DOCSLIB.ORG

A Review of Design Optimization Methods for Electrical Machines

Total Page：16

File Type：pdf, Size：1020Kb

Download full-text PDF Read full-text
A Review of Design Optimization Methods for Electrical Machines energies Review A Review of Design Optimization Methods for Electrical Machines Gang Lei 1,*, Jianguo Zhu 1, Youguang Guo 1, Chengcheng Liu 2 and Bo Ma 1 1 School of Electrical and Data Engineering, University of Technology Sydney, Ultimo 2007, Australia; [email protected] (J.Z.); [email protected] (Y.G.); [email protected] (B.M.) 2 State Key Laboratory of Reliability and Intelligence of Electrical Equipment, Hebei University of Technology, Tianjin 300131, China; [email protected] * Correspondence: [email protected]; Tel.: +61-2-9514-1268 Received: 30 September 2017; Accepted: 22 November 2017; Published: 24 November 2017 Abstract: Electrical machines are the hearts of many appliances, industrial equipment and systems. In the context of global sustainability, they must fulfill various requirements, not only physically and technologically but also environmentally. Therefore, their design optimization process becomes more and more complex as more engineering disciplines/domains and constraints are involved, such as electromagnetics, structural mechanics and heat transfer. This paper aims to present a review of the design optimization methods for electrical machines, including design analysis methods and models, optimization models, algorithms and methods/strategies. Several efficient optimization methods/strategies are highlighted with comments, including surrogate-model based and multi-level optimization methods. In addition, two promising and challenging topics in both academic and industrial communities are discussed, and two novel optimization methods are introduced for advanced design optimization of electrical machines. First, a system-level design optimization method is introduced for the development of advanced electric drive systems. Second, a robust design optimization method based on the design for six-sigma technique is introduced for high-quality manufacturing of electrical machines in production. Meanwhile, a proposal is presented for the development of a robust design optimization service based on industrial big data and cloud computing services. Finally, five future directions are proposed, including smart design optimization method for future intelligent design and production of electrical machines. Keywords: electrical machines; multi-level optimization; multi-objective optimization; system-level optimization; manufacturing variations; manufacturing quality; robust optimization; industrial big data; cloud computing 1. Introduction 1.1. Energy and Environmental Challenges Electrical machines, as the main drive sources, have been widely employed from industry to agriculture, from defence to community facilities, from domestic appliance to electronic products, etc. They are the foundation of the power industry and the core components of industrial machinery. Electrical machines consume about 46% of total electricity generated worldwide, resulting in about 6040 Mega-tonnes of CO2 emission. This is the largest portion of electricity use up to now, as shown in Figure1. Therefore, motor energy efficiency is crucial for the energy conservation, environment protection, and global sustainable development. Consequentially, high-efficiency motors will dominate the market development of electrical machines worldwide [1–3]. Energies 2017, 10, 1962; doi:10.3390/en10121962 www.mdpi.com/journal/energies Energies 2017, 10, 1962 2 of 31 Energies 2017, 10, 1962 2 of 31 Figure 1. GlobalGlobal electricity demand by sector and end end-use.-use. Besides the energy efficiency,efficiency, therethere areare manymany otherother specificationsspecifications and/orand/or requirements for the design optimization optimization of of electrical electrical machines, machines, such such as astorque, torque, power power density, density, volume volume and andweight. weight. For specialFor special applications applications like geologic like geologic or petroleum or petroleum drilling drilling engineering engineering and aerospace and aerospace engineering, engineering, more requirementsmore requirements and extreme and extreme constraints constraints should should be beinves investigatedtigated [4– [46]–6. ].For For example, example, the the environment environment temperature for motors used in petroleum and geology maymay reachreach 300300 ◦°CC,, or even to 500 °C.◦C. The high surrounding temperature will will result in severe performance degradation due due to to the increase of the copper loss and the demagnetization of permanent magnets (PM) for PM motors. ThusThus,, all performanceperformance specificationsspecifications and and requirements requirements including including energy energy efficiency efficiency are are vital vital for thefor theapplication application of electrical of electrical machines. machines. Consequently, Consequently, improving improving motor performance motor performance is of great significanceis of great signito bothficance the environment to both the environment protection and protection the energy and sustainability. the energy sustainability. To achieve this To goal, achieve optimization this goal, is optimialwaysz necessaryation is always [2,7–11 necessary]. [2,7–11]. 