Theses and Dissertations

Date of Award

5-2022

Document Type

Thesis

Degree Name

Master of Science (MS)

Department

Civil Engineering

First Advisor

Dr. Jungseok Ho

Second Advisor

Dr. Dongchul Kim

Third Advisor

Dr. Andrew Ernest

Abstract

Texas coastal communities are at constant risk of hurricane impacts every storm season. It is especially important to model and predict storm surge variations during hurricane and storm events. Traditionally, hurricane storm surge predictions have been the result of numerical hydrodynamics based simulations. This type of simulations often requires high amounts of computational resources and complex ocean modelling efforts. Recently, machine learning techniques are being explored and are gaining popularity in hydrologic and ocean engineering modelling fields based on their performance to model nonlinear relationships and low computational requirements for prediction. Advances in machine learning and artificial intelligence (A.I.) demand the application of these methods for the modelling of complex problems such as storm surge. This study gathers historical water level data from coastal buoy stations, uses gridded forecasted weather datasets, and builds a database of ADCIRC hydrodynamic simulations to create a machine learning based surrogate model to provide timely, non-computationally intensive and accurate storm surge predictions for the Lower Laguna Madre in Texas.

Comments

Copyright 2022 Cesar E. Davila Hernandez. All Rights Reserved.

https://go.openathens.net/redirector/utrgv.edu?url=https://www.proquest.com/dissertations-theses/machine-learning-based-surrogate-model-hurricane/docview/2699720602/se-2?accountid=7119

Share

COinS