IAEA-IWG-NPPCI-95/12 . LIMITED DISTRIBUTION
WORKING MATERIAL
INSTRUMENTATION AND EQUIPMENT FOR MONITORING AND CONTROLLING NPP POST-ACCIDENT SITUATIONS
Proceedings of a Specialists9 Meeting Organized by the International Atomic Energy Agency in co-operation with Research Institute of Atomic Reactors (RIAR) and held in Dimitrovgrad, Russian Federation, 12-15 September 1995
Reproduced by the IAEA Vienna, Austria, 1995
NOTE The material in this document has been supplied by the authors and has not been edited by the IAEA. The views expressed remain the responsibility of the named authors and do not necessarily reflect those of the government(s) of the designating Member State(s). In particular, neither the IAEA nor any other organization or body sponsoring this meeting can be held responsible for any material reproduced in this document. We regret that some of the
pages in the microfiche copy
of this report may not be up to the proper legibility standards,
even though the best possible
copy was used for preparing
the master fiche
Poor print: Sc J a. j FOREWORD
Accidents in nuclear power plants are postulated or un-postulated events that must not lead to a significant release of radioactive material to the environment. Accident management involves the actions taken by the nuclear power plant staff during the course of an accident to: prevent core damage, terminate progress of core damage, maintain containment integrity and minimize off-site releases.
Traditionally, nuclear utilities have relied on event-oriented emergency procedures, that is, planned, manually-controlled actions following reactor trip and other safety actions that are automatically performed. These procedures are based on design basis accident scenarios that are evaluated in the plant safety analysis. However, accident evolution may follow a route other than that predicted. Furthermore, combination of events and faults may be such that the accident cannot be clearly identified. To confront these problems, safety-function oriented emergency procedures have been proposed. In this approach there is less emphasis on accident characterization and more on restoring the plant to safe conditions based on the values of representative plant variables.
Due to the difficulties in determining cause and type of accident, fully automated safety systems are designed to mitigate the effects of postulated accidents. Operators need indications in the control room to verify that these safety systems succeeded in performing their intended safety functions. For example, measurements of neutron flux, control (or shutoff) rod position will reveal whether the shutdown system actuated successfully.
Ideally, accident and post-accident management should be conducted in the main control room by a special team composed operators and safety engineers. The team's ability to mitigate the consequences of the accident will strongly depend on the adequacy and reliability of the monitoring system.
Accident and post-accident monitoring requires special instrumentation that is independent of normal power plant instrumentation but qualified to survive the severest conditions associated with design-basis accidents. Normal power instrumentation may perform satisfactorily in the early phases of the accident but will gradually become unreliable and off-scale as the accident progresses and the measured quantities deviate significantly from the normal operating conditions. Instrumentation is also provided to assist operator actions involving safety-related systems that need to be manually started. Accident-monitoring instrumentation, and the variables being monitored, should be similar to those used in normal power operation. This will assure that the operator is familiar with the characteristics and behavior of the instruments.
ReUabiHty of the sensor signals is another serious concern because computer processing of the input data may sometimes mask sensor failures. Therefore some method of on-line signal validation is needed. Possible techniques include: consistency checks between redundant channels, noise detection for signal changes and process empirical modelling, i.e., the use of simple mathematical models that correlate the variables of the physical process.
1 Following the Three Mile Island accident several studies recommended the use of computerized aids to assist operators in monitoring plant safety-related information, correcting abnormal conditions and providing feedback to corrective actions. The idea has been embraced by nuclear utilities and computerized safety panels were installed in the main control rooms of many nuclear power plants. The safety panel may perform numerous functions including: monitoring the main safety variables, displaying the chronology of faults and the status of the safety actions and assisting the operator in identifying the procedure to be followed.
The use of digital technology in the safety panel has considerable advantages over conventional analog displays, among which are: lower costs, reliability, the ability to display information in different ways, use of mathematical models to interpret the ongoing phenomena, flexibility to modify the software and facility to interface with other digital devices. It is essential the operators participate in the development of these monitormg/diagnostics systems from the start. For example, operators can conduct preliminary tests on full scope simulators and provide the system designer with important feedback on system weaknesses and additional operational needs.
The Specialists' Meeting on "Instrumentation and Equipment for Monitoring and Controlling NPP Post-Accident Situations" was organized by the IAEA in co-operation with the Research Institute of Atomic Reactors (RIAR), Dimitrovgrad, Russian Federation. The meeting brought together experts on power plant operation with experts on application of today's instrumentation and control technology. In this way, a match can be made between those knowing the industry needs and requirements and those knowing the potentials of the technology.
The objectives of the Specialist's Meeting were:
To provide an international forum for presentation and discussion on experience with mstrumentation and equipment for monitoring and controlling NPP post-accident situations.
To share experience on the design and improvements in the subject area..
To identify and describe advanced features for safety and post-accident management improvements.
In order to facilitate a structured discussion and not to omit important problem areas, papers on the following subjects were considered to be within the scope of the Specialist's Meeting:
Analysis of existing post-accident instrumentation (in terms of range, qualification and redundancy) to support post-accident management.
2 3
Requirements for post accident mstromentation and equipment - operating experience, regulations, standards.
Experience in upgrading of the existing instrumentation.
Instrumentation and equipment for emergency management systems (emergency rooms and national centers).
Post-accident computerized aids to assist operators.
Post-accident operating procedures.
R&D in post-accident instrumentation, future trends.
The present volume contains: (1) the papers presented by national delegates, (2) programme of the meeting, (3) description of the RIAR research facilities, and (4) list of participants.
s
CONTENTS
PAPER PRESENTED AT THE MEETING
LI Improvement in Post-Accident Instrumentation for Spanish Nuclear Power Plants Rafael Cid, Spain
1.2 Safety Parameter Display System for Kalinin NPP V.l. Andreev,KN. Videneev, NN. Davidenko, G.I. S haft an, V.G. Dounaev and V.T. Neboyan, Russian Fed., J-C. Tissot, D. Joonekindt, France
1.3 Development of Transmitter with Hybrid-IC for Post-Accident Monitoring Instrumentation T. Ikeuchi and S. Watanabe, Japan
1.4 Relative Different Energy Neutron Radiometry in Reactors for Preventing of Accidents Caused by Uncontrolled Reactivity Variations S. V. Volkov, Russian Federation
1.5 A Stochastic Approach to Accident Identification in Nuclear Power Plants Kee-Choon Kwon, Soon-Ja Song, Won-Man Park, Joe-Chang Park and Chang- Shik Ham, Rep. of Korea
1.6 Accident Monitoring in Ventilation Stack V.Kapisovsky, V.Zbiejczukovâ, F.Gâbris, G.Belan, JZentan, J.Bukovjan, I.Rehak, Slovak Rep.
1.7 Experience with Neutron Flux Monitoring Systems Qualified for Post-Accident Monitoring KG. S Imgars and IF. Miller, USA.
1.8 Basic data of Emergency Response Centre O. Jenicek, Czech Republic
1.9 Applied Software of the Emergency Recording System for Reactor Facility Parameters under the Minor Statistics Conditions V.B. Ivanov, A.F. Grachev, O.M. Kinsky, RS.Makin, A.I. Ochrimenko, L.I. Demidov, V.l. Karpjuk, V.K. Afonin, RG. Iakanderov, Russian Fed.
1.10 Design Implementation of the Post-Accident Monitoring (РАМ) System for Wolsong NPP Units 2,3 & 4 in Korea San-Joon Han, Rep. of Korea
1.11 Electrical Characteristics of Nuclear Power Plant Cables at Accident Simulation V.l. Rogov, V.N. Ulimov, N.I. Filatov, V.S. Shestacov, Russian Fed.
1 6
1.12 Research Istitute of Atomic Reactors: General Presentation V.B. Ivanov, Russian Fed.
1.13 Creating Requirements for Control of Post-Accident Situations at NPP V.S. Dickarev, V.S. Ionov, Russian Fed.
1.14 Principles of Elaboration and Creation of Information-Analytical System "RI Operation Safety with SSC RIAR Research Reactors" V.B. Ivanov, A.F. Grachev, O.M. Kinsky, RS. Makin, A.I. Ochrimenko, L.I Demidov, V.l. Karpjuk, V.K. Afonin, RG. Iskanderov,Russian Fed.
1.15 Radiological Consequences Computer Code as Emergency Operator Help in Case of Fission Products Release During Reactor Accident V.D. Kizin, S.A. Efarov, KI. Stikokov, RK. Yakshin, Russian Fed.
2. PROGRAMME OF THE MEETING
3. DESCRIPTION OF THE RIAR RESEARCH FACILITIES
4. LIST OF PARTICIPANTS PAPER PRESENTED AT THE MEETING
INTERNATIONAL ATOMIC ENERGY AGENCY
SPECIALISTS MEETING ON INSTRUMENTATION AND EQUIPMENT FOR MONITORING AND CONTROLLING NPP POST-ACCIDENT SITUATIONS
12-15 September 1995 Dimitrovgrad, Russian Federation
IMPROVEMENT Ш POST-ACCIDENT INSTRUMENTATION FOR SPANISH NUCLEAR POWER PLANTS
Rafael Cid Spanish Regulatory Body (C.S.N.) Justo Dorado, 11 28040 Madrid, Spain 10
IMPROVEMENT ЕЧ POST-ACCIDENT INSTRUMENTATION FOR SPANISH NUCLEAR POWER PLANTS
Rafael Cid Spanish Regulatory Body (C.S.N.)
1. INTRODUCTION AND SUMMARY
Concerns about the adequacy of currently available instrumentation to withstand plant accident conditions and be available during the recovery from an accident raised in 1979 after the accident at Three Mile Island focused attention on post accident instrumentation and the emergency response capabilities. As result the US. Nuclear Regulatory Commission issued Regulatory Guide 1.97 Rev. 2, and the current issue, Rev. 3, that was published in May 1983. In addition was issued NUREG-0737 which contains requirements of the US. Nuclear Regulatory Commission related not only to instrumentation, but also other aspects of emergency response including emergency procedures, technical support center, Safety Parameter Display System, etc.
There are nine operating nuclear power plants in Spain eight of them are US. basic design (Westinghouse and General Electric), as a general rule, the safety technical requirement of the country which the basic design came from is considered applicable in Spain, in particular, Regulatory Guide 1.97 and NUREG-0737 apply to these plants and KTA-3502 for one Siemens-KWU design plant.
This regulation identifies and classified more than 100 variables to be monitored and specify the basic design and qualification requirements for each category. The impact for each plant of this regulation depend of the its generation. For new plants, (third generation) which operating license was issued (1987/1988) later that current regulation, the original design incorporate all requirements for post accident instrumentation. In the case of the second generation of pants ( those which operating licenses were issued in the period 1981-1984) current requirements were not included in its design basis. The Spanish Regulatory Body (C.S.N.) required to analyze discrepancies and provide a program to install new instrumentation or to modify the in-place one in order to meet this regulatory guide. For old plants, (first generation) which operating licenses were issued in 1968 and 1970, the design basis are far from current regulation, and a compromise must necessarily be made between the achievable safety benefits and the constraints on installing new systems of instrumentation in old plant and this modifications would be made as part of a general plant safety upgrade program.
This paper is focused in how the regulation issued as consequence of TMI-2 accident, has affected to the Spanish plant describing the main improvement that has been implemented in each plant and providing a general overview of the approach adopted and current status for the different generation of nuclear plants.
2. DISCUSSION OF POST-ACCIDENT INSTRUMENTATION REQUIREMENTS
Regulatory Guide 1.97 defines the mstnimentation needed to provide information in case of accident in five types. The types are:
Type A: Are those variables that provide primary information needed to allow operators to take specified manual controlled actions for which no automatic control is provided.
Type B: Are those variable that provide information to indicate whether plant safety functions are being accomplished. This safety function are: - Reactivity control - Core Cooling - Mamtaining Reactor Coolant System Integrity -Mamtaining Containment Integrity Type C: Are those variables that provide information to indicate the potential for breach of barriers to fission product release.
Type D: Are those variables that provide information to indicate the operation of individual safety systems.
Type E: Are those variables to be monitored as required for use in detemiining the magnitude of radioactive material releases and continuously assessing such releases.
The design criteria are separate in three categories that provided a graded approach to requirements depending on the importance to safety of specific variable. Category 1 provides the most stringent requirements and is specified for key variables, in general provides full qualification (environmental and seismic), redundancy and continuous display and onsite power. Category 2 provide less stringent requirements and applies to instrumentation for indicating system operating status, in general, the most important requirement for this instrumentation is the environmental qualification. Category 3 provide requirements that will ensure high-quality commercial-grade equipment and applies and backup and diagnostic instrumentation.
The essential of this regulatory guide is that identify the mmimum number of variables to be monitored, the range selection for each variable that would be sufficiently great to keep instruments on scale at all times and the design criteria for this instrumentation. Is important to remark that the environmental qualification requirements for temperature, pressure, humidity and radiation are set up in relation with the Design Bases Accidents and not specific requirement are considered in relation with Severe Accidents.
3 In relation with pervious review of this Regulatory Guide (Rev.l) it represent important changes in order to provide new instrumentation, increase range, (in the past, some instrumentation ranges have been selected based on the setpoint value for automatic protection) then wide range are required for monitoring degraded conditions associated with an accident and necessary environmental qualification. It is essential that instrumentation so upgraded does not degrade the accuracy required in normal operation.
KTA-3502 identify the post-accident variables and set-up the design requirements, in a similar way than KG. 1.97 does.
In addition, Supplement 1 to NUREG-0737, Clarification of ТМЗ Action Plan Requirements, Requirements for Emergency Response Capability, establish that a Safety Parameter Display System should provide a concise display of critical plant variables to the control room operators to aid them in rapidly and reliably determining the safety status of the plant. Although the SPDS will be operated during normal operations as well as during abnormal condition, the principal purpose and function of the SPDS is to aid the control room personnel during abnormal and emergency conditions in determining the safety status of the plant ad in assessing whether abnormal conditions warrant corrective action by operators to avoid a degraded core.
The basis for this requirement stems from the lack of centralized display capability in the TMI-2 control room. Control room personnel could not easily develop an overview of plant condition. This system must centralize all the variable for critical safety function in order to facilitate the comparison of variables or the integration of various symptoms within the same time frame. Also it could avoid some behaviors such as operator fixation on a limited set of plant anomalies while safety functions where in jeopardy.
3. IMPLEMENTATION APPROACH AND STATUS PER SPECIFIC PLANT
The general approach was to compare the existing instrumentation with Regulatory Guide 1.97 by the licensee and submit a report to the CSN that provides the following information for each variable:
- Instrument range - Environmental Qualification - Seismic qualification - Quality assurance - Redundancy - Power supply - Display
The submittal should identify deviation from the regulatory guide and provide supporting justification or alternatives such as installation new instruments or upgrade
4 13
existing instrumentation. The submittal should inform about schedule of implementation.
The application of this regulation depend of the date of design and construction for each plant. For an operating plant , is clearly difficult to implement such a large system, and it require flexibility on analyses justification of each deviation, takes into account the benefits obtained versus the impact to the plant in a general point of view.
Under this point of view, we can classify the Spanish NPP hi three generation:
3.1 First Generation
The first generation are two plants which operating licenses were issued before that current regulation and the original design is far from this requirement. Different approach have been taken in each case.
- José Cabrera NPP : Is a PWR, Westinghouse NSSS design, 160 Mwe, the operating licenses were issued in 1968.
As part of a general plant backfitting in safety systems, during a large shut down in 1985, it was adopted a different approach to the guidelines of R.G. 1.97. This approach consists in the selection of the minimal instrument necessary to provide aid to the operator, in combination with the emergency procedures, in diagnosis and recovery from an accident. Providing the operator with the capability to perform the manual actions specified in such procedures. Then a complete new Post-Accident Monitoring System was implemented with 21 variables to allow monitoring of the plant's Critical Safety Functions: Critical Safety Function Variable
SUBCRITICALITY Key: None Backup: Wide Range T. cold Core Exit Temperature RCS Boron Concentration
CORE COOLING Key: Core Exit Temperature Backup: Wide Range T. cold RCS Wide Range Pressure
HEAT SINK Key: S.G. Wide Range Level S.G. Narrow Range Level A.F.W. Flow Steam Line Pressure Core Exit Temperature Backup: None
RCS INTEGRITY Key: RCS Wide Range Pressure Wide Range T. cold Backup: Containment Pressure
5 Containment Radiation Secondary System Radiation
CONTAINMENT Key: Containment Pressure Containment Radiation Containment Sump Water Level Backup: None
RCS INVENTORY Key: Pressurizer Level Backup: Containment Sump Water Level S.G. Wide Range Level
Variables that provide information of operability of Safety Systems
- Containment Sump Water Temperature - RWST Level - SI Dovvncomer Flow - SI Pump Suction Pressure - SI Jet Pump Motive Flow - DWT Level - Spent Fuel Pit Level
The general functional requirements for these variables were: Redundancy (2 channels per variable); Electrical and physical separation; Powered from separated trains; Seismic and Environmental Qualification (DBA);
As second approach, in 1988 , CSN required a general review of the post accident instrumentation versus R.G. 1.97, identify deviation from the regulatory guide and provide supporting justification or others alternatives. As consequence new or upgrade instrumentation have been identified and installed during the following refueling outages (1990-1994).
- Neutron Flux (1996 refueling outage) - Degrees of Subcooling - Air Containment Cooling flow (one channel) - RHRS flow and temperature (sensor Quafification) - Containment Hydrogen Concentration (one channel) - Containment Area Radiation (High Range) - Containment Isolation valves Position (Qualification) - Steam Flow (Qualification) - Vent From SG Safety Relief Valves, Novel Gases and Vent Flow - CMmney vent discharging
Mam exception, accepted by CSN, versus a more modern Westinghouse plant are:
Reactor Vessel Water Level not provided
6 ( 5
- Redundancy for most variables is addressed by providing two channels per variable To strictly address information ambiguity a Third Channel for each variable would be required. - Same variables such Containment Hydrogen Concentration only one channel is provided.
The degree of implementation is at this time is almost complete , only the Nuclear Instrumentation System remain to be upgraded that is schedule for the next refuebng outage. Other important project to be implemented is a Safety Parameter System, now in phase of design.
- GARONA NPP :
Is a BWR-3, Mark-I , NSSS design by GE, 460 Mwe, the operating license was issued in 1970.
CSN required to this pant a initial review of plant instrumentation versus KG: 1.97 Rev. 2 (1982). Even deviations were important but not so large that José
Cabrera A A general backfitting (Systematic Evaluation Program) was made in the period 1984 - 1987. Improvement in instrumentation was made during this period ii relation with Reactor Protection System. In addition a Safety Parameter Display System was installed. In 1990 CSN required a systematic review of the post accident instrumentation versus R.G. 1.97 and correct or justify each deviation. As consequence important improvement have been made during the following refueling outage. Such as: - Reactor Vessel Water Level (range, qualification, recorder) - Temperature Drywell (range, environmental qualification) - Reactor Pressure (new) - Temperature of Suppression Pool (new) - Water L^vel in the Suppression Pool (physical separation) - LPCI Temperature exit of cooler - Temperature SW/LPCI exit of cooler - Water flow SW/LPCI exit of cooler - Containment Hydrogen Concentration - Containment Oxygen Concentration The degree of implementation is practically total except for the Containment Hydrogen and Oxygen concentration that is scheduled to install in the next refueling outage in 1996. 3.2 Second Generation The second generation are five plants which operating license was issued in the period 1980- 1985 : 7 16 - Almaraz I : PWR, NSSS (Westinghouse design), 930 Mwe, Op. License 1980. - Almaraz П: PWR, NSSS (Westinghcmse design), 930 Mwe. Op. License 1983 - Asco I: PWR, NSSS (Westinghouse design), 930 Mwe. Op. License 1982 - Asco П: PWR, NSSS (Westinghouse design), 930 Mwe. Op. License 1985 - Cofemtes: BWR, NSSS (G.E. design), 994 Mwe, Op. License 1984 Because of current regulation was issued during the final of construction of this plant, this new requirements are not included in its original design basis. However, CSN required to meet with this regulation and upgrade Post- Accident Instrumentation. The approach for these cases was more strict than old plants and very few exception were accepted. - ALMARAZ I & П ; ASCO I & П The four PWR Westinghouse pants are very similar and NSSS instrumentation is practically the identical. Existing qualify instrumentation (narrow range) for the Reactor Protection System is used as well as post-accident instrumentation. More relevant deviations (key variables) are in relation with: - Wide range requirements not provided * RCS Hot Leg Temperature *RCSCold Leg Temperature * Containment Sump Water Level * Containment Pressure * Containment Radiation * Pressurizer Level - Environmental Qualification * S.G. Water Level (wide range) * Neutron Flux * Containment Atmosphere Temperature * Core Exit Temperature * Containment Sump Water Level (narrow range) - Requirement of new instrumentation * Reactor Vessel Water Level * Degrees of Subcooling * Containment Hydrogen Concentration * Containment Sump Water Temperature * Containment Heat Removal * Primary System Safety Relief Valve Position 8 \ T- All these instrumentation have been upgraded to meet 1.97 Rev. 3 requirement. However, in relation with the Reactor Vessel Water Level instnunentation the utilities opposed to provide it. The utilities position was based in the high cost and relative benefits of this instrumentation which reliability and operability was in doubt and could mislead to operators and other existing alternative instrumentation (Core Exit Temperature and Degrees of Subcooling) could make its function. CSN disagree with this position considering it is a essential variable as was demonstrated in the TMI accident and the existence of commercial available instrumentation that provide warranty of unambiguous indication. Plants have changed obsolete process computer and provided Post-accident aids (SPDS) and other functions in the new one. - COFERNTES NPP During the final of construction most of the changes necessary were made in order to meet RG 1.97 Rev. 3 . Only a few variables with deviation remain after Commissioning: - Reactor Vessel Water Level (lack of qualification) - Suppression Pool Water Level (range) - Containment Hydrogen Concentration (range) These instrumentation was update in the first refueling outage. In addition a Safety Parameter Display System (ERIS) is operative since 1986. 3.3 Third Generation The third generation are two plants which operating license was issued in the period 1987- 1988 : - Vandellos П: PWR, NSSS (Westinghouse design), three loops, 930 Mwe, Op. License 1987. - Trillo I: PWR, NSSS (Siemens-KWU design), three loop, 1040 Mwe, Op. License 1988. In this cases the original design bases incorporate all post TMI requirements, and not significant deviation versus RG 1. 97 Rev.3 have been found. In relation with computerized aids to operators (SPDS) in Vandellos П NPP, the CSN review found deficiencies in time response (Slow displaying changes). SPDS is installed in a obsolete low capacity plant process computer with many function working at the same time. Plant is planing to changes this computer. Trillo NPP have not incorporated Safety Parameter Display System. 9 4. GENERAL OVERVIEW - Spanish current requirement in Post accident mstrumentation * US Design plants: RG 1.97 Rev. 3; NUREG-0737 * German plant: KTA-3502 - Severe Accident (non specific requirement to provide new instrumentation). - A great improvement and effort have been made in old plants (first generation) in Post-Accident Instrumentation to meet with the current regulation. Key variables have been addressed, however same exception have been accepted by CSN. - New plants (second and Third generation) provide all 1.97 Rev. 3 requirements. Important improvements and effort have been made in the second generation plants. - SPDS is provided in US design plants (Westinghouse & GE). Except one plant that is in the design phase. 5. REFERENCES 1. US Nuclear Regulatory Commission " Instrumentation for Light-Water-Cooled Nuclear Power Plants and Environs Conditions During and Following an Accident. RG 1.97 Rev. 3 (May 1983) 2. US Nuclear Regulatory Corrunission " Clarification of TMI Action Plan Requirements, Requirements for Emergency Response Capability, NUREG-0737 Supplement 1, December 1982 3 . Post Accident Instrumentation, KTA-3502 10 Safety Parameter Display System for Kalinin NPF V.l. Andrew, E.N. Videneev Concern '"'Rosenergoatom" Moscow, Russia J-C. Tissot, D. Joonefcmdt SEMA GROUP, Grenoble, France N.N.Davidenko, GXSUaftan Kalinin NPP V.G.Dounaev, V.T. Neboyan Consyst Moscow, Russia ABSTRACT Paper discusses the safety parameter display system (SPDS), which is being designed for Kalinin NPP. The assessment of the safety status of the plant is done by the continuous monitoring of six critical safety functions and the corresponding status trees. Besides, a number of additional functions are realized within the scope of KlnNPP, aimed at providing the operator and the safety engineer in the main control room with more detailed information in accidental situation as well as during the normal operation. In particular, these functions are: archiving, data logs and alarm handling, safety actions monitoring, mnemonic diagrams indicating the state of main technological equipment and basic plant parameters, reference data, etc. As compared with the traditional scope of functions of this kmd of systems, the functionality of KlnNPP SPDS is significantly expanded due to the inclusion in it the operator support function "computerized procedures". The basic SPDS implementation platform is ADACS of SEMA GROUP design. The system architecture includes two workstations in the main control room: one is for reactor operator and the other one for safety engineer. Every station has two CRT screens which ensures computerized procedures implementation and provides for extra services for the operator. Also, the information from the SPDS is transmitted to the local crisis center and to the crisis center of the State utility organization concern "Rosenergoatom". 1. INTRODUCTION The Russian State Utility Organization ROSENERGOATOM has been implementing a wide range of measures directed at improving nuclear safety, radiological protection and crisis management in nuclear industry. Among others, the following three items constitute basic part of those activities: - creation of a national crisis center - crisis center of the concern ROSENERGOATOM in Moscow, the local crisis centers on NPPs sites, as well as the system of data transmission from NPPs to crisis centers; - implementation at nuclear power plants the symptom-oriented emergency procedures; 1° 2 - implementation of safety parameter display systems (SPDS) on all NPP units. As can be easily seen from above, the SPDS plays a key role in this program since: 1. It is considered a main source of information supplied from NPPs to the crisis centers; 2. S"PDS plays a role of instrumentation tool for implementing symptom based emergency procedures on NPPs. Moreover, the implementation of SPDS meets the requirements of the Russian safety regulation [1]. The status of SPDS implementation work varies from one NPP to another. Rather promoted it is on unit two of Kalinin NPP. KlnNPP is a two unit nuclear power plant with WER- 1000 type of reactor of so called small series (design Y-338) with electrical capacity of950 MWe, four loops and traditional for PWRs engineered safety systems. The unit two has been in operation since 1987. The significant support in SPDS development for Russian NPPs comes from European Union through technical assistance programs TACIS. In 1994 in the scope of TACIS-91 program the French company Sema Group in cooperation with the Russian company Consyst developed an SPDS mockup. The reference plant for the mockup was unit two of KlnNPP. The recipient of a technical assistance was concern ROSENERGOATOM and the purpose of developing SPDS mockup was to support decision makers in the evaluation of the final design of the real SPDS system The work has been finalized successfully and in 1995 Sema Group was granted a new contract from the European Community for TACIS-93/2.03 project, aimed at development, testing and validation of a safety panel prototype (SPP) for WER type plants (Kalinin NPP unit two as a reference). Like in an earlier case, the Russian counterpart and a subcontractor to Sema Group is Consyst. SPP is not intended to be connected to a real plant under the scope of this project. However, the system shall be designed in such a way that no software modification is required to connect the system to a data acquisition system on the plant. The prototype of SPDS will utilize a Sema Group platform ADACS. In parallel, KtoNPP contracted Consyst for the tasks which are not in the scope of TACIS- 93/2.03. 2. PURPOSE OF SPDS Among the different types of operator support systems which are nowadays operational at NPPs, the safety parameter display system probably is the most widely used and well recognized facility. There is nowadays an international consensus on the minimal requirements for a safety parameter display system. Therefore, no wonder, that it has been taken into account by respect of the ШС-960 [2] and NUREG 696 [3] standards.. Though "SPDS" acronym is widely spread in technical literature, and its meaning (see NUREG-696 [2] and the above mentioned EEC-960) is accepted by designers and operators, the concrete realization of this operator support system might be different depending on MMI and supplementary functions, which the system is supposed to perform not only in case of emergency but in normal operating conditions as well. A quick progress in computer and information technologies gives a great opportimity to extend the scope of SPDS fimctionality. «ЯГ i 3 The primary purpose of SPDS is to provide information support to the main control room personnel in case of emergencies. Besides, in all operating modes SPDS presents concise information about the Critical Safety Functions (CSF). The operator support is ensured by: - display of critical safety functions vviiich assists operator in rapid and reliable evaluation of the safety status of the plant and advise the operator whether abnormal conditions require corrective action to avoid a degraded core or radioactive release (selection of emergency procedure); - a concise display of information important for farther safety evaluation and diagnostics; - evaluation of safety action and advise to the operator. Also, SPDS information is intended for utilization in the local crisis center and in the crisis center of the concern ROSENERGOATOM for evaluation of the safety status of the plants. 