1.21.2.. An An Overview of Design and Optimization of Electrical Machines Design optimization of electrical machines machines includes includes two two main main stages, stages, design design and and optimization optimization.. The main aim of the design stage is to findfind a feasible scheme (or several schemes) for a given application through the investigation of different materials and dimensions, motor typ typeses and topologies, multi multi-disciplinary-disciplinary analysis including electromagnetic analysis, and/or design design experience. experience. The analysis of this stage will provide information including motor parameter calculation and performance evaluation evaluation for for the the development development of ofoptimiz optimizationation model model that that will willbe used beused in the in next the stage next. Thestage. main The target main targetof optimization of optimization stage stageis to improve is to improve the performance the performance of the of motor the motor proposed proposed in the in designthe design stage stage through through some some optimization optimization algorithms algorithms and and methods. methods. As As the the outcome, outcome, an an optimal solution will be obt obtainedained for the the single single objective objective design design situation, situation, and and some some non non-dominated-dominated solutions (called Pareto optimal solutions solutions)) will be gained for the multi-objectivemulti-objective design situation after the completion of this stage. Figure Figure 22 illustratesillustrates aa briefbrief frameworkframework forfor thethe mainmain aspectsaspects coveredcovered inin thethe design and optimization of electrical machines.machines. There isis nono fixedfixed procedureprocedure for for the the design design optimization optimization of of electrical electrical machines. machines. However, However, there there are aresome some common common steps steps to be to followed.be followed. These These steps steps are are briefly briefly described described as follows.as follows. More More details details can can be beseen seen in thein the following following sections: sections: Step 1: Select/determine possible motor types and topologies, materials and dimensions according Step 1: Select/determine possible motor types and topologies, materials and dimensions to the requirements given by the applications and users. Requirements include steady-state and according to the requirements given by the applications and users. Requirements include steady- dynamic performances like efficiency, torque and torque ripple, material and manufacturing costs, state and dynamic performances like efficiency, torque and torque ripple, material and volume and others. The main aim of this step is to obtain a number of motor options which may be manufacturing costs, volume and others. The main aim of this step is to obtain a number of motor suitable for a specific applications. options which may be suitable for a specific applications. Step 2: Implement multi-physics design and analysis for each motor option. Due to the multi-physics Step 2: Implement multi-physics design and analysis for each motor option. Due to the multi- nature, many disciplines have to be investigated in this step, such as electromagnetics, structural physics nature, many disciplines have to be investigated in this step, such as electromagnetics, mechanics and heat transfer [2,12–19]. Moreover, power electronics and control should be included as structural mechanics and heat transfer [2,12–19]. Moreover, power electronics and control should be they are relevant to the dynamic responses of the machines, such as overshoot and settling time [20–23]. included as they are relevant to the dynamic responses of the machines, such as overshoot and This step targets to calculate some parameters for the evaluation of motor performance, including settling time [20–23]. This step targets to calculate some parameters for the evaluation of motor performance, including electromagnetic parameters like core loss, inductance and back electromotive force (EMF), and thermal parameters like temperature rise and distribution. Energies 2017, 10, 1962 3 of 31 electromagnetic parameters like core loss, inductance and back electromotive force (EMF), and thermal parametersEnergies 2017, 10 like, 1962 temperature rise and distribution. 3 of 31 Motor
Load more

Recommended publications
  • Njit-Etd2010-076
    Copyright Warning & Restrictions The copyright law of the United States (Title 17, United States Code) governs the making of photocopies or other reproductions of copyrighted material. Under certain conditions specified in the law, libraries and archives are authorized to furnish a photocopy or other reproduction. One of these specified conditions is that the photocopy or reproduction is not to be “used for any purpose other than private study, scholarship, or research.” If a, user makes a request for, or later uses, a photocopy or reproduction for purposes in excess of “fair use” that user may be liable for copyright infringement, This institution reserves the right to refuse to accept a copying order if, in its judgment, fulfillment of the order would involve violation of copyright law. Please Note: The author retains the copyright while the New Jersey Institute of Technology reserves the right to distribute this thesis or dissertation Printing note: If you do not wish to print this page, then select “Pages from: first page # to: last page #” on the print dialog screen The Van Houten library has removed some of the personal information and all signatures from the approval page and biographical sketches of theses and dissertations in order to protect the identity of NJIT graduates and faculty. ABSTRACT AUTOMATION OF ANATOMIC TORSION MONITOR FOR EVALUATION AND IMPROVEMENT OF LOW BACK DYSFUNCTION by Vishal Kumar Singh The existing Anatomical Torsion Monitor (ATM) to evaluate mechanical stiffness and viscoelasticity of the low back suffers from various inherent defects. This has to be replaced by an improved device. Also the existing ATM cannot provide oscillations to the low back.
