-port elementary circuits, in empty is composed by one parallel circuit with value Y = 0. D D 1 A A 1 Z A In this case the caracteristical matrix of circuit is: D 11 12 12 D21 D22 A22 1 A21 1 ZY 1 - Y 1 0 1 0 (14) A (7) Y 1 0 1 the parallel-serie matrix in short circuit F F 1 A A Y A The two-port elementary circuits, in short circuit is F 11 12 21 composed by one serial circuit with value Y = . F21 F22 A11 1 A12 1 - Z In this case the caracteristical matrix of circuit is: (15) 1 0 1 0 The line propagation constance is: A (8) Y 1 1 k ZY r jωlg jωc (16) The two-port Γ circuits, is the addition by two elementary circuits (series and parallel): The line caracteristical impedance is: 1 0 1 Z 1 Z Z r jωl A (9) Z (17) Y 1 0 1 Y 1 ZY c Y g jωc Those two-port Γ circuits (9) have following six The caracteristical matrix for transmission line by x matrics: length is: the caracteristical matrix chkx Z shkx U A A U c U 2 11 12 1 1 1 A A 1 Z shkx chkx 11 12 I2 A21 A22 I1 I1 A (10) Zc A21 A22 Y 1 ZY (18) the inverse caracteristical matrix In line case loss of energy (r = g = 0), obtain: T11 T12 1 A22 A12 1 1 ZY - Z k jω lc (19) T T21 T22 A A21 A11 A 0 1 And (11) the admittance matrix l Z (20) c c Y Y 1 A A 1 1 ZY A Y 11 12 22 Y Y A 1 A Z 0 1 1 21 22 12 11 If notate v (21) lc (12) The spead of propagation is: the impedance matrix v Z Z 1 A A 1 1 A λ 2π (22) Z 11 12 11 ω Z21 Z22 A21 1 A22 Y 1 1 ZY (13) Where λ is practically wavelength. In this case obtain: the serie-parallel matrix 74 Technical support by 2π k j (23) λ 2 Ux 2 2 ω lc Ux 0 x The caracteristical matrix of line is: (28) 2Ix ω 2lc Ix 0 U A A U 2 2 11 12 1 x I2 A 21 A 22 I1 2π 2π If , the (19) equation in case of loss of energy become of as ch j x Zc sh j x in the sistem: λ λ U1 1 2π 2π I1 sh j x ch j x 2 Zc λ λ Ux 2 k Ux 0 x 2π 2π (29) cos x jZ sin x 2 c U Ix λ λ 1 k Ix 0 2 j 2π 2π I x sin x cos x 1 Zc λ λ (24) In normal line (not loss of energy) for k and Zc is used (16) and (17) equation. λ For x , the caracteristical matrix become of: 4 The diferential equations in tis form have solutions in sinus funtion. This function can be writen by Euler equation, in complex space: 0 jZ c A11 A12 A j (25) 0 A21 A22 Zc kx kx Ux Ue Ue 1 (30) λ Ix U ekx U ekx For x , the caracteristical matrix is: Z 2 c A11 A12 -1 0 A (26) The value of k is calculated by (17). A21 A22 0 -1 The values U+ and U- resulted by the limit condition. DISTRIBUTED PARAMETERS TWO-WIRE LINE – MATLAB SOLUTION Let considered the transmission lines definited by infinite array of infinitesimale quadrupole network. In this case the For a two-wire line, in hear, whence is loss of energy, with liniar matrix equation become of a diferential matrix equation: two parallel conducting wire, with diameter d = 1 mm, and the distance of ducts D = 5 mm, with propagation sped v = 200.000 km/s and with frecvences f = 100 Mhz: Ux r jωl Ix 2D (27) Z 276lg (31) x Ix g jωc Ux c d Through the combination of this the diferential equations results two diferential equations by level two in V and I: 75 Technical support by CONCLUSION In those two aplications we conclusion that the wavelength λ not depended by line parameters, but depend by wave sped (the material parameters) and the wave frecvence. The caracteristical impedance depend by line parameters (dimensions). REFERENCES [1] Hee-Ran Ahn, “Asymmetric Passive Components in Microwave Integrated Circuits”, ISBN: 978-0-470-03695-2, WILEY, August 2006. [2] K. F. Zhang and X. Z. Dai, “Structural Analysis of Large-Scale Power Systems”, School of Automation, Southeast University, Nanjing, Jiangsu 210096, China. [3] David Thorson, “A Review of ABCD Parametrers. [4] Martin Grobler, “Determining transmission line parameters from time- stamped data. [5] 1. Emil Olteanu, Remus Joldeş, Emilian Ceuca – Matrix Shape for Rotations Operators in Graphics Media – 2006 IEEE–TTTC International Conference on AQTR – Cluj–Napoca – ISBN 973 713 114 2 – 2006. [6] EMIL OLTEANU – Proiectarea sistemelor informatice pentru ingineri, Editura Aetenitas, Alba Iulia, ISBN 978-973-1890-85-2 – 2010. 76 Technical support by Instrumentation to Study D.C. and A.C. Circuits Samoila Brana-Liliana, Arad Letitia-Susana, Utu Ilie Department of CECEEPE University of Petrosani, [email protected], [email protected], [email protected] Abstract— This paper refers to some of the virtual instruments This equation indicates the two variables that would affect we achieved in order to study the d.c. and a.c. circuits. These are the amount of current in a circuit. The current in a circuit is used in teaching and learning activities where they turned out to be directly proportional to the electric voltage across its ends and very helpful for students’ better understanding of the electric circuit inversely proportional to the total resistance of the external behavior under different conditions. We made them simple, easy to circuit. use and intuitive, being an alternative, more flexible, method of practice in the laboratories. Resistors can be connected in series or in parallel. Series means simply connecting the "out" side of one resistor to the Keywords— virtual instrument, d.c. circuit, a.c. circuit "in" side of another in a circuit. Parallel is when both of their terminals are respectively connected to each terminal of the I. INTRODUCTION other resistor or resistors LabVIEW is the acronym for Laboratory Virtual The formula for calculating a total of n number of resistors Instrument Engineering Workbench. Together with specific wired in series is: hardware devices, modern complex systems for data n acquisition and processing can be developed. It is a graphical R R programming environment that makes not necessary to know a S k (3) programming language itself, using algorithms designed as a k1 flowchart (diagram) instead of text instructions. The inverse of the equivalent resistance of two or more resistors connected in parallel is the algebraic sum of the Programs developed in LabVIEW are called virtual inverses of the individual resistances. instruments (VI) as they are like actual instruments [4], [6]. n Data entry can be provided in different ways: user input 1 1 (4) via keyboard or mouse; extracted from data files located on R R memory devices; received via video camera, network card, P k1 k data acquisition system, etc. Using these relations, we made a virtual instrument that Output data can be displayed on the screen or may be simulates a d.c. circuit with resistors in series and in parallel, saved in data files so that these will then be accessed by the shown in Fig. 1. user or by other programs. II. VIRTUAL INSTRUMENT FOR STUDYING A D.C. CIRCUIT BEHAVIOR We overview some elementary theoretical considerations which have to be taken into account in achieving virtual instruments for simulating d.c. circuits [1]. The predominant equation in the study of electric circuits is the equation given by the Ohm’s law: (1) In words, the electric potential difference between two points on a circuit (U) is equivalent to the product of the [6] Simulated d.c. circuit current between those two points (I) and the total resistance of all electrical devices present between those two points (R). We considered ideal instruments (ammeter and voltmeter) and wires with negligible resistance. The Ohm's law equation can be rearranged and expressed as: The front panel of the VI is shown in Fig. 2. U Using the switch K (Vertical Toggle Switch), we can study I (2) the two cases: R 77 Technical support by I. if K is out: I = 0 and U = E (false in the Case structure), Figure 2a; II. if K is in: I = E / (Rp + r) şi U = E - I·r (true in the Case structure), Fig. 2b. The VI displays the value of the voltage across the group of resistors, the current value as well as the power source voltage and the equivalent resistance. III. VIRTUAL INSTRUMENT FOR STUDYING AN A.C. CIRCUIT Study of RLC Series Circuit The simulated circuit is shown in Fig. 3. The impedance, Z, of a component or a circuit is defined [1] as: ZRjX (5) where R is the resistance, j is the imaginary unit, and X is the reactance. Capacitors and inductors are both components which can store energy: capacitors store it in an electric field and a) inductors in a magnetic field. Ideal capacitors and inductors are assumed to have zero resistance and so we have a pure imaginary impedance (reactance). 1 XL; X (6) L CC XXLXC (7) The total impedance in a RLC series circuit is given by: 2 2 ZRXLXC (8) When speaking about the phase difference, it can be calculated as: X R arctgarccos (9) R Z b) [7] Virtual Instrument for studying a DC circuit behavior Front Panel a) switch out; b) switch in. The current RMS value is: U I (10) Z The voltage U is: 78 Technical support by UURULUC (11) All these values are given in the virtual instrument, whose front panel is presented in Fig. 4. [8] Series R L C simulated circuit Virtual voltmeters indicate the RMS voltages across the three ideal elements: resistor, inductor and capacitor and the [10] Series resonance ammeter measures the current. A gauge type control gives the phase difference between the current and the total voltage. [11] RLC parallel simulated circuit With Kirchhoff’s first theorem, [1] we can write: IIrILIC (12) 2 U2 UU [9] Front panel of RLC series circuit VI I2I2II 2 I2 R L C or 2 (13) R XL XC There is a minimum value of the series impedance, when the voltages across capacitor and inductor are equal and From (13) it results: opposite. 2 I 11 1 1 UL = UC (14) UR2 XX Z In an RLC series circuit in which the inductor has no L C internal resistance it is possible to have a large voltage across The amount 1/Z is called admittance and it is noted with the inductor, an equally large voltage across capacitor but, as Y. Its components are the conductance G and the susceptance the two are 180° degrees out of phase, their voltages cancel, B. giving a total series voltage that is equal to UR. YGjB At resonance, the voltages across the capacitor and the (15) pure inductance cancel out, so the series impedance takes its The front panel of the VI which may be used to study a minimum value: Zo = R. Thus, if we keep the voltage constant, RLC parallel circuit is presented in Figure 7. the current is a maximum at resonance. A parallel circuit containing a resistance, R, an inductance, Such a resonance situation is represented in Fig. 5. L, and a capacitance, C, produces a parallel resonance (also called anti-resonance) when the resultant current through the Study of RLC Parallel Circuit parallel combination is in phase with the supply voltage. At resonance there will be a large circulating current between the The simulated circuit is shown in Fig. 6. inductor and the capacitor due to the energy of the oscillations. So, parallel circuits produce current resonance. Such a situation is shown in Fig. 8. 