3. FUNCTIONAL DESIGN 3.1. Critical safety functions monitoring The primary purpose of SPDS is to present concise information about the safety status of the plant. For this purpose, a limited number of safety status indicators should be defined. The safety is provided if the maximum number of the radioartivity barriers remain intact. The integrity of those barriers can be maintained if a fixed number of functions, called the Critical Safety Functions (CSF) are satisfied during and айет the emergency situation. Therefore, the required operator action is to monitor the states of CSF. The follow-up action is to maintain a safe or at least acceptable status of the CSF. For this purpose every CSF is associated with a number of symptom oriented emergency procedures. When a critical safety function is challenged, the operator proceeds with the implementation of appropriate emergency procedure. The SPDS must alert the operator in every situation when the critical safety function changes its status. Safety analysis of Kalinin NPP has shown that maintaining the integrity of the safety barriers can be achieved by maintaining the following critical safety functions: - Subcriticality - Core cooling - Heat sink - Reactor coolant integrity - Containment integrity - Inventory An important principle of the plant safety assessment by CSF is a prioritization of CSF according to their importance for safety. The order in the above list of CSFs just corresponds to the priority of every critical safety function. It means that, for example, a "Subcriticality" CSF has a maximum priority. 4 The status of CSF may vary depending on the severity of disturbance. Four states of a critical safety function are considered and a color coding is utilized for their indication: - CSF is operational -green; - CSF is partly degraded -yellow; - CSF is degraded -orange: - CSF is very degraded -red. The purpose of proposed hierarchy in CSF is to provide the operator with a rule: which CSF should be restored firsti n case when more than one CSF is degraded at the same time. The CSF monitoring is provided in SPDS on two levels: with a help of a primary CSF display, on which the status of all critical safety functions is presented, and on the second level with a help of status trees. 3.2. Monitoring of main parameters in normal operating conditions When operating the nuclear power plant, the operators respond to small disturbances by appropriate action in timely manner in order to avoid evolution of disturbances to serious incidents or even the accidents. In other words, by applying timely diagnostics and corrective actions operator must avoid degradation of CSFs. For this purpose SPDS displays in the normal operating condition the basic NPP parameters. The number of parameters is limited by 8-10 and the layout of display is such that the operator may react to parameters deviations easily and reliably. This format is displayed on SPDS screen by default in normal NPP operating conditions. 3.3. Safety action monitoring This function aims at showing the operator the current state and history of the safety actions. The function is organized in two levels. The firstleve l consists of a display that presents the status of all the safety actions on each train. The second level is a set of displays presenting the main circuits and their actuators. If there is a failure in the automatic operation of a safety action, an alarm is triggered. 3.4. Operator help displays Top level displays (main CSF display, status trees and "main parameters" display) are intended in the normal and abnormal modes of operation to aid the operator in the activation of the detection process. Third level displays, or operator help displays, aid the operator in the further stages of event identification, by providing more detailed mformation on the entire plant. It is also of utmost importance to detect event at an early stage and avoid its evaluation to emergency condition. The third level displays may be of a very good help in this process. Tree main types of displays may present additional mformation: mnemonic diagrams; operating status diagrams; trend curves. Groups of this kind of formats are associated with the formats of le"d one and two. Ä 3 5 3.5. Alarms An alarm is a failure in plant or system operation, which requires imrnediate manual or automatic action. There are alarms directly associated with equipment (level one alarms). Also, there are "elaborated" alarms which are associated with internal logic equation or process status (level two alarms). Each alarm has a gravity level There are two gravity levels. A color associated with each gravity level: red for gravity two and orange for gravity one. The "Alarm" function is configured in the AD ACS platform and provides for alarm status presentation (active or inactive, acknowledged, shelved, masked) and convenient operator interface. 3.6. Logbooks and archives These functions also are configured in the ADACS platform. Filtering is provided in the logbook for more efficient analysis of a sequence of events. Among others, the changes in CSF status may be selected or operator actions with the status trees or emergency procedures. Short, medium and long term archives are maintained. 3.7. Operator support by computerized procedures Critical safety function monitoring in SPDS and the mechanism of status trees provide a good environment for aiding operator in selecting appropriate emergency procedure when a critical safety function is degraded. Further implementation of the procedure may be done in a traditional way with "paper" procedures or with a sort of computerized procedures. Traditionally, computerized procedures are not considered within a scope of SPDS functions and normally represent a stand along operator support system. The reason why it was included in the scope of SPDS functions was a strong requirement of the customer - Kalinin NPP. SPDS computer capacity and data processing environment provide a good opportunity for this. Though different types of computerized procedures are operational on NPPs in the world with a different level of automation, the type of system chosen for KlnNPP utilizes relatively low level of automation. In fact, system performs two computerized functions: - computerized procedure presentation and - computerized check of conditions in the procedure. Every procedure can be presented in a form of a logic tree and in a "step by step" form. The first type of presentation provides a general overview and a convenient indication of the current status of the procedure. The second one is used for interactive work of operator with the procedure. The basic principle of computerized procedure operation is: it is the operator who confirms that the current step of procedure is finished and the operator can go to the next step but not the computer. 6 In order to implement this function \vithin SPDS, two CRT screens are provided for an operator station, from which one wul be used for procedure presentation. Procedure initiation can be done in a different way depending on a type of procedure. For example, after reactor shutdown or engineered safety system initiation a procedure A-0 is automatically stated. CSF restoration procedures are called on by operator by clicking on a corresponding branch of a status tree. 4. DATA PRESENTATION IN SPDS Data presentation in SPDS is based on a hierarchical structure of three levels. The upper Ievd. consists of two formats: the main CSF display (Fig. 1) and "main parameters" display. These displays present the most concise information about the safety status and the main parameters of the plant in normal and abnormal operating conditions. The next three levels provide for operator support on the further steps of event diagnostics by presenting more detailed information about the plant and its main systems and components. Level two consists of six displays with status trees corresponding to six critical safety functions. Fig.2 shows an example of such status tree. Every status tree display includes the title - the name of a corresponding CSF, a number of nodes representing logic "Yes - No" and two output branches. A brunch which corresponds to the current node condition "Yes" or "No" is considered active. A sequence of nodes and active branches constitute a path which teiminates at the right border of screen where a reference to an applicable restoration procedure is displayed. The active branches are highhghted by a specific color denoting the deterioration of the function. Of a principal importance is to inform the operator about the current status of CSFs regardless what a format is currently displayed on SPDS screen. For this purpose a low left comer of every format is dedicated for a display of a small replica of a main CSF display. By clicking on tins fragment operator can at any time switch to the main CSF display. Level three displays mostly are represented by fragments of technological schemes with a certain level of abstraction. Fig. 3 gives an example of this kind of displays. Computerized procedure displays also are associated with level three. An example of a procedure logic tree display is shown in Fig.4. 5. SPDS ARCHITECTURE Though SPDS is not classified as a safety grade system (see [2]), maximum precautions should be made to guaranty a reliable operation of SPDS in all operation conditions. It is a requirement of the regulatory body, that highly reliable hardware and software with significant operation experience be utilized. Standardizing on common, readily available hardware and software platforms is essential. Given the requirements of SPDS, the implementation will be based on the Sema Group ADACS platform Д5 / 5.1. ABACS platform ADACS is a Sema Group generic information and control system which operates with high flexibility. It is designed to operate continually and meets the requirements of several industries, including nuclear power plants The process monitoring is provided though a highly interactive man-machine interface using computerized operator stations giving access to the following function: Display diagrams Presented on graphical VDUs, the display diagrams present plant sub-systems showing both mdividual equipment and synthetic information. The operator is able to choose the appropriate level of detail, from an overall display of the whole plant right down to the detailed display of a single device. ADACS provides a library of graphical items that may be integrated in such a display (animated shapes, numeric values, bar-graphs, indicators and so on...). Operation windows By selecting an item of equipment on a VDU displays, the operator may access an operation window which allows him to perform various management and control functions (e.g. : testing, setting maintenance or testing mode, etc.) as well as accessing detailed information concerning the equipment (i.e.: a technical data sheet). Alarm notification and alarm handling Alarms are raised when a fault or an abnormal situation is detected in the production process (e.g: a threshold is crossed). The operator can directly access to the associated display diagram to exarnine and correct the fault. Event history The operator may consult the process archives which contain a log of all the significant events, sort them, display them and print them Plant parameters history The operator may consult the process archives which contain a log of plant parameters; these parameters may be displayed either as a tabular format or a graphical format. Equipment lists The operator may consult equipment lists which correspond to various criteria used in the monitoring and control processes (e.g.: list of sensors on calibration, list of invalid piece of equipment etc.). Reports The operator may write or consult reports. ADACS has been built using a modular object design approach which allows the customer's specific requirements to be closely met by the addition and modification of software functions without impacting on the software design. ADACS provides very high availability and immediate switch-over by allowing hot active redundancy which ensures that the computer system remains usable, and no data is lost during transitory phases (e.g.: loss or re-insertion of a computer). ADACS is based on several Sema Group software components: - Real time database BDT. The BDT is an open and configurable software component designed to manage a distributed real-time database. This tool manages data and the associated processing operated on the data. Processing can be both calculation and display presentation. 8 - Expression Interpreter IEX. LEX is a tool allowing expression calculation defined by the user using DGS without any knowledge of a programming language. - Data Generation System DGS. The offline Data Generation System manages object types, objects instances, synoptic and generic displays creation and modification, according to the concept of object oriented distributed database. DGS provides creation and maintenance facilities for a central master database describing an industrial process and its implementation on a target distributed data processing system The real time system may be split in applications mnning on the same computer or on different computers. DGS manages data projections for each application on each computer and assumes global consistency between all the projections. - EGIPTE graphical editor, - ARLIC Local Area Network This is an internal Sema Group network management product which is based on a dual redundant Ethernet 802.3 network. - System Administration ADS; - Operator Station PO; - Logbook, manager JDB\ - Archiving tool ANA. ADACS has been developed using high level programming languages (mainly Ada), and international standards (e.g.: X, Motif, Ethernet). As a result, ADACS is independent from any hardware and software suppliers, and is easily portable. 5.2. System architecture Two levels constitute the SPDS architecture: the data acquisition level 1, and the level 2, on which SPDS functionality is realized. The hardware components of level 2 are connected by SPDS network (Fig.5). The data sources and data acquisition components are connected by a local area network of level 1 (Fig. 6). Interface between two levels is provided by the front-end computer. The data sources for SPDS are: - plant process monitoring system "Complex Uran"; - the upper level of in-core instrumentation system SVRK; - the data acquisition level of a radiation monitoring system AKRB; - additional data acquisition modules. AU data to be displayed is validated on a real time basis. Data validation consists of cross-checking redundant measurements, consistency checks other algorithms. The data acquisition and processing of SPDS is characterized by the following figures: number of analog inputs - 450 number of discrete inputs - 500 number of derived variables - 750 The primary data base is maintained on a front-end computer, which sends data to level 2 network. The upper level of SPDS consists of a main computer, two operator workstations with the additional CRT monitors (one WS is intended for reactor operator and the other one for a safety 9 engineer), engineer workstation which serves the system maintenance (modifications in formats and data in the database, configuration of SPDS), network equipment and unmterruptable power supply. 5.3. Siting of SPDS Two workstations of a reactor operator and a safety engineer are sited in the main control room. Main computer, engineer workstation, front-end computer are located in the computer room. 5.4. Hardware specification The main SPDS hardware components are listed in table 1. Tabl.1 Component Type of computer Quantity Main computer DEC 3000/6001 1 Operator work station DEC 3000/300LX 2 Engineer work station DEC 3000/300LX 1 Front-end computer DEC 3000/300LX 1 Long-term archive computer PC Pentium 1 With every workstation a high resolution 1280x1024 color monitor .21" is used. The operator actions are applied by mouse and a functional keyboard. For the SPDS network a Sema Group product - local area network ARLIC is utilized. 6. VALIDATION AND TESTING The crucial point in SPDS validation process is a validation on an analytical simulator of WER-1000/338. The validation process of the SPDS will be carried out by means of interactive dynamic testing. For this purpose the SPDS is to be connected to a plant simulator. This simulator has been designed and manufactured by a Russian - American joint venture "Jet". It is based on CüiconGraphics hardware platform and utilizes the modeling technology of S3Technology. From the process simulation point of view technical performance of the simulator is very similar to full scope type of simulators. 7. CONCLUSIONS The Russian utility ROSENERGOATOM has been implementing a wide range of measures aimed at establishment of means to improve NPP operation, achieve plant surveillance and provide accurate information for support during emergency situations. Among them the important role plays the implementation of safety parameter display systems, which provide for a main source of data to the local and State Utility crisis centers and constitute an integral part in the 10 process of implementation of symptom based procedures. The work is significantly supported by European Union through TACIS programs. The concept of safety parameter display system has been demonstrated, evaluated and tested on a safety panel mockup, which is based on a nuclear proven technical platform ADACS and is a real step toward a prototype of the system. SPDS prototype (SPP) is under development for the concern ROSENERGOATOM and KlnNPP as a reference plant. SPP is not intended to be connected to a real plant under the scope of this project. However, the system shall be designed in such a way that no software modification is required to connect the system to a data acquisition system on the plant. The prototype of SPDS will utilize a Sema Group platform ADACS. In parallel, KlnNPP finance works which are not in the scope of TACIS-93/2.03. Development of the SPDS for KlnNPP is based on well established standards like IEC-960. The primary purpose of SPDS is to provide information support to the main control room personnel in case of emergencies. Besides, in all operating modes SPDS presents concise information about the Critical Safety Functions (CSF). The operator support is ensured by: display of critical safety functions which assists operator in rapid and reliable evaluation of the safety status of the plant and advise the operator whether abnormal conditions require corrective action to avoid a degraded core or radioactive release (selection of emergency procedure); a set of lower level displays will be provided as a constant information source for the operators in the follow-up to the restoration procedures. All safety systems, actions and parameters required by the recovery procedures will be monitored on these displays. SPDS information is also intended for utilization in the local crisis center and in the crisis center of the concern ROSENERGOATOM for evaluation of the safety status of the plants. Human factors engineering has been incorporated in the SPDS design to enhance the functional effectiveness of control room personnel. Computerized procedures are being developed as a supplementary function of SPDS with the aim to support operator in the process of implementation of symptom based emergency procedures. SPDS realization is based on a Sema Group generic information and control system ADACS. It is designed to operate continually and meets the requirements of several industries, including nuclear power plants. The validation process of the SPDS will be carried out by means of interactive dynamic testing. For this purpose the SPDS is to be connected to a plant simulator. Technical performance of the simulator is adequate for this purpose. The capacity of the system gives enough flexibility for modifications and new functions implementation in future. REFERENCES [1] General Regulations to Provide Safety at Nuclear Power Plants (OPB-88). Date of coming into force: July 1, 1990. *1 11 [2] Ihtemational Electtotecrmical Commission. Functional design criteria for a safety parameter display system for nuclear power stations, EC 960.1988. [3] U.S. Nuclear Regulatory Cornmission. Functional Criteria for Emergency Response Facilities, NUREG-0696.1981. КФБ Г Блок N |Тэо>Ф Режим: I \ Время с момента срабатывания ЙЗ: \ KP оз то цк го зт Активная ФУНКЦИЯ: I I Инструкция: I ~| Fig.l. Main CSF display 03 |Блок N IТЭ<Р<Р = Охлаждение активной зоны СОЗУ н*т да Нет HCT да шкал, реж. ДИАЛОГОВЫЙ РЕЖИМ ЗАВЕРШЕН:! TBC:. -I Р-Т запас ГЕ СЯОЗ J Темп.1к | Графики 12 3 4 5 6 Инструкция : Fig.2. CSF status tree ГЕ СЯОЗ Блок N I Тэ<рд> = THX2S22 TH13S22 TH13S23 ЧЧ TII12 S23 I НГЕ НГЕ THÜS22 1 Ргс РГЕ TH14S23 THJL2B01 TH13BQ1 TH14S22 TH11S23 H гс Нгс РГС Pre TH1JLBQ1 ТН14В01 VC00B01 + IX ^ / • THJL1S24 ч • • TH14S24 TH11S25 >Н6 TH12S27 f уГН13TH13SZ£ 7 TH13S26 TH14S25 • • П • -« X TH14S27 TH14S26 THJ-1S26 TH1JLS27 НГЕХ А НГЕ2 Ц НгсЗ НГЕ4 KP 03 то ик зт 0 25Q0 5000 7500 10080 Рн.ако 1 2 3 1 S 6 Р-Т запас Грэфики Fig.3. An example of level 3 format Поддержка инструкций С0АД. 3i/U2/\t994 23:55ЯЭ fl-0 Срабатывание аварийной оащиты или включение в работа систем безопасности Перейти к откатыз Аварийно реакторай Закрыть СРК по СРК от КУ месту I Произвести минимальные I ^u I дейетеня ТЦ при ftS-l | Перейти к "Отк&о отключение турбины по пару" 1 Выполнение I ИНСТРУКЦИИ Fig.4. Fragment of computerized procedure MCR Reactor operator workstation Safety engineer workstation Printer Шэд CD-ROM CD-ROM ,;,-|»«- •»r •• I — — тг»- « \ Z. \ ETHERNET SPDS network CD-ROM Uta CD-ROM DAT _l Long-term archive Main computer Engineer workstation /тол£ eW computer ETHERNET to Level 1 ETHERNET to local crisis center network Fig. 5 SPDS Level 2 architecture overview Analog and Radiation Additional digital paramètre parametrs inputs Analog and digital input Fig.6 SPDS level 1 architecture overview ЗА Development of Transmitter with Hybrid-IC for Post-accident Monitoring Instrumentation T. Ikeuchi and S. Watanabe" 1 Mitsubishi Heavy Industries. Ltd. 2 Mitsubishi Heavy Industries. Ltd. Vv'adasaki-cho. Hyogo-ku. Kobe 652. Japan Minatomirai. Nishi-ku. Yokohama 220-84. Japan ABSTRACT Basic design requirements of the РАМ instrumentation (ref. to After the TMI accident. Post-accident Monitoring (РАМ) Table 1) are specified in the following Japanese guideline: Instrumentation based on the U.S. Guideline (R.G.I.97) was JEAG4611 "Guide for Design of instrumentation & Control applied to Japanese PWRs. And we have back-fitted the РАМ Equipment with safety functions" Instrumentation to old plants step by step. Environment requirement under the accident condition is very Recently, new type transmitters arrive on the market. They severe and significant for the in-containment РАМ instrumenta have better accuracy, and stability than old type. However, they tion. The ordinary transmitters based on new technology cannot cannot be applied as the РАМ instrumentation, because new type be applied, and appropriate modifications and several qualification are insufficient in a qualification for the РАМ instrumentation and tests are needed for the appliance. a modification to endure in-comainment accident conditions. The in-comainmem РАМ Instrumentation in Japanese PWRs Hence, Japanese PWR utilities and Mitsubishi Heavy Industries are shown as follows: are developing a new type transmitter for РАМ instrumentation to • Neutron flux (source range) improve accuracy and stability in the period of 1994 through 1996. • Thot and Tcold (wide range) This paper describes nowadays results in this development of a • RCS pressure new РАМ transmitter. • Prcssurizer level • Steam generator level (narrow range) 1. INTRODUCTION • Steam generator level (wide range) • Containment temperature Transmitters are field equipment, and are generally installed in • Containment water level (narrow range) severer environments than the other instrumentation (ex. control • Containment water level (wide range) rack, control board). In ordinary industries, typical environmental requirements for installing a transmitter are temperature, pressure 3. ENVIRONMENT QUALIFICATION FOR РАМ and moisture. Therefore packaging material and airtightness are INSTRUMENTATION important for ordinary transmitters. In addition to ordinary environment requirements, in-contain- In-containment environment during and after accident is mem РАМ transmitters need radiation resistance to monitor the determined as a condition during design basis events (LOCA and post-accident plant condition. MSLB) and post accident aging. This environmental requirement Current movements of a transmitter are down sizing, intellec- is different plant by plant, because of in-containment volume, etc. tualizing and improving of accuracy and long term stability. Then in the qualification test for the РАМ instrumentation, the These movements are concerned with new detecting technology severest environmental condition in Japanese PWRs is applied. At and improvement of signal processing technology. However, in the qualification test for the РАМ instrumentation, we applied a general, transmitters based on new technology are gradually sequential test procedure based on IEEE-323. degraded by irradiation (alpha rays, gamma rays, neutron) and go to fail in radioactive field. Then new type transmitters for 4. DEVELOPMENT OF NEW TYPE TRANSMITTER FOR ordinary industries does not have enough endurance as the in- РАМ INSTRUMENTATION containment РАМ instrumentation. Hence, we are developing a new РАМ transmitter to grade up The development program of a new type РАМ transmitter is the operating РАМ transminer by performing a modification and shown as follows: qualification tests of a new type transmitter. (1) Study of detection technology (2) Study of structure technology 2. REQUIREMENTS FOR РАМ INSTRUMENTATION (3) Design and manufacturing of prototype samples (4) Execution of en-';'- -nment tests (radiation test, heat test) In Japanese PWRs, the РАМ instrumentation is required to (5) Manufacturing ^ .c- duct samples provide following information: (6) Execution of qu^.ncation tests program • minimum information for understanding plant status during and after accident Items (5) and (6) are planned to be executed in the period of • minimum information for the judgment of rapid manual 1995 - 1996. This paper describes the items (1) - (4) which have operation to assure safety been already performed. • main information of safety systems and components 1 Table 1 Importance classification of I&C system and design requirements I&C system will) safely functions redundancy independency environment seismic emergency testability record remark or diversity power supply I'.S-I ...... — — — .... ©,0:re<|tiiredn ©'«"" Q MS-1 Protection system © © © © © Л A:recommended System to directly control MS-1 system or component © © © © © © © Л X :nol required l'S-2 — —:not applicable system to supply waier to fuel pool X X О © © © © .... (grand directly prevent release of radioactivity о О X © — (*)eiiviroiiment control ©:aceident condition gjnoleJ MS-2 mitigate abnonnul condition © © © © © — O.anticipuled transient or notintil operation MS-2 system or sale shut down from out side of MCR X X о unlet © (*)seisinic component © О © ©.cln.stf Л or As Oitliis» С Information display system providing minimum required inlbnnution for understanding plunt status during and © © © © © © © © utter uccident Information displuy system providing minimum required information fur (lie judgment of rapid manual operation © © © © © © © lo assure safely (gyioteS Qtii»te5 Information display system providing main information © ntik> © © © А of MS-1 system und component © 0 PS-3 Syslcm tliut inuy cause unnormal condition X X о О X © © Л MS-3 System with sufety functions except MS-1,2 X X 0 О X © © А iiotel: necessary function during accident note2: required testing during operation nolo}: same requirement us system or component which (he I&C system control tiote4: functions reluted to sufe shut down from MCR note5: especittlly important information (maintain stib-crilieul, heul removal aller shut down, liCCS) 3 7 4.1 Study of detection technology (2) Results of studies (1) Targets to be studied 1) Electrostatic capacity type transmitter The following three detection methods have been studied, the It was confirmed that silicon oil used in the detection has no principle and features of which are shown as follows: problems in its heat resistance and radiation resistance, which 1) Electrostatic capacity type transmitter meant that i! has enough environment resistance. Fig. 1 shows the functional diagram of electrostatic capacity 2) Semiconductor type transmitter type transmitter. Its feature is that it has durability against It was determined that this type can hardly be used for the radiation peculiar to the nuclear power plant, since its detecting detecting section of the РАМ transmitter, because special product section is composed of metal plates. of high-temperature enforced type is now under research, and 2) Semiconductor type transmitter because ordinary products are ready to change their properties in Fig. 2 shows the functional diagram of semiconductor type high temperatures. transmitter. Its feature is that the characteristics, such as linearity, 3) Oscillatory type transmitter hysteresis, etc. are superior, since the repeatability and linearity of It was determined that it is difficult to use this type of the strain gauge of the semiconductor used for detecting section transmitter for the detecting section of the РАМ transmitter, are superior in its performance. because the microprocessor for special use of the signal processing 3) Oscillatory type transmitter section was destroyed in high radiation exposure. If we use this Fig. 3 shows the functional diagram of oscillatory type type, it must be necessary to develop a radiation-hard microproces transmitter. Its feature is that its resolution is high because the sor. detecting section gives an output signal by frequency, and that the oscillation element is composed of mono-crystal material, and is Judging from the above results, electrostatic capacity transmit precise and stable in operation. ter has been selected as the detecting section for the prototype. 4.2 Study of transmitter structure Variable electrode j By assuming that electrostatic capacity type, which is superior Fixed electrode in environment resistance, is selected for the detecting section, "remote signal processing" and "one body signal processing" were Sign.il Power compared with each other. process supply ing (1) Remote signal processing section By isolating the signal processing section from the detecting section in container vessel of plants, it has a merit of being free from influence by environmental conditions in accident. However, Signal processing it was learned difficult to apply this method, because it is affected Detecting section ( scciior by the floating capacity of the cable between the detecting section Transmitter and the signal processing section. (2) One body signal processing Eg. 1 Functional diagram of electrostatic capacity type transmitter The signal processing section must have an environment resistance. Here, a method may be applied to stand the environ mental conditions, by using the technology of environment resistance Hybrid-IC1*" as a means to improve the environment Pressure sensor resistance of signal processing section. Input 1 pressure *1 environment resistance Hybrid-IC (HlC) Ampli HIC whose radiation resistance and heat resistance have been fier improved by packaging electronic devices, such as ICs. etc. with radiation insulator. HIC operated normally after the following tests. 1) Radiation test: Total radiation dose 1 x 10s rad Detector section Amplifier section 2) Heat test: 155°C, for 5 hours continuously Transmission path 4.3 Design and manufacturing of prototype samples Fig. 2 Functional diagram of semiconductor type transmitter By judging from the study results of detection technologies and structure technologies, the following prototype samples have been manufactured: (1) Electrostatic capacity has been selected for detection. Diaphragm chip (2) One body signal processing section, in which environment resistance HIC is used for amplifier circuit, has been selected for Signal structure. process Power ing supply section Oscillation element Signal I processing Detector section section Transmitter Fig. 3 Functional diagram of oscillatory type transmitter 3 3 4.4 Envïronmenuests 2) The change of I/O characteristics before/after irradiation The radiation resistance test and then heat resistance . have was less than i^Vc of full span. been conducted with three units (No. 1 - 3) of prototype samples (differential pressure transmitter). What was done, and its results 4.4.2 Heat test are shown in the following paragraphs: This test was done after radiation exposure test for same samples. The samples were put into a tank of constant tempera 4.4-.] Radiation exposure test ture, and temperature was raised/lowered in the air. (Steam spray (1) Test conditions wasnotBsed.) J) Source of radiation: cobalt 60 (Co-60) у ray ll) Test conditions 7 2) Total radiation dose; 7-2 x 10 rad 190=C - 5 min. 152eC - 3 hrs. 135'C - 20 nr. and 123гС - 2 3) Dose rate: 1.5 x 106 rad/hr days were sequentially applied. (Fig. 6) 4) Time of radiation: 48 hrs (2) Test results 5) Measured data: Fig. 7 and 8 show the test results. Its summary is as follows: * 0% output error in irradiation (zero-point drift) 1 ) Max. error in test (zero-point drift) was 10.6% at most * I/O characteristics before/after irradiation However, the error was temporary, and the average in each (2) Test results temperature condition was 49c or Jess. (Table 2) Fig- 4 and 5 show the test results. Its summary is as follows: 2) In comparing I/O characteristics before and after test, that 1) Max. error in irradiation (zero-point drift) was 0.57% at after test tends to be smaller than that before lest. It is thought to most. be the result of effect by baking under high temperatures, in which the performances of electronic devices were recovered. (Table 3) 0.6 : : : : : : : : : : ....LLII.IL ^1ЬГ&а«.1.1.1Л.1.. 0.5 : : :::::: : : 0.4 : : :::::: : : — No.l 0.3 ....LLU-JI^^ri^ 0.2 1 j U|--;-H-I- лхш^.Л. uO&Ui - No.2 0.1 \ 4- .^^^RRXJ- I ! L.L1K1L • "* No.3 0 -0.1 -o.: 0.01 0.1 10 Total dose of radiation (xlO' rad) 36 —— t « « t • • * * : : : : : : |_ : : : \ \ : ! \ 34 32 RÏ> 1 30 \ - K^-; Р-Т-Р^ГГ 2S : : : / : : : : : : ПуГ J J J 26 24 : : •::::: I : :'::'::! : / : PIJL ': I : : : : : : î : : : : :::::: : / : : : J: : : : : ' : 20 : :::::: : : :::::: : 0.01 0.1 1 10 Total dose of radiation [x 10' rad) Fig. 4 Zero-point drift (0% output error), ала temperature of sample body during radiation exposure test 1.5r OI —= - - В -0.5b i -If~ -l.S> 0* 25» 50X 75* 100t 505, 100% Input pressure Input pressure [Before test] [After test] (Total dose of radiation 7.2 x 10' rad) raised lowered Fig. 5 Typical data of I/O characteristics before/after radiation exposure test 4 3 Ч Approx. 