    [Show full text]
  • Eesof EDA Advanced Design System
    Keysight Technologies Keysight EEsof EDA Advanced Design System The Industry’s Leading RF, Microwave, Signal Integrity and Power Integrity Design Platform 02 | Keysight | EEsof EDA Advanced Design System - Brochure Table of Contents Advanced Design System ����������������������������������������������������������������������������������������������������������������������������03 More than 30 Years of Enabling Innovation ������������������������������������������������������������������������������������������������04 Design with Confidence �������������������������������������������������������������������������������������������������������������������������������05 Complete Design Flow ���������������������������������������������������������������������������������������������������������������������������������07 Create Custom 3D Components for Simulation with ADS Layout Designs ����������������������������������������������� 10 Most Complete Solution ������������������������������������������������������������������������������������������������������������������������������ 11 ADS Provides a Cohesive Workflow for Signal Integrity and Power Integrity Analyses ��������������������������� 12 Integrated Solutions ������������������������������������������������������������������������������������������������������������������������������������ 13 Seamless Integration with Keysight’s EDA Tools and Measurement Instrumentation ������������������������������ 14 Making Your Job Easier with Worldwide Technical Keysight EDA Experts ������������������������������������������������
    [Show full text]
  • Machine Learning for Design Optimization of Electromagnetic Devices: Recent Developments and Future Directions
    applied sciences Review Machine Learning for Design Optimization of Electromagnetic Devices: Recent Developments and Future Directions Yanbin Li 1, Gang Lei 2,* , Gerd Bramerdorfer 3 , Sheng Peng 1, Xiaodong Sun 4 and Jianguo Zhu 5 1 School of Electronic and Information, Zhongyuan University of Technology, Zhengzhou 450007, China; [email protected] (Y.L.); [email protected] (S.P.) 2 School of Electrical and Data Engineering, University of Technology Sydney, Ultimo, NSW 2007, Australia 3 Department of Electrical Drives and Power Electronics, Johannes Kepler University Linz, Linz 4040, Austria; [email protected] 4 Automotive Engineering Research Institute, Jiangsu University, Zhenjiang 212013, China; [email protected] 5 School of Electrical and Information Engineering, The University of Sydney, Sydney, NSW 2006, Australia; [email protected] * Correspondence: [email protected] Abstract: This paper reviews the recent developments of design optimization methods for electromag- netic devices, with a focus on machine learning methods. First, the recent advances in multi-objective, multidisciplinary, multilevel, topology, fuzzy, and robust design optimization of electromagnetic devices are overviewed. Second, a review is presented to the performance prediction and design optimization of electromagnetic devices based on the machine learning algorithms, including artifi- cial neural network, support vector machine, extreme learning machine, random forest, and deep learning. Last, to meet modern requirements of high manufacturing/production quality and lifetime Citation: Li, Y.; Lei, G.; reliability, several promising topics, including the application of cloud services and digital twin, are Bramerdorfer, G.; Peng, S.; Sun, X.; discussed as future directions for design optimization of electromagnetic devices.