79 Technical support by [2] M. Pop, “Masurari electrice”, Universitas Publishing House Petrosani, 2010 [3] I. Utu, L. Samoila, Masurarea marimilor electrice, Universitas Publishing House Petrosani, 2010 [4] N. Patrascoiu, “Achizitie de date. Instrumentaţie virtuala”, Didactic and Pedagogic Publishing House, Bucharest, 2004 [5] I. Utu, L. Samoila, “Senzori si instrumentatie pentru sisteme electromecanice”, Universitas Publishing House, Petrosani, 2011 [6] * * * www.ni.com/labview/ [12] Front panel of a RLC parallel circuit simulation VI [13] Parallel resonance IV. CONCLUSIONS We presented three of the virtual instruments we achieved to study the direct current and the alternating current circuits. They proved to be very helpful during the teaching activities. They are simple, easy to use, flexible, allowing an intuitive laboratory practice. We made more VI-s in order to study Ohm’s law, calculation of equivalent resistance of resistors connected in series or parallel, resistance measuring methods such as: Wheatstone bridge, substitution method, comparison method, voltmeter and ammeter method [2], [3], study of power distribution in RLC a.c. circuits etc. All virtual instruments can be used as sub-VI-s in more complex applications. REFERENCES [1] L. S. Arad, “Electrotehnica”, Didactic and Pedagogic Publishing House, Bucharest, 2004 80 Technical support by Robots in Our Life. Support or Damage? Coroiu Adriana Mihaela University Babes Bolyai of Cluj Napoca, Department Computer Science, [email protected] Summary: The paper presents a few major domains, So what happens when robots begin manufacturing other important in human life, in which technologies from robots? While it may sound like a scene straight out of Artificial Intelligence are involved. Some examples of iRobot, I assure you, it is not. using AI in medicine, transport and business are Industrial manufacturing robots work on assembly displayed and also how these aspects make our life easier lines constructing other items, including vehicles and and why, we as humans, have to pay attention and to be electronics that are complex and have many interworking open to new results of research from this field. parts. If a robot can put together a television circuit board or Robots will continue to change us as they become both a Volkswagen, then why not another industrial robot? more common and more sophisticated. The real question Some see this as signs that manufacturing is moving away is: How much will we change? from actual manufacturing and into replication. However, these replicating ideas are currently only existent in the Keywords: artificial intelligence, intelligence applications, scientific testing phases. self-driving cars, monitoring systems, robots For an industrial robot to be able to fully replicate itself, it would have to be able to gather all the raw materials necessary to make itself autonomously. These industrial I. INTRODUCTION robotic systems, no matter how artificially intelligent or complex, do not have these abilities. The idea of this paper was to show how robots and Of course, it is possible that robot replication will artificial intelligence application are intrusive in our life. become an industry in the future as the need for more and Our ability to feel empathy toward an inanimate object may more industrial robot services become prevalent in society. lead to another use for robots: therapeutic companionship. A well-designed robot that can react to our emotional states B. Utility could serve as a companion during difficult times. In a Utilities are what keep our lives running smoothly at medical context, robots could gather data about a patient to home and at work every day – whether it is in the form of help doctors monitor the patient's health. gas, electric or water. Robots are starting to help to keep Robots may also become useful tools to help those utilities going strong, and at the same time, ensuring children develop social skills. As we increase our that there are no issues that could cause problems using understanding about autism and related conditions, we can these utilities on a daily basis. design tools to help people integrate into society and Everyone knows that power lines snap during wind understand social cues. Robots may teach us more about storms or when objects fall against them. However, power ourselves than we would learn otherwise. lines suffer from wear and tear on a regularly. So, there are Then there are the industrial and military uses for utility robotic systems that can run along electrical lines to robots. We're already relying on robots to work assembly detect this wear and tear, which makes it easier for electric lines. Military branches and police squads use robots for companies to replace and repair lines before they snap and bomb detection and other hazardous duties. We'll likely see cause outages for customers, or before they become a safety even more widespread use of robots in these industries in hazard for those on the road [6]. the future [1,3]. The gas companies also have robots that are designed to Robots will continue to change us as they become both move through their utility lines and ensure that those lines more common and more sophisticated. The real question is: are safe. Gas utility robotics are able to move through and How much will we change? around gas lines and mains to check for any weak areas that may be close to rupturing. A ruptured gas main will not only II. DOMAINS IN WHICH WE USE ROBOTS cut off natural gas to a large area, but it can also cause health issues for those in the immediate area. A. Manufacturing By using these types of robots, humans are able to continue to move throughout their home lives without Robots work in every facet of industrial manufacturing interruption, while also staying safe. This is similar to how today. These manufacturing robots are assembling industrial robots keep human workers safe every day. electronics, packaging food, cutting metal and welding cars. Industrial robots are able to keep workers out of hazardous work environments that may cause injuries or hardship. Like 81 Technical support by their utility robotic cousins, industrial robots take over these understand this page. Future computers may, in time, jobs that can be tough for humans, while also increasing achieve the key component of intelligence - human productivity and safety for the company overall. understanding. Understanding is a process of identifying, remembering and recalling the patterns in the interaction of C. Transportation objects [4]. In this age of automated welding, it is hard to find an Robots In Our Future - Humanoid Robots A Likely industry that doesn’t rely on welding cells to improve the Choice quality of their welds and the speed of their production. The The principal robots in the future are likely to resemble transportation industry is no exception. humans. History often mandates such choices. The cities Whether it is a car, a truck, a bus or a train, robotic and transportation systems of the world have been designed welding cells have probably touched them all. The to suit human movements. Trains, cars, airplanes, lifts, transportation industry turned to automated robot systems in staircases, doors, cupboards and rooms have been designed the 1980s to improve their production and keep up with the for humans [5]. increasing demand on their industry. While that may be true, the process of combination and As manufacturing changed, companies began to realize elimination are probably still the most effective ways to that they needed their production to not only work harder, perpetually improve welding cell applications. As cities but also smarter. That is where the welding cell came into grow and more transportation increases, the need for play. By combining and eliminating processes, vehicle vehicles, and the cells that weld them, will probably not go manufacturers were able to save even more money while anywhere for a long while. further increasing their production and keeping up with the competition. Even now, manufacturers are looking for ways to III. RESULTS improve their robotic welding cells. There is one school of This section contains some result of some dates collected thought that believes if welding cell equipment is related to uses of the robots over the world: standardized and vendors are streamlined, welding cells would be even more efficient and effective in their tasks [2]. TABLE I The vision of seamless human-robot interaction in our everyday life that allows for tight cooperation between Industry Value human and robot has not become reality yet. However, the Automotive industry 63200 recent increase in technology maturity finally made it Electrical/electronics industry 32700 possible to realize systems of high integration, advanced sensorial capabilities and enhanced power to cross this metal and machinery industry 13700 barrier and merge living spaces of humans and robot rubber and plastics industry 11400 workspaces to at least a certain extent. Together with the Food and beverage industry 4900 increasing industrial effort to realize first commercial service robotics products this makes it necessary to properly address one of the most fundamental questions of Human- Robot Interaction: How to ensure safety in human-robot TABLE II coexistence? The approach taken in this book focuses on the Year Sales biomechanical level of injury assessment, addresses the 2004 97000 physical evaluation of robot-human impacts, and isolates the major factors that cause human injuries. This assessment is 2005 120000 the basis for the design and exploration of various measures 2006 112000 to improve safety in human-robot interaction [6]. They 2007 114000 range from control schemes for collision detection, reflex reaction, and avoidance to the investigation of novel joint 2008 113000 designs that equip robots with fundamentally new 2009 60000 capabilities. By the depth of its analysis and exceptionally 2010 120000 salient experimental work, this monograph offers one of the most comprehensive treatments of the safety challenge in 2011 165000 the field. 