5mins 5 minsij' 'л 200 i- 1S0-- Fig. 6 Temperature sequence Ч-4..0 +3.0 •• - +;.0 i90°c o-,i5:°c|15 135°C МУС -5 min fhf" hr 20 hr - 2 unvs -1.0 T Fig. 7 Typical data of zero-point drift (0% output error) during heat test 1.5 .. — 1.0 0.5 nu n 1 0 E. -0.5Î- -1.0 -1.51- -1.5 0Я 25% 507c 75 100% 0% 25% 50% 75% 100? Input pressure Input pressure (Before test] [After test] (Total dose of radiation 7.2 x 107 rad) raised lowered Fig. 8 Typical data of I/O characteristics before/after heat test Table 2 Typical data of zero-point drift in heat test Table 3 Typical data of hysteresis error before/after test (max. value) Temperature 190eC 152°C 135°C 123°C Before test After test Amount of zero-point Error Under raising 1.2% 0.2% drift (max. value) pressure (0 -* 100%) +2.2% +3.7% +4.0% Average value +0.6% Under lowering 0.8% 0.2% Max. value +10.4% pressure (100 -+0%) 5 Le 4.4.3 Evaluation of test results 5- CONCLUSIONS AND FUTURE PLAN (1) Radiation exposure (est In the radiation exposure test results of 3 units of prototype In this paper we described nowadays results in the develop samples, it was confirmed that all the 3 units operate normally ment of the new type transmitter for the РАМ instrumentation1"'. under the max. total dose of 7.2 x 107 [rad] without damages. We have confirmed the excellent performance of highly radiation Though the error occurred due lo zero-point drift and I/O resistance and heat resistance for the prototype transmitter. characteristics, the amount of the dose was less than 1.5%. which Currently, we are manufacturing product samples of the new may well fall within the range applicable for РАМ transmitter. РАМ transmitter. And the qualification tests will be performed in Therefore, it is concluded that the target radiation resistance the period of late 1995 through 1996. for this study has been achieved as planned. We plan to apply this new РАМ transmitter to the operating (2) Heat test plants after the qualification tests. The amount of zero-point drift increased, though for a short time, under 150eC - 3 hrs, the I/O characteristics and amount of *2 This is a joint study - "Study for the Development of a New zero-point drift before/after test were small, and the transmitter Type Harsh Environment Resistive Transmitter" — performed by operated normally without damages. This gave a good result for a Japanese PWR utilities1'3' and Mitsubishi Heavy Industries. prototype. The subsequent examination showed that the max. drift error *3 Japanese PWR utilities; occurred in this test due to insufficient heat resistance of one of The Kansai Electric Power Co.. Inc.. amplifier circuit pans (not HIC). It is planned to apply improved Hokkaido Electric Power Co.. Inc., products in the future tests. Shikoku Electric Power Co.. Ltd., Kyushu Electric Power Co.. Ltd., The Japan Atomic Power Company. 6 It I RELATIVE DIFFERENT ENERGY NEUTRON RADIOMETRY IN REACTORS FOR PREVENTING OF ACCIDENTS CAUSED BY UNCONTROLLED REACTIVITY VARIATIONS S.V.Volkov SNUP-ASKRO, Moscow, Russian Federation The technology method based on measuring the different energy neutron flux from the coolant has been applied at nuclear reactors on NPP [1]. Further investigations have shown that the relative different energy neutron radiometry made it possible to consider and to solve new reactor safety problems during start-up, transient and extreme regimes. Almost fifteen years investigations were concentrated on methods & mstrumentation for preventing of accidents, arising from reactivity variations influenced by show neutron absorbers concentration changes. Physical essence of proposed approach consists in measuring the neutron spectral relative shift effects (fig.l) arising during reactivity changes influenced by slow neutron absorbers [2]. Being processed in pulse or current form, ratio signals for two neutron detectors, which have different spectral sensitivities (for example - cadmium ratio), are conversed to basic reactors characteristic functions. Each of these functions is controlled by two relatively switched pulse or current de tection units and detection assemblies (DU & DA), particularly - without any circuits for logarithm calculations. The DA measuring part structure is universe for any discussed method in-pile boris acid concentration measurements in the WWPR-440 & WWPR-1000 primary coolant have been carried out by the DA the DU of which [3] were installed in vertical positions into the ionization chambers channels in the biological shielding of the reactor during the first start-up and intermediate power operating regimes (fig3) and in similar positions in the tubes welded to the inner surface of the reactor core basket [4] during fuel reloading. The detection assemblies developed on the base of helium-3 proportional counters [1,6] were used for the first start-up experiments. The detection assemblies on the base of uranium fission chambers [6] were used for fuel reloading experiments. Summary data obtained in four runs, two for start-up and two for reloading, are plotted at fig.4. The relative neutron radiometry results highly correspond to chemical ones. Neutron WWPR power evaluation by relative radiometry method has been carried out during the first start-up (fig.5). Reactor power (in percentage of nominal value) has been scaled by readings of regular equipment for monitoring the neutron flux [5]. It appeared that the subcritical pile neutron power would be a linear function of cadmium ratio for chosen experimental conditions. Neutron power & its logarithm measurements, carried out at the V New-Voronezh NPP block up to 33 percents of nominal value has shown that being influenced only by boris acid concentration changes, the neutron power depends as logarithm on cadmium ratio (fig.6). Control rods moving disturbs this linear dependence. 1 Reactivity evaluation has been completed at the first Soum-Ukrainian NPP block during its first start-up. Reactivity-to-cadmium ratio dependence has appeared to be non-linear, almost hyperbolic (fig. 7). Period measurements during the first start-up have been carried out at the same power block. It has been found that period T was a linear function of cadmium ratio Red in the experimental error limits (fig. 8). Reactor period is a decreasing function of Red for three different positions of control rods. Using the obtained results, one would become capable to differ the reactivity effects caused by boris acid concentration changes and by rods moving. Boris acid concentration control results during fuel reloading allowed to discover the inversion of the cadmium ratio dependence on Boris acid concentration in comparison with the first start-up experiments. The result of the reactor critical condition determination must not depend on neutron spectra, since this is only reactor state when its material (bulk) parameter is equal to geometricalone. That is why two plotted dependencies intersection gives the critical value of neutron absorber concentration (fig. 9). Other results of relative neutron radiometry application have been obtained at research reactors. Reactor critical point determination during its first fueling has been carried out by inverse counts control and by plotting the corresponding curves (fig. 10). The inverse counts curves were plotted for thermal neutrons while loading nuclear fuel cassettes and for fast neutrons while loading the reflector ones. The latter curve shows "safe" and practically linear character, that therefore increases reliability of critical amount of cassettes detennination. Additional investigations allowed to propose a method for determination of thermal neutron radiometers sensitivity. It has appeared that existed the extremal dependence of cadmium ratio on pulse disciimination level for almost any radiometer with pulse detector based on boron-10, helium-3, uranium-235. The maximum of Rcd corresponds strictly to theradiometer maximum sensitivity and selectivity to thermal neutrons (fig. 11). The thinner is the neutron-sensitive detector radiator, the more evidently is shaped the maximum. This effect has made it possible to simplify the neutron radiometers calibration in neutron fields For the practical purposes of the relative neutron radiometry the start-up DU [3] with sensitivity of 10 pulses per second by unit thermal flux density has developed. The DU contents 5 corona counters [1,6], switched in parallel. The DU sensitivity is approximately equal to the initial value when fail occurs to one or two counters. This DU has 1200 times as much radiation resource in comparison to the DU, based on helium counters [1,3]. In addition to the start-up DA [3] with this DU, the wide-band pulse current DA has been developed for power and research reactors, capable to withstand post-accident conditions. The DA has been constructed using the industrial ionization chamber KNU-3[5]. This chamber contents two pulse fission chambers with the ratio of their sensitivities to the thermal neutrons equal 1:1000, current boron chamber and compensating current chamber, containing no neutron-sensitive radiator, in one case. The block-diagram of this DA is shown in fig. 12. The DA has been tested for a long time in the thermal column and in the vertical experimental channel (VEC) of the research reactor with the peak power equal to Mw. Figures 13-15 show basic characteristics of the DA. Two DA modifications have been investigated in which KNU-3 chambers had approximately 3,5 times different second pulse chamber sensitivity due to initial fission material dilution by uranium-238. It can be seen that more rigid neutron spectrum in VEC when compared to that in the thermal column, caused that the second fission chamber sensitivity (with uranium-238) became 2,6 times less than equivalent pulse sensitivity of the third, i.e. current volume of the KNU-3 (dependencies 2,2 and 3 in fig. 13). 2 A l Partial simulation of accident conditions for WWPR-1000 power operation has been conducted by three hours irradiation of the KNU-3 DU at lMw power (1, 88xlOn cm"2 s"1, 15,4xl0"2 A/kg (2,16x106 R/h)). After shut down the background counts in the first pulse channel of the wideband DA has been found to be 110 s"1 ; 14 s'1 three hours later and 3s"1 72 hours later. Using the lead screen 31 mm (approximately 12 inches) thick and compensating by the current of the fourth KNU-3 volume, practically full compensation of the reactor residual energy releasing current in the first two decades of the third control range (fig. 14), has become possible. Therefore it has become possible during the shut down to switch the channels over to the middle of the second pulse volume, inspire of increased background, thus proving the capability to withstand accident conditions. Figure 15 shows that such twin compensation is most effective in the intermediate range (0,1 - 103 W). As a result of investigations it is firmly established that startup and wide-band detection assemblies allowed to put into practice the methods of relative different energy neutron radiometry in reactors of different types in the range of neutron flux density changing no less than 14 decades (10"3 cm'V1 - 10u cm"2 s"1 ). REFERENCES 1. Volkov S.V., Zhemov V,S., Skatkin V.M. The detennination of process variables of the coolant and the reactor on the base of monitoring the neutron flux from the coolant. lAEA-NPPCI. Proceedings of Specialists Meeting on New Instrumentation of Water Cooled Reactors. Drezden, Zfk-568, October 1985, pp 112-124. 2. Филиппов E.M. Ядерная геофизика, т. II, Наука, Новосибирск, 1973,с.42. 3. Волков C.B., Михайлов Г.И., Прохоров Ю.Б. Экспериментальные исследования блоков и устройств детектирования аппаратуры контроля нейтронного потока АКНП. - В кн.: Вопросы атомной науки и техники. Сер. Ядерное приборостроение, Вып. I, М., ЦНИИ Фтоминформ, 1989, с.75-84. 4. Эксплуатация реакторных установок Ново-Воронежской АЭС. М., Атомиздат 1972, авт. Ф.Я.Овчинников, Л.М.Воронин, Л.И.Голубев и др., с.29- 30. 5. Pronjakin F.V., Borovik G.F., Baharev S.A et al. The complex of the equipment for monitoring the neutron flux and process variables of the control and protection system on water-moderated water cooled reactors seeref. 1, pp 149-159. 6. Каталог ионизационных камер для реакторной техники, М., ВНИИТФАД993. 3 kW о 2 ц s g ю iz щ ш 18 го Wßt % Рис. I. Зависимость изменения потока нейтронов в относи тельных единицах от содержания в пробах бора при регистрации тепловых (Т) надкадмяевых (Cet) нейтронов, а также нейтронов в области резонанса индия ( С/а ) и серебра (А $ ). Детектор быстрых нейтронов Регистратор Детектор тепловых нейтронов Аналпзатог. производно? отношения Регистратор Рис.З". Структурная схема рвдиомстрического канала МНИ л АКРБ повышенной эффективности для реактора ВВЭР, Рис.3. Устройство для непрерывного измерения концентрации поглощающего нейтроны вещества в теплоносителе ядерного реактора. КОРРЕЛЯЦИЯ ХИМИЧЕСКОГО МЕТОДА ИЗМЕРЕНИЯ КОНЦЕНТРАЦИИ БОРНОЙ КИСЛОТЫ В РЕАКТОРНОЙ ВОДЕ И МЕТОДА ОТНОСИТЕЛЬНОЙ РАДИОМЕТРИИ HB АЭС, 1(1980), ПЕРВЫЙ ПУСК ЮУ АЭС. Ï (1982), ПЕРВЫЙ ПУСК РО АЭС. 1 (1983), ПЕРЕГРУЗКА ЯД. ГОРЮЧЕГО РО АЭС. I (1983), ПЕРЕГРУЗКА ЯД. ГОРЮЧЕГО Ц8 g ? г 9 . -/о 42 *Ъ №. iS 16'-- &tà Рис. 5 Зависимость мощности (относительно показаний АКНП)подкритического реактора ВВЭР- -1000 ЮУАЭС от кадмиевого отношения блоков детектирования АПП со счетчиками CHM-I8. KW a 2.5 •y 2.0 / f /.5 1 10 05 / / о 2 ц б г /о /2 w /е « го РжссТ. Завжожмость логаржфма мощностж ВВЭР-IQOO У блока НВАЭО от кадмиевого отношения по урану-235 жощгзацжонных камер KHK-I5 (жзмервнжя 1980 г.), Рис.1 З&висгсмосттгъ реактибясгсшгс . -i i \ r ^— 6 7 8 9 Je R^J: Рис. 8 $а£цамостъ периода &BSP-iO0O fào/ca X Н)шно - yjcpauHCKv? 4ЭС or КааямееЬго отношения <5àc*v£ БЪАН- /7Р (âsxa/)pt> Ш Ca, г/кг ¥ше.3\. Инверсия завнсжмостж Бсс£ от 21а для новой активной зоны (I) ж облученной (при перегрузке) (2). Lfrotaf я/рреЭе/ге*/*//? /rpt/wt/vccrn/ ppax-/7rop#0t/ ус/пановги /?ри ее зсггрузха Рис..11 it , ni 41 / -— » A STOCHASTIC APPROACH TO ACCIDENT IDENTIFICATION IN NUCLEAR POWER PLANTS Kee-Choon Kwon, Soon-Ja Song, Won-Man Park, Jae-Chang Park, Chang-Shik Ham Korea Atomic Energy Research Institute P.O. Box 105, Yusong, Taejon, 305-600, Korea Tel. 82-42-868-2926 Fax. 82-42-868-8357 E-mail:[email protected] ABSTRACT Identification of the types of accidents and proper actions is required at an early stage of an accident in nuclear power plants. The accident of the plant can be identified by their symptom patterns related to the principal variables and operating status of major equipment. The patterns are identified by the Self-Organizing Feature Map (SOFM), unsupervised artificial neural network, for feature mapping algorithm and the Hidden Markov Model (HMM), a stochastic technique for solving the time series problem. The off-line data from a compact nuclear simulator are vector quantized by SOFM clustering algorithm. The HMM is created for each accident from a set of training data which are the result of vector quantization. The accident identification is decided by calculating which model has the highest probability for given test data. The system uses a left-to-right model including 6 states and 16 input variables to identify 7 types of accidents and the normal state. The HMM is trained by the maximum-likelihood estimation method which uses forward-backward algorithm and Baum-Welch re-estimation algorithm. The optimal path for each model at the given observation is found by Viterbi algorithm, and then the probability of optimal path is calculated. The simulation results show that the proposed system identifies the accident types correctly. It is also shown that the diagnosis is performed well for incomplete input observation caused by sensor fault or malfunction of certain equipment. I. INTRODUCTION The term diagnosis, as applied to an engineering system or process, means the determination of the cause which brought about an undesirable state or failure of the system. The diagnosis can be done at several different levels, e.g. component, subsystem, function or event.1 At the proposed accident identification system, diagnosis is made at event level to determine which accident has occurred. This system is intended to support operator's decision-making by interpreting major plant variables and operating status of the plant. The accident identification system has been developed using the techniques of rule-based system, artificial neural network, and fuzzy theory. But most of them are under test with test facility, not applied to operating Nuclear Power Plant (NPP).2,3,4 1 Accidents in a NPP are associated with unique patterns of major variables and equipment status; hence, diagnosis can be treated as pattern classification. A pattern is a quantitative description of objects, events, or phenomena. The classification may involve spatial and temporal patterns. Temporal patterns usually involve ordered sequences of data appearing in time. The goal of pattern classification is to assign a physical object, event, or phenomenon to one of the prespecified classes11. The same accident may occur under different conditions such as full power or half power, so it is possible to adopt a stochastic approach for classification of the patterns. Hidden Markov Model (HMM) is such a stochastic technique to solve the identification problems associated with time series, such as speech signal or plant process signal. II. DEVELOPMENT OF ACCIDENT IDENTIFICATION SYSTEM 1. An Outline of the System Identification of unknown patterns Л!" corresponds to finding optimal model W that maximizes the conditional probability P(W\ X) over the types of accident W. We can apply Bayes rule, ~ P(X\W)P(W) P(W\X)=D max—-—•— - The conditional probability P{X\W) comes from comparing shapes of the accident models with input observations while a priori probability P(W) comes from the accident model which represents how often the accident appears in NPP. Since P(JV) is independent of W, we get P(W\X) oc P'X]W)P(W) = max [P(X\W)P(W)] It is difficult to calculate a priori probability P(W) in NPP, so we can assume the probabilities of all accidents occurring are equal. In this paper, HMM is .used to estimate the conditional probability P(W\X). An HMM is trained for each accident from a set of training data. Incoming observations are recognized by calculating which model has the highest probability for producing that observation. The training and test data are provided off-line from a compact nuclear simulator. Major variables and equipment status are combined for input symptom vector in each accident when the simulator emulates an accident situation. The collected input symptoms are vector quantized for feature extraction, which means JV-dimensional measurement space transmitted to 1-dimensional feature space. Vector quantized code book is the input of the HMM identifier which is already trained. The final decision is the result of comparing with an HMM identifier to determine which model has the highest probability. 2 Accident 1 ; ATWS (Anticipated Transient Without Scram) Accident 2 ; LOC A (small break Loss Of Coolant Accident) Accident 3 ; SGTR (Steam Generator Tube Rupture) Accident 4 ; RCPT (Reactor Coolant Pump Trip) Accident 5 ; FWLB (FeedWaterLine Break inside containment) Accident 6 ; MSLB (Main Steam Line Break) Accident 7 ; PORV (Power Operated Relief Valve stuck open) The input symptom vectors are a collection of principal variables and status of major equipment from accident in the compact nuclear simulator. The following 16 major variables and equipment status are used to identify the seven different types of accidents and one normal state. (1) pressurizer pressure (2) pressurizer level (3) reactor coolant average temperature (4) reactor coolant flow (5) steam generator pressure (6) steam generator level (7) main steam flow (8) main steam pressure (9) reactor power (10) turbine power (11) main feedwater pump flow (12) main steam enthalpy (13) fuel temperature (14) status of reactor coolant pump (15) reactor trip signal (16) status of main steam isolation valve. 2. Self-Organizing Feature Map10'11 Feature extraction is an important task both for classification or recognition and is often necessary as a preprocessing stage of data. In this way data can be transformed from high-dimensional pattern space to low-dimensional feature space. Our goal is to identify a neural architecture that can learn feature mapping without supervision. The feature mapping algorithm is supposed to convert patterns of arbitrary dimensionality into the response of one-dimensional array of neurons.11 Suppose that an input pattern has N features and is represented by a vector x in an N-dimensional pattern space. The network maps the input patterns to an output space. The output space in this case is assumed to be one-dimensional array of output nodes, which possess a certain topological orderness. The question is how to train a network so that the ordered relationship can be preserved. Kohonen proposed to allow the output nodes to interact laterally, leading to the Self-Organizing Feature Map (SOFM), in other words Kohonen network. A simple configuration of the SOFM is illustrated in 3 slowly decreases with each iteration. A more detailed description of the training phase is provided below: 1) First, a winning neuron is selected as the one with the shortest Euclidean distance j| x - - min I x - w,\ ' between its weight vector and the input vector, where w< denotes the weight vector corresponding to the rth output neuron. 2) Let i* denote the index of the winner and let /* denote a set of indices corresponding to a defined neighborhood of winner i'. Then the weights associated with the winner and its neighboring neurons are updated by Aw; = T](x - wj) for all the indices ye/*, and 77 is a small positive learning rate. The amount of updating may be weighted according to a preassigned "neighborhood function", л(/,0. Aw,- = 77 Л(у',/*)(х - wj) for all j. For example, a neighborhood function A(J,i') may be chosen as A(j,i*)= exp(-|r,-nf/2cr2) where rj represents the position of the neuron j in the output space. The convergence of the feature map depends on a proper choice of 77. One plausible choice is that 77 = 1 / /. The size of neighborhood should decrease gradually. 3) The weight update should be immediately succeeded by the normalization of Wi. In the retrieving phase, all the output neurons calculate the Euclidean distance between the weights and the input vector, and the winning neuron is the one with the shortest distance.12 3. Application of Hidden Markov Model 3.1 Hidden Markov Model6'7'8 The basic theory of HMM was introduced in late 1960s and implemented for continuous speech recognition in mid 1970s. After this application, HMM has been successfully applied to real problems which can not be solved by conventional Markov models. HMM has advantages that provide proper solutions by modeling and learning by itself even if it does not have exact knowledge about problem solving. HMM is represented by a graph structure which consists of iV nodes called state and arcs that means a directional transition between nodes. In a graph, the observation symbol probability distribution which models spatial characteristics and initial state probability distribution stored in a node, and state transition probability distribution which models time characteristics stored in an arc. HMM states are not directly observable, and can be observed only through a sequence of observed symbols. 4 ь з То describe the HMM formally, the following model notation for an HMM can be used. N : the number of states in the model L : the number of distinct observation symbols per transition, denote the set of individual symbols V = {vl,v1,—3vL} T: the lengths of the observation sequence, Ол,02,—,0Т S = { St} : a set of states, st s{l,2,--,^}, t = 1,2,—, Г, state / at time / may be denoted by st = i A = {au} : the state transition probability distribution, i,j = 1, 2,.... N, cr5 = P(stMÏ = j\st = 7") : the transition probability from state / to state j, the а parameter should satisfy the stochastic constraint ^ ч ~ I В = {bj(k)} : the observation symbol probability distribution, k = l,2,-,L, Ь,{к) = Р(ук\5м=у) : observation probability of kth symbol vk in transition atj, the parameter should satisfy the stochastic constraint ^bj(k)= 1 к 7Г — {7Tj} : the initial state distribution, where яг, = Pr(s; =/), / = 1,2,-~,N satisfying ^ л-,. =1 i S j : a set of initial states, SF : a set of final states, Nj : the number of initial states, NF : the number of final states. An HMM can be represented by the compact notation Л = (А,В,я). Specification of an HMM involves the choice of the number of states, N, the number of discrete symbols, L, and the specification of three probability densities with matrix form, A,B and яг. 3.2 Training and Identification Training means that the characteristics of input patterns to be modeled by the parameter of Я = (А,В,тг). An HMM is applied to an identification problem under the assumption that we can precisely determine the model parameters for given observations. But it is difficult that this assumption is exactly realized because of the complexity of problem and having local optimal not global optimal. At present, we are satisfied to find local optimal in parameter optimization methods. In this paper, we use the maximum likelihood estimation for training. This method maximizes the following equation given input observations O. 5 6k S l-\ It calculates the probability of given observation symbols at all paths from initial to final state. The model parameter which maximizes the above equation is efficiently computed by forward-backward algorithm and Baum-Welch re-estimation algorithm. The forward variable at (/) can be defined : at(i) = P(.0lt02,-0tyst=i\Ä) This is actually the probability of the partial observation sequence to time t and state i which is reached at time t for a given Л. We can calculate at(i) by forward algorithm as follows:7 Step 1 : аг (/') =я"Д (Ol), for all states /" (if / eS, 7r,= — ; otherwise к~ 0 ) Ni Step 2: Calculating a() along the time axis, for t=2,---,T, and all states j, compute: i Step 3 : Final probability is given by: P(0\X) = £ ^(O is St The backward variable ßt (г) which is used to optimize the model parameter with forward variable, can be defined as : ß,(i) = P{Ot_„Ot+2,~<0T\st=i,X) i.e. the probability of the partial observation sequence from t+l to the final observation T, given state /' at time r and the model Я and /?,(/) can be calculated by backward algorithm as follows:7 Step 1: ßr(i) = — , for all states / ZSF, otherwise ßr(J) = 0 NF Step 2: Calculating ßQ along the time axis, for t = T-1, T-2,-• • ,1 and all states j, compute: г Step 3: Final probability is given by: p(0\Д) = n b >{01) ß -.(i) t<=Sl The most difficult problem in HMM is how to adjust the model parameters (A,B,7t) to maximize the probability of the observation sequence given a model. The iterative algorithm used in HMM-based recognition is known as the Baum-Welch algorithm. The a posterior probability of transitions, yt(j,j), will be defined as the probability of a path being in state /' at time t and making a transition to state j at time /т 1, given the observation sequence and the particular model. It can be computed as : 6 6Ь Р(0\Я) 5>г<*) Similarly, a posterior probability of being in state /' at time /, у, (/), given the observation sequence and model is IceS, At this point, a,j , bj , 7it of re-estimated new model Л can be computed as: T-1 r-i l^Y&j) 2>r(U) » r-i r-i YLr&ji 2>,(o 2>,0) f=l ^ =ri(0 Thus, if Л is iteratively to replace Л and repeat the above re-estimation calculation, it can be guaranteed that Р(0\Л) can be improved until some limiting point is reached. Identification or recognition means to find the best path in each model and select the one which maximizes the path probability for a given input observation. There are several possible ways to find the optimal state sequence associated with the given observation sequence. One possible optimality criterion is to choose the states, st, which are in the best path with the highest probability, i.e. with maximum P(0,S\l). A formal technique for finding this single best state sequence is called the Viterbi algorithm, which works as follows:7'8 Step 1 : Initialization. For all states i, yriii) = 0 7 CG Step 2: Recursion. From time t=2 to T, for all states j, i у/ = arg max[S t - i(i)a,j] i Step 3: Termination. (* indicate the optimized results) P* = Mzx[Jr(5)] s'r = arg maxfj T(S)] seSr Step 4: Path (state sequence) backtracking. From time T-l to 1 S*t = у/ t + l(S*r + l) 4. Design of a Prototype Accident Identification System At first, we collect 10 training and test data per accident from the compact nuclear simulator. The collected training and test data are vector quantized by the SOFM clustering algorithm as shown in are 0.5, to 1 way like last state, the initial value of atj is 1.0. By assuming that the observation symbol probability is equivalent to each state, observation symbol probability £.(vt) =0.0119 when satisfy the equation L> «r=i Initialized HMM is illustrated in 8 6 7 The probability, P(P\ Я), is calculated by the optimal path which is obtained by Viterbi algorithm for given input observations in each model. We identify the accident by examining which model has the highest probability for given input observations. 5. The Result of Experiment The experiments were carried out offline in HP715/33 workstation the programming done with "C" language. Ш. CONCLUSION We proposed a prototype accident identification system based on the stochastic modeling approach of HMM. We can identify accidents by recognizing the patterns of accidents, which is a new attempt to expand the application area of HMM to accident identification. After a proper training using the train vector, the prototype system exactly identifies the accidents from input observations. At present, this prototype system is implemented off-line, but it should be implemented on-line to accept continuous time series data in real time. There are a lot of problems which should be solved before its actual application to an operating plant. However, this system have advantages such as an easy expansion of accident types and observation symbol sequences, and a relatively short time for training. Further effort is being made to improve the system. ACKNOWLEDGMENTS We acknowledge the financial support of the Ministry of Science and Technology (MOST) for this work. REFERENCES 1.1. S. Kim, "Computerized Systems for On-line Management of Failures: A State-of- the-Art Discussion of Alarm Systems and Diagnostic Systems Applied in the Nuclear Industry," Reliability Engineering and System Safety, Vol. 44, pp. 279- 295, 1994. 2. S. W. Cheon, A Study on the Application of Expert Systems and Neural Networks for the Development of Operator Support Systems in Nuclear Power Plants, Ph. D Thesis, Korea Advance Institute of Science and Technology, 1993. 3. A. Ikonomopoulos, et al., "A Hybrid Neural Network-Fuzzy Logic Approach to Nuclear Power Plant Accident Identification," A191 Frontiers Innovative Computing for the Nuclear Industry, Jackson, Wyoming, pp. 217-226, Sep. 15- 18, 1991. 9 4. S.W. Cheon, S.H. Chang, "Application of Neural Networks to a Connectionist Expert System for Accident Identification in Nuclear Power Plants," Nuclear Technology, Vol. 102, pp. 177-191, May 1993. 5. K.C. Kwon, et al., "An Application of Hidden Markov Model to Transient Identification in Nuclear Power Plants," ANS Embedded Topical Meeting on Computer-Based Human Support Systems, pp. 275-280, Philadelphia, June 25-29, 1995. 6. J. Y. Ha, Unconstrained Handwritten Word Recognition with Interconnected Hidden Markov Models, Ph.D Thesis, KAIST Department of Computer Science, 1994. 7. X.D. Huang, Y. Ariki & M.A. Jack, Hidden Markov Models for Speech Recognition, Edinburgh University Press, 1990. 8. L.R. Rabiner, "A Tutorial on Hidden Markov Models and Selected Application in Speech Recognition," Proceedings of the IEEE, Vol. 77, No. 2, pp. 257-285, 1989. 9. P. Smith, "Hidden Markov Models for Fault Detection in Dynamic Systems," Pattern Recognition, Vol. 27, No. 1, pp. 149-164, 1994. 10. T. Starner, Visual Recognition of American Sign Language Using Hidden Markov Models, MS Thesis, MIT, Feb. 1995. ll.J. Zurada, Intriduction to Artificial Neural Systems, West Info Access, 1992. 12. Y. Kung, Digital Neural Networks, Prentice Hall, 1993. 13. J. Freeman, D. Skapura, Neural Networks Algorithms, Applications, and Programming Techniques, Addison-Wesley, 1991. 