    [Show full text]
  • State of the Art of Generative Design and Topology Optimization and Potential Research Needs
    NordDesign 2018 August 14 – 17, 2018 Linköping, Sweden State of the art of generative design and topology optimization and potential research needs Evangelos Tyflopoulos1, David Tollnes Flem2, Martin Steinert 3, Anna Olsen 4 1 Department of Mechanical and Industrial Engineering, NTNU [email protected] 2 Department of Mechanical and Industrial Engineering, NTNU [email protected] 3 Department of Mechanical and Industrial Engineering, NTNU [email protected] 4 Department of Mechanical and Industrial Engineering, NTNU [email protected] Abstract Additive manufacturing allows us to build almost anything; traditional CAD however restricts us to known geometries and encourages the re-usage of previously designed objects, resulting in robust but nowhere near optimum designs. Generative design and topology optimization promise to close this chasm by introducing evolutionary algorithms and optimization on various target dimensions. The design is optimized using either 'gradient-based' programming techniques, for example the optimality criteria algorithm and the method of moving asymptotes, or 'non gradient- based' such as genetic algorithms SIMP and BESO. Topology optimization contributes in solving the basic engineering problem by finding the limited used material. The common bottlenecks of this technology, address different aspects of the structural design problem. This paper gives an overview over the current principles and approaches of topology optimization. We argue that the identification of the evolutionary probing of the design boundaries is the key missing element of current technologies. Additionally, we discuss the key limitation, i.e. its sensitivity to the spatial placement of the involved components and the configuration of their supporting structure. A case study of a ski binding, is presented in order to support the theory and tie the academic text to a realistic application of topology optimization.
    [Show full text]
  • Design Thinking for User-Centric Website Optimization
    Paper to be presented at the DRUID Academy Conference 2018 at University of Southern Denmark, Odense, Denmark January 17-19, 2018 Design thinking for user-centric website optimization Vella Verónica Somoza Sanchez University of Southern Denmark Department of Marketing and Management/CI2M [email protected] Stefan Daniel Suivei University of Southern Denmark The Maersk Mc-Kinney Moller Institute [email protected] René Chester Goduscheit SDU Department of Marketing and Management [email protected] Peter Schneider-Kamp University of Southern Denmark IMADA [email protected] Abstract Title: Design thinking for user-centric website consequently, improve their user experience. The optimization. A case study. case considered in this study focuses on a Name: Vella Somoza company from the retail banking industry offering Affiliation: CI2M, SDU loans through comparison websites. Year of enrolment: 2017. First, to find a solution to this wicked problem we Expected final date: 2019 adapt the process of design thinking to the E-mail address: [email protected] context studied, i.e., to the design optimization of retail banking web sites. Second, we provide insights in how big data can aid the development This paper presents results from a pilot study, of a website design that optimizes the user which is part of a larger research project where we experience of the customers. As the creation of study the process innovation of website these user-centric websites is a never confronted optimization as a service through the use of big task for this type of organization, a variety of data. Over a process of design thinking we methods is applied.