2012 159000 So far, we have the next aspects: 2013 162000 Robots In Our Future - Understanding Is The Key The robots in our future will, one day, be able to read and 82 Technical support by And the diagrams corresponded of these tables are: Fig. 2: New installations of industrial robots worldwide in 2013 (by industry) (Table I) Fig. 3: Worldwide sales of industrial robots from 2004 to 2013 TABLE III TABLE IV Percent of all Perc Industrial Robots by Sector Robots Used Jobs ent Automotive 33.20% 35.4 Handling of Materials and Processes 0% Unspecified 25% 28.9 Electrical and Electronics 9.90% Welding and Soldering 0% Chemical, Rubber, and Assembling and disassembling 13% Plastics 9.40% 8.60 Other 9.20% Unspecified % Machinery 4.30% 7.90 Other % Metal Products 3.70% 3.80 Communications 2.50% Dispensing and Painting % Food 1.50% 2.50 Cutting, milling, and other processing % Precision Optics 0.80% The result diagrams for these tables are: Fig. 4: Percent of all robots used in industrial Robots (by sector) Fig. 5: Percent of robot used in different jobs 83 Technical support by IV. CONCLUSIONS The purpose of this paper was to present few domains in 84 Technical support by A New Method to Estimate Reegineering Efficiency in Distributed Network Measurements Vintea Adela, Șchiopu Paul Ghita Octavian Mihai University Politehnica of Bucharest University Politehnica of Bucharest Faculty of Electronics, Telecommunications and Faculty of Electrical Engineering Information Technology Bucharest, Romania Bucharest, Romania [email protected] [email protected], [email protected] Abstract— The generalized use of new energy sources require also a redesign of today networks via reengineering methods. In order to establish how efficient this methods are, distributed measurement techniques and a new generation of sensors are required to be implemented. To achieve this goal, supplementary costs, resources and efforts will be required. This paper present a new method to estimate efficiency of reengineering electrical networks taking account of several parameters. Keywords— distributed measurements, mathematical model, Fig. 1. Model for an Independent Power Grid electrical networks. C3I-IPG is a system implemented through several classes I. INTRODUCTION of tangible resources, with systemic organization and advanced sensors for retrieval and transmission of functional The management of electrical networks in Independent data. Organizational and structural relationships ensure the Power Grids (IPG) for establishing new decisional solutions integrity of driven systems and the driving system (C3I) is means Command-Control-Communications (C3I) techniques capable to influence effectively the driven system to ensure to be implemented. For proper functionality, C3I need fulfilment of activities. distributed measurement data and a mathematical model. Based on this, the system could perform certain functions Given the fact that connections between these two systems within a process and take into account influencing parameters, will be determined by the actual circumstances or due to the such as: associated services, distributed sensors and human need to use shared information, not only by a dispatcher, but resources. also by individuals or legal entities, perhaps even by the local government, it can appreciate that research exceeds the field The C3I– IPG logistic reengineering is needed to establish, of independent power grids. define and determine the qualities (basics characteristics) of such a structure . It is difficult to estimate the efficiency of this reengineering process without a mathematical model. Otherwise, time, funds and efforts can be spent in vain, without clear outputs or benefits. The main difference between today classic networks and future power grids consist in the higher amount of equipment that need to be used. For example, a wind generator that transfer power to the network should be accustomed by smart sensors for wind speed estimation, more powerful and flexible inverters, robust power transformers and an optical fiber network to allow data traffic. Beside this, human decisions must be taken, knowing also additional information about customers, climate conditions and several other factors.[1] It is very difficult to let an automatic hierarchical chain to control such networks, because there are many parameters that cannot be estimated correctly. In such conditions, the total Fig. 2. C3I – IPG system overview effort to reengineer these networks become a burden without having a tool to estimate the efficiency of the process (see Figure 1). 85 Technical support by II. MODELS USED FOR REEINGINEERING 17. knowledge maintenance to maintain adequate Reengineering is the simultaneity of all processes of knowledge base about mode of operations; renewal and consists of rethinking followed by radical 18. relevancy to identify and exploit the most useful redesign of business processes, aiming to radical improvement information; in performance, regarding costs, technology and product quality, service and speed of operation. E. Timeliness to provide sufficient early warning and execution time This process implies re-design of sensors or direct replacement, as an example: due to large quantity of energy 19. early warning to sufficiently forewarn actors; manipulated, electromagnetic fields are present and can 20. execution time to provide sufficient time for actors to influence current and voltage sensors. In order to decrease the make the best decision; influence of those fields, a new generation of sensors based on optical fibers for current and voltage should be installed. For 21. reliability in providing high-quality, complete better understanding the complexity of reengineering, a C3I information. model similar to these used in centralized systems should be taken into account. III. MATHEMATICAL MODEL C3I – IPG system reengineering is the establishment, This paper propose a mathematical model to calculate the definition and design of a new integrated system, which – cost modifications of reengineering process through all steps broadly – it is aimed to optimizing the levels of the following involved into: five fundamental features, by maximizing the performance of Design costs of the system; services associated with each [3, 5], as follows: A. Dispersion to support decentralised operations Management & IFD costs; 1. decentralisation to maintain C3I during fluid and Maintenance costs; crisis; Human resources costs; 2. flexibility to support actions; Economical costs; 3. independence to operate in stand-alone mode; Decision and communication costs. B. Invulnerability against active and passive attack The mathematical model for tracking efficiency of C3I – 4. indeterminacy to make information vague to attackers; IPG system is as follows [3, 4]: 5. information security (INFOSEC) to kept adversaries CCCCCCCTOTTOTTOTTOTTOTTOTTOT CIIPGI3design C 3management menten C 3 I humaneconomicdecision & from exploiting friendly information; &IFD resourcefinancial communication flexibility securization (1) 6. survivability against loss or degradation of information; and C. Mobility to support the tempo of crisis CTOT – total expenses 7. modularity to effectively interconnect C3I parts; where 8. redundancy to ensure access to information; C – expenses for each of the functionalities: design & C3I- 3 i 9. self-reparability to correct C I failures; flexibility; management & IFD; maintenance C3I-security; 10. good technical design for supportability and human resource; economic-financial and decision & interoperability; communications. 11. homogeneity among means to acquire and transfer Each term from relation (1) can be described as: information; CCCCTOT TOTTOTTOT designCI2 technicalforecast risc D. Responsiveness to needs of actors flexibility scientificprognosiscatastrophe research chaos (2) 12. adaptability to contingencies and unforeseen needs; TOTTOT TOTTOT CmanagementCcontrol&communicationsCreengineeringCmarketing 13. data transformation to manipulate raw data into &IFDMaster&SlaveSystem&IFDpromotion advertisement information for decision-making; (3) TOTTOTTOTTOTTOTTOT C2CCCCC 14. connectivity for prompt communications among users mentenanceCImentenancesensorsoptimalTotalINFOSEC &securizationMaster&SlaveequipmentsregimeforQuality SystemrepairementfunctioningManagement 15. decision support to aid actors in formulation and (4) testing viable courses of action; TOTTOTTOTTOTTOT ChumanCtrainingCergonomyClevelofCprocedures 16. direction/monitoring to help actors issue orders and resourceimprovement empatypaymentstandardisation monitor implementation; (5) 86 Technical support by TOTTOTTOTTOT theory. Calculations and determinations begin with estimation CeconomicCaccountingCfinancialClevelof financialevidencefraud of the percentage of the five fundamental features of interest (6) in C3I-IPG [3]. By allocating certain analysed percentage (in TOTTOTTOTTOT this case the system Responsiveness and Opportunity are each Cdecision&CearlyCdecisionCsynthesize40%), one can choose which of the system features are communicationwarningsupportconsulting review (7) relevant in the business of obtaining and storing the IFD for use. For each term in Relation (1), there are other relations (2) – (7), that represents the mathematical expression. It is In turn, the five fundamental features are differentiated by preferred to have a initial estimation for each term to start the the percentages of the services associated within each. Thus, it numerical determination of (1). All these requirements should estimates the service percentages so that their sum, of all, to be be validated in order to understand how logistics monitoring exactly 100% (e.g, in Dispersion, “decentralization” can be of the C3I – IPG systems contributes to ensure the success of rated to 40%, the same as the “independence” of the leading any reengineering activity. system, also 40%, leaving 20% for “flexibility”). Relation (1) can be completed using ratio coefficients that At the end of estimates, it must allocate the appropriate balance one or another step in the process. This can be percentages to Human and/or Technique influences within performed if one is establishing monitoring priorities for this each service, so that the sum of these two influences to be system. 100% within each service. In order to establish a clear priority for the order to start and run the entire operation/activity of The following paragraphs explain the associated services, reengineering the integrated C3I-IPG system, it will use 3 in terms of what a C I– IPG should and should not do for the measurements determined by laborious calculations. decision-maker. The associated services are listed in alphabetical order in Table 1 [2]. 