唐天宁博士来访实验室并做学术论坛

发布日期：2023-12-19

报告题目

时间

2023年12月19日

地点

木兰船建大楼B808会议室

报告人

唐天宁博士

 

报告人简介

 Tim Tang is currently working as an Eric and Wendy Schmidt Al in Science Postdoctoral Fellow at the University of Oxford. He graduated with a BEng from the University of Nottingham coming top in each year of the course, He moved straight to a DPhil at Oxford publishing 5 journal papers including four JFMs. He won first place in the Osborne Reynolds Day competition for the UK's best DPhil student in fluid mechanics.His thesis covered the analysis of field data, experiments, numerical modelling using two different models and the application of data science methods to ocean engineering problems. His current research focuses on extreme events in fluid mechanics with an emphasis on machine learning, including data-driven predictions on extreme waves and extreme structural loading, considering leading order physics such as nonlinear wave dynamics and instabilities, breaking waves, and long-short wave interaction.

报告摘要

Model testing is common in coastal and offshore engineering. The design of such model tests is important such that the maximal information of the underlying physics can be extrapolated with a limited amount of test cases. The optimal design of experiments also requires considering the previous similar experimental results and the typical sea-states of the ocean environments. in this study, we develop a model test design strategy based on Bayesian sampling for a classic problem in ocean engineering -- nonlinear wave loading on a vertical cylinder. The new experimental design strategy is achieved through a Gp-based surrogate model, which considers the previous experimental data as the prior information. The metocean data are further incorporated into the experimental design through a modified acquisition function. We perform a new experiment, which is mainly designed by data-driven methods including several critical parameters such as the size of the cylinder and all the wave conditions. We examine the performance of such a method when compared to traditional experimental design based on manual decisions. This method is a step forward to a more systematic way of approaching test designs with marginally better performance in capturing the higher-order force coefficients, The current surrogate model also made several `interpretable` decisions which can be explained with physical insights.