10 HMM Vector Identifier Quantizer vector Input quantized Normal Model 1 Vectors observations Trained Accident Model 2 Accident Description clusteringSOF M identification* SOFM HMM parameter Accident Model 7 Training rs-€stimatfon training Aif clusters Accident Model 8 training Output Neurons input 11 30- 1 5 3 13 17 21 25 29 33 37 41 45 43 53 57 61 65 69 73 77 81 0.333 0.333 0.333 0.333 0.5 l.O 0.334 0.334 0.334 0.334 Likelihood Likelihood Accident Model Probability Accident Model Probability Normal 0.0 Normal 0.0 ATWS 9.506e-30 ATWS 0.0 LOCA 3.547e-35 LOCA 6.699e-43 SGTR 2.232e-39 SGTR 2.153e-35 RCPT 1.398e-42 RCPT 2.675e-38 FWLB 1.447e-43 FWLB 2.209e-39 MSLB 8.668e-43 MSLB 2.131e-41 PORV 3.283e-44 PORV 5.993e-43 12 IAEA Specialists' Meeting on "Instrumentation and Equipment for Monitoring and Controlling NPP Post-Accident Situations" Dimitrovgrad, Russia, 12-15 September 1995. Accident Monitoring in Ventilation Stack V.Kapisovsky, V.Zbiejczukova Nuclear Power Plants Research Institute (VÙJE) Trnava, Slovak Republic F.Gâbris, G.Belan, J.Zeman, J.Bukovjan, LRehâk Slovak Metrological Institute Bratislava, Slovak Republic 1 Introduction A ventilation stack of WER 440 NPPs is serving to both of two reactor units constructed as twins. The air flow through the stack is approximately 162 mV1 (5.109 nxVyr) affected by the reactors operation mode (nominal power or refueling). The hermetic zone (enclosure of the primary system components) at an accident is not vented directly to the stack due to automatic changes in the ventilation system airflows. Hence no uncontrolled radioactive matter should appear in the stack. Nevertheless some scenarios have been considered when this may happen (e.g. fuel damage at transport operations, leakages to vented areas at accidents). Standard gaseous effluent monitoring system RKS 2-03 provides the monitoring of the noble gases up to 2,2.109Bqm"3. For the monitoring at accident conditions this may be inadequate. A typical value for a range of monitoring the radioactive matter in the air of common ventilation stack is 5.104 - 5.I013 Bqiu3 [1]. The purpose of the monitoring is to detect and evaluate significant releases and to enable the long-term monitoring after an accident. Another typical values suggested for monitoring noble gases in a ventilation stack [2] are: up to 1012Bqm"3 at air flow of 200000 nrV1 up to 1013 Bqm"3 at air flow of 20000 mV To fulfil the requirement for accident monitoring in the ventilation stack at Bohunice NPP a type of detector used in the environmental monitoring system has been considered for ease of installation and connection to accident monitoring network. The DC-4D-84 (Si) detector is available in two versions - low range and high range. Only the later one is considered in the paper. Calculations The basic parameters of the DC-4D-84/V detector provided by the manufacturer are: Range of measurement for gamma rays: 0 to 1000 R/h 2 J > Probe basic sensitivity: 10 imp.s'VR.h*1 for 137Cs Basic count-rate error of output pulses: ±10% for the range 0 to 500 R/h -20% to +5% for the range 500 to 1000 R/h Detector own background: max 0.5 imp.s"1 Since the manufacturer provided the range and sensitivity in units of exposure, these have been converted to dose in air [3]. Then the response of the instrument can be expressed in terms of dose rate dividing an instrument response [imp.s*1] by mstrument sensmvity expressed in pmp.s'VGy.s"1]. Further if the response of the monitor is required in volume activity [Bq.m"3] the conversion factor К [Gy.s^/Bq.m"3] is needed. In this work the conversion factor has been calculated using code AMOC-K a version for geometry of WER ventilation stack. The computing method is based on solving Boltzmann photon transport equation by adjoint Monte Carlo calculation [4,5]. As a result of the calculation the photon faiences and the rate of kerma in air which approximates the dose rate in air, are obtained. As an example of the parametric calculation performed Fig. 1 shows the dependence of dose rate on height of the measuring position along the stack axis. The dose rate versus radial position of measuring point at the height of 15 m is presented at Fig.2. In both cases the source assumed is 137Cs homogeneously distributed in the air. Experimental The angular and energy characteristics of the detectors have been determined with detectors inserted into a Pb shielding as used in the environmental monitoring system. All measurements have been performed in the laboratory conditions. The built-in check source was dismantled except in kerma rate measurements. The angular characteristics have been measured at the energy of 118 keV and 662 keV (u7Cs). A massive turn-table has been used enabling detector positioning parallel to the photon flux at 0 7 k and 180° and with detector head in the mrning center. An example of the angular detector response is presented in Fig.3. The dependence of detector response on rate of photon kerma in air was measured from, the level of laboratory background to 4,56 Gy.h'1. An example of measurements for two detectors are plotted in Fig.4 as relative response error vs. dose rate. The measurements of the detectors energy characteristics have been performed in the gamma irradiation laboratory and in the X-ray laboratory. Fig.5 presents an example of measured characteristics. Discussion From the calculations performed as illustrated by Fig. 1 and Fig.2 it can be concluded that the response of the monitor is not very sensitive with respect to its position within the ventilation stack. Hence there is not any serious reason to prefer any particular position disregarding potential possibility of surface contamination of both the stack inner wall and the probe mounting construction. An estimated operational range of detector considered is 2,4.107 to 4,8.1011 Bq.m"3 137Cs or 137Cs equivalent. Taking into account the air flow through the stack (approximately 570000 пЛг1) the upper value of the range is corresponding to the values suggested in [2]. The results of the measurements of the angular characteristics well corresponds to the expectations. For the energy of 662 keV the detector response is independent to the direction of photon flux up to the 70-80°. The effect of shielding can be observed at higher angles. The angular dependence is more pronounced at lower energies .At 118 keV the response does not changes only to 50°. Systematic underestimation of the response with dose rate can be observed with average value about 10%. The influence of check-source is dorninant at small dose rates (not plotted in Fig.4). The energy characteristic is difficult to interpret. The data plotted at Fig.5 are normalized to the response of 137Cs. After sharp rise at approximately 100 keV the response is overestimated. The influence of photons scattered from the wall, floor and ceiling is considered to be negligible due to precise flux coUimation. The estimated contribution does not exceeds 10%. 4 7 5 Conclusions The experimental measurements have been performed with greate care with several detectors, all with shielding designed for application in environmental monitoring system. The shielding affects mainly angular dependence of the response which in the calculation was considered isotropic. From the measurements can be derived that the data provided by the manufacturer should be accepted with a caution. The probe performance, although not ideal, enables its application in accident monitoring. However its recalibration would be strongly recommended. Based on experimental angular and energy characteristics the response of the detector can be calculated and conversion factor derived to express the response in volume activity of most probable nuclides composition in releases. References [ 1] mstrumentation for hght-water-cooled nuclear power plants to asses plant and environs conditions during and following an accident,Regulatory Guide 1.97, Rev.3, 1983 [2] Radiation monitoring equipment for accident and post- accident conditions in nuclear power plants. Part 2: Equipment for continuously monitoring radioactive noble gases in gaseous effluents . ГЕС Publ. 951-2 (1988) [3] CSN 01 13 08, 1986 [4] Irving D.C. NucLEng. Des. 15,273, 1971 [5] Koblinger L., KFKI-76-57, Budapest, 1976 5 D [ Gy/s ] ( E-15 ) / / / ...I I I I. - I 10 20 30 40 50 60 height [ m ] Fig. 1 DOSE RATE AT DIFFERENT HEIGHT OF MEASURING POSITION D [ Gy/s ] ( E-15 ) M' 0 1 2 3 distance from the wall [ m ] Fig. 2 DOSE RATE AT DIFFERENT RADIAL MEASURING POSITION response 1.2 О 50 100 150 200 angle I degree ] Fig. 3 ANGULAR DEPENDENCE OF THE DETECTOR DC-40-84/V relative response error [ % ] 20 r" - О doserate 1 mGy/h J ---880102 1-910063 Fig. 4 RELATIVE RESPONSE ERROR OF THE DETECTORS AT DIFFERENT DOSERATES 1.6 1.4 10 100 1000 energy [ keV J Fig.5 ENERGY DEPENDENCE OF THE DETECTOR <3 v EXPERIENCE Wm№UTRON FLUX MONITORING SYSTEMS QUALIFIED FOR POST-ACCIDENT MONITORING H. Gordon Shugars, Pi., and James F. Miller, P.E. GAMMA-METRICS 5788 Pacific Center Boulevard San Diego, California 92121 USA Abstract Category 1 systems require the full environmental and seismic qualification required for reactor safety systems. These In this paper we discuss the environmental requirements for ex- requirements are invoked by other NRC Regulatory Guides and core neutron flux monitors that are qualified for use during and documents (R.G. 1.89, R.G. 1.100, NUREG-0588), which afterpostulated accidents in Pressurized Water Reactors (PWRs). endorse industrial standards such as ШЕЕ 323-1974 [2] We emphasize PWRs designed in the United States, which ere (Environmental Qualification), IEEE 344-1975 [3] (Seismic similar to those used also in parts of Western Europe and Eastern Qualification), tad пипк-rous supporting standards addressing Asia. We then discuss design features of the flux monitoring independence and isolation, criteria for type testing, application systems necessary to address the environmental, functional, and of single-failure criterion, criteria for periodic testing, general regulatory requirements, and the experience with these systems. principles for reliability analysis, etc. I. INTRODUCTION Moreover, Category 1 systems also require that the system be qualified, manufactured, installed, and maintained under a quality In 1979 the Three Mile Island Unit 2 reactor experienced a Loss assurance program, specified by the legal requirements of the of Coolant Accident (LOCA). For several days after the accident, U.S. Code of Federal Regulations (specifically 10 CFR 50 experts tried to assess the damage to the core and to predict if the Appendix B) and by various NRC Regulatory Guides that endorse core would once again become critical. Their efforts demonstrated industrial standards such as ANSI/ANS N45.2 and ASME the importance of the information they obtained from the neutron NQA-1. flux monitoring channels, some of which failed after the accident. Ex-core neutron flux detectors in U.S. PWRs are typically In December 1980 the U.S. Nuclear Regulatory Commission installed in vertical wells that are exposed to the temperature, (NRC) published Revision 2 to Regulatory Guide (R.G.) 1.97, pressure, and humidity environment inside the containment Most which addresses instrumentation for post-accident monitoring. of these detectors are also exposed to chemicals that would be This revision included the requirement that neutron flux sprayed inside containment after a Loss of Coolant Accident monitoring systems be qualified for the accident and post-accident (LOCA) or Main Steam Line Break (MSLB), although some well environment and that they cover the range fromful l power to 10"6 designs protect the detectors from this spray. Some detectors are percent power. As discussed below, the guide also specified the installed inside wells in tanks of water (Primary Shield Tanks), systems to be "Category 1," which requires full environmental but their associated cables may still be exposed to the post- and seismic qualification, redundancy, continuous real-time accident environment. Thus the most common designs of the display, and on-site standby power. Quality Assurance detectors and the other in-containment components (e.g., cables, requirements were to be the same as for safety systems. penetrations, connectors, etc.) must withstand and operate during the harsh, accident and post-accident environment In May 1983 the NRC issued Revision 3 of the regulatoiy guide [1], which modified some of the earlier requirements for radiation This environment depends on the specific plant design. However, monitors and meterological measurements. Requirements for a limiting, composite environment that envelopes the most severe neutron flux monitoring were not changed. conditions of typical PWRs is included for reference in Appendix A of Reference 2. A typical LOCA/MSLB qualification test П. REQUIREMENTS profile, based on the reference, is shown below. A. Environmental Requirements The detectors must also operate through and after a Safe Shutdown Earthquake (SSE), the motion induced at the detector's Reference 1 classifies neutron flux as a "Type B" variable, one location by the design basis event. The magnitude and frequency that provides information to indicate whether plant safety systems of this motion depends both upon the specific plant design and its are performing their safety functions. One of these functions is location. reactivity control. Neutron flux monitoring is essential for this function, and carries the classification "Category 1." Other Again however, a seismic response that envelopes the most functions are cooling the core, maintaining reactor coolant system severe earthquake has been developed, and a typical seismic integrity, and maintaining containment integrity, including qualification test profile based on this response is shown below. radioactive effluent control. PWR LOCA/MSLB Test Profile 70 _ i г / \ , Pressure Profile 500 о / Te Tipera ure О 3file 400 3 ) *•* Щ 300 ев / ф \ О. к - 200 E а fChe m cal Spray per IEE E 323-1 £37 4 f- 1300 0 jэр т Boron 10.00C3 ppm Sodium Sufitl e 100 Add Sodium Hydroxide to make pH 10.5 1Flow 0.5 gal/sq ft on Horizontal Surface S 0 Start of Chemical Spray Tîme (Seconds) Seismic Profile - SSE 100 1 3 6 10 30 60 100 Frequency (Hz) Tested Profile, Horizontal and Vertical Directions, 1% Damping Less well defined is the neutron flux and gamma radiation • Design features that increase flexibility in choosing a following an accident In general, the thermal neutron flux in the location to install the new cables and electronics. For ex-core wells at full power is between 3 x 10' to 1 x Ю10 n/cm2- example, a system that permits the first stage of electronics sec. The actual value depends on the core and fuel design, the to be as far as 300 m (of cable length) from the detector burnup, and to some extent the physical design of the well. simplifies the plant-specific engineering needed to install the new system. Gamma radiation at full power is approximately 10s R/hr. Gamma radiation after an accident can be assumed to be higher, and at • Design features that minimize the risk of damage during least one system design assumes 106 R/hr. Measuring neutron flux handling, especially given the concern about speed when accurately in gamma radiation this high poses problems for the installing the equipment system designer, as discussed below. • Design features to rninimize the time required to connect B. Functional Requirements cables inside containment, yet allow for secure, reliable connections The neutron flux monitoring system must cover at least a flux range corresponding to an upper level of mil power and a lower E. Reliability and Maintainability Issues level of 10"6 percent power. Systems typically provide an output proportional to the logarithm of the neutron flux. In general, Closely related to plant integration are issues of reliability and accuracy, drift and response times are not denned specifically for maintainability. These include design features to minimize the post-accident monitoring; however, when the post-accident moni number and frequencyo f adjustments, to rnmimize drift (closely toring channel replaces the original channel, the functional design associated with inmimizing adjustments), and to nrniimize the of the original channel apply. Typical accuracy requirements for amount of time needed to perform those tests and calibrations that original Intermediate Range instrumentation (channels with range are necessary. Such features include self monitoring, built-in test comparable to that required by Reference 1 ) was ±5 percent equi and calibration circuits, and careful component selection. valent full scale linear. Advances in technology incorporated into modem systems have improved that accuracy to as great as ±1 Component selection not only can improve reliability and drift percent equivalent full scale linear. performance, it also can extend the life of the system by avoiding components that are likely to become obsolete. Many of the Response times vary, depending on the application of the channel plants' original fluxmonitorin g channels contain components for in reactor safety systems. In general, response times at low flux which spare parts are either not readily available or which have levels are long, to allow sufficient time for "meaningful" readings unacceptable delivery times. (i.e., slow enough for the operator to read but fast enough to follow expected changes in flux). At higher flux levels the Ш. SOLUTIONS response times are short, compatible with requirements of protec tion functions that may depend on signals fromth e channels. The basic approach taken to date has been (1) to discriminate strongly against signals fromgamm a radiation; (2) to seal the in- C. Requirements of Other Industrial Standards containment equipment against the harsh environment; and (3) to place electronics outside containment. Original plant equipment Other industrial standards not written specifically for Nuclear often could not meet the third requirement, but advances in Power Generating Stations are now frequently applied to modern technology have allowed some system designs to place the first neutron flux monitoring systems. These include EEC 801-3 for stage of amplification 300 m (of cable length) from the detector, immunity to electromagnetic radiation and DEC 801-2 for easily meeting this requirement. immunity to electrostatic discharge. Others are sometimes applied to establish predicted reliability. Systems use uranium-coated fission chambers to provide the following advantages: D. Integration into Plant Designs and Operations • The fission process provides high discrimination against In most U.S. PWRs, the original design of the Intermediate Range signals caused by gamma fromthos e caused by neutrons. or Safety Channel neutron fluxmonitor s was not and could not be qualified for the post-accident environment, for reasons discussed • The integration of the above detector property with the below. Most plants had been designed, and many were in design of the electronics permits some channels to cover the operation before post-accident qualified systems were required. entire operating range of neutron flux with only one detector Therefore the new systems had to be integrated into existing plant assembly. Using the counting and Campbelling (mean square designs and operating practices. voltage) modes of detector operation avoids having to use two monitoring channels (one for each of two detector types Considerations for installing new channels into an operating plant commonly found in earlier operational channels), which was include: the traditional practice and a result of limitations of the tech nology at that time, as discussed in Referenced. The • Design features to rninirnize the time that construction combined range exceeds the requirements of Reference 1 and workers must handle the equipment inside containment, to reduces overall costs of equipment replacement. reduce radiation exposure to workers • Counting and Campbelling, both A.C. phenomena, are years), and as in-core detectors in Boiling Water Reactors exploited to avoid the D.C. signal changes caused by the (BWRs) and some P WRs. effects on cables (e.g., decreased insulation resistance) of high temperature and gamma flux. There is uncertainty in Physical features included to meet the post-accident requirements accounting for these changes in D.C. signals, especially at have allowed ihe channels to survive mistakes during plant main low levels of neutron flux. (Counting and Campbelling are tenance and refueling, mistakes that have damaged detectors and discussed extensively in References 5 through 9.) cabling of earlier, non-qualified designs. These problems include water in detector wells and extremes of temperature following • The high inherent discrimination in the fission chamber refueling and maintenance. eliminates the need for compensating voltages (which are required for boron-lined compensated ion chambers). There A result of the detector and physical design features discussed is uncertainty on how to adjust the compensation for boron- above is that the post-accident channels have proved to be more lined chambers to account for the effect of both normal and reliable than the original channels. Since 1986, almost every post-accident gamma flux, and the compensating voltage PWR that has installed post-accident flux channels has used these adjustment adds to the complexity of the system's design, channels to replace the original equipment. This results in fewer maintenance, and operation. channels to be maintained, more reliable equipment, and constant operator involvement with the new channels. Other benefits Minimal degradation of sensitivity with use of the detector. include reduced maintenance costs, elimination of noise on Depending on the design of the detector, sensitivity changes Source Range signals (especially during refueling outages, where are minimal over forty years of normal plant operation. Also noise in the Source Range can halt fuel movement, often a with proper design, saturation characteristics change critical-path activity), and readily available replacement parts. minimally for a specified range of high voltage, and detector's gas properties are chosen so that they also change Although there may be concern in general about using post- rninimally during operation. These latter two characteristics accident monitoring instrumentation as the normal preclude the need for high voltage adjustments (e.g., the instrumentation, for neutron flux monitoring the post-accident "plateau curves" common with Source Range proportional instrumentation can meet or exceed the original performance counters). Instead, the value of the high voltage can be requirements and historical reliability. Moreover, there are monitored to ensure it stays above a level that might affect advantages to using the channels for both purposes: Reference 1 system accuracy. states in part • When the post-accident system also replaces the Source Normal power plant instrumentation remaining Range channel, these detector characteristics also preclude functional for all accident conditions can provide the need to replace the Source Range every three to five indication, records, and (with certain types of years (or more frequently in some cases). instruments) time-history responses for many variables important to following the course of the accident. In summary, with one replacement channel, the performance of Therefore, it is prudent to select the required accident- the original Source Range channel (typically using a boron-lined monitoring instrumentation fromth e normal power plant or boron-gas proportional counter) and the Intermediate Range instrumentation to enable operators to use, during channel (typically using boron-lined CICs) can be matched or accident situations, instruments with which they are exceeded. The detector properties allow the balance of the design most familiar. Since some accidents could impose to be less complex, which helps to improve reliability and reduce severe operating requirements on instrumentation costs. components, it may be necessary to upgrade those normal power plant instrumentation components to However, the choice of fission chambers as the sensing element, withstand the more severe operating conditions and to combined with keeping the electronics outside containment, pre measure greater variations of monitored variables that sents other challenges to the designer. Cabling and penetrations may be associated with an accident. It is essential that must be designed and proven to prevent electromagnetic fields instrumentation so upgraded does not degrade the from corrupting the small signals that come from the detector. accuracy and sensitivity required for normal operation. And the detector itself must show reliable and consistent In some cases, this will necessitate use of overlapping performance for all modes of operation employed (as a minimum, ranges of instruments to monitor the required range of counting and Campbelling but at times also including linear the variable to be monitored, possible with different current operation). performance requirements in each range. IV. EXPERIENCE Experience has shown that separate post-accident monitoring instrumentation is sometimes not adjusted as frequently as the Fission chambers have extensive operating experience, as normal power plant instrumentation (to maintain the correspon discussed in Reference 4. One U.S. PWR manufacturer originally dence of neutron flux to indicated reactor power), nor are employed fission chambers for the Intermediate Range and, in operators as well aware of the significance of the indications of their later plants, also for the Power Range. Fission chambers had the supplementary instruments as they are of the operational also been employed in the U.S. in high temperature gas-cooled channels. When the post-accident channels replace and become reactor, research reactors (where some have operated for 30 the operational channels, these issues disappear. In summary, experience has shown: Generating Stations," Institute of Electrical and Electronics Engineers (Piscataway, New Jersey) • Systems using fission chambers have been designed and qualified by testing to meet the requirements for post- [4] Miller, James F., and Shugars, H. Gordon, "Applying Ex-Core accident monitoring Fission Chamber Technology to Reactor Protection Systems in European Pressurized Water Reactors: A Comparative Analysis " • The performance of these systems can meet or exceed the Presented at the 1992 ШЕЕ Nuclear Science Symposium, Sessions performance of the original Intermediate Range channels on Nuclear Power Systems, October 27-30,1992, Published in the Conference Record. • With careful design, the performance of these systems can also meet or exceed the performance of the original Source [5] Gwinn, David A., and Trenholme, William M., "A Log N and Range channels Period Amplifier Utilizing Statistical Fluctuation Signals from a Neutron Detector," Vol. NS-10, ШЕЕ Transactions on Nuclear • The ruggedness of the systems, intended for the post-accident Science. April. 1963. environment, also benefits the owners during unusual events during or after maintenance. [6] Popper, Glenn F., and Harrer, Joseph M., "The Performance of a Counting-Mean-Square Voltage Channel in the EBR-JJ," Vol. V. REFERENCES NS-15. ШЕЕ Transactions on Nuclear Science. February. 1968. [1] U.S. Nuclear Regulatory Commission Regulatory Guide 1.97, [7] Thomas. HA., and McBride, A.C., "Gamma Discrimination and Revision З.Мау 1983, "Instrumentation for Light-Water-Coolcd Sensitivities of Averaging and RMS Type Detector Circuits for Nuclear Power Plants to Assess Plant and Environs Conditions Campbelling Channels," Vol. NS-15, ШЕЕ Transactions on During and Following an Accident" Nuclear Science. February, 1968 [2] IEEE Standard 323-1974, "Qualifying Class IE Equipment for [8] Hsu, Chun, and Popper, Glenn F., "An Analysis of the Differences Nuclear Power Generating Stations," Institute of Electrical and Between True-Mean-Square and Average-Mean-Squarcd Detector Electronics Engineers (Piscataway, New Jersey) Circuits for Use in Campbclling Neutron-Monitoring Systems," Presented at the 16th Nuclear Science Symposium, October, 1969. [3] ШЕЕ Standard 344-1975, 'TEEE Recommended Practices for Seismic Qualification of Class IE Equipment for Nuclear Power [9] Knoll, Glenn F., Radiation Detection and Measurement, Second Edition, John Wiley & Sons, 1989 (New York). INTERNATIONAL ATOMIC ENERGY AGENCY SPECIALISTS MEETING ON INSTRUMENTATION AND EQUIPMENT FOR MONITORING AND CONTROLLING NPP POST-ACCIDENT SITUATIONS 12-15 September 1995 Dimitrovgrad, Russian Federation BASIC DATA OF EMERGENCY RESPONSE CENTRE Oldoich Jenièek State Office for Nuclear Safety (SONS) Slezskâ 9 120 29 Praha 2, Czech Republic BASIC DATA OF EMERGENCY RESPONSE CENTRE Oldrich Jenicek STATE OFFICE TOR NUC1EAR SAFETY IAEA Speciab'sts Meeting Sep. 12-15,1995 DimitTovgrad BASIC DATA OF EMERGENCY RESPONSE CENTRE Introduction Emergency response centres have been established in the countries with developed nuclear technologies as a qualified support for the government (or a government- approved institution) and regional organs of the State Adrninistration to offer a qualified help in their decision making in cases of nuclear emergency situation occurring on NPP. Such centres are equipped not only by the necessary methodical, technical and information means but they employ high-level professional personnel. The justification for creating such centres follows from the comparison of economic and social consequences of nuclear emergencies and the costs required for the estabUshment and running of these centres. The mentioned centres under the standard situation help to create a professional structure assuring permanent preparedness of a professional background and its continuous qualitative up to date development. Emergency Response Centre (ERC) of Czech Republic is a highly specialised institution belonging to Nuclear Safety State A(lmmistration (SONS), which assures its activities both organisationally and technically. Main function of the ERC in the case of nuclear emergency is to fulfil the needs of SONS, Governmental Committee for Nuclear Emergencies in ER (GCNE ÈR) and the regional organs of State Authorities concerning the emergency planning and preparedness, evaluation of nuclear emergency consequences, including the emergency management and response. In the case of major failure or accident on NPP, the ERC carries out the performance analysis and review of a given NPP. It also monitors the dosimetric situation and transfers the recommendation to GCNE ÈR, Regional Emergency Management Committees and to NPP. The Emergency Response Centre is established as a professional workplace with permanent preparedness. It is set up by Nuclear Energy State Administration (SONS). Professionally, the ERC cover the area of nuclear safety and radiation protection, namely: • operational and emergency aspects of NPP • distribution of radionuclides in the environment • decontamination processes and emergency management The function of ERC predorninantly consist of: • coordination of information flow • acquisition, correlation and interpretation of technical and organisational data on emergency preparedness or on the emergency which already occurred. Evaluation of emergency consequences • notification or activation of selected bodies in the case of imminent nuclear emergency -2- BASIC DATA OF EMERGENCY RESPONSE CENTRE Oldrich Jenicek STATE OFFICE FOR NUCLEAR SAFETY IAEA Specialists Meeting Sep. 12-15. 1995 Dimitrovgrad • preparation of proposals for protective measures and emergency management which are submitted to GCNE to enable their decision making in the case of emergency In the course of its standard activity the ERC assures permanent preparedness, technically it fulfills the role of a liaison body and in cooperation with other organizations it supplies GCNE with information necessary for emergency management. The following aspects are of particular interest: • creation and development of the database system concerning the emergency preparedness of the country • proposal of preventive measures for the population, environment and property protection in the case of emergency, • the time schedule of their implementation. The role of ERC and the cooperating bodies in the case of nuclear emergency is seen in information acquisition for GCNE to enable the fulfillment of GCNE functions. These are in particular: • decisions on population protection measures. If possible, these should be made before the occurrence of escape of radioactive substances into the environment of nuclear power plant • assessment of efficiency of these measures from the viewpoint of prevention or reduction of emergency consequences upon population and environment There are proposed the following system of ERC organisation: • managing group • information logistics group • group evaluating the condition of nuclear energy equipment • group for prediction, evaluation and emergency response The managing group coordinates the activity of emergency response groups and on this basis it draws conclusions that are finally submitted to GCNE for making decisions. The information logistics group ensures the data acquisition from the nuclear energy equipment under emergency. It is responsible for communication with other bodies involved in the emergency preparedness network, data transfer and their statistical evaluation. Technically, it assures the operation of ERC including the protection against unauthorised access to data. The group evaluating the condition of nuclear energy equipment monitors the situation on the equipment under emergency state, carries out its safety diagnostics and makes prognosis about further development of the emergency. In the case of radiation emergency danger, the group makes predictions as to the extent of radiation source. The activities of the group dealing with prediction, evaluation and emergency response will depend on the stage at which the emergency is occurring. If there is only a danger of radioactive contamination of the environment, the group makes a prediction of its consequences based upon the information given by the group evaluating the condition -3 - BASIC DATA OF EMERGENCY RESPONSE CENTRE Oldridi Jenicck STATE OFFICE FOR NUCLEAR SAFETY IAEA Specialists Meeting ^ Sep. 12-15, 1995 Dimitrovgrad of nuclear energy equipment. If the escape already occurred, it analyses short and long term consequences of the escape and makes a more precise estimate about the source potential based on the results of dosimetric results. In more advanced stages the group deals mainly with the emergency response, namely by selecting suitable procedures for the elimination of consequences. The assurance of operational activity of the ERC represents a multicomponent task involving methodology, software, technical means including the computing, telecommunication and audiovisual equipment, technical and software documentation, etc. It will be necessary to work out the guidelines for individual ERC teams which would inform about the mode of utilization of given communication, computer equipment and software tools. Each software package will have detailed documentation which will describe the programme and its use to reach a required target. Methodical and software backing will be concerned namely with the following tools: • Means used for obtaining the information on safety condition of nuclear energy equipment • Means for evaluation and prediction of radiobiological consequences of accident • Logistic, methodical and software tools In the following text a basic specification of this software is given from the viewpoint of its utilisation for assuring the activity of ERC. Means of Providing Information on NPP Safety The following methodical and software means belong to this category: • software for monitoring the nuclear equipment condition, • software for monitoring the radiation situation within the nuclear facility, • software enabling early detection of emergency situation, • software for evaluating the condition of barriers and critical safety functions, • software for emergency diagnostics and its further prediction. The software enabling the monitoring the condition of nuclear equipment is an important tool for a detailed diagnostics of emergency and its localisation. This tool enables to display all the measured technological values from the primary and secondary circuit in a clearly arranged manner. The software used for monitoring the on-site radiation situation should enable to display all the data from the dosimetric control system. Its purpose is: • to localise the leakage points and the direction of contamination spreading, • to assess the realistic operational possibilities which would prevent further aggravation of the situation and which would lead to the minimization of emergency consequences, • to determine the escape routes for the evacuation of nuclear plant personnel, • to determine the way of personnel protection during the repairs and rescue work. -4 - BASIC