3 TABLE I. C I–IPG CHARACTERISTICS & SERVICES Basics Characteristics Associated Services a Decentralization 1. DISPERSION b Flexibility c Independence a Indeterminacy SE 2. INVULNERABILITY b Info security c Survivability a Modularity b Redundancy 3. MOBILITY c Self-reparability d Good tech design e Homogeneity a Adaptability b Data transformation c Connectivity d Decision support 4. RESPONSIVENESS Fig. 3. Model of influencing parameters over reengineering efficiency e Direction/monitoring Knowledge f maintenance IV. CONCLUSIONS g Relevancy The need to introduce reengineering is represented by the a Early warning simultaneity of all command and control processes, if aiming 5. OPPORTUNITY b Execution time to a radical improvement in performance, regarding costs, c Reliability technology and product quality, service and speed of operation. The analysis of reengineering described later is based on Cost estimation remain a problem and covering these costs 3 consumer (C I– IPG) behaviour and utility theory. Because the is not an easy option. But without a mathematical model that 3 concepts of consumer (C I– IPG) behaviour rest on some can estimate at least the level of costs, reengineering remain arguable assumptions, it is important to discuss the just a challenge, not an implementation. assumptions here as they relate to top level structure decisions 3 (Informatisated Center for Independent Power Grids The C I – IPG system monitored, can give an idea about Management, Figure 2). the needed resources, using control, planning, directing, coordinating technical and human resources (equipment, These assumptions probably fit the defence decision- communications, facilities and procedures) in achieving one, maker better than they fit the classic consumer in economic two or simultaneously three of the following goals: 87 Technical support by Optimization of energy consumption; Response capacity in different critical events over the network; Better monitoring the quality of energy distribution; Readiness of networks to absorb future energy sources. Acknowledgment The work has been funded by the Sectoral Operational Programme Human Resources Development 2007-2013 of the Ministry of European Funds through the Financial Agreement POSDRU/159/1.5/S/132397. REFERENCES [1] A. Vintea, P. Schiopu, O. M. Ghita, “Optimal Model for Efficiency of Reegineering the Independent Power Grids”, IWSSS 2014 “2nd International Workshop on Systems Safety & Security”, to be published, 23-25.10.2014, Bucharest, Romania. [2] A. Vintea, P. Schiopu, “Reengineering for optimized control of DC networks”, The 7th edition of the International Conference "Advanced Topics in Optoelectronics, Microelectronics and Nanotechnologies", 21 - 24 August 2014, Constanta, Romania. [3] Bjorklund, R., C., “The dollars and sense of Command and Control”, Institute for National Strategic Studies Press, Washington, DC, 2005. [4] A. Carabulea, “Tratat de inginerie industrial”, U.P.B., 2006. [5] Vintea, D., “The projection of Communications System targeting the preservation of the Statistic and Informational Flux of Decision”, International and National Conference on Sustainable Industrial Systems, Bucharest, 2005. [6] Mihoc G., Muja, A., Diatcu, E., “Bazele matematice ale teoriei fiabilităţii”, Editura Dacia Publisher, Cluj-Napoca, 1988. [7] Boisrobert, C., Coutaz, J., L., “Materiaux Optiques, Photonique et Systemes”, Universite Paul Verlaine – Metz, France, 2004. [8] IEEE Std 1100-1999 "IEEE Recommended Practice for power and Grounding Electronic Equipment", IEEE Press, 1999. [9] Agustoni, A., Borioli, E., Ferrari, P., Mariscotti, A., Picco, E., Pinceti, P., Simioli, G., “LV DC Networks for distributed energy resources”, Proc. of CIGRE, Athens, 2005. [10] Hammer, M., Champy, J.: “Reingineria întreprinderii”, Editura Tehnică Publisher, Bucharest, 1997. 88 Technical support by Cracking the SIP Authentication with a Hybrid Password Cracking Method Specified for SIP Applications İsmail Melih Taş1, Hüseyin Taş2, Hüseyin Bahad r Öz3 1 Netaş Telekomünikasyon A.Ş, İstanbul, Türkiye [email protected] 2İstanbul Gelişim Üniversitesi M.Y.O., İstanbul, Türkiye [email protected] 3Bahçeşehir Üniversitesi, İstanbul, Türkiye [email protected] Abstract— Today internet telephony, multimedia conference, (SDP). For the transmission of media streams (voice, video) distance learning, instant messaging and many more similar SIP typically employs the Real-time Transport Protocol (RTP) application are done over Internet Protocol (IP). On the basis of or Secure Real-time Transport Protocol (SRTP). For secure the provision of these services, Voice over IP (VoIP) protocol is dealing. Although VoIP protocols provide similar mechanisms, transmissions of SIP messages, the protocol may be encrypted Session Initiation Protocol (SIP) is the most popular one because with Transport Layer Security (TLS). of its flexible structure. SIP employs design elements similar to the HTTP In this study, the anatomy of password cracking attack which request/response transaction model. Each transaction consists we developed to crack SIP authentication is mentioned and a of a client request that invokes a particular method or function method automate the password cracking attack specified SIP applications is introduced. on the server and at least one response. SIP reuses most of the header fields, encoding rules and status codes of HTTP, Keywords—VoIP, SIP, SIP Authentication, SIP Registration, providing a readable text-based format. SIP Security, Password Cracking If internet environment is in question, the weakest point of I. INTRODUCTION these kinds of protocols is security. One of the main issues for Today internet telephony, multimedia conference, distance SIP security is the authentication process between user agent learning, instant messaging and many more similar application and server. SIP security mechanism uses different protocols are done over Internet Protocol (IP). On the basis of the for different OSI layers. provision of these services, Voice over IP (VoIP) protocol is dealing. Although VoIP protocols provide similar For transport layer it uses TLS (Transport Layer Security), for mechanisms, Session Initiation Protocol (SIP) is the most network layer it uses IPsec (IP security), for application layer popular one because of its flexible structure. it uses HTTP authentication. In HTTP authentication, instead of sending user name and password in clear text format, it uses Originally designed by Henning Schulzrinne and Mark MD5 (Message Digest algorithm 5) encryption algorithm to Handley in 1996, SIP has been developed and standardized in encrypt the data. On the destination side, the same algorithm RFC 3261 under the auspices of the Internet Engineering Task should be used to decrypt the data. Force (IETF). It is an application layer protocol designed to be Whenever a client tries to contact with a server, in order to independent of the underlying transport layer; it can run on verify the client, server sends “401 Unauthorized” response to Transmission Control Protocol (TCP), User Datagram client; whenever a client tries to contact with a proxy server, Protocol (UDP) or Stream Control Transmission Protocol proxy server sends “407 Proxy Authentication Required” (SCTP). It is a text-based protocol, incorporating many response to client. elements of the Hypertext Transfer Protocol (HTTP) and the Simple Mail Transfer Protocol (SMTP). SIP works in Bypassing the SIP authentication, data transfers in SIP-based conjunction with several other application layer protocols that environments can be accessed and manipulated which may identify and carry the session media. Media identification and results eavesdropping, denial of service and many other negotiation is achieved with the Session Description Protocol malicious activities. In this study, the anatomy of password 89 Technical support by cracking attack which we developed to crack SIP Via: SIP/2.0/TLS client.biloxi.example.com:5061;branch=z9hG4bKnashds7 authentication is mentioned and a method automate the Max-Forwards: 70 password cracking attack specified SIP applications is From: User-Agent SIP/2.0 401 Unauthorized Via: SIP/2.0/TLS client.biloxi.example.com:5061;branch=z9hG4bKnashds7;received=192.0.2. 201 From: User-Agent Authorization: Digest username=" user-agent", REGISTER message details that User Agent sent to realm="atlanta.example.com" SIP server; nonce="ea9c8e88df84f1cec4341ae6cbe5a359", opaque="", uri="sips:ss2.biloxi.example.com", REGISTER sips:ss2.biloxi.example.com SIP/2.0 90 Technical support by response="dfe56131d1958046689d83306477ecc" network. Then it filters the SIP registration message flows to Content-Length: 0 the SIP server and takes out the hashed data along with some other useful data in order to use in cracking processes. As shown below, needed data in SIP REGISTER message is “200 OK” message details in order to verify user’s shown as bold, response value is the MD5 (Message Digest 5) hash values that needs to be cracked. password data that SIP server sent to User Agent; REGISTER sips:ss2.biloxi.example.com SIP/2.0 Via: SIP/2.0/TLS client.biloxi.example.com:5061;branch=z9hG4bKnashd92 SIP/2.0 200 OK Max-Forwards: 70 Via:SIP/2.0/TLS From: User-Agent testing usage. It sniffs the network packet traffic over 5060 port in promiscuous or monitor mode in the target network and filters the SIP OPTIONS messages and responses to that messages Kali: Kali Linux is a Debian-derived Linux distribution which allows detecting the SIP servers and clients on the designed for digital forensics and penetration testing. It is maintained and funded by Offensive Security Ltd. It was 91 Technical support by developed by Mati Aharoni and Devon Kearns of Offensive after Digium requested that its developers cease the use of the Security through the rewriting BackTrack, their previous word "Asterisk"; the renaming was further justified by the fact forensics Linux distribution. that the product at that time consisted of much more than just Asterisk. Kali Linux is pre-installed with numerous penetration-testing applications, including nmap (a port scanner), Wireshark (a Softphone: Zoiper, Gizmo, MicroSIP, xline, express talk packet analyzer), John the Ripper (a password cracker), and business, iaxComm, phoner, 3CXPhone, Blink, PhoneLite. Aircrack-ng (a software suite for penetration-testing wireless Softphone allows making VoIP calls through a PBX or a SIP LANs). Users may run Kali Linux from a hard disk, live CD, provider. or live USB. It is a supported platform of the Metasploit Project's Metasploit Framework, a tool for developing and Ngrep: Ngrep strives to provide most of GNU grep's common executing security exploits. features, applying them to the network layer. Ngrep is a pcap- aware tool that will allow you to specify extended regular or Eclipse IDE: In computer programming, Eclipse is an hexadecimal expressions to match against data payloads of integrated