Session: 01-06-01: AI and Machine Learning in Acoustics and Vibrations

Paper Number: 166661

AI Based Free Vibration Analysis of Sandwich Shells With Viscoelastic Core 

Sandwich structures with viscoelastic cores have gained significant attention in engineering applications due to their superior damping characteristics, lightweight nature, and high structural efficiency. The accurate prediction of their dynamic response is crucial for ensuring their reliability in aerospace, automotive, and civil engineering applications. However, the complex interaction between the viscoelastic core and the isotropic face layers poses significant challenges in understanding their vibrational behaviour. The primary motivation of this research is to tailor the dynamic response of sandwich shells with viscoelastic core and to develop efficient predictive models for their vibration characteristics.

This study presents a comprehensive investigation into the free vibration response of a three-layer sandwich shell with isotropic face layers and a viscoelastic core. The research aims to bridge the gap between experimental and numerical approaches for determining the natural frequencies and modal loss factors of these structures under various boundary conditions. The contribution of this work lies in its integration of experimental modal analysis, finite element modelling, and artificial intelligence-based predictive techniques to provide a robust framework for analysing and optimizing the dynamic behaviour of sandwich shells. By integrating AI-driven methodologies, this research advances the field of structural dynamics and contributes to the development of intelligent structural health monitoring systems.

The methodology employed in this study combines both experimental and numerical approaches. Experimental modal analysis is performed using the impact hammer method to capture the dynamic response of the sandwich shell. Concurrently, finite element modelling is utilized to simulate the vibrational behaviour and validate the experimental results. The numerical model incorporates various geometrical and material parameters to assess their influence on the natural frequencies and damping characteristics. Additionally, a parametric study is performed to analyze the effect of core thickness, face layer properties for different boundary conditions on the free vibration response of the sandwich shell.

To further enhance the predictive capabilities, artificial neural network (ANN)-based models are developed. The ANN model efficiently estimates the natural frequencies and modal loss factors of the sandwich shell, reducing the need for extensive computational simulations. The predictive accuracy of the ANN model is validated against both experimental and numerical results, demonstrating its effectiveness in capturing the complex vibrational behaviour of viscoelastic sandwich structures.

The preliminary results indicate a strong correlation between the experimental and numerical findings, confirming the reliability of the finite element model in predicting the dynamic response of sandwich shells. The parametric study reveals that the viscoelastic core plays a crucial role in damping enhancement, while variations in core thickness and face layer properties significantly influence the natural frequencies and modal loss factors of the sandwich shell. The ANN model exhibits high accuracy in predicting the modal parameters, proving its effectiveness as a computationally efficient alternative to traditional numerical simulations.

In conclusion, this research highlights the potential of AI-driven techniques in the field of structural dynamics, particularly for the analysis of complex sandwich structures. The integration of experimental, numerical, and artificial intelligence approaches provides a powerful framework for understanding and optimizing the vibrational behaviour of sandwich shells. The findings of this study contribute to the advancement of structural health monitoring systems and the development of high-performance materials for engineering applications.

Presenting Author: Sukesh Chandra Mohanty National Institute of Technology Rourkela

Presenting Author Biography: Dr. Sukesh Chandra Mohanty currently works as a professor of Mechanical Engineering Department in National Institute of Technology, Rourkela, Odisha, India. His area of research includes structural dynamics, Composite materials and Gear Dynamics. He has thirty years of experience in teaching and research.

Authors:

Sukesh Chandra Mohanty National Institute of Technology Rourkela
Jagesh Kumar Prusty Madanapalle Institute of Technology & Science