DATA OF EMERGENCY RESPONSE CENTRE Oldrich Jenicek STATE OFFICE FOR NUCLEAR SAFETY IAEA Specialists Meeting Sep. 12-15,1995 Dimitrovgrad The software enabling a timely detection of emergency will help to ERC to identify the kind of emergency on the nuclear equipment. In some cases it will be possible to identify also the cause of the emergency. Methodically, the program will be based on the EFD (Early Fault Detection) approach which enables during a very short time period to carry out analysis of mdividual components of nuclear equipment or even to recognise its failure utilizing the real-time data. The software used for the evaluation of condition of barriers and critical safety functions will evaluate the real-time technological and radiation data from the viewpoint fulfillment of critical safety functions and barrier ability to prevent the spread of nuclear fission products in the environment. Means of Evaluation and Prediction of Radiobiological Consequences of Accident Methodical and software tools for the evaluation and prediction of radiobiological consequences of accident will fulfil the following functions: • display and updating the radiation situation in the vicinity of nuclear energy equipment, • prognosis of activity concentration and radiation doses in the vicinity of nuclear energy equipment, • calculation and display of radioactive cloud movement trace at changing meteorological conditions, • reconstruction of the radiation source potential by using the dosimetric data within the nuclear facility and in its environment. Logistic Methodical Tools and Software The logistic part of software package will deal with all the necessary information and data serving to the optimization of emergency response in order to protect the population, property and to enable the decisions leading to the elimination of accident consequences. The logistic software will yield the information about: • notification of organisations and bodies that are connected with the system of notification and activation of rescue teams, • population warning in connection with the alarm systems, • the measures planned to be taken within the framework of internal and external emergency plans of the nuclear power plant and the territory, • organisational assurance of the preparation, operation and activity termination of the operating teams, • measures and actions taken, specified in connection with time, space, personnel and communication modes, • information acquisition, processing and distribution concerning the meteorological situation, air and water contamination, water level in rivers and other environmental data. A part of activities in connection with SW will involve also the preparation of procedures and mechanisms within the ERC itself. These will determine the principles -5 - BASIC DATA OF EMERGENCY RESPONSE CENTRE Oldridi Jenicck STATE OFFICE FOR NUCXEAR SAFETY IAEA Specialists Meeting Sep.12-15, 1995 Dimitrovgrad and procedures to be observed by the ERC personnel in order to reduce the probability of errors and meaningless improvisation under the stress situations. Hardware The ERC hardware will consist of combination of servers and workstations operated under the UNIX system and the personal computers of Pentium class operating under the NT Windows and Novell operation system. The hardware used in ERC should be powerful enough to be able to store the processed data and to provide the printing services mcluding the coloured printing facilities (namely the printing of the maps). The hardware used will also assure the data communication with other participants and for this reason it will be selected accordingly. Telecommunication Means Because of its importance and significance in the case of nuclear emergency, the ERC must be equipped with reliable telecommunication means enabling not only the standard verbal and fax communication among all the participants of the emergency preparedness network but also with those who are not commonly accessible. The following means of corrununication are considered: • selected lines for fax transmissions and verbal communication • solid data circuits (metallic, optical, RRS, SAT) • line selected exclusively for verbal communication with Czech Energy Works (ÈEZ) • activation connection (estabhshing a solid connection within an agreed upon time) • preferential connection (at the overload of telephone exchanges, this connection remains operative also for the external communication) • standard lines (direct) • standard lines (through the telephone exchange) • special lines enabling the access to the networks of Ministry of Interior, Defence and Transport) All the verbal means of telecommunication at the disposal of ERC should be equipped with a recording facility. Audiovisual Equipment The Emergency Response Centre will need the audiovisual equipment for presenting the results of its work to the members of GCNE and also for monitoring the activity of mass information media. To present the results of ERC activities it will be necessary to equip the Centre by standard audiovisual tools such as e.g. an overhead projector which would enable the projection of computer displays. For monitoring the public information about the course of accident and to follow the response of mass information media it will be necessary to equip the ERC by standard audiovisual technique (radio and television sets, tape recorder and video recorder). BASIC DATA OF EMERGENCY RESPONSE CENTRE Oldricfa Jenicek STATE OFFICE FOR NUCLEAR SAFETY IAEA Specialists Meeting Sep. 12-15,1995 Dimitro'.grad Documentation The ERC will need extensive documentation for its activity. This will include predominantly: • ERC operational documentation • drawing documentation of NPP • operational regulations of NPP • internal and external emergency plans of NPP Archival Storing of Data and Programs An archive which will be used in ERC should enable the storage and reinstallation of ERC software as well as archival backup of important data files. This archive should be specially protected against theft and fire in order to prevent loss in case of burglary or fire in the ERC premises. -7- 11 Applied Software of the Emergency Recording System for Reactor Facility Parameters Under the Minor Statistice Conditions Ivanov V.BJ, Grachev A.F., Kinsky O.M., Matin R.S-, Ochrimenko A-I-, Demidov L.I., Karpjuk V.l., Afonin V.K., Iskanderov R.G. The State Scientific Centre of Russian Federation Research Institute of •Atomic Reactors DimitrovgradT Russia Abstract The requirements to computer-aided system of accidents analysis, based on symptom-oriented approach to initial events are given in this paper. The purpose, composition and realisation of regression analysis algorithms of fast process measurements results under the Minor statistics Conditions are described. The description of information compacting methods for its long storing, on the base of adaptive digitisation methods is given. The results of simulation on the base of computers elaborated technical means and applied software are described. ПРИКЛАДНОЕ ПРОГРАММНОЕ ОБЕСПЕЧЕНИЕ РЕГРЕССИОННОГО АНАЛИЗА В УСЛОВИЯХ МАЛОЙ СТАТИСТИКИ СИСТЕМЫ АВАРИЙНОЙ РЕГИСТРАЦИИ ПАРАМЕТРОВ РЕАКТОРНОЙ УСТАНОВКИ С ИССЛЕ ДОВАТЕЛЬСКИМ РЕАКТОРОМ ВК-50. Иванов В.Б., Грачев А.Ф., Кинский О.М., Макин P.C., Охркменко А.И., Демидов Л. И., Карпюк В.И., Лашева Н.В.. Искандеров Р.Г. Государственный На.учный Центр Российской Федерации Научно—Исследовательский ИНСТИТУТ АТОМНЫХ Реакторов ДИМИТРОВГРАД, Россия. АННОТАЦИЯ Приведены требовс-.м;.;?! К автоматизированной системе для анализа аварийны;-: ситуации на основе сипптомно—ориентированно— гс подхода к исходным событиям. Описаны назначение, состав и реализация алгоритмов рег рессионного анализа результатов измерений быстропротекающих процессов в условиях малой статистики. ПРИВОДИТСЯ также описание приемов сжатия ИНФОРМАЦИИ ДЛЯ ее длительного хранения на основе методов адаптивной дискрети зации. Описаны результаты макетирования системы на основе ПЭВМ, разработанных технических средств и пакета прикладных прог рамм. 1. ВВЕДЕНИЕ Дальнейшее развитие информационных Функций ЛСНИ на реак торных установках института планируется осуществить путем раз работки и создания Информационно—Аналитической Системы "Безо пасность эксплуатации РУ с исследовательскими реакторами ГНЦ НИИАР" С13. Б случае установки) с реактором ВК-50 система долж на содержать около четырнадцати подсистем, решающи:: информаци онные 1Л аналитические задачи, а также задачи поддержки персо нала. Одной из важнейших является подсистема "Аварийность", которая решает информационные задачи и задачи анализа (Таблица 1). Функции подсистемы предполагается реализовать на базе двух персональных компьютеров в составе Распределенной сети, один из КОТОРЫХ оснащен устройствами сопряжения с объек том и решает информационные задачи. На ВТОРОМ компьютере реша ются задачи анализа с помощью Функционального треназгера и ряда вспомогательных программ. Наиболее жесткие требования предъявляются к компьютеру, решающему информационные задачи. \8 2. ОСНОВНЫЕ ТЕХНИЧЕСКИЕ ТРЕБОВАНИЯ, СХЕМА ФУНКЦИОНИРОВАНИЯ И АЛГОРИТМЫ ПЕРВИЧНОЙ СТАТИСТИЧЕСКОЙ ОБРАБОТКИ ИЗМЕРЯЕМЫХ ПАРАМЕТРОВ. . Требуется в течение компании реактора ВК—50 (время между двумя частичными перегрузками активной зоны ~ 300 суток) заре гистрировать на тзврдом диске персонального компьютера измене ния группы непрерывны;: аналоговых сигналов при возникновении аварийны:-: режимов для последующего оперативного или неопера тивного iriani/'ji.. О&ъем регистрируемы:-: параметров должен быть получен h<à осhoes спмптомно—ориентированного подхода, удовлетворять тре бованию минимальности и являться значениями критическим Функ ций. 3 настоящее брэмя путем экспертный оценок на реакторной установке ВК—50 определен относительно небольшой объем пара метров, необходимый для длительного хранения во внешней памяти компьютера (Тзьл. 2) . Современные операционные системы реального времени лозвс— ля»т решить такую задачу путзм создания системы. Функционирую щей по принципиальное, схеме, пр еде таз ленной на рис.1. После инициализации всей системы диспетчер делает актив ной задачу ввода исходной информации от оператора. Задача ВВОД выходит на диалог с оператором, который задает режим работы системы (в общем случае, это тип измерительного модуля, стра тегия измерений, шаг дискретности регистрации, и т.п.). В про цессе работы задачи ВВОД входная информация регистрируется е отведенной области памяти комлютера (на рис.1 не показана), к которой будут обращаться все задача системы. Окончание ввода сигнализируется посылкой сообщения ое этом диспетчеру. Диспет чер запускает задачу управления измерениями (УИЗ) аналоговых сигналов, которая выполняет измерения и регистрирует их ре зультаты б отведенной области памяти компьютера (буфере ре гистрации результатов измерений). По окончании одной серии из мерений задача УИЗ Еыдает об этом сообщение диспетчеру, кото рый запускает в работу задачу обработки (ОБРАБОТКА) , которая также регистрирует результаты е отведенной области памяти (ёу- seps регистрации результатов обработки). Далее диспетчером за пускается задача регистрации Результатов обработки на твердом диске для длительного хранения и последующего анализа. Регистрация результатсЕ измерений в памяти компьютера осуществляется по стандартной схеме измерительного тракта (де тектор - линия связи — нормирующий преобразователь - коммута тор - программируемый усилитель — аналого-цифровой преобразо ватель - память компьютера) . Управление измерениями состоит в указании: номера линии связи детектора, подключаемого к входам коммутатора, промежут ка времени At Ciaara дискретности) , через который необходимо та кие подключения производить, а также ряда других параметров (коэффициента усиления, порядка измерений и т.п.). В случае быстропротекающих процессов возникает ряд проблем по выбору шага дискретности &t . Очевидно, что если £t относительно велико, то не все существенные изменения измеряемой непрерыв ной переменной x(t) будут Фиксироваться и произойдет потеря информации. При необоснованно частом съеме информации ( д-с ма ло) возрастает избыточность и требуется.увеличение оперативное и внешней памяти компьютера. Известно, что нижний предел шага дискретности V рений реализации случайной Функции у. Б.А.Котельникова C23i (t) со спектром. находящемся в интервале (0,F), полностью определяется последовательностью ее значений в точках. отстоящих на время t = 1/2F секунд ДРУГ от'друга. Указанная теорема имеет практическое значение для случай ных процессов, модель которых известна (известна спектральная плотность или корреляционная Функция) и спектр частот сигнала ограничен. В случае неограниченного спектра применяется пред варительное Фильтрование (цифровое или аналоговое) сигнала. Последнее учитывается nFn проектировании измерительного трак та. Однако дахе ПРИ устранении ПОСТОРОННИХ шумов случайный ха рактер сигнала сохраняется и может оказаться полезным. Типичным примером являются Флуктуации тока ионизационной камеры вокруг среднего значения ПРИ работе реактора ВК-50 на постоянном уровне мощности. Построение автокорреляционной Функции по измеренному значению тока- ионизационной камеры че рез равные промежутки времени з условиях стационарности и ЭР ГОДИЧНОСТИ процесса позволяет дать количественную оценку бли зости реактора к границе нейтронно-Физической устойчивости. Для этого найденные в результате измерений и вычислений значе ния автокорреляционной Функции аппроксимируются выражением -•ft К CT? > =г - cos cat; (1) где if - показатель затухания автокорреляционной Функции; О - резонансная частота; - временной сдвиг. ПРИ малых значениях "£ (""0,2) реактор близок к границе ли нейной . устойчивости. По этой причине показатель нейтрснно—Фи зической устойчивости включен в состав критических Функций ре актора ВК-5С (Таблица 2). 3 случае переходных и аварийных процессов ""РИ ПРсведении ксяг- ЕЛ яцлс:-;яс- регг>Егсс;."имогз анализа остальных значений перед ня критически.: ;у:-!К.ций Р'У- ЭК—50 (Таблица 2> достаточно ограни— читься вычислением математического ожидания случайней ФУНКЦИИ, зарегистрирован:-.ой з дискретные моменты времени ?5г?scСйоммьа1. анализ) <л спектр КОТОРОЙ заранее неизвестен. В гтих условиях ВЫБОР шага дискретности At осуществляется из условий со- дзржательности и КОРРЕК ГИС.-ТИ МЯТО.ЦСЗ КОРРЕЛЯЦИОННО—РЭГРЭССИ-- ÙHHO.'O =нализа; ! ai rSi;'.:ïva:'K /.зм5?£н»'.1л должны эыть сох э~тически -«зави симы^::/:» е) аеличинз условной дисперсии зависимой ЛЭРЭМЭННОЙ X ПРИ изменении аргумента t должна оставаться неизменной или быть, пропорциональной некоторой известной Функции; в) ПРИ каждом Фиксированном значении аргумента • t зависи мая переменная X должна быть распределена по нормальному зако ну. Проверка указанных исходных посылок корреляционно-регрес сионного анализа в случае быстропротекающих процессов осложня ется малым объемом выборочных значений переменной xi,X2.' хп (п^25), которые должны использоваться для проверки. В этих условиях для проверки результатов измерений xi, Х2»»••* *п на стохастическую независимость рекомендуется критерий "восходя- [ob 4 плодящих" сэрий, а также критерий квадратов последо- ^. отклонений Г.ЗЗ. Для проверки однородности Ряда дис— лерсий рекомендуется "критическая статистика" основанная на распределении "V_2 ЕЗЗ. Нормальность распределения должна про веряться по критерию у2 который получен К.СаРкади. При выполнении немодных предпосылок регрессионного анали за стандартный метод оценки математического ожидания х£ t ) состоит в построении модели сигнала в виде полинома некоторой степени г, коэффициенты которого вычисляются по методу наи меньших квадратов. 3 случае неопределенности степени полинома г в качестве базисных Функций времени t в разложении математи ческого ожидания удойно использовать ертонормированные полино мы. Тогда применение метзда наименьийк hibsspstob для определе ния коы?-гициентов разложения приводит к удойным рекурентным соотношениям Форсайта, возможности оценки остаточной суммы квадратов и дисперсии Г.43. Последнее делает удойным и обосно ванным применения .критерия Фишера для определения коэффициен тов разложения и степени полинома С43. Приведенные методики проверки исходны:: предпосылок рег рессионного анализа и построения Регрессионной кривой с по мощь» ортонормированных полиномов и критерия Фишера реализова ны в зиде пакета прикладных программ персонального компьютера. Пакет программ может быть использован для экспериментального определения шага дискретизации &t , определения объема выборки n и степени полинома г при изучении быстропротекающих про цессов. Такой пакет программ должен входить в состав приклад ного программного обеспечения системы аварийной регистрации параметров, но его практическое применение в составе задачи реального времени ОБРАБОТКА \рис.1) является проблематичным из-за сложности процедур. Поэтому в результате изучения быст- рспротекающих процессов на РУ ВК-50 с." помощью указанного паке та программ предлагается другой способ построения математичес кого ожидания, основанный на результатах измерений дискретных значений случайной величины xi, *2> в моменты време ни t;, t 2? • - • ! tm. При этом г..=оМежутг^ зрзмэни t^-t ; считает ся до : ': оточно длительным -..?сз>. П.1 „'sг:анен1'!ь предлага'.-м-."- :.• зосос • г гсыг и? r-v-- —os. :;^зас!/1Ц1-1гм гы полинол.й с г; -и.с; > млиг-ггчь—и:; г-. — » . . л где ni - ширина окна слева от значения >:j; г.2 — ширина окна справа от значения x j ; bfc - коэффициенты линейной формы, получаемые методом на именьших квадратов и независящие от значений Xj. loi Формула (3) последовательно применяется ("скользит") по всем моментам времени tj( i=l,2, ...m ) в результате чего опре деляются сглаженные значения xj, 5?2» - - • хт. На этом первый этап получения математического ожидания заканчивается. ВТОРОЙ этап состоит в организации итерационного процесса по дальнейшему уточнению дискретных значений математического ожидания. ПРИ ЭТОМ измеренные значения переменной Х± считаются началом итерационного процесса (х^°}= Xi>, а результаты приме нения Формулы СЗ) — его первым этапом (xf^=Xj). На ВТОРОМ шаге процесса значения переменной подменяются результатами по лученными на первом шаге и процедура повторяется. Условием окончания итераций по уточнению математического ожидания явля ется выполнение неравенства: ; -<р>_ -<р-1) • max (4) -Cp-I)~ где - заданная точность сходимости процесса, определя емая погрешностью измерений; р - номер итерации. Необходимо заметить, что в операциях по получению матема тического ожидания в моменты времени tj, t2, • ••» tm участвуют результаты ni измерений полученных ранее и г>2 последующих измерений. Результатом описанного алгоритма первичной статис тической обработки являются сглаженные значения случайной Функции ;~i в моменты времени tj (i = l, 2,..., m). Если xj - электрические сигналы детектора, то значение Физического пара метра восстанавливается по градуировочной характеристике де тектора. ПРИ запуске системы аварийной' регистрации в начале кампа нии реактора задача управления измерениями (Рис.1) заполняет буфер регистрации. После набора ni+m+r>2 результатов измерений диспетчером запускается задача ОБРАБОТКА, которая по изложен ному алгоритму получает m значений математического ожидания: *1» * 2»---?*m. Задача УИЗ ПРИ ЭТОМ СВОЮ работу продолжает. В следующем цикле своей работы задача УИЗ регистрирует в буФере т+п2 результаты измерений, которые являются исходными данными для задачи ОБРАБОТКА по изложенному алгоритму. ПРИ ЭТОМ обра ботка результатов ведется с учетом nj значений математического лжидания полученных в предыдущем цикле обработки. Далее проце дура повторяется. Завершив один цикл, задача ОБРАБОТКА должна сообщить об этом диспетчеру, КОТОРЫЙ запустит задачу регистрации m значе ний математического ожидания xj, *2>--'!>*т в отведенный облас—. ти памяти на твердом диске (РИС.1). Однако, определением математического ожидания в одном цикле обработка результатов не заканчивается. Требуется полу чить Физическое значение измеренного параметра с помощью гра дуировочной характеристики, а также решить вопрос об изменении параметров реакторной установки в одном цикле измерений. Если параметры установки в пределах одного цикла измерений (или его части) не изменяются (стационарное состояние), то требуется провести корреляционный анализ (оценить автокорреляционную Функцию) тока ионизационной камеры в условиях ЭРГОДИЧНОСТИ И loi 6 стационарности. Па результатам оценки автокорреляционной Функ ции и аппроксимационной зависимости (1) методом наименьших квадратов должны выть определены: показатель нейтРонна-Физи- ческой устойчивости и Pf. .гчансная частота Для определения ус.т~сл*~-. с неизменности параметров реак торной установки динамический диапазон каждого из ни:: Схт£П, 1:тах"Т числом n разбивается на интервалы Cxj — i х^З ((j=l,2,..., п> (рис.2). Если определенные в одном цикле измерений значе ние математического ожидания величины X находятся внутри j-ro интервала Lx ^..-i :•• j2 казбигния динамического диапазона (j=B на рис.2), го Функции, \(f> (определяемую значениями ïïi , x^«, ...,;"-.> следует призна t ь постоянной на заданном ctpeskf вре мени (отрезок Г - 1 )• ДТ, гй'ДТЗ рис 2). При этом стационарное-, ь параметра X опред г л чете я длиной ин тервалов развиениу-. динамического диапазона А:;=-"max--"-r^.in^/'п : которая, а сбою ачагедь, определяет параметр с- s условие - 3 условиях стациона?*чс-_-:и параметров реакторной установки нет смысла в рер-лстрацип м«> тзердои диске их математических ожида ний ix«, х'2, • - • 5 х"-.-.ученных в одном цикле измерений .сли- тельностью ДТ (р:.-;с.2>. Достаточно только зарегистрировать : одно значение г.стг'^чичгского ожиданиц vis отрезка L x j — x x .; 3 , длину K.I-.t i: ;- u. ала ДХ, a T5-cïs моменты времени начала и конца"реализации стационарного состояния. Если napsi'.ü ïfo.-«. X является ток ионизационной камеры, то следует дополнительно провести корреляционный анализ и зарегистрировать на твердом диске значения *jj и cD . Такой прием позволяет сократить объем регистрируемой на твердом диске информации е случае стационарного состояния ре акторной установки i--. hcï=t быть обобщен на случай нестационар ных состояний. 3. ПРИЕМЫ СЖАТИЯ РЕГИСТРИРУЕМОЙ ИНФОРМАЦИИ В НЕСТАЦИОНАРНЫХ СОСТОЯНИЯХ РЕАКТОРНОЙ УСТАНОВКИ Полагая, чте неслучайная Функция x(t> полностью определя ется набором своих дискретных значений у. i, ~?! • * • ! и может быть восстановлена с помощью процедур интерполяции (линейной, сплайн), s задаче ОБРАБОТКА при каждом значении параметра xj(j—С, 1, 2, г.! следует решить на заданном отрезке Бре мени (отрезок C.Ti-ДТ, (г: - 1 )• ДТ) на рис.2? уравнение X (t)=*:: j=0, 1, . . . , П (5) относительно времени t. Полученные решения следует Располо жить e порядке возрастания. Тогда получится"конечная последо вательность пар -Ct .j Xfj> корней уравнений (5) и граничных точек интервалов разбиения динамического диапазона величины X (Ctj х7>, 7 Необходимо заметить, что к решению уравнений (5) сводится так же, процедура поиска стационарных значений xft). Указанный прием сжатия информации носит название метода адаптивной дискретизации <РИС.2) и должен быть реализован Б составе задачи ОБРАБОТКА. Реализация метода в одном цикле из мерений состоит из трех этапов: - интерполяция значений iTj, х"2? • - - > - гелгемйэ уравнений (5) ? •- упорядочивание корней уравнения. (5). 4. СОСТОЯНИЕ РАЗРАБОТКИ СИСТЕМЫ АВАРИЙНОЙ РЕГУСТРАЦИИ ПАРАМЕТРОВ. Предложенные АЛГОРИТМЫ статистической обработки сигналов в условиях мапой статистики, совместно с методами адаптивней дискретизации реализованы в виде прикладного ПРСГГа?'.много обеспечения и cT."asEHJ в автономном режиме на технических средства»:-: макета системы аварийное1: регистрации г.а.= »ME-POÖ . Технические c-a.".:Tbi- макета .'. пс-З) содержа.г.' -• герсональный : .o:-*.r.b!CTeF FC/ftT~3B£ стандартной ком-аиг ура- -- Раз работами УК. плату сопряжения с шиной ввода-вывода ISA пеР сона л ь но го компь ютер a î - разработанные устройства сопряжения с объекте?! 'комму таторы, АЦП, контроллеры)? - разработанный .-юдуль РАСШИРИТЕЛЯ ШИНЫ езодз -выгода, г-.-эг - сона.ь но г с к--_!: тьютера, к КОТОРОЙ подключены чеггз ко-т- ГОЛЛЕРЬ: измерительные модули. Степень готовности разработанных технических средств ма кета, v. его характеристики представлены на РИС.3. Макет нахо дится а стадии опытной эксплуатации и ориентирозан на перечень критических Функций РУ ВК—50 (Таблица 2>. Для решения задачи анализа дополнителен-.:; используется модуль детектирования диск- }-;UÎI ^игг.алоз «'Таблица 3). ЗАКЛЮЧЕНИЕ. ПРе.д*"ожэнный алгоритм первичной статистической обработки i=ss>f!t.~àTCE измерений обладае— следующими достоинствами: - вариацией степени полинома г и выбором окна (г. 1,г-2- можно г.ол учить асе известнее ФОРМУЛЫ ПО сглаживанию экспери,— hgrtla.'iuHSlX i äSL'iC. Hf.UC ! â'Jl ï - использование вместо полиномов (2) ортогональных поли номов, дает возможность автоматизировать процедуру выбора сте пени полинома г? - организация итерационной процедуры по сглаживанию дает возможность последовательного уточнения результатов измерения в одни и те же моменты времени» - использование условия (4) для окончания итерационного процесса дает возможность связать результаты обработки с пог решностью измерений. Указанные достоинства предложенного алгоритма могут быть использованы для целей оперативного контроля. Для этого деста- Л о Ц о точна ограничиться одним ша: n итерационного процесса по сгла живанию. Тогда быбоь JM степени г- ортогонального полинома и ок на г. I, г>2) можно организовать оперативную обработку и представление ее результатов оператору реакторной установки по "1"г"п2+^ измерениям хранимы:; в буфере регистрации результатов измерений. Предложенные методы адаптивной дискретизации позволяют дг;ект!ЛЕНс рааитс зада-: регистрации ?=эрийных параметров ре— 01;тсркой уи-:=:чс1Ь;-.п м-=. я-осты:: средств av. вычислительной техники ^ условиях IV.- длительной .^з^оты ssa er-еаательств? оператора. ЛИТЕРАТУРА . 1« Пеанов В.-Б. и ДР. ПРИНЦИПЫ разработки cecs-ü^a ИНФОР МАЦИОННО— Аналитической Системы "Безопасность эксплуатации РУ с исследовательскими реакторами ГНЦ НИИАР". Доклад s настоящем ей. ДимиТРОВГt-s.р- 1995 ^. 2- оаскакзе СИ. Радиотехнические цепи ы г;л--'-алы. М. * Sscus. шк. , l^BS г. 3- Айвазян С.А. Статистическое исследование заэисимостей. :и..5 Металлургия, 1?£8 г. 4. Худсон Д. Статистика для Физиков. М-: Мир, 1970 г. 9 Таелица 1 Задачи подсистемы "Аварийность". Информационные задачи. 1.1. СБОР И «ранение в базе данных информации о всех аварийны:-: ситуациях на АЗС s виде ПРОТОКОЛОВ ава рий, переведенных с яе-'Л С UK ъ i-.es-'t онс..-.^пы-Г-. ком.п*--- йТ£рй зарегчлстриРоьсдг1иь.:-. =г г-езулатвте йнтг.мгтигм- 'Ч:««НЧОГО KOt-T Î---0." Я. 1.2. Возможность пе-ос мотрз ПРОТОН, о г СЕ с помощью п»ог- яамм- визуализации на .г. и с- г .с s? е -персональна:-:: комль- «-•:сРа. 1.7. Рсз--!Ожнссть Е^ДЭНГ.? учет л ст^тигт-.'-^ зт':азоз лс 4 • .г-^н'лтой cräf-ÄäP тис;! тсг.пке рг.-г. Ts,zos-=.i-:-:л аза- V.'! СИТУ К..;.-1." - :. '.. Ком: - ,..•:*>•-. выпсйнгний ме>- ог»».?яти#. намеченных -.к : i.Mi- РЗС1г.г.2дс.ааи«я Ссрс-кп. ответственные) . Зада-':---, анализа. 2.1. Возможность моделирования аварийных ситуаций на «.^-национальных тренажерах систем с цельк onpese- ~ =_-?-*-•!*= альтернативных путей развития it л и ~~гдотв- ращения аварии, опреде nsHv.r недостатке.» с-зорудо- вания или производственной /-окументации. 2.2. П? с-дстав ление наглядное информации (к ои^С гРаФи- '•.ci.f диагр£:;м и т.д.) с е.-' о-, казах по принятой на классификации (г r.--.::.i.ция но отказам гис-гум, si/i- ü c-ii€»p-ôTHbf.c • i--î^i -•.••>-с=:сл.':-^"е-с лёРссг-»л5, недо- •,-.-.:-ток прсе«--тм1-'.: cp-r«vv.?u'::-ä s-, т.д.> . 2. Т. .-.-s ;ic-s TOP Р.е'-'.сс т"т^агж *ÏC- зинг однотипного .-дивак1-57:. еяеми агг==-ата, смены i.: л и FS- •. _\. -<;J'i эгиг*-.д:ПОДР-S»делений РУ, времени года и [об 10 Таблица 2. Перечень критических функций РУ ВК-50. Номер функции Наименование параметра * i 2 Давление в реакторе 3 4 5 Уровень в реакторе б 7 Давление пара перед 8 турбиной 9 Расход питательной воды на охлаждение СУЗ 10 Расход воды по линии секционного питания 11 Давление пара перед турбиной 12 13 Расход пара из реактора 14 по 1-*4 паропроводам 15 16 Расход пара через ГПЗ 1 и 2 на 17 турбину 18 Уровень в деаэраторе 19 Перепад давления на регулирующих клапанах 20 Уровень в гидроемкости САОР 21 Расход продувочной воды 22 Индикатор давления в реакторе 23 Расход питательной воды 24 Расход питательной воды через клапан А-117 25 Расход питательной воды на эжекторы 26 Давление пара перед деаэратором 27 Уровень мощности ПО I 28 Период нарастания мощности каналу 29 Уровень мощности по II 30 Период нарастания мощности каналу 31 Уровень мощности по III 32 Период нарастания мощности каналу 33 Реактивность 34 Показатель нейтронно-фиэнческон устойчивости 35 Запас до кризиса теплообмена 36 37 Температура среды в шахте реактора 38 39 Температура воды реактора 40 Температура корпуса реактора . 41 Температура воды на механизмы СУЗ 42 Температура воды в деаэраторах 43 Температура воды в САОР lï От оператора 5адача ВВОД аг На диск Диспетчер Задача Задача Задача управления регистрации на диске обработки измерениями с (РЕГИСТРАЦИЯ) (ОБРАБОТКА) (У ИЗ) 7\ V 3Z Буфер регистрации Буфер регистрации результатов результатов обработки измерений Рис.1. Вариант схемы функционирования системы регистрации аварийных параметров Рис.2. Иллюстрация метода адаптивной дискретизации. О - значения функции x(t), полученные в результате адаптивной днскрет'яации. I 04 13 Некоторые Быстрый АЦП параметры m а МУ илт ( 10 изм / сек) минимальный период измерений s отдельном канале, (/•%Л- Выход- токи сек 0 025 VJ"} Память Встроенный плектр.Л 0 от датчиков число одновременно i каналов ! коммутатор 3 типа измеряемых токовых ''"Сапфир". сигналов 32 ' 3 Токовые ! МУ ИШ на 32 канала U ! (модуль управл.нзмер.пост.токов) сигналы от АКНМ компоненты собственной г Импульсный блок питания "Карпаты". рслргботси для гальванической развязки с компоненты, соторыс РС/АТ-336/387 прелпслага&.ся разработать в Память Встроеный Встроенный Термо будущем каналов интегр. коммуттатор метры сопротив готовые изделия АЦП на герконаа ления готовые изделия, МУ ИТП на 8 каналов которые (модуль управл.измер.темпер.помещ.) предполагается заменить элементами Токовые собственной 1 логичес разработки мдс (модуль детектирования событий) кие » PC/AT -35 сигналы J 386/387 MMP <§ ПС ——j (модуль магистральн. расширителя) Vi' с шиной Встроенный ISA таймер PC/AT п 386/387 МУ иэт (модуль управления измерениями Некоторые параметры ЭДС термопар) МУ ИТП ! МУ ИПР минимальный период измерений в отдельном «•-*( (модуль управления измерениями канале, сек -12 8 i положений регуляторов) число одновременно измеряемых aimanoB-8 Рис.3. Архитектура системы аварийной регистрации параметров РУ ВК-50 \\o 14 Таблица 3. Перечень дискретных сигналов (событий) системы J Сброс AK 17 09-31 2 Сигнал A3 3 Ввод борного раствора при срабатывании A3 и исчезновении 10Гц 4 Разрыв трубопровода i 5 Отключение MB генератор*. 1 e Повышение температуры в шахте реактора ! 7 Исчезновение напряжения =220в в цепях управления Сигналы, \ вызывающие i s Исчезновение напряжения =45в на СП-1.2,6.7 глушение ! 9 Исчезновение напряжения на РУ бкв реактора Гю Отключение турбины, закрытие стопорного клапана ! ii Понижение уровня в реакторе, по 2-м из 3-х приборов ! позиции 179/1,2.3 12 Повышение давлениг" перед турбиной по 2-м из 3-х приборов позиции 680/l-f3 13 Повышение давления в реакторе по 2-м из 3-х приборов позиции 100/1*3 14 Снижение периода нарастания мощности по 2-м из 3-х каналов 15 Повышение уровня мощности по 2-м из 3-х каналов 16 Нажата кнопка A3 (любая из 3-х) 1 Сброс АК-19 07-31 2 Сброс АК-18 11-37 3 Монжюс впрыска бора 4 Монжюс СУЗ Сигналы 5 Привода 63/ 1*3 контрол я 6 Клапаны А-108/1*4 готовности 7 БРОУ систем S Техническая вода безопасности и 9 Турбина автоматического 10 Насос ПЭН-4 включения Ii Насос ПЭН-3 резервного 12 Насос ПЭН-2 оборудования 13 Насос ПЭН-1 14 Дренчерная система 15 Предохранительные клапаны 16 Насос Т-10/140-4 17 Насос Т-10/140-3 18 Насос Т- Ю/140-2 19 Насос Т-10/140-1 20 Контур реактора Hl Design Implementation of the Post-Accident Monitoring (РАМ) System for Wolsong NPP Units 2.3&4 in Korea SPECIALISTS' MEETING on INSTRUMENTATION AND EQUIPMENT FOR MONITORING AND CONTROLLING NPP POST-ACCIDENT SITUATIONS 12-15 September 1995 Dimitrovgrad, Russian Federation Sang-Joon Han Korea Atomic Energy Research Institute P.O-Box 105, Yusong Taejon, Korea Titie : Design Implementation of the Post-Accident Monitoring (РАМ) System for Wolsong NPP Units 2,3&4 in Korea Summary Wolsong NPP Units 2.3&4 (hereinafter "Wolsong 2.3&4") are unique CANDU-type reactors originally designed by Atomic Energy Canada Limited (AECL), and now being jointly designed in collaboration between Korean engineers and their Canadian counterparts, and constructed in Korea. The units are referenced to the design implemented for the existing Wolsong NPP Unit 1 which has been in service since 1983. except that many design improvements are incorporated to enhance plant safety as well as operational performance. The post-accident monitoring (РАМ) system is part of these improvements, and plays a vital role providing the operator with the necessary information to manage the outcome of the event by monitoring & displaying critical safety parameters after plant accidents. The fundamental design concept of the РАМ system for Wolsong 2.3&4 conforms to the requirements of the Canadian Standard which was established after TMI Accident in 1979. The functional requirements for the РАМ system are to provide sufficient mformation for 1) verification of reactor shutdown, 2) verification of reactor heat removal, 3) verification of a barrier to reactivity release, 4) evaluation of radiological conditions, and 5) assistance in carrying out recovery actions and for the operator to monitor the post-accident state of the plant. This paper is intended to introduce the actual design implementation of the post-accident monitoring system and to provide a good opportunity to understand what issue items are at the design implementation stage for Wolsong 2.3&4. 