development environment (IDE). It contains a base packets. It currently recognizes IPv4/6, TCP, UDP, workspace and an extensible plug-in system for customizing ICMPv4/6, IGMP and Raw across Ethernet, PPP, SLIP, FDDI, the environment. Written mostly in Java, Eclipse can be used Token Ring and null interfaces, and understands BPF filter to develop applications. logic in the same fashion as more common packet sniffing tools, such as tcpdump and snoop. Wireshark: Wireshark is a free and open-source packet analyzer. It is used for network troubleshooting, analysis, software and communications protocol development, and V. HYBRID PASSWORD CRACKING METHOD AND ITS education. IMPLEMENTATION Wireshark allows the user to put network interface controllers In this part of the study, the working principle of the cracker that support promiscuous mode into that mode, in order to see tool we developed is introduced. In order to be more efficient all traffic visible on that interface, not just traffic addressed to and effective, a hybrid password cracking method specified one of the interface's configured addresses and for SIP application has been developed. After gaining the broadcast/multicast traffic. hashed values from SIP signaling, in an offline side, However, when capturing with a packet analyzer in concurrent password cracking methods are started to run and promiscuous mode on a port on a network switch, not all of automate the processes. the traffic travelling through the switch will necessarily be sent to the port on which the capture is being done, so In the first cracking method, the statics of common passwords capturing in promiscuous mode will not necessarily be in SIP applications was pre-defined and hash values of those sufficient to see all traffic on the network. passwords are listed in a file which we use to call while during cracking process. The hash values took out from the signaling Port mirroring or various network taps extend capture to any are compared to the hash values from that pre-defined list point on the network. Simple passive taps are extremely whether there is any match. resistant to tampering. This crypto-analysis method is commonly known as rainbow table method; in this study we customize this method specific SIPp: SIPp is a free Open Source test tool / traffic generator for SIP implementation environments considering our for the SIP protocol. It includes a few basic SipStone user experiences. In terms of performance and speed, rainbow table agent scenarios (UAC and UAS) and establishes and releases method is one of the most successful methods. multiple calls with the INVITE and BYE methods. The other method which will work parallel is a type of brute- force cracking method. It generates a password according to Asterisk: Asterisk is an open source framework for building pre-defined criteria specific for SIP implementation communications applications. Asterisk turns an ordinary environments and calculates the hash values of those computer into a communications server. Asterisk powers IP passwords and compares with the hash values took out from PBX systems, VoIP gateways, conference servers and other the signaling whether there is any match. custom solutions. It is used by small businesses, large businesses, call centers, carriers and government agencies, The pre-defined criteria are results of our experiences also worldwide. Asterisk is free and open source. considering the default passwords policies of well-known vendors’ products. So, in terms of performance and speed, it is Trixbox: Trixbox was initially released under the name designed to work more efficient than a classic brute-force Asterisk@Home. In October 2006 it was renamed to Trixbox cracking methods. 92 Technical support by successfully and gave the results to use cracked password for Both cracking methods are working parallel, if any of them malicious activities. cracks the password then both the processes stop and provides their statistical results in a log file and they automatically All processes are tested in our VoIP security laboratory initiate new cracking processes if they catch a new hash value environment. from signaling till anyone stops them manually. These both cracking methods are successfully tested in our VI. CONCLUSION laboratory environment. During testing phases, before the cracking process, while This study is used as a piece of a commercial project titled as system is working normal working condition, the users “VoIP Vulnerability Tool” which is developed at Netas. The registered to Trixbox SIP-PBX server are shown in Figure 2 cracker tool can be used to audit the registration process along with its IP addresses. security to evaluate the authentication mechanism and security/password policies in SIP-based VoIP systems. After the cracker codes are run successfully and cracked passwords are used for malicious activities, if we check the This study (1139B411401727) is supported by TUBITAK same screen on Trixbox SIP-PBX web interface, most under 2241-A program which is given for the selected probably we will see the same users with different IP graduation projects that provide solutions to the several addresses. problems of industries. As result, using the cracker code we developed, SIP registration mechanism can be cracked and registration The “VoIP Vulnerability Tool” project (3130514) is supported hijacking attacks are easily possible to perform. by "The Scientific and Technological Research Council of Turkey (TUBITAK)" under "Technology and Innovation Funding Programs Directorate (TEYDEB)" program targeting to increase research-technology development capability, innovation culture, and competitiveness of Turkish companies. REFERENCES [1] http://www.sscwa.com/voice-over-internet-protocol-voip/ [2] http://tools.ietf.org/html/rfc3261#page-8 [3] Chad Carter, Microsoft XNA Unleashed: Graphics and Game Programming for Xbox 360 and Windows, Sams (August 5, 2007) [4] http://tools.ietf.org/html/rfc3261#page-9 [5] http://tools.ietf.org/html/rfc3261 [6] http://tools.ietf.org/html/rfc3665 [7] Franks, et al. http Authentication: Basic and Digest Access Authentication. RFC 2617, June 1999 [8] Qi Qiu “Study of Digest Authentication for Session Initiation Protocol (SIP)” Fig 2. Registered users and IP addresses shown inTrixbox SIP-PBX web interfaceResults [9] http://tools.ietf.org/html/rfc3665#page-5 [10] http://eaglenet.edgren.pac.dodea.edu/student%20parents/Eaglenet/Extrac As the nature of SIP protocol, it is vulnerable to signaling urricular/JROTC/Cyber%20Patriot%20SY%202012%20%202013/8Pass wordSecurity.pdf manipulation attacks. [11] Goode, B., “Voice over Internet Protocol (VoIP)”, Proceedings of the Although authentication is activated, SIP registration IEEE, vol. 90 (6), pp. 1495-1517, (2002) mechanism can be cracked and this situation can be used for [12] Henning, S., ve Rosenberg, J., “The Session Initiation Protocol: Internet- various malicious activities by crackers. centric signaling” IEEE Com. Magazine, vol. 38(10), pp.134-141, (2000). [13] http://tools.ietf.org/html/rfc3261 In this study, we captured the SIP registration signaling, and [14] http://tools.ietf.org/html/rfc3665 took out the hash value of the password and in an offline side [15] Taş İ. M., Özbirecikli O. ,Çağal U. ,Taşk n E. ,Taş H. “SIP Kay t Silme our cracker code used that data to crack the password Sald r s n n Anatomisi ve Savunma Stratejileri”, 22. Sinyal İşleme ve Uygulama Kurultay , pp. 1-4, Trabzon, Turkey, (2014). 93 Technical support by [16] Zheng X., Jin J. “Research for the application and safety of MD5 [17] Özbirecikli O., VoIP, “SIP Sinyalleşmeye Yönelik Sald r Uygulamalar , algorithm in password authentication” 2012 9th Int. Con. on Fuzzy Zafiyet Analizleri ve Güvenlik Önlemleri”, Kocaeli Üniversitesi Lisans Systems and Knowledge Discovery (FSKD), pp. 2216 – 2219, Sichuan, Tezi, Mühendislik Fakültesi, Kocaeli 2013. (2012). [18] Endler D., Collier M., “Hacking Exposed VoIP:Voice Over IP Security Secrets & Solutions”, McGRAW-Hill/Osborne, (2007). 94 Technical support by Emerging Areas of Photonics and Optoelectronics A Survey on Organic Electronics Vlădescu Marian “Politehnica” University of Bucharest Optoelectronics Research Center Bucharest, Romania [email protected] Şchiopu Paul “Politehnica” University of Bucharest Optoelectronics Research Center Bucharest, Romania Abstract— The photonics and optoelectronics education at - Visible Light Communications (VLC); Faculty of Electronics, Telecommunications and Information Technology of “Politehnica” University of Bucharest started in - Organic Electronics / Optoelectronics. early '90s, trying to address the growing demand of engineers with a complex optoelectronics profile and to meet the increased A. Free Space Optical Communications requirements of microelectronics, optoelectronics, and The concept of transmitting information through the open nanotechnologies, in order to provide a high level of theoretical space by means of a modulated light signal is quite old. background combined with experimental platforms in Although significant advances have been made over the past laboratories and also with advanced simulation software tools. Based on our accumulated experience, we would like to highlight 10 years, the basic concept is simple: a narrow beam of light is nowadays emerging areas of this domain and to propose new launched at a emmission station, transmitted through the topics to be included in curricula for master programs at the atmosphere, and received at the receiving station. The universities aiming to bring highly qualified personnel to advances, which have led to FSO, have come in response to optoelectronics R&D and industry as well. the need for greater bandwidth and efficient communications systems. Both FSO and fiber-optic transmission systems use Keywords—education; photonics; optoelectronics; organic similar infrared (IR) spectrum and have similar transmission electronics; nanothechnology bandwidth capabilities. FSO is often referred to as “fiberless optics” or “optical wireless” transmission. I. INTRODUCTION Most FSO systems use simple on-off keying (OOK) as a Photonics is studying the generation, emission, modulation format, the same standard modulation technique transmission, modulation, signal processing, switching, that is used in digital fiberoptics systems, data being amplification, and detection/sensing of optical radiation. It transmitted in a digital format. This simple modulation type covers all technical applications of optical radiation over the allows FSO systems to be designed as bandwidth- and whole spectrum from ultraviolet (UV) over the visible to