Modal Analysis Based Machine Learning Method Structure Using Ansys Software

Chin, Mark Ashley (2026) Modal Analysis Based Machine Learning Method Structure Using Ansys Software. Final Year Project (Bachelor), Tunku Abdul Rahman University of Management and Technology.

Text MARK ASHLEY CHIN_Full Text.pdf Restricted to Registered users only Download (16MB)

Abstract

The combination of machine learning (ML) techniques with finite element analysis (FEA) has become increasingly popular due to its ability to improve computational efficiency without sacrificing accuracy. In this project, the focus is on creating an ML-based framework to predict the modal characteristics of structures using ANSYS simulations. ML models will be trained with data generated through various sampling methods, such as Latin Hypercube Sampling, Optimal space filling design and Box Behnken Sampling. The models performance is then assessed by looking at its predictive accuracy and computational speed, comparing the results to traditional FEA outcomes. The goal is to determine the best pairing of ML method and sampling strategy that can reduce simulation time while still ensuring reliable results, ultimately contributing to advancements in modal analysis for applications like digital twins and structural health monitoring