Projection outlines in hybrid reliability-based design optimization
Distinguished Research Seminar Series
-
Date
29 Aug 2022
-
Organiser
Department of Industrial and Systems Engineering, PolyU
-
Time
10:30 - 12:00
-
Venue
Online via ZOOM
Speaker
Prof. Liang GAO
Remarks
Meeting link will be sent to successful registrants
Summary
Traditional reliability analysis (RA) and reliability-based design optimization (RBDO) methods often require a lot of evaluations of performance functions. However, the evaluations of performance functions often involve time-consuming computing processes, such as finite element analysis, which result in high time cost during analysis and optimization. To reduce the number of evaluations of performance functions, surrogate model-assisted approaches get high attention, where the modelling accuracy and efficiency of crucial approximate region directly influence the performance of RA and RBDO methods, such as the limit-state surface in random RA and RBDO. For hybrid RA and RBDO with random and interval uncertainties, we deduce that the crucial approximate region is the projection outlines on the limit-state surfaces. Based on projection outlines, different active learning mechanisms for the cases with independent variables, correlated variables, small failure probabilities and multiple failure modes are proposed, which show high efficiency and accuracy for hybrid RA and RBDO.
Keynote Speaker
Prof. Liang GAO
School of Mechanical Science and Engineering,
Huazhong University of Science and Technology,
Wuhan 430074, China.
Liang Gao received his B.Sc. degree in mechatronic engineering from Xidian University, Xi’ an, China, in 1996, and the Ph.D. degree in mechatronic engineering from Huazhong University of Science and Technology (HUST), Wuhan, China, in 2002. He is a Professor of the Department of Industrial and Manufacturing Systems Engineering (IMSE), the Deputy Director of State Key Laboratory of Digital Manufacturing Equipment and Technology and Chairman of School of Mechanical Science and Engineering, HUST. He was supported by the Program for New Century Excellent Talents in University in 2008 and National Science Fund for Distinguished Young Scholars of China in 2018. His research interests include intelligent optimization algorithms, big data, deep learning with their application in Design & Manufacturing. He published more than 440 papers indexed by SCIE, authored 13 monographs. His citation is 11000 times and H is 55 according to Web of Science. He is Highly Cited Researcher 2020 and 2021 by Clarivate Web of Science and 2020 Highly Cited Chinese Researchers by Elsevier. He currently serves as co-Editor-in-Chief of IET Collaborative Intelligent Manufacturing, Associate Editor of Swarm and Evolutionary Computation, Journal of Industrial and Production Engineering, etc.
You may also like