protocol-transparent physical layer connections. In addition infrared (IR), near-, mid-, and far-IR. the fact that the optical spectrum is unlicensed with Optoelectronics is considered a sub-field of photonics frequencies of the order of hundreds of terahertz is an which is studying the electronic devices that produce, detect advantage. and control the optical radiation, and their applications. In order to examine FSO performance, we must take into Optoelectronic devices are electrical to optical and optical to consideration several system parameters. In general, these electrical convertors. The equipment that uses such devices in parameters can be divided into two categories: internal their operation is also considered to belong to optoelectronics parameters and external parameters. area. Internal parameters are related to the design of a FSO system and include optical power, wavelength, transmission II. CURRENT AREAS OF INTEREST IN OPTOELECTRONICS bandwidth, divergence angle, and optical loss on the transmit side and receiver sensitivity, bit-error rate (BER), receive lens We identified the areas of interest in photonics and diameter, and receiver field of view (FOV) on the receiving optoelectronics to be nowadays the followings: side. - Free Space Optical communications (FSO); 95 Technical support by External parameters, or non-system-specific parameters, new manufacturing sites, but also new jobs. Some of the key are related to the environment in which the system is operating areas of the value chain are: materials, organic lighting, and include visibility and atmospheric attenuation, turbulence, sensors and manufacturing equipment. scintillation, deployment distance, window loss, and pointing European Union (EU) leading organizations have loss [1-3]. collaborated to form Commercialization of Organic and Large B. Visible Light Communications Area Electronics (COLAE), an European funded project under the Seventh Framework Programme (FP7) between 2007-2013 VLC is using the visible light portion of the [8]. electromagnetic spectrum to transmit information. This is in contrast to classic forms of wireless communication such as Organic and Large Area Electronics (OLAE) has received Wi-Fi which uses radio frequency (RF) signals to transmit support in earlier EU framework programmes and priorities data. related to this field are now included in various parts of the Horizon 2020 programmes. Within VLC, data is transmitted by modulating the intensity of the light in such a way that it is not perceptible to EU has is ahead of the rest of the world in many aspects of the human eye. Data is received by a photo-sensitive detector OLAE and probably has the most robust vertical integration of which demodulates the light signal into electronic form. efforts in OLAE. The players in OLAE field from EU are several big companies, universities and research institutes VLC is a category of Optical Wireless Communications from various European countries, corporate and academic (OWC). OWC includes IR and UV communications as well as spin-offs and also start-ups as well. All of them have visible light. However, VLC is unique in that the same visible contributed to the growth of OLAE which is widely light energy used for illumination may also be used for recognized to have a high commercial potential [9-11]. communication. OLAE based products have some important advantages: The operation principle of VLC is connected to LED light. they are generally thin, lightweight, energy-efficient, robust When a constant current is applied to an LED light bulb a and flexible. Current and future innovations in existing constant stream of photons are emitted from the bulb which consumer (handheld) or automotive products, mainly because we observe as visible light. If we vary the current up and down of form factor benefits and user interface possible slowly the output intensity of the light we see dims up and improvements, prove that OLAE technology has the potential down. Because LED bulbs are semi-conductor devices we can to provide new products, which are not achievable with actually vary the current, and hence the optical output, at traditional electronics. extremely high speeds which are imperceptible to the human eye but which can be detected by a photo-detector device. We may consider that OLAE is a new scientific and Using this technique high speed information can be technological field of Nanotechnology having many potential transmitted from an LED light bulb. applications. An overview of OLAE applications is presented in Fig. 1 [12]. Compared with RF, VLC has a number of benefits: capacity (bandwidth, data density, high speed, planning), There are four main groups of potential applications of effiviency (low cost), safety (non-hazardous) and security OLAE, each of them including specific applications, among (containment, control) [4-6]. which we can mention the followings: Display (display, lighting, signage) C. Organic Electronics / Optoelectronics Organic electronics (also called printed electronics or Power (solar cells, batteries) polymer electronics) refers to electronics based on carbon Sensor (sensors, actuators) chemistry, instead of conventional silicon [7]. Memory (transistors, smart tags) Organic electronics is using widely available, cheaper and less toxic materials than traditional electronics. It is based on a Most new technical and business opportunities are scalable manufacturing technology, environmental friendly, perceived to be in energy (organic photovoltaics - OPV), often using additive methods such as printing on flexible displays, signage and lighting (organic light emitting diodes - substrates. OLED) and sensors (organic thin film transistors - OTFT). Our survey is focusing on organic electronics, highlighting the applications connected to photonics and optoelectronics. III. ORGANIC ELECTRONICS Based on organic and oxide semiconducting materials instead of traditional silicon semiconductors, organic electronics is an emerging technology, well suited for applications that need flexible substrates like paper, plastic films or metal foils, having a business potential in many application domains and also the potential of creating not only 96 Technical support by The breakthrough of OLED displays, which happened in 2012 with small displays for smartphones, is followed now by large displays for TV. Fig. 1. Overview of potential applications of OLAE [12]. Fig. 2. Organic sollar cells [14]. Organic electronics is a rapidly evolving field with vast Organic-based optoelectronic devices, including solar cells number of applications having high potential for commercial (OSCs) shown in Fig. 2 and light-emitting diodes (OLEDs) success. Although a great progress has been made, many shown in Fig. 3, hold great promise as low-cost and large-area organic electronic applications: organic light-emitting diodes electro-optical devices and renewable energy sources. (OLEDs), organic field-effect transistors (OFETs), organic However, further improvement in efficiency remains a solar cells, etc; still require further optimization to fulfill the daunting challenge due to limited light extraction or requirements for successful commercialization. For many absorption in conventional device architectures. applications, available at this time organic materials do not provide satisfactory performance and stability, which hinders the possibility of a large-scale production. Therefore, the key ingredient needed for a successful improvement in performance and stability of organic electronic devices is in- depth knowledge of physical and chemical properties of molecular and polymeric materials. Since many applications encompass several thin film layers made of organics, and often also inorganic materials, the understanding of both organic-organic and hybrid interfaces is yet another important issue necessary for the successful development of organic electronics, the energetics of organic-organic and hybrid interfaces being still not completely understood [13]. {Research opportunities arise from the challenges that humanity faces today, including the growing energy demand, the depletion of fossil fuels, and the possible climate changes Fig. 3. Organic light emitting diodes [14]. Possible solutions are provided by materials science and, in particular, require intensive research efforts in organic and Organic light-emitting diode is one of the main applications in macromolecular materials [14]. which novel properties of organic materials are utilized. The early prototype device was build of inorganic electrodes IV. OPTOELECTRONIC APPLICATIONS OF ORGANIC ELECTRONICS (ITO and Mg/Ag alloy) and two subsequently deposited Organic based optoelectronic devices, such as organic light molecular layers. emitting diodes (OLED) and organic solar cells (OSC), are The first layer comprised of aromatic diamine, which served very promising regarding the low cost and the large as holetransporting layer (HTL), while the second included 8- deployment. Examples of high volume applications are OLED hydroxyquinoline-aluminium, which was the emissive layer. for both rigid and rollable displays, OLED for lighting and The discovery led to extensive research devoted to the organic photovoltaics (OPV). improvement of OLEDs operational conditions and performance. 