1 . Introduction In the event that a nuclear power plant accident occurs, initial protecting and mitigating actions are taken automatically by the plant safety systems. As the duration of the accident progresses, the operator will tend to have an increasing role in managing the outcome of the event. The post accident monitoring system provides the monitoring and display of necessary information for the plant operator to manage the event. The system has been designed to meet the requirements of applicable Codes and Standards which are issued by Canadian Standards Association (CSA) N 290.6 which has a similar design approach to other types of nuclear reactors. The functional requirements for this system are to provide sufficient mformation for 1) verification of i 1 i reactor shutdown, 2) verification of reactor heat removal, 3) verification of a barrier to radioactivity release, 4) Evaluation of radiological conditions, and 5) assistance in carrying out recovery actions and for the operators to monitor the post accident state of. the plant. Table 1 shows some examples of РАМ Variables and the safety function as recommended by CSA 290.6. The design approach for this system is to ensure the information is available to the operator in either the main control room (MCR) or the secondary control area (SCA), which is used as a backup in case MCR is uninhabitable. This РАМ system is not an independent system but a process to ensure that the instrumentation required for post accident monitoring is systematically identified and meets specific requirements. It makes maximum utilization of existing instrument loops of process, safety and safety-related systems for obtaining the necessary information. The components of instrument loops for РАМ are essentially seismically and environmentally qualified in order to perform their own safety functions during the mission time as required by the applicable codes and standards (Reference 2). Where the existing loops with the necessary information are not covered for the sensing ranges to meet the РАМ monitoring, these loops were upgraded to include post accident conditions. The upgraded loops are identified for each system designer to confirm and implement these new РАМ requirements during detail design stages. Where the coverage of the sensing ranges and qualification requirements is not within existing equipment, some additional loops and instruments are installed to meet these requirements. These are in addition to as the referenced existing Wolsong Unit 1. Furthermore, always-available voice and site communication links are required by the operator to carry out the specific management of post accident activities. These necessary links utilize the plant telephone system inside and/ or outside the plant, and then follow the plant emergency procedures. As compared with American type PWR plants of which applicable codes and standards are based upon US NRC Reg. Guide 1.97, ANSI/ANS-4.5 and IEEE 497, the requirements for CAMDU reactors are essentially the same approach but the detailed design implementation is slightly different due to the different Canadian reactor systems and reactor safety design principles. 2. Applicable codes and standards CSA N290.6 (Reference 2) provides the basic requirements to provide rules and recommendations for the design, manufacture, installation and qualification of components for the РАМ system. The general design principles and rules recommended by this standard are summarized below : - The system design should be kept simple. - Equipment accuracy and availability should be verified. - Ensure that the system is not rendered inoperative by the specific design basis events. Il k - Design should facilitate maintenance. - Same equipment as normally used. - Design to avoid giving anomalous indication. - Design to be clearly distinguishable to the operator. - Design man-machine interface using ergonomie principles. - Design to provide sufficient information chains to ensure independence, reliability and redundancy. - Measuring ranges of instruments should be adequate to cover all the possible range of during and following the accident situation. - Instruments should be qualified to remain in operation during aiid following design basis events. - Design target of availability should be better than 99.0% on a parameter basis for information chains and power supply sources . - No single information chain component failure should incapacitate any of the РАМ parameters. - An alternate information chain should always be available. - The alternate chain components should be physically and functionally independent of each other. - Verification of proper operation by means of on-line, off-line testing and/or comparing with alternate measurements at all time including post-accident should be provided. - The ability foreplacement, repair, adjustment and calibration of components at full power should be considered. - Easily distinguishable identification from other process system for components and modules in the form of nameplate tagging, color coding, physical positioning, etc. should be provided. - Indications of displays and panel meters should be located close to the area where the operator performs his role, and always be visible to the operator. 3- РАМ Design Basis Events The design basis initiating the events for which РАМ is required are based on the safety analysis reports carried out by plant safety design group for defined postulated events during plant operation. These design basis events are categorized as It-5 follows depending on the total or partial failure and consequent threat to the integrity of 1) reactor fuel sheath, 2) primary heat transport system, and 3) the containment system, which provides the physical barrier preventing the release of radioactivity to the environment: - Loss of coolant accident (LOCA) which directly affects on reactor safety. - Loss of secondary side accident inside and outside containment which directly affects the loss of coolant heat sink : . Loss of feedwater accidents . Main steam pipeline breaks accidents - LOCA plus Loss of emergency core cooling system which is the most serious postulated accident affecting reactor safety. - LOCA plus Loss of normal electric power (called "class VI power") which could seriously affect reactor safety. - Single fuel channel event which might be one of the following : . Stagnation feeder break accident . Pressure tube rupture accident . Channel flow blockage event . End fitting failure - LOCA plus site design earthquake - Design basis earthquake It is noted that normal process upsets and minor events such as loss of pressure and inventory control of primary coolant, loss of reactor regulation, loss of forced flow and normal electric power are not considered as design basis events because these kinds of events are basically anticipated during normal operational plant transient and do not require post accident monitoring beyond what is already provided by normal existing monitoring means for these transients. In addition, common mode events such as fire, tornado, flooding, etc. are not treated as РАМ events since the plants are normally designed for these common mode events by complementary safety design principles and guidelines, such as grouping and separation requirements and fire protection requirements. The event sequences which are used to develop each of the РАМ initiating events are also characterized chronologically for the predicted post accident conditions and thus establishes the determination of the nature and course of the accident As an example of event sequences, a typical sequence of events of a large LOCA is shown on the Table 2. To meet this type of event scenario, the РАМ mission time and parameters are selected and essentially determined. From the РАМ event sequences, the credited safety and safety related systems define its function following any specific accident. The system function to mitigate the accident are credited in safety analysis design by using analytical methodology. l\6 In addition to the identification of the required safety and safety-related systems, the predicted РАМ event sequences also reveal the required action to be undertaken by the operator. There are approximately 12 main courses of generalized operator actions to lead the plant to the final safety state even if most of plant safety systems are actuated automatically by the sensing of initiating conditions (Table 3). These generalized operator actions are also considered as the basis for the identification of information grouping associated with РАМ parameters since for any operator action, only specific parameters may be required to monitor the accident trends. It is noted that not all instruments are required to be environmentally and seismically qualified for all events but are qualified for each credited event to ensure its functional capability following each of the design basis events. 4. РАМ Design Implementation As discussed above for the selection of РАМ parameters, the main design factors are how to identify parameters properly by the indication of the conditions that pose a threat to the integrity of fuel sheath, primary heat transport system and containment following postulated accidents so that the operator would be able to adequately verify reactor shutdown, fuel cooling and heat removal, and containment radioactivity. The reactor shutdown states are indicated by the reactor power measurements by means of in-core and out-core flux detectors. Fuel sheath failures are prevented by maintaining adequate fuel cooling which depends on the reactor power, the circulation of the coolant and the heat sink. The effectiveness of coolant circulation and heat removal is contingent on the operating condition of the primary heat transport system and verified via coolant pressure and temperature. The effectiveness of the heat sink would be mainly verified by the steam generator level and pressure or alternately by the shutdown cooling system when used for primary heat removal. The integrity of the primary heat transport system boundary can fail as a result of overpressure, overheating or metallurgical degradation from postulated ' initiating events. The critical РАМ parameters used to verify fuel cooling and heat removal, the integrity of primary heat transport system are the coolant pressure and temperature. Metallurgical degradation is usually a much longer term effect and should be treated as an operational issue. The reactor building pressure and radioactivity parameters are used to button up the containment envelope. Radioactivity in the effluent streams reveals containment leakage or bypass. Monitoring of containment isolation status provides additional assurance of containment integrity. In summary, the main РАМ parameter sets are selected to verify the safety functions following postulated events via systematic coverage as derived from the РАМ event sequences (see Table 1.) : - Verification of reactor power. - Verification of primary heat transport system pressure and temperature, - Verification of steam generator pressure and level, - Verification of reactor building pressure and radioactivity. РАМ design implementation consists of identification and verification of the selected parameters as required by applicable codes and standards. As already discussed, the design basis events and event sequences are categorized by the safety analysis design group and the identification of selected parameters are verified in order to make maximum utilization of existing instrument loops. These loops consist of signal sensing devices, signal transmitters (or signal conditioning devices), and indicators (panel meters) or display devices (computer CRT display) available in main control room and/ or secondary control area. Each instrument loop is identified with the specified safety function of the instruments, the specified ranges for the instruments, and the specified postulated design basis scenarios and mission times for instruments and to which they must be qualified. The channelization of the information chains are finally examined with care to provide redundancy so that single component failure of each instrument will not deprive the operator of the РАМ information. As a minimum, physical and functional separation are maintained between two redundant channels and if possible, diverse parameters for redundancy are preferred. Operator-Machine interfaces are also part of design implementation of the РАМ design. Colour-coded bezel of indicators (or display devices) around the meter in the main control room provide a contrast in appearance. Where the separate РАМ indicators are adjacent to each other, a colour-coded group bezel with cutouts are provided around the indicators. Several plant communication systems such as the telephone and paging telephone system are designated to provide the main control room personnel with access in a diverse and redundant fashion to the various relevant plant and external personnel to enable actions requested b6y the operator to be carried out. The plant control computer, called as "Digital Control Computer", which mainly performs the plant control function as well as the display functions, provide the means to improve the Operator-Machine interfaces. Information is displayed on the operator console utilizing colour CRT screens. The computers can be interacted with by depressing either a dedicated function pushbutton or by using the "call any function" pushbutton so that an interrupt in the computer can execute the operator's request for the following functions: - call up a display, - revise some attribute of display, such as modify bar chart or trend display, - enter numeric information. - cancel entry or request, - making hard copy of any information. In the event that the main control room becomes uninhabitable due to such events as earthquake, smoke, fire, toxic gases, flooding, etc., adequate special safety back-up parameters are also displayed in the secondary control area to allow plant monitoring. H* 5. Operation of РАМ Under normal operation when all information chains are functioning properly, the operator can compare the redundant indications of each parameters. In the post accident situation, the appropriate emergency procedures would be followed for operators to take action for maintaining the plant in a safe state. The РАМ indications must assist the operators in minimizing possible plant damage, and to allow the operator to return the plant to the normal state. In case of degraded operation, when one or more information chains are impaired, the readout from the failed chain will usually be irrational. If a discrepancy between two readings occurs, the operator could test the information chains or compare the readings with related variables to determine which one is correct. The maintenance action would be undertaken to repair the defective chains. In case of both plant computers failure (one is normal and the other is back-up operation), the fail-safe action would result in the loss of one of the redundant chains for РАМ parameters. For this eventually a redundant analog meter indication is provided as backup. 6. Conclusion From the above discussions, the РАМ design implementation on Wolsong 2.3&4 is weH established from the point of view of the design basis, approaf operational requirements, and the applicable codes and standards. This РАМ system, "v .st provide information which enables the operator to have an increasing role in managing the outcome of the event by continuously monitoring the plant critical safety parameters as the event progresses. At the start of an accident, it may be difncult for the operator to determine immediately what accident has occurred or is occurring, and therefore to détermine the appropriate response. For this reason, plant safety systems are designed to perform automatically during the initial stages of any accident. Judging from various aspects of design implementation, the post-accident monitoring(PAM) system for Wolsong 2.3&4 is considered to effectively incorporate basic concepts of the applicable codes and standards, and to have an important role which enables the operation of manually initiated safety systems and other appropriate operator actions involved with systems important to safety. 7. References 1) 86-68930-DM-001. Design Manual for Post Accident Monitoring System of Wolsong NPP 2.3&4 2) CSA CAN3-N290.6-M82. "Requirements for Monitoring and Display of CANDU Nuclear Power Plant Stations in the Event of an Accident", National Standard of Canada 3) Wolsong Nuclear Power Plant Units No. 2/3/4, Final Safety Analysis Report (VOL. 5), 1995 May. Korea Electric Power Corporation in Table 1. Examples of РАМ Variables and Safety Functions 1 VARIABLES A D E В j с Primary Heat Transport Coolant System X Pressure Primary Heat Transport Coolant System j x i Flow 1 i Primary Heat Transport Coolant System X Channel Outlet Temperature Primary Heat Transport Coolant System X Pressurizer Level Primary Heat Transport Coolant System X Storage Tank Level Neutron Flux X Neutron Flux Rate of Change X Shutdown System 1 Status X Shutdown System 2 Status X Containment Pressure X Containment Temperature X X Containment Activity X Containment Isolation Status 1x Containment Hydrogen Concentration X X Containment Dousing Tank Level 1 X Emergency Coolant Injection Heat X Exchanger Outlet Temperature Emergency Coolant Injection Sump Level X Emergency Coolant Injection Pressure X and Flow Moderator Flow X Moderator Temperature X Steam Generator Pressure X Steam Generator Water Level X Deaerator Storage Tank Water Level X Steam Generator Steam Relief Valve X Status Emergency Water Supply Tank Level and X Emergency Power Supply Status i 1 Standby Electric Generator Status X X ! Shutdown Cooling System Status X X I Where A : Verification of reactor shutdown В : Verification of reactor heat removal С : Verification of a barrier to radioactivity release D : Evaluation of radiological conditions Б : Assistance in carrying out recovery action. 1 X\ Table 2. Event Sequence for Large LOCA Time after Iniüating EVENTS Event 0-2 s Power pulse Sharp increase in R/B Pressure [65 'kPa(g)] SDS(s) activated Dousing spray system starts [14 kPa(g)] Containment isolation 2-10 s Fission power decays below 10 %FP Fuel damage occurs PT/CT contact 10-900 s Loop isol. activated - feed/bleed, pressurizer isolated LOCA signal activated - HTS pressure < 5.5 MPa(g) HPECC - Pressurize accumulators Open injection valves Open isolation valves MPECC - Open dousing tank isolation valves Open MPECC injection valves Start ECCS pumps Close HPECC valves when accumulator is empty Class III standby generator started by ECCCS SG crashcool activated HTS pumps tripped R/B pressure decrease 15 min - LPECC 1 day R/B cooldd by local air coolers Post-LOCA Instrument Air started by Operator > 1 day Post accident R/B depressurization (controlled release to environment) Table 3. List of Generalized Operator Actions No. Operator Actions 1 Start EWS and EPS after LOCA/SDE or DBE 2 Establish Heat Sink (ECCS, SG, Moderator) 3 Open MSSVs after MSLB (or DBE) 4 Close Instrument Air to R/B 5 Start Shutdown Cooling System 6 Start D2O Recovery System Dryers .7 Control Depressurization of Containment 8 Use HPECC to Break Rupture ECCS Discs for EWS after DBE 9 Initiate Controlled SG Cooldown 10 Terminate EWS Make-up Mode and Initiate Recirculation Mode 11 Maintain SG Level Manually 12 Provide EWS Flow to ECCS Heat Exchangers for LOCA/SDE When in EWS Mode I U3 ELECTRIC CHARACTERISTICS OF NUCLEAR POWER PLANT CABLES AT ACCIDENT SIMULATION Rogov V.l., Ulimnv V.N., Filatov N.I., Shestacov V.S. Reseach Institute of Sointifio Instruments (RISI) of RF Ministry of Atomic Energy is the Centre of accelerated radiation tests of Nuclear Power Plant (NPP) electrotechnical materials, components and equipments. The main objectives of the Centre: 1. Carrying out of research works and development of electric insulation aging models at long-term simultaneous exposure of radiation and NPP operating factors (temperaure, humidity, pressure, electrical and mechanical stresses and etc.). 2. Development and certification of accelerated test tecniques, qualification test performance of electrotechnical products and assessment of its 1 if time at NFP opeating and accident conditions. 3. Creation of technical state nondestructiv control methods and monitoring means for assessment of electric insulation aging degree and residual life. 4. Accumulation and analysis of investigation and test results, arrangement of Data Bank on radiation resistance and reliability of electrotechnical products and materials for NPP. Creation of complex automated system for prediction of service life and residual life of NFP electrotechnical products. 5. Development of standads and instructions on establishment of general requirements for NPP electric components resistance to radiation and operating stresses; on methodology of qualification tests and lifetime assessment, monitoring and electrotechnical product residual life. The experimental base of Centre includes the steady-state gamma-neutron reactor, gamma-radiation isotopic sources and climatic chamber complex enabling to carry out tests and investigations under irradiation with dose rates of 1 to 100D0 Gy/h under exposure to constant and cyclic operating factors in a wide range of temperature, humidity, pressure, voltage and etc. Centre's technique permits to realize complex diagnostics of actual state of electrotechnical products and construction materials during investigations and tests by measuring: - electric characteristics: insulation resistance (IR), dielectric strength and partial discharge parameters, capacity (dielectric constant) and angle tangent of dielectric losses; - mechanical properties: tensile strenght and relative elongation at break, resistance to bending and vibration; - structural parameters by methods of electron-positron annihilation, differential scanning microcalorimetry, infra-red spectroscopy, differential thermal arid thermogravimetric analysis. The Centre's research programmes are carried out in cooperation with leading institutes developing electrotechnical products, institutes of KF Academy of Sciences, Nuclear Power Plants, Rosenergoatom Concern and Gosftomnadzor organisations. \ хв - Л ~ RISI is currently conducting: research on representative samples of NPP cables bouth during; long"-term aging and accident simulation, since insulated electrical cables are used to provide instrumentation signals, power, or control to virtually all remotely operated power plant equipment. The objectives of this program are to determine the suitability of these cables for extended life (beyond 40-year disign basis) and to assess various cable condition monitoring techniques for predicting; remaining- cable life. The cables are being aged for long times at relatively mild exposure conditions with various condition monitoring techniques being emploed during the aging process. Following the aging process, the cablfes are being exposed to a sequential accident profile consisting of high dose rate irradiation followed by simulated desing basis loss-of-coolant accident (LOCA) steam exposure. This paper presents some results of a study of NPP cables electrical characteristics at normal and accidental conditions. The polymeric cable insulating and jacket materials include cross-linked polyethylene (XLFE), polyvinyl chloride (PVC) and ethylene-propylene rubber (EPR). XLPE During long - term aging electrical characteristics of these cables (Figure 1) change as a function of the irradiation dose, the dose rate and the temperature. The higher the temperature and lower the dose rate, the greater the degradation for a given dose. Insulation resistance decreases at the LOCA simulation (Figure 2). The higher the agind irradiation dose, the greater the degradation of IR at the LOCA. For an equal aging irradiation dose, degradation after LOCA is higher for aged cables at lower dose rates. PVC and EPR During long-term aging electrical characteristics of these cables (Figure 3,4) change as a function of the irradiation dose and independ of the dose rate and the temperature. Jacket and insulation crack at the LOCA simulation (Figure 5,6). Insulation resistance decrease lower of the limit IR. OR, Ohm-m 101A\ I OCA and POST LOCA p^oflies T 1 i I I !_ J I L 0 2 А 6 qtfi 0 2. A 6 & iO 12 H i,dQLjb IRtQhm-m ' о ° о о о o-Q-o его 10* Q -Э s О. 1011 г Q i 7 У / <у iO1С Сое ее - VP Я Jnsuêation - XLP£ о - D =0 . г i о - D - 3'JO 6vatJ00°C 10* s о - D -5 40 64 at m*e 10* / 4 / iO7 Limit Iß Г7777 III11/ I / / / / / / / /// J L_ J 1 I 1 I ! L Q ci 6 û 6 â 10 iZ 14 1 \ 3L* JR, Qhm Caèêe - KW& JnsulatCon - PvC и iO 10 о - 0,5 6yfS at 25°C - O.Q Gy/S at ês°û - QM 6 u/S at 25°C - 0,04 &y/5 at 65°e -0.04 ôy/s at i00°0 iff 10 10'' U1 Caßfe - KA/R • Jnsu?a icon - EPR A- э iö1 Д Оо io' О - 0,5 G.,/,5 at г5°С » - 0.2 б у/S at 20ö£ • - 0,04 P,Bazs T • it t * li i Bt,hz 0 £ 4 £ & 10 12 14 i^oijb 77 Caêfe - KVVG Jûsutat Con - РУС о - J) -Û 2>--ö'lOA 6g at 100°C V limit JR -ггтттттгтт О О A • 9 9 • ' ' ' ' I i 'it' 6 s Uz о г a 6 • Ô lo ie u 4jayS LOCA and POST LOCA peofites P,8QZS I I » » I 0 2 4 6 8 £thz 0 S 4 6 & 10 1£ 14 todays 3R, Ohm-m 1D 10 о Caêie - К A/M о Jnsutatùor} - В PR io- о - D =• О л • - v =3 40*6% oit dOO°C 10* во •7m- о 105 -Po- л / \• 10* __l L t ! I i t i t I L 0 £ 4 6 &l,hi 0 3. 4 в 8 10 it Щ ±ßoy% \ ъ 3 Director Директор Иванов 8.Б., Prof. Valentin B. Ivanov доктор технических наук, corresponding member of ATS RF член-корреспондент ÄTH РФ HISTORICAL REFERENCE ИСТОРИЧЕСКАЯ СПРАВКА In 1956 in accordance with the decision of the Soviet Government the В 1956 г. по решению Советского Правительства было начато строи construction of RIAR was started. The aim of RIAR creation is engi тельство НИИАР. Цель создания института - инженерные и научные ис neering and scientific investigations in the field of atomic energy. следования в области атомной энергии. Firstling of RIAR is the intermediate neutron SM-2 reactor with ther Первенец НИИАР - реактор СМ-2 с потоком тепловых нейтронов mal neutron flux of 5»1015 cnvz s-1. It was put into operation in 5 • 1015 см2 er1 вступил в строй действующих в октябре 1961 г. С целью October, 1961. To ensure safety and to meet demands of повышения безопасности и приведения в соответствие с требованиями Gosatomnadsor, in 1991—1992 the reconstruction of the reactor Госатомнадзора в 1991-1992 гг. была проведена реконструкция реак SM was undertaken. торной установки СМ-2. The following reactors were subsequently commissioned: ARGUS Последовательно вступали в строй реакторы: АРБУС (после реконструк ции АСТ-1), ВК-50, МИР, в декабре 1969 г. - пуск реактора БОР-60, в (after redesign of AST-1), VK-50, MIR, in Decerr.der, 1969 — BOR- 1975 г. - РБТ-6, в 1983 г. - РБТ-10/1 и в 1984 г. - РБТ-10/2. 60, in 1975 — RBT-6, in 1983 — RBT-10/1 and in 1984 — RBT-1072. В январе 1964 г. введен в эксплуатацию материаловедческий отдел, а In January, 1964 the Material Science Department was established в 1965 г. - радиохимический. Оба эти отдела имеют прекрасно обору and in 1965 — the Radiochemical one. Both these departments дованные радиационно-защитные камеры (защитные камеры), позво possess well-equipped radiation-shielded chambers (shielded ляющие работать с активностью до 100 тыс. кюри. Институт располо chambers), allowing operation with activity up to 100,000 Ci. The жен в 6 км от жилого массива и занимает площадь 17 кв. км. Institute is situated 6 km from dwelling houses and its area is 17 square km. ЭКСПЕРИМЕНТАЛЬНАЯ БАЗА ИНСТИТУТА включает в себя: ис следовательский реактор СМ тепловой мощностью 100 МВт и макси EXPERIMENTAL BASE OF THE INSTITUTE includesмально: the high-fluй плотностьx ю потока нейтронов (с энергией > 0.1 МэВ) 2 • 1015см- research SM reactor of thermal power 100 1Ш and maximum neu 2с1; петлевой испытательный реактор МИР тепловой мощностью до 100 tron flux density (E>0.1 MeV)2-10«cm-2s-1; the loop testing MIR МВт и максимальной плотностью потока тепловых нейтронов reactor of thermal power up to 100 MW and maximum neutron flux 5 -1014 см-2^1 ; три реактора серии РБТ (бассейнового типа, топливо - density 5'101«cnr2s-1; three reactors of the RBT type (pool type, отработавшее в реакторе СМ); опытную энергетическую установку ВК- spent fuel of the SM reactor); the experimental power station with 50 с корпусным реактором кипящего типа (теплоноситель - вода) элек the WK-50 boiling type water cooled/water moderated reactor of 50 трической мощностью 50 МВт; опытную энергетическую установку MW and the BOR-60 sodium cooled fast breeder reactor of 12 MW; БОР-60 с реактором на быстрых нейтронах и жидкометаллическим на design division; complex of "hot" material science laboratories for триевым теплоносителем электрической мощностью 12 МВт; конструк торский отдел; комплекс "горячих* материаловёдческих лабораторий investigation of the BN and WER irradiated fuel elements and fuel для исследования облученных твэлов и TBC реакторов типа БЫ, ВВЗР и assemblies; experimental workshops for production of unique, исследовательских; опытно-экспериментальный цех по изготовлению non-standard equipment used in reactors, experimental plants and уникального, нестандартного оборудования для реакторов, опытных ус facilities. тановок и устройств для проведения исследований в институте. BASIC TRENDS OF INSTITUTE ACTIVITIES ОСНОВНЫЕ НАПРАВЛЕНИЯ ДЕЯТЕЛЬНОСТИ НИИАР Reactor Material Science and Methods for NuclearРеакторное Power Plant материаловедение и методики испытания материалов (NPP) Materials and Elements Testing incorporate:и developmen элементовt of ядерных энергетических установок (ЯЭУ), включаю theoretical basis for reactor material science methods; methodical щие: разработку теоретических основ реакторных методов материаловеде support, software and hardware of material science investigations, ния; методическое и аппаратурное обеспечение материаловёдческих ис design and manufacture of devices for testing, metrological certifi следований; проектирование и изготовление устройств для испытаний; ме cation of systems applied for real measurements; acquisition of трологическую аттестацию систем, применяемых для реальных изменений; data on physical-mechanical properties of irradiated materials получение данных о физико-механических свойствах облученных материа under different loading character and effect of factors modelling лов при различном характере нагружения и воздействии факторов, модели operation conditions; vessel structural material investigation, in- рующих условия эксплуатации; исследования конструкционных материалов vessel devices, the WER and BN fuel element claddings and fuel прочных корпусов, внутрикорпусных устройств, оболочек твэлов и чехлов assembly jackets; fuel absorbing materials, moderators and reflec TBC реакторов ВВЭР и БН; топливные, поглощающие материалы, замедли tors, TRI) pure metals and their alloys, ceramics for nuclear and тели и отражатели, трансурановые чистые металлы и их сплавы, керамика fusion power plants. для ядерных и термоядерных энергетических установок. Физико-технические проблемы ядерных реакторов и вопросы безопасности, включающие: физические, теплофизические, теп- лопедравлические и прочностные расчеты реакторов в обеспечение ре конструкции действующих реакторов института, разработок и предло жений по новым реакторам; получение экспериментальных данных по физике, теплофизике, теплогидравлике, выходу и распространению продуктов деления, поведению материалов твэлов и TBC, параметрам химических взаимодействий, необходимых для верификации расчетных программ и обоснования безопасности действующих и проектируемых ЯЭУ и исследовательских реакторов; разработку методов и технических средств для исследований TBC, твэлов и их фрагментов в аварийных условиях в реакторах и защитных камерах. Получение радионуклидов и изделий из них, а именно: научные и технические вопросы технологии реакторного производства радиону клидов; исследование свойств трансплутониевых элементов (ТПЭ) в обоснование технологии их выделения и изготовления источников; ме трологическое обеспечение, паспортизация источников и препаратов, пооперационный контроль технологического процесса, автоматизация егп; получение актинидов в металлическом состоянии, исследование их свойств применительно к изготовлению источников; радиометрический и масс-спектрометрический методы анализа образцов радионуклидных Physical and technical issues of nuclear reactors and safety препаратов и источников. aspects inccrporate: physical, thermal-physical, thermal-hydraulic and strength reactor calculations to provide reconstruction of the Ядерный топливный цикл, включающий: электрохимический про RIAR operating reactors, developments and proposals for innova цесс переработки облученного топлива и получения гранулированных tive reactor concepts; acquisition of experimental data on physics, уран-плутониевых оксидных композиций; разработку конструкций и thermal physics, thermal-hydraulics, fission product release and технологии изготовления твэлов и TBC методом виброуплотнения; ме propagation, fuel element and assembly behaviour, chemical inter тодическое и аналитическое обеспечение процесса переработки и пас action parameters which are necessary for code verification and портизации топлива; создание, испытание и эксплуатацию опытных ус also for safety validation of the operating and designed NPP and тановок по переработке и подготовке гранулированного топлива; сме research reactors; development of methods and technical means шивающие и дозирующие установки и устройства; трансмугацию мино- for fuel element and assembly investigations under emergency con рактинидов и долгоживущих радионуклидов деления; короткий топлив ditions in reactors and in hot cells; development and testing of tech ный цикл ядерных реакторов. nical means for NPP diagnostics and for its operating safety. Технические проблемы экологически чистых технологий, раз Production of radionuclides and radionuclide products includes: работки для других отраслей народного хозяйства включают в scientific and technological issues of reactor radionuclide produc себя: разработку и внедрение технологий и устройств для интенсифи tion; investigation of TPE properties for technology validation of кации процессов смешивания, разделения, дегазации сред с различны their extraction and source manufacture; metrological provision, ми физико-химическими свойствами; разработку и внедрение экологи source and preparation certifications, step-by-step operation moni чески чистых технологий очистки различных технологических отходов toring of technological process, its automation; production of metal производств (газов, жидкостей, теплоносителей) и поверхностей обору lic actinides, their property investigation as applied to source pro дования от отложений; разработку и изготовление различных первич duction; radiometric and mass- spectrometry methods for analysis ных преобразователей (термопар, датчиков, измерительных устройств). of radionuclide preparation and source specimens. Radionuclide sources for technical and medical application Nuclear fuel cycle includes: electrochemical process of irradiated fuel reprocessing and production of uranium-plutonium oxide gran ulated compositions; development of constructions and technolo gies of fuel element and assembly production by vibropac method; methodical and analytical support of fuel reprocessing and certifi cation; creation, testing and operation of pilot plants for reprocess ing and preparation of granulated fuel; mixing and measuring facil ities and devices; transmutation of actinide and long-lived fission radio-nuclides; short-term nuclear reactor fuel cycle. Technical issues of ecologically safe technologies, develop ments for other branches of national economy involve: develop ment and introduction of technologies and devices to intensify the processes of mixing, separation and degassing of. mediums with different physical and chemical features; development and adoption of ecologically pure technologies for decontamination of technological wastes (gas, liquid, cooiant) and equipment sur faces from deposits; development and production of different pri mary transducers (thermocouples, detectors, measuring devices). Радионуклидные источники для технических и. медицинских целей 433510 Dimitrovgrad-10, Ulyanovsk region, Russia, Telephones (84235) 32021,32727,36620 Telex 263854 VELA RU Fax (84235)35648 E-mail: [email protected] Director Valentin B. Ivanov 35280 Deputy director on science Vladimir A. Tzykanov 32158 Chief engineer Alexey F. Grachev 32530 433510, Димитровград-10, Ульяновская обл., Россия Телефоны (84235) 32021,32727, 36620 Телекс 263854 VELA RU Телетайп 263711 "ЧАЙКА" Факс (84235)35648 E-mail: [email protected] Директор института Иванов Валентин Борисович 35280 Первый заместитель директора Цыканов Владимир Андреевич 32158 Главный инженер Грачев Алексей Фролович 32530. M РиЫ'ьПсй оу INTERSVIAZ 1 J> J Creating of requirements for control of postaccident situations at NPP V. S. Dickarev, V. S. Ionov INP RRC Kl Moscow, Russia Requirements of legal documents to control of postaccidents situations based at existing nuclear safety requirements and assumed for monitoring of situation and control of postaccident actions using of normal operation equipment and safety systems, supposed by design of NPP. That means are corresponded to requirements for design accidents and unnormed accidents, when violated limits and conditions for design accidents. Presents main topics and problems which arise during creating of regulatory requirements for post accident situations at NPP, for example: 1) classification of plant conditions by consequence severity, possibility of sequences events and prevent of them, and possibility of return to normal operation or decommission; 2) determine of safety functions for conditions which not included existing requirements. The acceptance of the new safety functions allow prepare requirements to post accident monitoring, control situations and technical means for mitigation of accidents; 3) determine of requirements and criteria to using nonplant means for postaccident conditions, in particular to connection with damaged unit of plant and quality of safety function executions. That requirements allow to solve problem extreme requirements to safety systems and retard "safety system race"; 4) classification of severe accidents by - obviously symptoms of situation severity and of severe accidents for depository too; -normal operation equipment and safety system conditions (degree of possibility supply safety functions); - state and localization of radioactive materials and so on. This should help to prepare of severe accident list for addressing to design. 