    [Show full text]
  • The Role of Multidisciplinary Design Optimization (Mdo) in the Development Process of Complex Engineering Products
    21ST INTERNATIONAL CONFERENCE ON ENGINEERING DESIGN, ICED17 21-25 AUGUST 2017, THE UNIVERSITY OF BRITISH COLUMBIA, VANCOUVER, CANADA THE ROLE OF MULTIDISCIPLINARY DESIGN OPTIMIZATION (MDO) IN THE DEVELOPMENT PROCESS OF COMPLEX ENGINEERING PRODUCTS Papageorgiou, Athanasios; Ölvander, Johan Linköping University, Sweden Abstract The work presented in this paper explores several concepts related to the design of complex engineering products and emphasizes on the effects of considering Multidisciplinary Design Optimization (MDO) in the development process. This paper is by no means a comprehensive literature review, but instead, the aim is to discuss some key points through theory and references to common MDO applications. In this respect, the central topics which are addressed herein are the enhancement of the generic product development process, the road towards a better integration of the organization’s functions, the methods to manage complex system architectures, and finally, the shortcomings of the MDO field. As a link to more tangible industrial applications, Unmanned Aerial Vehicles (UAVs) are chosen as an illustrative example due to their technical complexity as well as the demanding requirements of the corresponding market. Overall, the paper shows that despite the current state-of-the-art limitations, MDO can be a valuable tool within the “traditional” design process that has the potential to enable products of better quality while simultaneously reducing the total development time and effort. Keywords: Design methods, Design process, Optimisation, New product development, Multidisciplinary Design Optimization Contact: Athanasios Papageorgiou Linköping University Management and Engineering Sweden [email protected] Please cite this paper as: Surnames, Initials: Title of paper. In: Proceedings of the 21st International Conference on Engineering Design (ICED17), Vol.
    [Show full text]
  • From Design Automation to Generative Design in AEC
    The Journey from Design Automation to Generative Design in AEC Dieter Vermeulen Technical Sales Specialist AEC Generative Design & Engineering @BIM4Struc © 2020 Autodesk, Inc. DESIGNS TODAY @BIM4Struc TRADITIONAL DESIGN PROCESS @BIM4Struc @BIM4Struc @BIM4Struc @BIM4Struc GENERATIVE DESIGN: A form of artificial intelligence, dedicated to the creation of better outcomes for products, buildings, infrastructure and systems. @BIM4Struc GENERATIVE DESIGN PROCESS PEOPLE COMPUTER PEOPLE @BIM4Struc STADIUM DESIGN CONSTRUCTION PLANNING SITE DRAINAGE NEIGHBORHOOD PLANNING SPACE PLANNING FAÇADE DESIGN @BIM4Struc Generative design is a methodology and a process more than a singular product or tool @BIM4Struc Benefits of Generative Design @BIM4Struc What level of design progression ? @BIM4Struc Traditional Design @BIM4Struc Sketching @BIM4Struc Computer Aided Drafting @BIM4Struc Parametric Design @BIM4Struc PARAMETRIC DESIGN Designer/engineer uses computer as passive machine one one limited human + computer = design @BIM4Struc PARAMETRIC MODELING @BIM4Struc Parametric Modeling a = 2 b = 1 a – b = c @BIM4Struc Parametric Modeling a = 2 b = 1 a – b = c f(x) @BIM4Struc Parametric Modeling a = 2 b = 1 a – b = c @BIM4Struc Conceptual Tower Mass – Design Model Assign Parametric Create Geometry Modify Parameters Document the Idea Constraints @BIM4Struc Design Automation @BIM4Struc input design output automation @BIM4Struc Dynamo Ecosystem to increase capabilities REVIT API Difficulty REVIT DYNAMO Expressive Power @BIM4Struc Dynamo to simplify things PROGRAMMING
    [Show full text]
  • Numerical Experience with Two-Point Adaptive Nonlinearity Approximation for Design Optimization
    INTERNATIONAL DESIGN CONFERENCE - DESIGN 2004 Dubrovnik, May 18 - 21, 2004. NUMERICAL EXPERIENCE WITH TWO-POINT ADAPTIVE NONLINEARITY APPROXIMATION FOR DESIGN OPTIMIZATION G. Magazinović Keywords: design optimization, function approximation, TANA-3 1. Introduction During the last decade numerous approximation techniques have taken place for application in optimization. Therefore, function approximation techniques emerge as one of the most important fields of interest. A thorough survey on this topic can be found in [Barthelemy and Haftka 1993]. The main objective of various approximation techniques in design and structural optimization is the reduction of repetitive evaluation of cost and constraint functions. In general, more precise function approximation provides faster convergence and more efficient achievement of optimum design. Consequently, the quality of function approximations emerge as one of the most important issues in the field of design and structural optimization. The main purpose of this paper is to investigate and quantify the accuracy of the Two-Point Adaptive Nonlinearity Approximation, TANA-3 [Xu and Grandhi 1998]. The result could be a valuable tool in comparison with other approximation schemes such as the Generalized Convex Approximation, GCA [Chickermane and Gea 1996], numerically evaluated in [Magazinović 2000]. The outline of the paper is as follows: In Section 2, a brief overview of the Two-Point Adaptive Nonlinearity Approximation algorithm is presented. In Section 3, a description of performed numerical tests is given and finally, in Section 4, numerical results are presented. 2. Two-Point Adaptive Nonlinearity Approximation 2.1 TANA-3 Basics The Two-Point Adaptive Nonlinearity Approximation, TANA-3, is the last and probably the best approximation scheme in a well known series of four function approximations: TANA [Wang and Grandhi 1994], TANA-1 and TANA-2 [Wang and Grandhi 1995], and finally TANA-3 [Xu and Grandhi 1998].