97 Technical support by The simple structure of the early OLEDs has gradually Since OLAE technologies are becoming ready for evolved to the multi-layered devices. The incorporation of the incorporation into products of all types there is a need for HTL and electron-transporting layers (ETL) provides balancing technology transfer and research commercialization. of charges entering the emissive layer [13]. REFERENCES [1] S. Bloom & al., “Understanding the performance of free-space optics”, Optical Society of America JON 2330 June 2003 / Vol. 2, No. 6 / JOURNAL OF OPTICAL NETWORKING, 2003. [2] Y.B. Song, “Optical Communication Systems for Smart Dust”, Thesis, 2002. [3] X. Zhu and J. M. Kahn, “Free-Space Optical Communication Through Fig. 4. Schematic representation of small molecule-based OLED [13]. Atmospheric Turbulence Channels Atmospheric turbulence”, IEEE TRANSACTIONS ON COMMUNICATIONS, VOL. 50, NO. 8, Schematic representations of small molecule-based OLED AUGUST 2002. and polymer flexible PLED are presented in Fig. 5 and Fig. 6. [4] ***, “Visible Light Communication (VLC) - An introductory guide”, PureVLC, 2012. [5] ***, “Differences Between Radio & Visible Light Communications A technical guide”, PureVLC, 2012. [6] D. O’Brian et all. “Indoor Visible Light Communications: challenges and prospects”, Free-Space Laser Communications VIII, edited by Arun K. Majumdar, Christopher C. Davis, Proc. of SPIE Vol. 7091, 709106, (2008). [7] J. Helliwell, “Plastic and Organic Electronics – The Production Challenge”, PETEC, 2011. [8] M. Chachamidou, S. Logothetidis, “Commercialisation of Organic and Large Area Electronics – COLAE Project”, 2012. [9] Photonics21, “A European strategy for Organic and Large Area Fig. 5. Schematic representation of polymer flexible PLED [13]. Electronics (OLAE) - Vision paper”, http://www.photonics21.org/download/Brochures/AEuropeanStrategyon OrganicandLargeAreaElectronicsOLAE_2013-06-2622.pdf, June 2013. V. CONCLUSIONS [10] The FP7-ICT Coordination Action OPERA and the European OLAE applications offer substantial advantages and the Commission's DG INFSO Unit G5 "Photonics", “An Overview of possibility for low cost large-scale manufacturing processes in OLAE Innovation Clusters and Competence Centres”, http://cordis.europa.eu/fp7/ict/photonics/docs/reports/eu-report-olae- high volumes. clusters-sept2011_en.pdf, October 2011. The applications of OLAE are already penetrating every [11] ***, “Building a European strategy on Organic and Large Area commercial and industrial field, and it is expected to dominate Electronics (OLAE) – A concept paper, http://cordis.europa.eu/fp7/ict/photonics/docs/olae-strategy-paper- many aspects of life worldwide. horizon2020_en.pdf, January 2013. One can easily see that five out of the total of twelve [12] ***, “Organic Large Area Electronics (OLAE)”, Factsheet. OLAE applications presented in Fig. 1 belong to photonics [13] S. Braun, “Studies of Materials and Interfaces for Organic Electronics”, and optoelectronics domain, which means almost 42%. Thus, Duissertation thesis, 2007. we may conclude that an important part of organic electronics [14] H. Frauenrath, “Organic electronic materials”, (class 11). is dedicated to photonics and optoelectronics, and also that an emerging area in photonics and optoelectronics is based on organic electronics. 98 Technical support by Development of Testing Methods for Specific Mechanical Parameters of Electric Mining Cables and Accessories Vătavu Niculina, Părăian Mihaela, Jurca Adrian, Păun Florin Adrian, Lupu Leonard, Muntean Berzan Florina, Gabor Dan Department D.S.I.E.A., National Institute for Research and Development in Mine Safety and Protection to Explosion – INCD- INSEMEX Petrosani, Hunedoara / Romania, [email protected], [email protected], [email protected], [email protected], [email protected], [email protected] Abstract—Having in view risk mitigation and increasing conductors in protective tubes or conduits, and rods with safety, health and explosion protection in the firedamp or non sealed capsulation, explosion-proofed by a type of protection, firedamp underground mining industry, as well as in quarries, according to the intended group of explosion and the zone. when cable networks are employed it's imposing to use performant electric cables, certified for this purpose, having a The electric cables and their accessories shall be mounted robust structure specific for mechanical use stresses, also such as to prevent their exposure to mechanical damage, containing the elements required to fulfill electric and thermal corrosion or chemical influences (for example solvents: protection, and protection against flame propagation. ethanol, methanol a.s.o.) as well as heat effects and flame Using non performant material and an improper construction propagation.[2], [3] of electric cables used in Ex dangerous areas may lead to particularly dangerous short-circuits that can develop heat, If exposure to highly corrosive factors cannot be avoided, electric arcs and metallic droplets which may become an then protective measures should be adopted, such as mounting imminent hazard for fires and/or explosions. the cables into conduits or selecting adequate cables (for From the point of view of mine safety, the task of the electric example using armored, screened cables, with Aluminum cables used in electric mine cable networks is to preventing sheath or cables having mineral insulation or semi-rigid electrocution, fire and explosion hazards. This consideration sheath). determined the cable manufacturers, in collaboration with INSEMEX Petroşani to design and manufacture a special TECHNICAL SAFETY REQUIREMENTS FOR MINE category of industrial cables constructed for the mining industry, ELECTRIC CABLES AND THEIR ACCESSORIES [1], [2], resistant to mechanical stresses applied when posing the cables, at mounting or in normal operation (for example dragging the [4], [7], [8], [9] cables on the mine workings floor, rolling or unrolling the cable The safety requirements for mine electric cables address on the drum). the electric, mechanical and flame propagation performances. Keywords—firedamp atmosphere, explosion protection, electric Taking into consideration the mechanical stresses in the mining cables and accessories mining industry (dynamic and static crushing, simultaneous torsion and flexion, alternate torsion and flexion), the flexible I. INTRODUCTION mine electric cables with natural or synthetic rubber insulation, as well as the mine armored cables with If for the electric and non-electric equipment or components and protective components used in areas with polyethylene or PVC insulation and PVC sheath, intended for explosion hazards specific types of protection had been use in electric installations and equipment for powering, designed according to the dangerous zone the equipment is signaling, telecommunications and telemechanics, in the intended to be used and to the chosen category of equipment, underground and quarries mine workings have to assure a in the case of electric cables especially in large electric strength appropriate to these stresses. networks where particularly dangerous short-circuits may The technical safety requirements the mining cables have to occur, since short-circuits develop heat, electric arcs, and fulfill are imposed through the technical requirements in metallic droplets that may represent an imminent danger of STAS 10411: occurrence of unwanted events: fires or explosions and environment pollution. [2] As result of the dynamic crushing test, in order for the cable to pass the test, insulation faults, short-circuits or In all the areas with explosion hazard (groups I, II or III) phase breaks shall not occur, and the preventive disconnection due to the high vulnerability of the cables employed in electric coefficient shall be lower than the unit. networks, only certified cables should be used, insulated 99 Technical support by ' diminished compared to the standardized values, and to n1 k ' pressing forces also 20% diminished. n2 Presently, modern cable jointing is practiced with cold cast where: resins, in cast iron or plastic casings. These are specific to armored cables and, as consequence, are posed stationary in 5 the mine workings. n1i n' i1 Flexible cables connecting require the most light and as 1 5 flexible jointing possible. The 3M cable boxes are considered to be the most performant ones, as they are: flexible, light, ' n1 - represents the mean value of the number of strikes flame retardant, and cold cast in underground mining when an insulation fault occur, signalized by the relay for conditions. The 3M jointing technology is a modern one, it insulation resistance control; follows perfectly the cable structure, overpasses with only very little the diameter of the cable on which it's applied and 5 has very good electrical and mechanical performances. As result of the simultaneous torsion and flexion test, n2i n' i1 in order for the cable to pass the test, insulation faults, short- 2 5 circuits or phase breaks shall not occur, and the disconnection coefficient shall be in accordance with the product standards. ' n2 - represents the mean value of the number of strikes '' when a short-circuit between the phases or a phase break '' n1 k '' occur n2 As result of the static crushing test, in order for the cable to pass the test, insulation faults, short-circuits or phase where: breaks shall not occur, and the preventive disconnection coefficient shall be in accordance with the product standards. 3 n 1i ' i1 P n1 ' 1 3 k ' P2 n1 - represents the mean value of cycles when a when an insulation fault occur, signalized by the relay for insulation where: resistance control; 3 5 n2i P i1 1i n2 ' i1 3 P 1 5 n2 - represents the mean value of the cycles when a short- ' circuit between the phases or a phase break occur P1 - represents the mean value of the crushing force at which an insulation fault occur, signalized by the relay for When tested for the resistance to flame propagation, in insulation resistance control; accordance with SR EN 60332-1-1:2005 and SR EN 60332-1- 2:2005, the insulated conductor or the cable successfully 5 passes the resistance to flame propagation or increased P resistance to flame propagation if the distance between the 2i lower edge of the upper support and the limit of the ' i1 P2 carbonized zone is higher than 50 mm, and in the case when 5 the carbonized area spreads down on a length higher than 540 mm starting from the lower limit of the upper support, then the ' test is considered to be failed and as consequence the cables P2 - represents the mean value of the crushing force at are not allowed to be used in the underground of firedamp which a short-circuits or phase break occur. mines, non-firedamp mines or quarries.. The dynamic and static crushing tests apply also for the cable boxes connected on the most representative cables, and these have to withstand successive strikes from heights 20% 100 Technical support by TESTS OF MINE ELECTRIC CABLES AND THEIR - 0,75 m for the cables with phase conductors cross area up 2 ACCESSORIES HAVING IN VIEW CONFORMITY to 16 mm ; CERTIFICATION IN THE VOLUNTARY DOMAIN [2], [4], - 1,5 m for the cables with phase conductors cross area [5], [6], [7], [8] above 16 mm2. In order to assess conformity with the specific The static crushing test is performed with the apparatus requirements for use in the mining industry, the electric cables and according to the main circuit diagram shown in fig. 1, by have to fulfill the requirements in the standards SR 11388, SR means of a rig consisting in a hydraulic press with the force of EN 60811, SR EN 60332, and for certification in the voluntary 108 kN, a rail having the unit mass of 24kg/m and a half domain having in view cables admittance in mine workings trolley (see fig.3). both at surface and underground, the cables have to fulfill the requirements in STAS 10411. The live/powered electric cable is submitted to a crushing