Session: 06-03-01 Fluid-Structure, Multi-body and Wave-body Interaction I

Paper Number: 102682

102682 - Data Informed Sequential Model Test Design With Machine Learning – an Example in Nonlinear Load on Vertical Cylinder. 

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.

Presenting Author: Tianning Tang University of Oxford

Presenting Author Biography: Tianning Tang is currently working as a postdoc researcher at University of Oxford. He graduated with a BEng from the University of Nottingham coming first in his year in each year of the course. He moved straight to a DPhil at Oxford publishing 5 journal papers including two JFMs. His thesis covered analysis of field data, experiments, numerical modelling using two different models and the application data science methods to ocean engineering problems. His current research focuses on extreme events in fluid mechanics with an emphasis on rogue waves in the ocean, including data-driven predictions on extreme waves and extreme structural loading, considering leading order physics such as: nonlinear wave dynamics and instabilities, breaking waves, slamming events on structures.

Authors:

Tianning Tang University of Oxford
Haoyu Ding University of Bath
Saishuai Dai University of Strathclyde
Xi Chen University of Bath
Paul H. Taylor University of Western Australia
Jun Zang University of Bath
Thomas A. A. Adcock University of Oxford