工程防灾减灾学术讲坛
题目：Application of Bayesian Networks and Parameter Learning Techniques to Integrity Management of Oil and Gas Pipelines
讲座人：Wenxing Zhou
时间：2018-5-30（周三），下午3:30-4:30
地点：灾后重建与管理学院C313
摘要
Bayesian networks (BNs) are directed graphical models expressing the joint probabilistic distribution of a set of random variables by the local conditional probability tables. Given the intuitive graphical nature and the ability to perform Bayesian inference between a large set of random variables efficiently, BNs have received increasing attentions in the structural reliability community. This talk will discuss the application of BNs and parameter learning techniques to the integrity management of buried oil and gas pipelines based on inspection data and observation records. Two case studies involving two major integrity threats, i.e. corrosion and third-party damage, are used to illustrate the development of the BN structures, model parametrization based on parameter learning and validation of the developed models.
报告人简介
Wenxing Zhou, PhD, PEng, is an Associate Professor and Associate Chair – Undergraduate in the Department of Civil and Environmental Engineering at the University of Western Ontario (UWO) in London, Ontario, Canada. He obtained his BEng, MEng, and PhD from Tongji University, Tsinghua University and UWO in 1993, 1996 and 2001, respectively. His main research interest is the reliability- and risk-based structural integrity management of oil and gas pipelines. Before joining UWO, Dr. Zhou had extensive practical experience in the Canadian pipeline industry. He has published more than 100 papers, including over 60 peer-reviewed journal papers and 40 peer-reviewed conference papers. He is a Vice Chair of the Technical Committee of the Canadian Standards Association (CSA) standard Pipeline Safety Metrics, CSA Z260.
新葡萄8883官网AMG中国西部抗震中心
新葡萄8883官网AMG
新葡萄8883官网AMG灾后重建与管理学院
欢迎广大师生参加
