From 3rd to 6th March, Eptisa will have the honour to welcome Dr. Zhen Sun at its headquarters. We will have the pleasure to enjoy several presentations related to AI Technology use on bridge testing, also testing and supervision market at Portugal.
About the Speaker
Zhen Sun is a postdoctoral researcher in the Department of Civil Engineering at the University of Porto, Portugal. He got his PhD from Yokohama National University in Japan. His main research interests include vehicle-bridge interaction analysis, machine learning-based structural condition assessment, structural monitoring and signal processing, and damage detection/load-carrying capacity evaluation of bridges. He published over 40 academic articles with over 400 citations. He also holds six registered invention patents, and has participated in three standards/codes related to the condition assessment of bridges. He has contributed to condition assessment and load-carrying capacity evaluation of bridges in several countries including China, Japan and Portugal.
He is the supervisor of one PhD student and co-supervisor of three master students. He was the technical committee member at ICSBOC (International Cable Supported Bridge Operators’ Conference) 2022 in Kobe, Japan. He also serves as a guest editor for a special issue “Failure mechanism and prevention of civil infrastructure under operational and extreme conditions” in Engineering Failure Analysis (Elsevier, Impact factor: 3.634) and is the reviewer for more than 10 international journals.
Bridges play an essential role in the transportation infrastructure network, and their safety ensures traffic functioning and economic development. The collapse of Morandi Bridge in Genoa, Italy in August 2018 brought about deep concerns among management agencies of highway and railway bridges. It is imperative to keep bridges safe and reliable for our society.
Nowadays, visual inspection and structural health monitoring have provided a large volume of data on ageing bridges, and such big data has posed a challenge for traditional structural analysis approaches. Thanks to the advancement in AI technology, sophisticated machine learning and deep learning methods have been developed and applied more and more in the asset management of bridges.
This talk will present the research on bridge condition assessment with physics-based machine learning methods. Firstly, a condition rating method is introduced for highway bridges based on natural language processing and machine learning, which is verified with inspection reports of 263 bridges. Secondly, machine learning-based approaches are developed to categorize the trainloads and estimate the fatigue damage in the truss girder of a suspension bridge in Portugal. Thirdly, truck weight limit determination is carried out for bridges with a reliability-based method considering the influence of the stochastic traffic flow. At last, the talk will describe the future trends regarding intelligent infrastructure management.