    [Show full text]
  • To Optimize the Design of a Magnetic Motor Chandana M1, Charan N2, Dikshith Gowda R3, Kiran R4, M
    ISSN (Online) 2581-9429 IJAR ST International Journal of Advanced Research in Science & Technology (IJARST) Volume 7, Issue 4, July 2020 To Optimize the Design of a Magnetic Motor Chandana M1, Charan N2, Dikshith Gowda R3, Kiran R4, M. J. Chandrashekar5 UG Students, Department of Electrical and Electronics Engineering1,2,3,4 Associate Professor, Department of Electrical and Electronics Engineering5 SJB Institute of Technology, Bangalore, Karnataka [email protected], [email protected], [email protected] [email protected] and [email protected] Abstract: The main objective of this work is to study the free energy permanent magnet motor, where the natural repulsion or attraction characteristics of magnet poles creates a perpetual motion which can be harnessed by the magnet motor. The essential idea of this work has started by taking advantages of permanent magnet properties in these machines when replacing electric coils by permanent magnets, to build a free energy magnetic motor. The design of rotor drum plays a vital role in these magnetic motor where the magnetic field and repelling force is when the magnets is separated by a definite distance is considered. The simulation is done using a matlab software where the electrical parameters are simulated using the magnet properties and electrical equations. The materials were selected according to the characteristics and influencing factor such as temperature, weight and stored energy. The permanent magnets have invisible and continuous power where it can attract the iron or other specific kind of metal with the energy flow which allows the permanent magnet to defy the gravity for years. Therefore, it is believed that the free energy can be extracted from the permanents magnets by arranging the magnets in special configuration.
    [Show full text]
  • International Journal of Innovative Technology and Exploring Engineering
    International Journal of Innovative Technology and Exploring Engineering ISSN : 2278 - 3075 Website: www.ijitee.org Volume-8 Issue-9S3, JULY 2019 Published by: Blue Eyes Intelligence Engineering and Sciences Publication xploring En E gi d ne an e r y in g g lo o n h c e T IjItEe e I n v i t t e E a X r v P N n o L O a O I n T t R A i o I V n N O I G N n IN f a o l l J a o r n u www.ijitee.org Exploring Innovation Editor-In-Chief Chair Dr. Shiv Kumar Ph.D. (CSE), M.Tech. (IT, Honors), B.Tech. (IT), Senior Member of IEEE Professor, Department of Computer Science & Engineering, Lakshmi Narain College of Technology Excellence (LNCTE), Bhopal (M.P.), India Associated Editor-In-Chief Chair Dr. Dinesh Varshney Professor, School of Physics, Devi Ahilya University, Indore (M.P.), India Associated Editor-In-Chief Members Dr. Hai Shanker Hota Ph.D. (CSE), MCA, MSc (Mathematics) Professor & Head, Department of CS, Bilaspur University, Bilaspur (C.G.), India Dr. Gamal Abd El-Nasser Ahmed Mohamed Said Ph.D(CSE), MS(CSE), BSc(EE) Department of Computer and Information Technology , Port Training Institute, Arab Academy for Science ,Technology and Maritime Transport, Egypt Dr. Mayank Singh PDF (Purs), Ph.D(CSE), ME(Software Engineering), BE(CSE), SMACM, MIEEE, LMCSI, SMIACSIT Department of Electrical, Electronic and Computer Engineering, School of Engineering, Howard College, University of KwaZulu- Natal, Durban, South Africa. Scientific Editors Prof.
