Abstract

Bayesian intelligent structural health monitoring (BISHM) integrates Bayesian inference and machine learning to develop an effective and efficient framework for reliable structural health monitoring (SHM). Bayesian inference provides a rigorous scheme for utilizing the available information to determine the optimal estimates and quantify the associated estimation uncertainty in the form of probability distribution. On the other hand, machine learning offers a powerful computational environment to automate analytical modeling and handle tremendous amount of dynamic data streams. By exploiting the remarkable features of Bayesian inference and machine learning, BISHM offers a promising direction to tackle challenging issues encountered in SHM investigation. In this seminar, our recent developments on BISHM and the applications to benchmark SHM projects will be presented.

Biography: Dr. Kuok is an academic visitor in the Department of Engineering Science at University of Oxford. She received her Ph.D. degree in Civil Engineering from University of Macau in 2015 and worked as a postdoctoral scientist at Cornell University afterward. Since 2018, she has been being employed as a lecturer at University of Macau and she is currently under a two-years collaborative research program in UK. Her research interest focuses on developing reliable structural health monitoring methodologies via Bayesian inference and machine learning. The implementations of her developed algorithms cover model class selection, optimal sensor network configuration design, signal pre-processing, outlier cleansing, system identification, seismic attenuation modeling, structural health indicator interpretation, and wind-structural interaction analysis.