Efficient Uncertainty Quantification for Complex Systems Analysis【土建学院“建苑论坛”2019年第12讲】
题 目：Efficient Uncertainty Quantification for Complex Systems Analysis
时 间：2019年4月3日 上午 9:30
In our developed societies, engineering systems are characterized by a rapid growth in scale and complexity. The amount of information needed to model these systems with their complexity is, thus, growing as well. In contrast to this increasing need for information the available information remains almost at the same level. Hence, with increasing scale and complexity the gap between required and available information is growing quickly, so that uncertainties and risks are involved in our models and analyses to a greater extent than ever before. This challenge can be addressed well with concepts of imprecise probabilities for reliability assessment of engineering systems when only limited information is available. In order to achieve high numerical efficiency, in particular when dealing with large complex systems, the concept of survival signature is adopted for reliability assessment. Based on the developments of a survival analysis and importance analysis of systems with multiple types of components it is shown how imprecise probabilities help to reveal the most critical components of the system and the most critical uncertainties, as well. Novel pathways to capture interdependencies between systems while estimating their reliability efficiently in a time dependent manner are discussed. Engineering examples are presented to demonstrate the capabilities of the approaches and concepts. This includes diagnosis and maintenance of complex machinery at high computational efficiency. Conclusions on a targeted reduction of imprecision are drawn.
Michael Beer is Professor and Head of the Institute for Risk and Reliability, Leibniz Universit?t Hannover, Germany, since 2015. He is also part time Professor at the University of Liverpool and at Tongji University, Shanghai, China. Dr. Beer is Editor in Chief (joint) of the Encyclopedia of Earthquake Engineering, Associate Editor of the ASCE-ASME Journal of Risk and Uncertainty in Engineering Systems, Associate Editor of the International Journal of Reliability and Safety, and Member of thirteen Editorial Boards including Probabilistic Engineering Mechanics, Computers & Structures, Structural Safety, Mechanical Systems and Signal Processing, and International Journal for Uncertainty Quantification. He has won several awards including the CADLM PRIZE 2007 – Intelligent Optimal Design and a Certificate for Highly Cited Research in Structural Safety. His publications include a book, several monographs and a large number of journal and conference papers.