Reliability analysis using hybrid advanced machine learning models

报告题目：Reliability analysis using hybrid advanced machine learning models

报告人：Behrooz Keshtegar教授/博导/国际杰青

报告时间：2025年1月3日（周二），上午9：30

报告地点：明志楼B504

报告人简介：Behrooz Keshtegar is a staff member in Department of Civil Engineering at University of Zabol. He received his Ph.D. from Sistan and Baluchestan, Zahedan, Iran in 2013. His research fields focused on the multidisciplinary topics such as; structural reliability analysis, reliability–based design optimization, machine learning methods and optimization -based population approaches. He has published more than 150 peer-reviewed papers in scientific journal. He received a prize of university of Zabol for four most cited papers published in JOF and CMAME journals in 2022. He got distinguished top researcher form university of Zabol at 2018 to 2020. He received young researcher of civil engineering from the academy of Science of Iran, IRAN at 2022 and he cited in list of world’s top 2% scientists since 2019。

报告内容摘要：The accuracy, efficiency, stable results are major issues of the computational reliability methods. The performance-based accuracy and efficiency with robust searching scheme for determining the most probable point (MPP) is proposed by hybrid strategies given by first order reliability method (FORM) and hybrid support vector regression (SVR) coupled by population-based optimization method of global best particle swarm optimization (GPSO). The PSO and GPSO combined with SVR and also applied for MPP search are compared using traditional FORM iterative algorithms for accuracy, robustness and efficiency through several engineering problems.

主办单位：机电工程学院

石油天然气装备教育部重点实验室

皇家永利深层地热能装备技术工程研究中心

皇家永利地热能开发与利用工程技术研究中心

石油天然气装备技术皇家永利科技资源共享服务平台

成都市机械工程学会

科学技术发展研究院