5) creating system analysis concept to postaccident situations for monitoring and controlling systems of facilities, preparing general for NPP principles of monitoring and requirements, and regiment requirements for analysis, monitoring and reduce of accident consequences of particular facility. I 3 * Principles Elaboration and Creation of Information-Analytical System "HI Operation Safety with SSC RIAR Research Reactors'*. Ivanov V.B., Grachev A.F. , Kinsky O.M., Макin P.S., Ochrimenko A.I., Demidov L.I., Karpjuk V.l., Afonin V.K., Iskanderov R.G- The State Scientific Centre of Russian Federation the Research Institute of Atomic Reactors Dimitrovgrad, Russia. Abstract In this paper an approach is described, which is accepted at elaboration and creation of computer-aided control system of technological process (CCS TP) at the installations with research reactors. The tasks and the main technological requirements to elaborated information-analytical system, are formulated, based on the accepted approach, experience of computer-aided systems and analysis of technological processes at reactor installations (RI) of SSC RIAR. The system includes the following installations: the SM-3; the VK-50. the RET-10, the 'BOR-60 and the MIR. Based on the given example there is a classification and the purposes of the modern system of information personnel support of research reactors are formulated as well as approaches to its creation, including creation of determined models of the processes, which are realised in simulators and statistic methods of time series. According to the accepted approaches the results of systematic- technical synthesis and modern states with system simulation are described. -2- 1. Introduction Nowadays a series of systems, which are united under the common name "Computer Operator Support System" (COSS) is elaborated. The main functions of COSS are determined by the international standard. They are representation of critical functions (CF) and safety parameters (SP). The first requirements to COSS are given in USA standards NUREG-0737, NUKEG-0696 after accident at the Atomic Power Plant "Tree Mile Island-2". They are called SPDS. In many,, countries SPDS is used as a system of information about normal and emergency modes of NPP power blocks work in the national emergency Centers. The example of such a system can be SICOEM. The similar work is carried out in Rassia in frame of this approach. In measure plan, which is directed into the RI operation safety increase, the elaboration of actions in cases of emergency is foreseen. The system is based on the new symptom-oriented type of proceedings. This system is based on computer-aided estimation of safety critical functions state, other words, on SPDS base. Above given tasks determine the requirements to functional content of SPDS (fig.l). Information representation in compact form with minimum set of technological parameters is common in requirements to SPDS in the given systems. 2. The Main Technical Requirements to SPDS Performance. Existing system of branch safety control from the point of view of acceptance-transfer of information is a successive structure (fig.2). The main requirements to SPDS computer complex from the control system in cases of emergency are based on accepted safety principles (Table 1). These principles contain requirements of the main norm documents of Russia Gosatomnadsor 0ПБ-88 and ПБЯ-89. Based on the analysis of control room (CR) operating personnel work, the following requirements to functional characteristics of SPDS can be presented. (Table 2). Above mentioned technical requirements to SPDS performance are used at suggestion elaboration of information analytical system (IAS) creation under the name " Operation Safety of RI with RIAR Research Reactors". Besides, the following was taken into account: -3- - RI with research reactors specification in comparison with NPP; - Creation experience of computer-aided system of information collecting, registration and representation at reactor installations; - accepted method of researches in computer-aided systems development at the reactor installations. 3. Brief Characteristics of Automatics Objects and-the Main Principles of IAS Creation. All Institute .reactor installations are neutron resources and intended to realize the appropriate experimental researches. RI VK-50 and B0R-60 are intended, in addition, to elaborate electric power and heat. RI VK-50 and BOR-60 have the intermediate position between installations with research reactors (SM-3, RBT, MIR) and industrial APP. At that, installation VK-50 is closer to commercial NPP with reactors WER. In 1985 the BOR-60 and the VK-50 started CSRD operating computer-aided scientific researches system (CSRS).They were the first in the field. Further CSRS of the RBT-10 SM-3 and water loop of the MIR started operating. Above mentioned GSRS are build on different technical means complexes (TMC) of this and other countries productions and differ from each other in composition of common systems and application software. But they fulfil functions of Institute scientific subdivisions information providing and contain the elements of computer-aided technological control. The following development of CSRS at reactor installations is realized in two directions: - information providing of operation personnel. Institute and field authoritj»- about different aspects of RI operation; - elaboration of partial RI control transfer to computer means of CSRS. The first direction is determined by the necessity to meet the requirements to CSRS performance and to Institute including into the given structure of RI safety control system (fig.3). The second direction is determined by the necessity of effective approaches elaboration to realise the first direction. The final aim of the second direction is transformation of CSRS into Computer-aided Control System of Technological Process ( CCS TP) at reactor installations with functions, given in table 3. Work at creation of reactor installations simulators take the particular place in researches. It is necessary to bring out an= -A— check the most reasonable control, as well as information representation. The last is important to meet COSS main requirement - minimisation of technological parameters set. Specification of RI with research reactors, CSRS creation experience and method of researches carrying out, which are directed to CCS-TP creation, are used at IAS elaboration "Safe Operation of RI with Research Reactors of SSC RIAR". 4. Purpose, Functions, Main Requirements and Composition of InfQrmation-analjtical System. As the level of CCS-TP elaboration of industrial NPP is much higher than of installations with research reactors, the succession of automatics objects investigation is the following: the VK-50, the BOR-60, the SM-3, the MIR, the RBT. Information-analytical system "Operation" unites 14 subsystems in present variant (table 4), which embrace the most part of questions of operative work, operation organisation repair service, connected with RI control aspects in routine work. A group of users of this system is defined with configuration of RI present information net work. Degree of access into acceptance or- input of information is defined by the volume of users" duties and can be controlled by different program means. The detailed content of the tasks, which are solved by the subsystems Analysis of the main technological process at the other Institute installations gives a clear picture of subsystems composition and their tasks 5. Modern State of IAS Elaboration. Math models of neutron-physical and heat-hidraulic processes in reactor were elaborated. Elaborated numeral models•are realised by application software of functional-analytical simulator model (the FAS of the VK-50), which is situated on the input stage (industrial operation date is the end of 1995). Standard configuration is used as a technical means complex of the FAT VK-50. The model of the FAS of the VK-50 should solve the problems of information support of CR subsystem and tasks of subsystem "Accident Rate" analysis. I 43 -5- Structural scheme of system technical means location and their connections is given on fig.4. Math models and imitators of. reactor installations are elaborated according to tbe schedule of functional—analytical and full-scope simulators equipment of SSC EIAR training center (TC). Conclusion. Above mentioned approaches and method of researches carrying out of IAS ,- "Operation Safety of RI with Research Reactors of SSC RIAR" creation lets hope for earlier realisation of safety control system of RI and research reactors. I UM -6- Control system of branch safety 1 ! i I Functional SPDS content System of actions in Information system cases of emergency of RI level (sympto-oriented proceedings) Fig.l. Forming of Requirements го SFDS Functional Providing. U5 -7- Existing Structure of Safety Control System. Operation proceedings RI information system. Operation regulation The main control board Permitted modes table RI control room Emergency proceedings CCS TP RI RI operating personnel Subjective perception Work experience Training Operating personnel; Information perception from RI personnel Work experience Training Branch experts Subjective perception Work experience Absence of operating commun i ca t i on Fig.2. The Structure of Information about Control Branch System Transfer. I и 6 -8- SPDS CorvLro l Ro<э т Enterprise Local ВГс inch Authority Crisis Centre Crisis Centre 1 I Local Branch Authority Authority Fig.3. The Given Structure of Control System of RI Safety in the Part of Operating Technological Information Transfer. I ЦТ- -9- Table 1. Safety Principles 1. Safety guarantees core operation, which are determined by absolute control of possible output out of permitted limits of neutron flow density and reserve till hea"t-exchange crisis. The following operation conditions should be realized: fi) heat removal from the coolant and construction elements of the first circuit should be realized with the help of the main and secondary systems of the second circuit in all modes, past emergency mode including. (ii) holding up of given reserve and coolant level in the main and secondary core cooling systems in limits, which are foreseen by operation proceedings. 2. Prevention of protective barrier elements destruction, including vessel, and as a result of it,, of possible radioactive products output out of their limits. -10- ТаЪ1е 2. Principle Realization of Minimum Quality of Parameters. (i) It is necessary to use hierarchy structure of system creation of critical functions and safety parameters. (ii) Information presentation is subdivided into levels according to the principle, in which general state ox RI safety can bè divided into the assessment, according to the critical functions. (iii) Parameters control, connected with reactivity, with purpose ' of necessary reactivity providing in any moment for reliable stop of fission reaction and reactor holding in subcritical state with given reserve of subcriticality. (iv) The given degree of fuel assembly cooling with the help of coolant with purpose of preventing of their features change from the point of view of shell embrittlement, fuel elements, as well as change of fuel state of aggregation. (iiv) Every safety function can contain parameters, according to which they judge about the reasons of critical function status change. (xi) The user should have opportunity to look through the new parameters, which characterise general safety and dynamic of their changes, perhaps, with forecast of these values. If it is necessary to have detailed information about expanded set of pax"ameters and systems, connected with safety, the possibility of transfer to the second and third hierarchy levels should be provided. -11- Table 3. CCS TP Functions: - Collection and processing of information about technology state; - Information about different kinds of signaling, registration, etc. ; - Organization of anti-emergency work as well as protection of technological object from damage; - Regulation of. processes in the given modes; - Logic-program control of object executive devices in dependence on current state of technology; - Optimisation of the process according to the given criteria; - Technical-economical figures account of the technology; - Technical diagnosis of separate parts and a whole system; - Operating connection with upper levels of control system hierarchy. [ Ъо -12- Table 4. Composition of the Main Subsystems and their Tasks. 1. CR subsystem (information tasks, and analytical tasks, personnel support tasks). 2. "Current Repair" subsystem (information tasks, analytical tasks). 3. "Technical-economical Figures" (information, tasks; analytical tasks, personnel support tasks). 4. "Report" subsystem (information tasks, analytical tasks). 5. "Accident .Rate" subsystem (information tasks, analytical tasks). 6. "Operation Documents" subsystem (information tasks, analytical tasks, personnel support tasks). 7. "Personnel" subsystem (information tasks, analytical tasks). 8. "Safety" subsystem (information tasks, analytical tasks, personnel support tasks). 9. "Faults" subsystem (information tasks, analytical tasks, personnel support tasks). 10. "Radiation Safety" subsystem (information tasks, analytical tasks). 11. " Safety" subsystem (information tasks, analytical tasks, personnel support tasks). 12. "Equipment Reliability Control" subsystem (information tasks, analytical tasks, personnel support tasks). 13. "Registration and Control of Repair" subsystem l'information tasks, analytical tasks). 14. "Experiments at RI" subsystem (information tasks, analytical 'tasks, science researches support tasks. \ 5) -12- Reference 1.Functional criteria for Emergency response facilities. NUREG- 0696, 1981. 2.Johnson S. DASS - A Decision Aid Integrating Safety Parameter Display System and Emergency Functional Recovery Procedures. - EPRY-NP-3595, EPRY, 1984. 3.Felkel L/ The STAR Concept, System to Assist the Operator During Abnormal Events. - Atomenergie, 1984, bd 45, N 4, p«.252. 4-Fukumoto S. . Naito N., Takisawa Y. An Integrated Operator Decision Aid System for Boiling Water Reactor Power Plant. - Nuclear Technology, v.99, Jul, 1992, p.120-132. 5.NUREG-1342/ Lapinsky В., Eckenrode R. , Gootman P., Correia R. A Status Report Regarding Industry Implementation of Safets' Parameter Display Systems. - U.S. Nuclear Regulatory Commission, 1992. 6.Little J., Wooods D. A design Methodology for the Man-Machine Interface for Nuclear Power Plant Emergency Response Facilities - WCAP-1987. 7.Acero M., Martin A., Villota C. Emergency Response Data System Linking the Spanish Nuclear Plants and the Nuclear Safety Goncil (CSN), Tokio, 1992. S.Shachrezy E. , Grof I. Calculating System EC-1011. 9-Kish P. Microprocessor of Videoton Plant for Technology Control System / Devices and Control System, 1980. Edition 8. lO.Gergey I. Computer EC—1011 Software / Devices Control System. 1980. Edition 8. ll.Leschenko U.I et.all. In-pile Control System of Energy Extraction with Mobile Detectors of Direct Loading. Preprint RIAR-2K380), Dimitrovgrad, 1979. 12.Sadulin V.P. et. all. Energy Release Account in Boiling Water Reactor on Indication Base of Mobile Rodium Detectors of Direct Loading. 13.Karacuba U.M. Kinetics Model of Direct Loading Rodium Detector in Two-group Neutron Field and Algorithm of Flew Density Restoration. Preprint RIAR-23(431), Dimitrovgrad, 1980. 14.Mitelman M.G. et all. In-pile Energy Release Measurement Detectors. M: Atomizdat, 1977. 15.0smachkin V.S., Borisov V.D. Hydravlic Resistance of fuel rods in the Boiling Water Flew. Preprint IAE—1957, Moscow, 1977. 16.Ayvazjan C.A. Statistic Investigation of Dependence. MrMetallurgy, 1968. 17.Ivanov V.B. et. all Application Software of Regression Analysis in Conditions of Small Statistic System of Emergency parameters Registration of Reactor Installations with Research Reactor VK-50. Report in Present selection Dimitrovgrad, 1995. I 5 J IAEA Specialists' Meeting on "Instrumentation and Equipment for Monitoring and Controlling NPP Post-Accident Situations" Dimitrovgrad, Russian Federation, 12-15 September 1995. RADIOLOGICAL CONSEQUENCES COMPUTER CODE AS EMERGENCY OPERATOR HELP Ш CASE OF FISSION PRODUCTS RELEASE DURING REACTOR ACCIDENT Kizin V.D., Efarov S.A., Shkokov E.L, Yakshin E.K., State Scientific Centre Research Institute of Atomic Reactors, Dimitrovgrad, Russian Federation Seven research nuclear reactors are situated on the territory of SSC RIAR. Permanent work is carried out to improve the safety of reactor operation and to correct somehow actions of radiation accidents. This report is dedicated to ideology of the computer code, the main purpose of which is the prediction of radiation consequences of fission products release into atmosphere. The principles of mode don't depend on type of reactors. Therefore, it is not necessary to speak about physical characteristics and scenarios of supposed accidents. It is important to have knowledge about range of parameters values, which are important for consequences. On agree with existent possibilities first stage of work is only carried out now. One may propose the complex code, which is to control all accident processes from beginning to end ( from initial event up to radiation damage ). Design of such integrated code is planned in future. Authors of the report believe, as the first step it is useful to have automatic operator code, which would base on real characteristics weather and fission product release into atmosphere, which are measured continuously. Such code is described in the report. Institute's nuclear reactors ( except BOR-60 ) use light water for cooling their cores. It is necessary to note, that radiation harm will be define gaseous fission products ( GFP ) and volatile compounds of iodine until core is not damaged. In case emergency at BOR-60 a large quantity of high activity sodium aerosols may be released. However in accordance with the results of calculations, the influence of radioactive sodium aerosols will not be essential. Such tape of accidents has been taken into account in reactors project, and protective measure had been provided. The value of discharge activity depends on leakages of coolant, activity concentration, failed fuel rods number, leakage from the boxes. Respective date for nuclear reactors are given in table 1. Table 1. \ 5^ Some Characteristics of SRIAR's nuclear reactors Parameter Name of reactors VK-50 BOR-60 MIR-1M SM-2 RBT-10 Power, MW 170 60 100 100 10 Coolant РЬО Na H20 H20 H20 Coolant volume, 50 25 76 45 67 m Pressure, MPa 7 <0.5 5 0.18 Température, °C 274 320-520 83 86 60 Volume of germetic 4500 SOO 2000 22520 4S0 boxes, m3 Activity concentra 6.29-107 5.5-101° 7.03-107 7.4-108 6.66-106 tion in emergency boxes, Bq/m3 If the core is damaged, the gaseous and volatile fission products will determine the primary radiation harm also. Long time radiation consequences will be determined by Cs, Sr, Ru, Ce, Ba, Pu and other radionuclides. The contribution of other radionuclide core components will be least of all. Such tape accidents are referred to overprojective ( severe ) accidents. Countermeasures are recommended in special (manual's ) guidance. They come into operation in case when the event project accident may become to severe accident. In case of severe accidents with core damage the most essential factors are the reactor power, duration of company ( work period ), corium temperature. Research reactors computers code that would calculate consequences of severe accidents including vapor explosion, core melting and core- concrete interaction are probably not designed. Therefore, our code do not take into account fission product release of such accidents. Presented code are destined for task, which independent of different reactor's peculiarity and accident scenario. Code is based on direct measurements of some physical values, that improves reliability of results. The algorithm ( procedure ) of calculation has a speedy decision, and all time of procedure is defined by initial date's input time. As a result the time of taking protective measures increases. That is just the aim of code and operator of institute central control office in case of the first accident features are emerged. The FP escape is possible out of emerged building ( if is has damages ) in a few seconds after accidents beginning. The FP release from high altitude stack may occur.in 1-3 minutes after accidents. The duration of accidental release may be going from 4 up to 10 hours. The time of FP expansion to settlements is from 0.15 to 1.5 hours in dependence on weather conditions. This time is that for taking measures to protect population. Model of code takes into account: I 5 5 - gaseous fission product released contents; - contents of iodine volatile compounds in discharge; - contents of fission product's aerosols in discharge; - escape of the radioactive air out of the building; - discharge of radioactive air out of ventilation stack; - weather conditions; - relief of the region; - structure of region economy; - contamination and use of food products; - secondary wind lifting. The results of code calculation: • at early stage of accident is: - external irradiation from the radioactive cloud; - internal irradiation as a result of radionuclides inhalation; - irradiation dose in settlements; - irradiation dose on the axis of the cloud track; - track picture of the cloud on the map; - picture of zones with different contamination; - recommendation to protect the population and choose action in contaminated zones; • at middle stage of accident is: - levels of surface contamination on cloud motion path; - surface contamination from secondary wind lifting; - use of contaminated vegetation by agriculture animals; - use of contaminated foods by people. The structural scheme of code is presented on Figure 1. CODE Constants Input data • weather category; • wind direction; • type of precipitation; • and etc. Operator Fig. 1. I 5 Я The picture shows, that code uses two data types : - constants; - initial input data. . The constants are: - characteristics of radionuclides; - coordinates of settlements; - data of region relief; - radionuclides source ( coordinates, geometric dimensions, temperature of released air, etc. ). The input data are: - value and duration of discharge; - wind direction; - type of precipitation; - precipitation intensity; - weather category; - dosimetric characteristic of emergency reactor. The integration users code environment allows: - to input quicly data in dialog mode; - to view and correct all input data; - to obtain a short information about emergence reactor, accident danger degree and etc.; - to carry out the calculation of radiation consequences; - to view the results of calculations in graphic ( maps, scheme ) and textual ( tables ) forms; - to output the results of calculations in form of specific certificate. A prognosis of radiological consequences, based on real activity measurements of radioactive discharge and surface contamination, does allow to short the spade-work period and increase the effectiveness of activities. Совещание специалистов МАГАТЭ "Программно-технические средства контроля и управления послеаварийными ситуациями на АЭС" г.Димитровград, Российская Федерация 12-15 сентября 1995 года Компьютерная программа поддержки оператора для оператив ного прогнозирования радиационных последствий аварий на АС с выбросом продуктов деления в атмосферу. Якшин Е.К., Ефаров С.А., Кизин В.Д., Шкоков Е.И., ГНЦ НИИАР г.Димитровград, Российская Федерация На территории ГНЦ НИИАР располагаются и работают семь исследовательских и опьггно-промьшшенньгх реакторных установок. Проводится ' систематическая работа по совершенствованию безопасности их эксплуатации и уточнению действий при радиацион ных авариях. Настоящий доклад посвящен идеологии ( принципам ) компьютерной программы, главной целью которой является прогнозирование ( предсказание ) радиационных последствий в случае выхода продуктов деления в атмосферу. Принципы программы не зависят от типа реакторной установки. Поэтому нет необходимости говорить о физических характеристиках и сценариях предполагаемых аварий. Нужно лишь иметь представление об интервалах тех величин, которые имеют значение для последствий. В настоящее время выполнен только первый этап работы, который соответствует существующим возможностям. Можно представить себе более сложную "комплексную" программу, которая должна контролировать весь процесс развития аварии от начала до конца ( от начального отклонения технологических параметров до радиационного ущерба ). Составление такой программы или комплекса программ - дело будущего. Авторы доклада считают, что очень полезно, как минимум иметь автоматизированную программу для оператора, которая в ка честве исходных данных имеет непрерывно измеряемые характе ристики погоды и выбросов продуктов деления в атмосферу. Такая программа и докладывается. Реакторные установки института ( кроме БОР-60 ) охлаждаются водой, которая циркулирует по замкнутому контуру. Существенно то, что в случае аврии радиационный ущерб будет определятся газообразными продуктами деления ( ГПД ) и летучими соединениями йода, если активная зона не разрушена. И только при аварии на БОР-60 возможно выделение большого количества аэрозолей натрия. Однако, как показывает расчет, влияние аэрозолей натрия так же не является определяющим. Этот класс аварий учтен при проектировании и предусмотрены меры защиты. Количество выброшенной активности зависит от расхода протечек, концентрации активности, числа негерметичных твэлов, плотности помещений. Соответствующие сведения по реакторным установкам приведены в табл.1. Таблица 1. Характеристики реакторных установок ГНЦ НИИАР Параметры Реакторные установки ВК-50 БОР-60 МИР-1М СМ-2 РБТ-10 Мощность 170 60 100 100 10 тепловая, МВт гЬО Na Н20 Н 0 Теплоноситель н?о 2 Объем теплоноси 50 25 76 45 67 теля, м3 Давление, МПа 7 <0.5 5 0.18 Температура, °С 274 320-520 83 86 60 Объем герметич 4500 800 2000 22520 480 ной зоны, м3 Концентрация PB 6.29-107 5.5-1010 7.03-107 7.4-10s 6.66-106 в аварийном по мещении, Бк/м3 При разрушении активной зоны первоначальный радиационный ущерб также определяется упомянутыми выше продуктами деления. На долговременные последствия будут влиять радионуклиды Cs, Sr, Ru, Ce, Ba, Pu и др. Другие компоненты активной зоны не дадут заметного вклада в радиационное загрязнение. Все аварии такого рода отнесены к запроектным. Меры управления рекомендованы в инструкциях по управлению авариями. Они активизируются при угрозе перехода проектной аварии в запроектную. Для запроектных аварий с повреждением активной зоны самым существеным фактором является мощность реактора, длительность его работы, температура кориума. Для исследовательских реакторов нет лицензированных компьютерных кодов расчета последствий тяжелых аварий, сопровож даемых взрывом активной зоны, проплавлениями корпуса, взаимо действием расплава с бетоном. Предлагаемая программа поэтому также не учитывает последствия выхода продуктов деления в атмосферу при таких авриях. Рассматриваемая компьютерная программа решает задачу, не зависящую от множества особенностей разных реакторных установок и сценариев аварий. Она основана на прямых измерениях физических величин, что определяет ее достоверность. Заложенный в нее алгоритм решения практически мгновенно, а время процедуры зависит только от времени ввода исходных данных. В результате 1 6° высвобождается драгоценное время для принятия мер защиты населения. Это и есть главная задача программы и оператора центральной диспетчерской института при первых признаках аварии. Утечка радиоактивных продуктов из аварийных зданий может происходить спустя несколько секунд после начала аварии, и через 1-2 минуты - в высотную трубу. Длительность выброса при авариях до 10 часов. До населенных пунктов радиоактивный шлейф может двигаться от 0.15 до 1.5 часов ( в зависимости от погодных условий ). Это время является временем для принятия решений по выполнению мер по защите населения. Алгоритм программы учитывает: - состав ГПД в выбросе; - состав летучих соединений йода в выбросе; - состав нелетучих продуктов деления - аэрозолей в выбросе; - истечение радиоактивного воздуха в высотную трубу; - истечение радиоактивного воздуха из зданий; - погодные условия; - рельеф местности; - инфраструктуру местного хозяйства; - загрязнение и потребление пищевых продуктов; - вторичный ветровой перенос. Результатом расчета компьютерной программы являются: • на ранней стадии аварии: - внешнее облучение от радиоактивного облака; - внутреннее облучение от ингаляции радионуклидов; - доза облучения в населенных пунктах; - доза облучения на оси следа выброса в заданных точках; - изображение следа выброса на карте местности; - изображение зон с различными уровнями загрязнения; - рекомендации по защите населения и действиям в - зоне радиоактивного загрязнения; • на средней стадии аварии: - уровни загрязнения местности на пути движения облака; - загрязнение местности от вторичного ветрового переноса; - потребление загрязненной растительности с/х животными; - потребление загрязненных пищевых продуктов человеком. Структурная схема программы представлена на рис.1. \ 6 ПРОГРАММА Данные-константы • Вводимые данные • погодные условия; • параметры выброса; • тип осадков; • ДР- Оператор Рис.1 1 6X Как видно из рисунка программа опрерирует с двумя типами данных: - константы; - вводимые данные. Данными-константами программы являются: - характеристики радионуклидов; - координаты населенных пунктов; - данные о рельефе местности; - параметры источника выброса ( координаты, геометрические размеры, температура рабочей среды и т.д. ). Вводимыми данными программы являются: - величина и продолжительность выброса; - направление выброса; - скорость ветра; - тип осадков; - интенсивность осадков; - категория погоды; - дозиметрические характеристики аварийного реактора. Сервисная оболочка программы позволяет: - вводить информацию в диалоговом режиме в течение короткого промежутка времени; - просматривать и корректировать все введенные данные; - получать краткую справочную информацию об аварийном реакторе, о степени опасности аварии и т.