    [Show full text]
  • Generative Design by Reinforcement Learning: Enhancing the Diversity of Topology Optimization Designs
    Generative Design by Reinforcement Learning: Enhancing the Diversity of Topology Optimization Designs Seowoo Jang Department of Electrical and Computer Engineering Seoul National University [email protected] Soyoung Yoo Department of Mechanical Systems Engineering Sookmyung Women’s University [email protected] Namwoo Kang* Department of Mechanical Systems Engineering Sookmyung Women’s University [email protected] *Corresponding author Abstract Generative design refers to computational design methods that can automatically conduct design exploration under constraints defined by designers. Among many approaches, topology optimization-based generative designs aim to explore diverse topology designs, which cannot be represented by conventional parametric design approaches. Recently, data-driven topology optimization research has started to exploit artificial intelligence, such as deep learning or machine learning, to improve the capability of design exploration. This study proposes a reinforcement learning (RL) based generative design process, with reward functions maximizing the diversity of topology designs. We formulate generative design as a sequential problem of finding optimal design parameter combinations in accordance with a given reference design. Proximal Policy Optimization is used as the learning framework, which is demonstrated in the case study of an automotive wheel design problem. To reduce the heavy computational burden of the wheel topology optimization process required by our RL formulation, we approximate the optimization process with neural networks. With efficient data preprocessing/augmentation and neural architecture, the neural networks achieve a generalized performance and symmetricity-reserving characteristics. We show that RL-based generative design produces a large number of diverse designs within a short inference time by exploiting GPU in a fully automated manner.
    [Show full text]
  • Building Optimization Through a Parametric Design Platform: Using Sensitivity Analysis to Improve a Radial-Based Algorithm Performance
    sustainability Article Building Optimization through a Parametric Design Platform: Using Sensitivity Analysis to Improve a Radial-Based Algorithm Performance Nayara R. M. Sakiyama 1,2,*, Joyce C. Carlo 3, Leonardo Mazzaferro 4 and Harald Garrecht 1 1 Materials Testing Institute (MPA), University of Stuttgart, Pfaffenwaldring 2b, 70569 Stuttgart, Germany; [email protected] 2 Institute for Science, Engineering and Technology (ICET), Federal University of the Jeq. and Muc. Valleys (UFVJM), R. Cruzeiro, 01-Jardim São Paulo, Teófilo Otoni 39803-371, Brazil 3 Architecture and Urbanism Department (DAU), Federal University of Vicosa (UFV), Av P. H. Rolfs, Viçosa 36570-900, Brazil; [email protected] 4 Laboratory of Energy Efficiency in Buildings (LabEEE), Federal University of Santa Catarina (UFSC), Caixa Postal 476, Florianópolis 88040-970, Brazil; [email protected] * Correspondence: [email protected] Abstract: Performance-based design using computational and parametric optimization is an effective strategy to solve the multiobjective problems typical of building design. In this sense, this study investigates the developing process of parametric modeling and optimization of a naturally ventilated house located in a region with well-defined seasons. Its purpose is to improve its thermal comfort during the cooling period by maximizing Natural Ventilation Effectiveness (NVE) and diminishing annual building energy demand, namely Total Cooling Loads (TCL) and Total Heating Loads (THL). Citation: R. M. Sakiyama, N.; C. Following a structured workflow, divided into (i) model setting, (ii) Sensitivity Analyses (SA), and Carlo, J.; Mazzaferro, L.; Garrecht, H. (iii) Multiobjective Optimization (MOO), the process is straightforwardly implemented through Building Optimization through a a 3D parametric modeling platform.
    [Show full text]