Crack Detection in A Cantilever Beam Using Correlation Model and Machine Learning Approach

Main Article Content

Vikas KHALKAR
Pratik OAK
A MOSHI
Pon HARIHARASAKHTISUDHAN
Lalitkumar JUGULKAR
Raman BANE

Abstract

Crack in a structural member alters local stiffness that affects the dynamic response, such as natural frequency and mode shapes. The purpose of structural health monitoring is to diagnose and predict structural health. In this paper, a correlation model is developed to detect crack parameters, i.e., crack location and crack depth, in the beam. To evaluate the authenticity of the developed correlation model, the Artificial Intelligence-based approach is used to predict the crack parameters. Twenty-three Artificial Intelligence algorithms were used to predict the locations and depths of the crack in a cantilever beam. The developed correlation model used the first two normalized natural frequencies to predict the crack parameters. On the other hand, the first three normalized natural frequencies were used to input the machine learning models to predict the crack parameters. In this research study, V-shaped and U-shaped open edges cracks were considered on the cantilever beam. FEA software, ANSYS, is used to do the modal vibration analysis of various cracked cases of beams. The data set of V-shaped and a U-shaped cracked case obtained from finite element analysis (FEA) were used to develop the correlation model and machine learning models. The results for crack locations and crack depth obtained from the correlation model and machine learning models agree with the actual results. In the future, the proposed correlation model of crack detection can be used to detect cracks in more complicated structures.


 

Downloads

Download data is not yet available.

Article Details

How to Cite
[1]
2023. Crack Detection in A Cantilever Beam Using Correlation Model and Machine Learning Approach. Romanian Journal of Acoustics and Vibration. 19, 2 (Mar. 2023), 121–133.
Section
Articles

How to Cite

[1]
2023. Crack Detection in A Cantilever Beam Using Correlation Model and Machine Learning Approach. Romanian Journal of Acoustics and Vibration. 19, 2 (Mar. 2023), 121–133.