force of: According to SR 11388 and SR EN 60811, the electric cables intended for use in the mining industry shall be - 14,7 kN for the cables with phase conductors cross submitted to verifications and type tests addressing the area up to 16mm2 and rated voltage up to 1,2kV; following: - 29,4 kN for the cables with phase conductors cross - constructional characteristics and dimensional area over 16mm2 and rated voltage up to 1,2kV; verifications; - 49 kN for cables with rated voltage over 1,2kV. - electric tests and verifications (insulation resistance, electric resistance, cable voltage test, dielectric The simultaneous torsion and flexion test is performed strength test etc); with the apparatus and according to the main assembly diagram shown in fig.1. - mechanical tests (cable sheath tearing test, insulation adherence at flexion test, temperature For the needs of ensuring the rotary and translation shock test). motions for the cable samples submitted to this test the rig in fig. 4 had been designed. The rig ensures through its According to the standard series SR EN 603321-1, SR EN translation device, a number of 7...8 coming and going cycles 60332-1-2, SR EN 60332-1-3, SR EN 60332-2-1, SR EN with the speed of 20...30 m/min, and through its torsion device 60332-2-2, and according to the way of posing (individually at each cycle 3 rotations (6 radians) on one side and 3 or in a group of cables), the electric cables intended for use in rotations (6 radians) on the other side are generated. the mining industry shall undergo fire tests, to be able to prove flame propagation resistance or high flame propagation The distance between sample the locking devices, in their resistance. wide spaced position shall be 2,5 m, and the stroke length of the translation motion shall be 1,5 m. The test samples locked According to STAS 10411 the flexible mine electric cables in these locking devices is connected to the control apparatus with natural or synthetic rubber insulation, as well as the mine shown in fig. 1. armored cables with polyethylene or PVC insulation and PVC sheath, intended for use in electric installations and equipment The system monitors and acquisition the following for powering, signaling, telecommunications and parameters: telemechanics, in the underground and quarries mine workings - the insulation resistance in order to detect the insulation shall undergo verifications and type tests. It was found out that flaws in practice the mechanical stresses in the mining industry are - the currents on the three phases in order to detect short- mostly accurate reproduced by the following tests: dynamic circuits or splitting of conductors inside the cable crushing test; static crushing test; simultaneous torsion and - the number of strikes in case of dynamic crushing or the flexion test and alternate flexion test. number of twists in case of torsion test In order to perform these tests, in the ENExEMEIP - the pressure applied in case of static crushing test laboratory had been conceived and carried out test stands with The current on its three phases R, S, T is measured by software interface for measurement/ signaling/ control/ means of current transformers. This current in case of normal recording of the test cycles of static and dynamic stresses. operation without faults is identical for each of the three The apparatus and the circuit diagram are shown in fig.1. phases. The current increases when a short-circuit occurs and decreases when a phase is split. The dynamic crushing test is carried out by means of a rig provided with a ram weighing 20 ± 1kg, falling upon the test sample, which is placed on a 80 mm diameter Copper- Zinc cylinder from the height of: 101 Technical support by to the data monitoring and aquisition system 1 - cable test sample 6 - current transformers 2 - cable screen 7 - digital manometer 3 - insulation control relay k1 ... k5 - switches 4 - signaling lamps T1 - 380/127 V transformer 5 - 250 W / 220 V lamps for control of conductors continuity T2 - 127/380 V transformer Fig. 23. The circuit diagram from the current COMPUTER transformers from the monitoring device for DATA AQUISITION UNIT insulation resistance CHANNEL MULTIPLEXOR from the strike or twist counter from the manometer test stop when fault occurs Fig. 24. The measuring / signaling / control / recording system for the testing cycles in the dynamic and static crushing tests The insulation resistance is permanently monitored and in 190 case an insulation fault occurs, a signal is transmitted to the data acquisition unit, in order to stop the test (if case). An electric signal is taken over from a digital numerator 0 and then recorded together with the other data for setting out 8 the number of rotations or twists up to occurrence of fault. The pressure applied to the dynamic crushing test stand is 5 4 displayed and monitored by a digital manometer that transmits ~ the data to the data acquisition unit, through the multiplexor. 50 Fig. 25. Ram's profile 102 Technical support by outer sheath and the inner sheath or when it is separated as an individually insulated conductor disposed between the insulated phase conductors.. 3 5 5 7 9 7 1 1 1 CONCLUSIONS R R 20 R The electric cable network in underground are carried out with mine electric cables that, together with the assembly of electric equipment and device in the electromechanical installations, have the role of ensuring protection against 8 R electrocution, fire and/or explosion hazards. Having this in view, the cables shall be conformed with the 60 legislation in force and comply, together with the general 100 applicable safety requirements regulated through GD 1049/2006, with the standardized laboratory tests, applicable to the mine electric cables and their accessories, for the purpose of their certification in the voluntary domain. The safety specific technical requirements for mine cables are not regulated at international level, such as each country Fig. 26. The trolley wheel profile applies its own national regulations / standards. In Romania the requirements in STAS 10 411 apply. For the need of performing laboratory tests on electric 2 1 3 cables for certification in the voluntary field, a modern test stand had been carried out, for performing mechanical stresses tests on mine electric cables and their accessories (cable boxes) provided with a software interface for measurement / signaling / control / recording of testing cycles at dynamical and static stresses. REFERENCES 1 - cable test sample; 2 -device generating rotary motion; 3 - device generating translation motion [1] N. Vătavu ş.a.: Considerations regarding admittance criteria and risk factors when using electric cables in areas with explosion hazard, having in view environment protection, International symposium – Fig. 27. Installation for the simultaneous torsion and flexion test Occupational health and safety – SESAM 2013, Sibiu, 2-25 oct. 2013 [2] N. Vătavu: Studiul performanţelor de securitate şi dezvoltarea unei When applying the translation motion, the cable sample tehnologii inovative pentru testarea cablurilor electrice miniere şi shall be twisted in one way (until it forms loops) and at the accesoriile acestora – PSDCE, Contractul de cercetare nr. 45N/2007, act stroke back motion the sample shall be untwisted. At the next adiţional nr. 1/2014, PN 07 45 02 50/2014; going stroke the sample shall be twisted in the other way (until [3] HG nr. 1049 / 2006 privind cerinţele minime pentru asigurarea securităţii şi sănătăţii lucrătorilor din industria extractivă de suprafaţă it forms loops), and the cycle shall be repeated three times. sau subteran; In order to successfully pass the above mentioned tests, the [4] STAS 10411-89: Electrical cables for mining purposes. Test methods electric power cables as well as the signaling and control [5] SR 11388 – 2000: Common test methods for electric cables and cables used in underground mine workings, along with being conductors manufactured with materials adequate from a point of view of [6] SR HD 629.1 S2 – 2006: Test requirements on accessories for use on their mechanical, thermal, electric characteristics and flame power cables of rated voltage from 3,6/6(7,2) kV up to 20,8/36(42) kV. Part 1: Cables with extruded insulation retardancy, the cables shall have a proper constructional form [7] SR EN 60332 - 1-2: 2005 - Tests on electric and optical fibre cables with a special protective conductor, designed for connection to under fire conditions. Part 1-2: Test for vertical flame propagation for a earth in the general network. single insulated wire or cable. Procedure for 1kW pre-mixed flame After performing tests and assessments was found out that [8] SR EN 60332 - 1-1: 2005 - Tests on electric and optical fibre cables under fire conditions. Part 1-1: Test for vertical flame propagation for a the most adequate power cable construction is the one having single insulated wire or cable. Apparatus the protective conductor evenly distributed around each phase. [9] SR EN 60332 - 2-1: 2005: Tests on electric and optical fibre cables For these cables the risk of phase breaking or occurrence of a under fire conditions. Part 2-1: Test for vertical flame propagation for a short-circuit before insulation faults is significantly lower than single small insulated wire or cable. Apparatus in case when the protective conductor is placed between the 103 Technical support by Author Index Lupu Leonard, 99 A M Aldea Mihaela, 21 Arad Letitia-Susana, 77 Magyari Mihai, 39 Marc Gheorghe, 3, 17, 43, 47, 61 B Micaciu Alexandru Cătălin, 57 Mițǎ Oana, 65 Barosan Ion, 3 Moldovan Lucian, 39 Boca Maria Loredana, 43 Muntean Berzan Florina, 99 Bostan Cazimir-Gabriel, 65 Muntean Maria, 61 Burian Sorin, 25, 39 Muntean Nicolae, 51 C O Cioflica Dumitru, 69 Olteanu Emil, 73 Ciortea Elisabeta Mihaela, 21 Colda Cosmin, 25 Ö Coroiu Adriana Mihaela, 81 Croitoru Bogdan, 3 Ömer Meşedilci, 43 Csaszar Tiberiu, 25 P D Părăian Mihaela, 99 Darie Marius, 25 Păun Florin Adrian, 99 Dobra Remus, 29, 47 Pocan Ioan, 69 F R Fotău Dragoș, 39 Rad Daniel, 39 Friedmann Martin, 39 Rîşteiu Mircea, 3, 17, 61 Rotar Corina, 9 G Rus Daniela Carmen, 35 Gabor Dan, 99 S Ghicioi Emilian, 35 Ghita Octavian Mihai, 85 Samoila Brana-Liliana, 77 Şofâlcă Ionuţ, 17 H Sorandaru Ciprian, 51 Hüseyin Bahad r Öz, 89 Ş Hüseyin Taş, 89 Şchiopu Paul, 65, 85, 95 I T Ileană Ioan, 3, 17, 61 Ionescu Jeana, 25 Topor Marcel, 51 Ionica Andreea, 29 Tulbure Adrian, 69 İsmail Melih Taş, 89 U J Utu Ilie, 77 Jurca Adrian, 99 V K Vătavu Niculina, 99 Kovacs Attila, 35 Vintea Adela, 85 Vlădescu Marian, 95 L Vușcan Gheorghe Ioan, 57 Leba Monica, 29 104 Technical support by