д.; - производить расчет радиационных последствий; - просматривать результаты расчетов в графическом ( карты, схемы ) и текстовом ( таблицы ) виде; - выдавать результаты расчетов в виде формализованной справки. Таким образом использование программы позволяет в кратчайший срок оценить степень опасности возникшей аварии и принять меры по защите населения от воздействия радиоактивных веществ. SPECIALISTS MEETING on INSTRUMENTATION AND EQUIPMENT FOR MONITORING AND CONTROLLING NPP POST-ACCIDENT SITUATIONS DIMTTROVGRAD RUSSIAN FEDERATION 12-15 September 1995 PROGRAMME Tuesday, 12 September 1995; 09:00 - 10:00 Registration: Culture Center of RIAR Dimitrovgrad The contact person in SSC RIAR: Mr. O.Kmsky, Mr. .R.Makin, Mr. M.Pozdeev Dimitrovgrad 433510 Ulianovsk Region Russia Federation Tel :+(84235) 3-27-27 Telex: 224854 VELASU Fax :+(84235) 3-56-48 10:00 - 10:30 Welcoming addresses: Mr. V. Kalygin, SSC RIAR, Russian Federation Mr. A. Kossilov, IAEA COFFEE 11:00 - 12:45 Session 1 Chairman: Mr. H.G. Shugars, USA 1 Improvement in Post-Accident Instrumentation for Spanish Nuclear Power Plants, Rafael Cid, Spain 2 Safety Parameter Display System for Kalinin NPP, V.l. Andreev, EN. Videneev, NN. Davidenko, G.I. Shaftan, V.G. Dounaev and V.T. Neboyan, Russian Fed., J-C. Tissot, D. Joonekindt, France \6ti 3 Development of Transmitter with Hybrid-IC for Post- Accident Monitoring Instrumentation, T. Ikeuclti and S. Watanabe, Japan LUNCH 14:30 - 17:00 Session 2 Chairman: Mr. V. Neboyan, Russian Federation 4 Relative Different Energy Neutron Radiometry in Reactors for Preventing of Accidents Caused by Uncontrolled Reactivity Variations S. V. Volkov, Russian Federation 5 A Stochastic Approach to Accident Identification in Nuclear Power Plants Kee-CItoon Kwon, Soon-Ja Song, Won-Man Park, Jae- Chang Park and Chang-Sttik Ham, Rep. of Korea 6 Accident Monitoring in Ventilation Stack V.Kapisovsky, V.Zbiejczukovâ, F.Gabris, G.Belan, IZeman, J.Bukovjan, LReliâk, Slovak Rep. 7 Experience with Neutron Flux Monitoring Systems Qualified for Post-Accident Monitoring H.G. Shugars and IF. Miller, USA. 18:00 COCKTAILPARTY Wednesday, 13 September 1995 09:30 - 12:30 Session 3 8 Basic data of Emergency Response Centre O. Jenicek, Czech Republic 9 Applied Software of the Emergency Recording System for Reactor Facility Parameters under the Minor Statistics Conditions V.B. Ivanov, A.F. Grachev, O.M. Kinsky, RS.Makin, A.I. Ochrimenko, LJ. Demidov, V.l. Karpjuk, V.K. Afonin, R.G. Iakanderov, Russian Fed. 10 Design Implementation of the Post-Accident Monitoring (РАМ) System for Wolsong NPP Units 2,3 & 4 in Korea San-Joon Han, Rep. of Korea 11 Electrical Characteristics of Nuclear Power Plant Cables at Accident Simulation V.l. Rogov, V.N. Ulimov, N.T. Filatov, V.S. Shestacov, Russian Fed. LUNCH 14:00 16:00 Session 4 Chairman: Mr. R. Cid, Spain 12 Research Istitute of Atomic Reactors: General Presentation V.B. Ivatiov, Russian Fed. 13 Creating Requirements for Control of Post-Accident Situations at NPP VS. Dickarev, VS. Iotwv, Russian Fed. 14 Principles of Elaboration and Creation of Information- Analytical System "RI Operation Safety with SSC RIAR Research Reactors" V.B. Ivanov, A.F. Grachev, O.M. Kinsky, R.S. Makin, A.I. Ochrimenko, L.I. Demidov, V.l. Karpjuk, V.K. Afonin, RG. Iskanderov,Russian Fed. 15 Radiological Consequences Computer Code as Emergency Operator Help in Case of Fission Products Release During Reactor Accident V.D. Kizin, S.A. Efarov, EI. Shkokov, EK. Yakshin, Russian Fed. COFFEE 16:30 - 17:30 Session 5 Chairman: Mr. A. Kossilov, IAEA General Discussion, Conclusions and Recommendations 18:00 SOCIAL EVENT Thursday, 14 September 1995 09:00 - 16:00 Technical Visit to SSC RIAR 18:30 DINNER PROVIDED BY SSC RIAR 16é \ 64 DESCRIPTION OF THE RIAR RESEARCH FACILITIES \ GS GENERAL INFORMATION 1. Location: State Scientific Center "Research Institute of Atomic Reactors" is situated 150km away from the city of Samara and 100km from Ulyanovsk in a regional center the town of Dimitrovgrad, Ulyanovsk region. 2. Detailed address: 433510, Dimitrovgrad-10, Ulyanovsk region, Russia, SSC RF RIAR. 3. Telephone: (84235) 32021 4. Fax: (84235) 35648 5. E-mail: [email protected] 6. Full name of the director of the organization: Valentin B. IVANOV 7. Full name of persons responsable for the international contacts: Victor A.KUPRIENKO - Deputy Director Klaudia N. VINOGRADOVA - Head of Bureau for International Relations DESCRIPTION CARD OF THE UNIQUE RESEARCH FACILITY Main Scientific and Technological Characteristics 1. Name: "SM Research Reactor" of the State Scientific Center "Research Institute of Atomic Reactors". 2. Commissioning: 1961 - start-up; 1993 -restart after reconstruction. 3. Scope of knowledge: Physics, material science. 4. Scientific trends: - Radiation testing of different materials (fuel and structural) for atomic technology. - Radiation testing of thermonuclear reactor materials and elements. - Accumulation of the heavy transuranium elements. - Production of radionuclides with high specific activity. 5. Main scientific topics: - Testing of different materials to enrich the fundamental knowledge of their behaviour under irradiation. - Testing of materials, fuel element samples, absorbing elements and other objects for the new generation of power reactors with higher safety (500-New Generation,600-Water High Safety Power Reactor, and others). - Testing of materials and elements of International Thermonuclear Experimental Reactor ITER. - Production of weight amounts of curium (Cm), berkelium (Bk), californium (Cf), einstainium (Es) used in fundamental research, when investigating transmutation processes of long-lived wastes of atomic power industry and also as radiation sources in industry and medicine. - Production of cobalt-60, iridium-192, selenium -75 and other radionuclides used in medicine and industry. 6. Basic technical parameters of the facility. Thermal power, MW 100 Maximum thermal neutron flux density, n/cm'V1 51013 Maximum fast neutron flux density, n/cm'V1 21015 Water parameters of the first circuit: temperature,0 С up to 95 pressure, MPa 5 Amount of places for irradiation: in neutron trap 1 in the core 6 in the reflector 30 7. Primary advantages of the facility (substantiation of its singularity) The SM reactor is a unique one because it is one among all the reactors all over the world with the highest flux. It permits to carry out tha shortcut material testing up to the high values of neutron fluence and obtain above-named heavy transuranium elements, inaccessible to obtain in other reactors. Loss of such a reactor means the loss of corresponding research trends associated with its unique possibilities. 8. Current investigations All the investigations listed in the point 5 are carried out in the reactor at present. 9. Additional investigations that can be performend at present The investigations set out in the point 5 may be extended because the existing potencialities of irradiation are used only on 70% today. 10. Brief description of the main scientific results The main result of the reactor operation is the provision of one or another irradiation object by necessary fluence of neutrons. In this case the change of object properties is determined both in the process of irradiation and during further special material science testings. The overall research conclusions are obtained after the completion of the whole complex of works, the result of which is the radiation basis of new design of fuel, absorbtion elements and other core elements, thermonuclear reactor components and so on. New date of the transuranium element properties, obtained with the help of the reactor are also important results of its work. DESCRIPTION CARD OF THE UNIQUE RESEARCH FACILITY Main Scientific and Technological Characteristics 1. Name: "Loop type testing reactor MIR" of the State Scientific Center "Research Institute of Atomic Reactors". 2. Commissioning: 1966 - start-up, 1976 - restart after reconstruction. 3. Scope of knowledge: Physics, material science. 4. Scientific trends: -Radiation testing in loop facilities of the prototipe fuel assemblies in designed operation conditions. - Simulating accident and other special experiments with fuel elements and assemblies. - Radioactive nuclides production. 5. Main scientific topics: - Testing of the prototipe fuel elements and fuel assemblies for the cores of the reactors of new generation with the increased safety (500-New Generation,600-Water High Safety Power Reactor and others) in conditions correlating the design. - Simulation of the accident conditions of a "small leak", "great leak" types of the WWER reactor in loop channel conditions for fuel element and fuel assemblies studies. - Studies of the behaviour of fuel elements with deep bumup when operating in power cycling and ramp conditions. - Studies of the deviation from the standard operating conditions (waiter chemistry, convective heat transfer crisis and others) influence on fuel elements and fuel assemblies state. - Production of Co-60, Ir-192 and other radionuclides. - Testing of new fuel assemblies for research reactors. 6. Basic technical parameters of the facility: Thermal power, MW 100 Maximum thermal neutron flux density, n/cm2s 51014 First circuit water parameters: temperature, 0 С up to 85 pressure, MPa 1,2 Number of loop channels 11 Number of experimental loops 7 One loop power, kW 2000 i 11 Loop water parameters: temperature, 0 С up to 350 pressure, MPa ..up to 20 7. Primary advantages of the facility (substantiation of its singularity) Reactor MTR in total of its experimental potantialities is one of the largest research reactors in the world, allowing to carry out the experimental checking of a new design of promising power reactor fuel assemblies. Prototjpe fuel assemblies testing in the reactor MTR is a necessary stage of the experimental basis of the national alaborations core new design. It is also the only one research reactor in Russia where are carried out simulating emergency and other special experiments with fuel assemblies, including up to 20 one meter long elements. 8. Currant investigations. All the investigations listed in the point 5 are carried out in the reactor at present. 9. Additional investigations that can be performed at present Investigations set out in the point 5 may be extended because the existing potencialities of irradiation are used only on 70% today. 10. Brief description of the main scientific results. The main result of the reactor operation is the provision of fuel elements and fuel assemblies irradiation in the given neutron-physics, thermohydrolic and water-chemical conditions, including special conditions, simulating emergency, transient and other situations. Fuel element and fuel assembly states changes are studied as in the process of irradiation so during further material science investigations The overall scientific conclusions are obtained after the completion of the whole complex of works, the result of which is the radiation basis of the fuel assembly new design for the promissing reactors. Simulating emergency experiments allow to study the behaviour of fuel elements and fuel assemblies in designed and severe accident, and varify the corresponding calculation codes. DESCRIPTION CARD OF UNIQUE RESEARCH FACILITY Main Scientific and Technological Characteristics 1. Name: "Research Material Science Complex" as a part of State Scientific Centre "Research Institute of Atomic Reactors" 2. Commissioning: January, 1964. 3. Scope of knowledge: material science, physics (radiation solid-state physics). 4. Scientific trends: - Material science and thermal treatment; - Experimental investigations of irradiation effect on materials; - Nuclear physics. 5. Main scientific topics: - Material science of nuclear power facilities; - Investigation of radiation effect on nuclear material properties. 6. Basic technical parameters: Material Science complex incorporates three interconnected research trends: - non-destructive examinations of structural elements of nuclear reactor cores; - investigations of physical-mechanical characteristics, structural and elemental composition of irradiated materials; - development of methods, specimens and devices for irradiation in the research reactors of Institute. Non-destructive examination of fuel, fuel elements and assemblies are carried out in seven shielded chambers, situated at a separate building and allowing investigation of irradiated fuel assemblies and in-vessel devices up to 4000 mm length. The planned scope of investigations is 20 fuel assemblies/year. The material properties are studied in detail with unique remote equipment mounted in 39 shielded chambers providing for safe operation within 10000 -100000 Ci activity range. The maximum size of фесйпега does not exceed 1200 mm. 7. Primary advantages (substantiation of singularity) As compared to other Russian and foreign hot laboratories the SSC RIAR unique material science complex allows for a complete cycle of investigations of any type of reactor materials in shielded chambers, in-pile reirradiation at specified flux density and neutron spectrum as well as investigation of the WER, BN and PWR full- scale fuel assemblies. Engineering and methodical support of complex allows investigation of core elements under emergency- conditions including investigations of molten spent fuel. 8. Current investigations: - all reactor types fuel assemblies irradiated up to high bum-up; - control and scram rods; - physical-mechanical properties of all nuclear fuel types and its certification ; - radiation resistance of promising materilas for nuclear and thermonuclear engmeering; - behaviour of core elements and their properties under different emergency conditions; - certification of structural materials of cores, vessel nuclear and thermonuclear reactors; - experimental validation of safe operation parameters of fuel elements and assemblies. Development: - welding methods of irradiated materials; - methods and equipment for irradiated material investigations; - technology for manufacture of experimental fuel elements, fuel assemblies, irradiation capsules and targets for accumulation of transuranium elements and investigation of their physical-mechanical properties. 9. Additional investigations that can be performed at present: - fuel, fuel elements, structural and absorbing materials on order of foreign companies. 10. Brief description of the main scientific results: During last years nuclear fuel and structural materials of fuel assemblies irradiated in the WWER-400, WWER- 1000 up to the maximum bumup of 64 MW days/kg U, and in the BOR-60 up to the maximum bumup of 26% h.a. were investigated. The properties of neutron absorbing materials (boron carbide, hafnium, dysprosium titanate) were investigated after service life testings in the WWER cores. Copper, vanadium alloys and chromium-manganese steels were investigated according to ITER project. Mock-up assemblies were investigated after testing in regime of "Small leakage"in the MIR reactor (SSC RIAR), as well as after testing in regime of severe fuel damage at the CORA (KFK) facility. The results of investigations were used as the basis for the international standard problem on verification of calculation codes (ISP- 36). The experiments were prepared and the fuel bumup was investigated after testing under emergency conditions of the RIA type. The investigations carried out at the material science complex were used in licensing the WWER fuel including the substantiation of the WWER-400 and WWER-1000 change-over to four- and three-year fuel cycle, respectively, and replacement of steel structural elements of fuel assemblies with zirconium alloys. DESCRIPTION CARD OF THE UNIQUE SCIENTIFIC FACILITY The most important technical-scientific characteristics 1. Name: "Pilot-research complex of the chemical-engineering department (PRC CED) being a part of State Scientific Centre "Research Institute of Atomic Reactors". 2. Commisionmg: 1977 - bringing into operation, 1989- start-up after reconstruction. 3. Held of knowledge: Physics, radiochetnistry, reactor material study. 4. Scientific trends: - Development of the ecologically pure and safe closed fuel cycle for utilization of Pu and minor actinides on the basis of "dry" pyroelectrochemical nuclear fuel reprocessing methods and vibropacing technology. - Development and improvement of the production technology for granular oxide fuel on the basis of Pu and minor actinides. - Design and technology of fuel rods on the basis of vibropac granular oxide fuel including designs providing recyde of plutonium, minor actinides and long-lived fission products. - Preparation of automatic remotely-controlled processes and equipment for production of granular fuel, fuel rods with vibropac fuel rods and fuel sub-assemblies. - Study of transuranium elements behaviour under irradiation in the reactors with different physical characteristics. - Investigation of content, properties and methods for localization and storage of radioactive wastes being formed at all stages of fuel recycle. 5. The most important scientific topics: - Development of the scientific and technological background of the pyroelectrochemical nuclear fuel reprocessing; - Calculatim-experimental and technological investigations related to the development of a fuel rod with vibropac UPuOz fuel capable to achieve super high bumups; - Development and validation of methods for transmutation, utilization and storage of minor actinides and radioactive wastes of pyroelectrochemical production. 6. The main technical parameters of the facility: Production of granular UPu02 or TJ02 fuel, kg/year up to 1500 Manufacture of the BN-600 type fuel rods, fr/h up to 10 Manufacture of the BOR-60 type fuel rods, fiVh up to 6 Production (design output) of the BN-600 F As, FA/y ..up to 200 Manufacture of the BOR-60 fuel sub-assemblies, FA/y up to 50 7. The major advantages of the facility (substantiation of its unique capabilities): Pilot research complex of the SSC RIAR CED - is the largest in Russia and the only complex in the world where the most safe and ecologically pure "dry" pyroelectrochemical method of nuclear fuel reprocessing in salt melts and automatic remotely-controlled production process for power reactor granular fuel rods and sub-assemblies. The complex comprises a line of three shielded cells equipped by the remotely controlled systems for nuclear fuel reprocessing and (UO2, PuCb, UPuCs) granulate production, line of heavy boxes for analytical control of fuel and two large-scale cells for automatic production and monitoring of fuel rods and also fuel sub-assemblies for the fast neutron power reactors of the BN-800 type. The complex also comprises a line of heavy boxes intended for production of experimental fuel rods and sub-assemblies practically for all reactors of Russia. 8. Current investigations: Currently the facility is used for investigations listed in Clause 5. 9. Additional investigations that can be performed at present It is possible to expand the investigations on all the trends of Clause 5 because only 50% of the existing potential is bang used presently. 10. Brief description of the main scientific results obtained with this facility: The main scientific results are development of the semi-industrial reprocessing technology of nuclear fuel reprocessing into granulate, mastering the remotely-controlled automatic production of vibropac fuel rods and sub assemblies and development of the fuel rod design capable to achieve super high bumups (>25%); the data on the production technology and properties of fuel compositions containing transuranium and transplutonium elements (Np,Am). DESCRIPTION CARD OF THE UNIQUE SCIENTIFIC FACILITY The most important techmcal-sdentific characteristics 1. Name: "Experimental Fast Reactor BOR-60" of the State Scientific Center "Research Institute of Atomic Reactors" 2. Commisioning: Reached the designed power in 1970. 3. Scope of knowledge: Physics, nuclear power. 4. Scientific trends: The reactor facility BOR-60 was designed for testing and investigation of all the fast neutrons with sodium coolant. 5. Main sdentific topics: •f Ik - mass testing of different design fuel elements with promising fuel compositions; - testing of absorbing elements with different absorbing materials; - investigations on radiation material science and on fast reactor safety; - investigations on sodium engineering and technology; - testing of experimental equipment (pumps, steam generators, heat exchangers, valves), diagnostic and protection systems. 6. Basic technical parameters of the faculty: Reactor power, mermal/electric, MW. 60/12 Maximum neutron flux density, cm2 s4 3,7 10° Average neutron energy, MeV 0,45 Neutron fluence per year, cm"2 5 10" Fuel UQz or UOz+PuCs mixture Coolant sodium Outlet sodium temperature, °C 530 Damage dose rate, dpa/year up to 25 Fuel bumup rate, %/year 6 7. Primary advantages of the facffirv(substantiation singularity). Singularity of the BOR-60 reactor facility, apart from the implementation of the major srientific trends according to item 5, is that electrical and thermal power is being produced. Besides that provision is made for the possibility to produce radionuclides preparations when irradiating appropriate targets in the core. 8. Current investigations: All investigations listed in item 5 are carried out in the reactor at present. 9. Additional investigations that can be performed at present: Investigations set out in the item 5 may be extended because only 70% of the existing irradiation potentiaht is used at present. 10. Brief description of the main scientific results: Testing results of the fuel assemblies contaning prototype fuel elements of the BN-350 and BN-600 reactors allowed for improvement of the technigue of fuel elements manufacturing for these reactors. Successful testings of the vibropac uranium and mixed uranium- plutonium dioxide fuel assemblies in BOR-60, which bum up has reached 26%, permitted to manufacture in RIAR and to deliver tens of fuel assemblies to BN-350 and BN-600 reactors. A lot of samples of structural materials of various types of reactors have been irradiated in the devices with natural and forced circulation of sodium and without it. Among those materials there were both of home and foreign firm production, for examle, such steels as AISI 316 (USA), 316 (France), alloys 1,4988 (BRD), 12R72ESR (Sweden), PE-16 (GB), etc. Prototypes of the steam generators for the reactor BN-600 of the home production, and for the reactor BN- 350 designed and manufactured in Czeck together with the diagnostic and safety systems have been tested. Si » 4um circuits equipment testing are carried out (mechanical and electromagnetic pumps, heat exchangers, valves). The technology of sodium coolant cleaning from different impurities, including radionuclides is developed. The in-pile investigations on fast reactor safety are carried out. DESCRIPTION CARD OF ТВОЕ UNIQUE RESEARCH FACILITY Main Scientific and Technological Characteristics. 1. Name: "Experimental Power Facility with the Vessel-Type Boiling Water Reactor - VK-50" in the State Scientific Center "Research Institute of Atomic Reactors". 2. Commissioning: October 1965. 3. Scope of knowledge: physics, atomic energy. 4. Scientific trends: - study of operational parameters and performance of one-circuit nuclear power stations (NPP) with steam directly supplied from a reactor to a turbine. 5. Main scientific topics a wide range of works is performed at the VK-50 reactor: - study of the neutron-physics and thermohydraulic characteristics of the natural-circulation vessel type BWR; I 7-5 - performance of the main equipment; - radiation and technical safety, water conditions; - corrosion resistance of various structural materials; - performance of fuel assemblies and others. 6. Basic technical parameters of the facility: Operation pressure, Mpa 5 Thermal power, MW 200 Power density, kW/1 40 Thermal flux, kW/m2 average 296 maximum 856 Core outlet mass of steam, % 8,7 7. Primary advantages of the facility (substantiation of its singularity) VK-50 is the only facility with vessel-type nurlear reactor in Russian Federation. Its singularity is determined by a wide spectrum of possible working pressure (1.5 " :; MPa), possibilities of loading different cores, large scale testing of experimental equipment. 8. Current Investigations Investigations on optimization of the fuel cycle with the increase of water-fuel ratio from 2,2 to 3,0 and loaded fuel enrichment from 2,4 to 3 %. 9. Additional investigations that can be performed at present The reactor modernization is planned with installation of in-vessel separation devices and improvement of safety system. It will allow for conversion of the VK-50 reactor into a prototype reactor facility for underground NPPs and provision of characteristics corresponding the SBWR type reactor which is under development now. 10. Brief description of the main scientific results The results of investigations performed during long-term operation of the VK-50 reactor facility showed its main advantages in comparison with the other types of reactors: - ease and reliability of the direct cycle of the reactor facility operation (without using steam generator loops); - ease of core cooling at the expense of coolant natural circulation both in ordinary and in emergency operating conditions; -perfect self-regulation and power self-restriction properties due to high negative values of steam and temperature reactivity effects; - ease and reliability of water-chemical (without corrections and with oxygen) condition, providing high corrosion resistance of structural materials and widening the use of carbon steel in NPP circuits; - good radiation environment of the equipment and indoors of the primary circuit as a consequence of nongaseous radionuclides limited transfer from water to steam (primary circuit equipment and piping as a rule do not require decontamination before the repair); - value of gaseous fission products normal effluence is at the level of the best foreign NPP with BWR. The range of the investigations and thechnical tests performed shows the vessel-type boiling-water reactor VK-50 to be a reliable source of energy for low - and average power NPP (up to 600 MW thermal). DESCRIPTION CARD OF THE UNIQUE RESEARCH FACILITY Main Scientific and Technological Characteristics 1. Name: "Radiochemical Complex of Radionuclide Sources and Preparations (RCCRSP) as a part of the State Scientific Centre "Research Institute of Atomic Reactors" (SSC RF RIAR). 2. Commissioning: 1964 3. Scope of knowledge: Radiochemistry 4. Scientific trends: - Investigations on the radionuclides accumulation in the reactors; - Development of technological processes for radionuclides separation from the irradiated targets and their refining; - Investigation on the physical-chemical properties of actinides and their compounds; - Development of production technologies for ionizing radiation sources (1RS) and radionuclide preparations. 5. Main scientific topics: - Calculations and experimental investigations on the accumulation of transplutonium elements (TPE), other actinides and radionuclides from P up to Ir in the RIAR reactors; - Investigations on the TPE metallic status (production and investigation of metals, alloys and intermetallics); - Synthesis and investigation of TPE solid compounds; - Investigation on the TPE separation processes and those of other elements by different methods (extraction, sorption, chromatography, cementation, distillation and precipitation); - Optimizing production conditions of standard preparations; - Production of new materials for 1RS; - Development of 1RS based on TPE, REE and other radonuclides to.- industrial and medical application: - Investigations on validation of analytical cont'd methods for radionuclide products production processes and products certification. 6. Basic technical parameters of the facility: In the radiochemical building there are 16 shielded hot cells, providing work with radionuclides having activity up to 100000 Ci, and 4 chains of different boxes (glove- and manipulator ones). The RCCRSP comprises 4 research laboratories and 2 pilot-industrial facilities for radionuclide preparation and 1RS production. The divisions are provided with uptodate research, production and technical equipment. 7. Primary advantages of the facility (substantiation of its singularity) The radiochemical complex singularity is that it is not an independent department but an integral structural part of the powerful institute comprising 7 different reactors, material science laboratory and special engineering arrangement complex which provide efficient radiation safety for the workers of the institute and population of the nearest area. The availability of the SM reactor with 5xl015 s~2 neutron flux allows for production of TPE (including 252Cf) as well as lighter radionuclides having specific activity which is not achievable in the other reactors. In RIAR it is possible to implement the whole cycle of work with radionuclides: in-pile production, removal from irradiated targets, properties investigation, standard preparations production, 1RS production, products delivery to customers, reception of spent products and waste disposal. 8. Current investigations - Production and investigation of metals, alloys and TPE intermetallics of platinum group and other elements (Ni, Al and Si). - Development and improvement of production processes for TPE, 32P, 55Fe, 63M, 109Cd, шВа, ,53Gd, ,8SW, etc. - Production of alpha-sources based on ^Cm alloys and compounds for space application. -Development of gamma-sources based on ^Co,75 Se, 1Ä2Ir, I33Gd, etc. for medical and technical application. - Improvement of methods for radionuclide accumulation in the reactors. - Study of the possibility for extension of the field of radionuclide practical application. 9. Additional investigations that can be performed at present Additional work in the following directions can be carried out at present. - Investigation on ecological issues concerning TPE. - Development of the refining processes of TPE waste solutions. - TPE transmutation by the in-pile irradiation. - Production of target materials for the in-pile irradiation by the thermal destruction method of radionuclide extracts. - Improvement of the Am-Cm separation methods. - New investigations on the properties of metal Cm and 23SPu. - Production and investigation of intermetallics and alloys in the Am(Cm)-Ni(Fe,Co,Al,Pd,Ge), Np-Ce and Am-Pu systems. - Creation of the measurement complex based on the neutron sdector and BOR-60 reactor fast neutron horizontal channel for the urne-of-flight investigations. 10. Brief description of the main scientific results The results of investigations, performed over the period of the Radiochemical Complex existence, are published in several hundreds of scientific works (articles, preprints, papers presented at different conferences including international ones). As for the topics the works present investigations on the kinetics of oxidation-reduction reactions of actinides, complex formation, oxidation-reduction potentials, extraction and sorption processes, radio chemical processes, synthesis of different solid compounds and investigations on their properties. Some of the actinides were obtained in the out-of-ordinary valency states (curium(TV), curium(VT), californium (V). Such metals as - ^Cm, M9Bk and M9CF, intermetallics - Am, Cm, Bk, and Cf with the metals of platinum group (some of them for the first time) were produced and investigated. Technologies were developed for TPE separation using nitric acid solutions and stainless steel equipment, phosphorus-33 with a molar activity up to theoretically possible value (5000 Ci/mmole). Detailed investigations of the in-pile inadiation conditions and chemical reprocessing allowed for production of gadolinium-153 preparation of high quality (specific activity >100 Ci/g, active impurities content <0.0005%). Production processes of 252Cf neutron sources and ^Co, 75Se, I53Gd, 169Yb, 170Tm and 192Ir gamma- sources were developed. The sources are manufactured on the customer's order basis. \1* SPECIALISTS MEETING ON INSTRUMENTATION AND EQUIPMENT FOR MONITORING AND CONTROLLING NPP POST-ACCIDENT SITUATION DIMITROVGRAD RUSSIAN FEDERATION 12-15 September 1995 LIST OF PARTICIPANTS CZECH REPUBLIC Mr. O. JENICEK State Office for Nuclear Safety Prague 2 Slezska9,120 19 Tel: 42 2 2417 2458 Fax: 42 2 2417 2055 JAPAN Mr. SATOSHIWATANABE Mitsubishi Heavy Industries, Ltd. 3-1 Mmatomirai 3-Chome, Nishi-KU Yokoham 220-84 Tel: 045 224-9689 Fax: 045 224-9969 JAPAN Mr. TAKESHI DXEUCHI Mitsubishi Heavy Industries, Ltd. 1-1, Wadasaki-Cho 1 Chôme, Hyogo-Ku Kobe City 652 Tel: 078 672 3326 Fax: 078 672 3268 REP. OF KOREA Mr.S.-J.HAN Korea Atomic Energy Research Institute P.O.BOX 105 Yusung, Taejon 305-600 Tel: 042 868 8180 Fax: 042 861 4859 REP. OF KOREA Mr. K. KEE-CHOON Korea Atomic Energy Research Institute P.O.Box 105 Yusong Taejon 305-600 Tel: 82-42-868-2926 1 Fax: 82-42-868-8357 RUSSIAN FEDERATION Mr. V. NEB О Y AN Consyst Ferganskaya str. 25 Moscow 109507 Tel: 376 13 14 Fax: 376 13 14 Mr. V.L. SMUTNEV Novovoronezh NPP 396072 Novovoronezh Voronezh Region Tel: (07364) 7-31-79 Fax: (07364) 7-33-02 Mr.V.S.IONOV Moscow AAC "KI" INR 123182 Kurchatov St., Russia Phone: (095) 196-73-76 (095) 196-71-75 Fax: (095)196-61-72 Mr A.B. BALUNOV Atomenergoproekt St.- Peterburg mstitute 193036 St.-Peterburg Suvorovsky pr. 2 A, Russia Tel: (812) 277-06-53 Fax: (812) 277-07-03 Mr. S.V. VOLKOV SNHP- ASKRO, Moscow 123060 Raspletm street,5 Tel: (095) 198-97-20 Fax: (095) 943-00-63 Mr. N.L FDLATOV 140061 Lkkarino Research Iasutute of Scientific mstruments Tel: 095 552-42-11 Ms. H.V. BUDDLINA 140061 Li&arino Research bstitute of Scientific Instruments Tel: 095 552-42-11 2 Mr. V.V. KALYGIN Dmütrovgrad State Scientific Centre Research mstitute of Atomic Reactors Phone:(84235)35830 Fax: (84235)35648 Mr. V.K. RIZVANOV Dimitrovgrad 433510 State Scientific Centre Research Institute of Atomic Reactors Phone: (84235) 5-1052-4105 Fax: (84235)35648 Mr. A.L OCHMMENKO Dimitrovgrad 433510 State Scientific Centre Research mstitute of Atomic Reactors Phone: (84235) 5-1052-4105 Fax: (84235)35648 Mr. O.M. KE4SKY Dimitrovgrad 433510 State Scientific Centre Research Institute of Atomic Reactors Phone: (84235) 5-1052-4105 Fax: (84235)35648 Mr. R.S. MAKTN Dimitrovgrad 433510 State Scientific Centre Research Institute of Atomic Reactors Phone: (84235) 5-1052-4105 Fax: (84235)35648 Mr. A.A. M3NAKOV Dimitrovgrad 433510 State Scientific Centre Research Institute of Atomic Reactors Phone: (84235) 5-1052-4105 Dr. A.Yu. TOKAREV Dimitrovgrad 433510 State Scientific Centre Research mstitute of Atomic Reactors Phone:+7 (84235) 32324 3 ( g X Mr. A.A. KASHKIROV Dimitrovgrad 433510 State Scientific Centre Research Institute of Atomic Reactors Phone:+7 (84235) 32324 Mr. V.N. PRTDACHIN Dimttrovgrad 433510 State Scientific Centre Research Institute of Atomic Reactors Phone:+7 (84235) 32324 Dr. V.A. KACHALJN rÄnitrovgrad 433510 State Scientific Centre Research Institute of Atomic Reactors Phone: +7 (842235) 32324 Dr. A.B. MURALEV r^mitrovgrad 433510 State Scientific Centre Research mstitute of Atomic Reactors Phone: +7 (84235) 32324 SLOVAK REPUBLIC Mr. V. KAPISHOVSKY Vyskumny ustav jadrovych elektxarni Trnava Inc. Okruzna 5 918 64 Trnava Tel: + 42 805 91217 Fax:+ 42 805 91264 SPAIN Mr. R. СП) Consejo de Seguridad Nuclear Justo Dorado, 11 28040 Madrid Tel: 34-1-346.02.44 Fax: 34-1-346.05.88 USA Mr. H.GORDON S SUGARS Gamma-metrics 5788 Pacific Center BLVD San Diego, CA 92121 USA Tel: (619) 450-9811 Fax: (619) 452-9250 IAEA Mr. A, KOSSBLOV International Atomic Energy Agency Wagramerstrasse 5 A-1400 Vienna Tel: +43 1 2060 22796 Fax: +43 1 20607 shows the best path probability of each model which is the result of Viterbi algorithm when the input observation is the same as training data. In this case, model #2 accident, ATWS, has the highest probability. In the case of one sensor fault or equipment malfunction, the accident identification system also exactly identified the accident as shown in
for model #4 accident, SGTR. However, it could not correctly identify the accident when more than 2 sensor faults or equipment malfunctions have occurred.
Likelihood probability (Case I)
Likelihood probability (Case II)