Author ORCID Identifier

https://orcid.org/0000-0002-8881-2366

Author Linked-In Account

www.linkedin.com/in/venkatesan94

Biosketch

I’m a Simulation Engineer with a Ph.D. in Mechanical Engineering, focused on delivering simulation-driven solutions for industrial and robotic products. I’ve led the structural and dynamic simulation of agricultural harvesting systems and developed a cycloid reducer from concept to production—including FEA, multi-body dynamics, material selection, and test rig validation. I’ve also worked on the design and prototyping of an axial flux motor and explored control system development using physics-based models and simulation tools. My background also includes applying machine learning and optimization algorithms to solve engineering problems. With a strong foundation in both academic research and industrial engineering, I’m passionate about solving mechatronic challenges in electric mobility, automation, and robotics.

Date of Award

31-8-2024

Document Type

Thesis

School

School of Mechanical Engineering

Programme

Ph.D.-Doctoral of Philosophy

First Advisor

Dr.M.Karthikeyan

Keywords

Delamination, Ply Drop Off, Natural Frequency, Mode Shape Curvature, Genetic Algorithm

Abstract

Fibre-reinforced polymer (FRP) composite plates have been widely used in various engineering applications due to their high strength-to-weight ratio, corrosion resistance, and durability. Despite their many benefits, Delamination is a separation of the layers in composite material and can occur due to various factors such as impact, fatigue, and manufacturing defects. Identifying delamination in composite structures is crucial for ensuring safety and reliability. In this study, model-dependent vibration methods, which are an effective non-destructive technique, were utilised for delamination identification. The primary objective is to investigate the dynamic behaviour of delaminated thickness-tapered plates and develop a reliable methodology for delamination identification.

Initially, this research aims to address the effect of delamination and form a surrogate model to solve the forward problem in tapered composite plates. Initially, the design points for delamination resistance during the design of the sharp plate were considered. The numerical model incorporates the layerwise theory (LWT) and the "constrained mode" delamination approach. The eigenfrequency analysis was solved using the Finite Element Method. The accuracy of the numerical model is correlated with experimental modal analysis measurements. The surrogate model is developed using the Response Surface Methodology (RSM), Multi-output Artificial Neural Networks (ANN), and single-output ANN model.

The surrogate model predicts the natural frequency shift of the delaminated tapered plate. Two algorithms, such as the Surrogate Assisted Real-coded Genetic Algorithm (SARGA) and direct measurement using ANN, are employed for solving inverse problems to identify the delamination's size, in-plane location, and interface. The literature shows limited research has been conducted to identify damages in tapered laminated composites with ply drop-off using the damage identification methods (Modal Curvature).

Therefore, this study aims to address these gaps by experimentally analysing the damage detection capabilities on a glass fibre-reinforced polymer (GFRP) composite tapered plate under varying severity of delamination. In this study, the modal curvature and its derivative parameters, such as the Damage Index (DI) based on Mode Shape Curvature (MSC), Gapped Smoothing Method (GSM), Mode Shape Curvature Square (MSCS), and Gapped Smoothing Method Square (GSMS), as well as the Curvature Damage Factor (CDF) based on MSC, MSCS, GSM, and GSMS.

Download

Share

COinS