Damage Detection in beams using Vibration analysis and Artificial Neural Network

Abstract

The investigation of cracks in various structures is vital in mechanical applications. Detection of cracks becomes an important task since failure to do so may result in strength reduction which may lead to failure of a system. The detection of cracks is broadly carried out using NDT methods, vibration-based methods, and various mathematical models. An alternative way for detecting cracks that can be used is Artificial Neural Networks (ANN) based modelling. The evolution in the ANN has brought up various new potential in the arena of complex problems. To recognize the cracks in flat plates using non-destructive intelligent identification, Artificial Neural Networks is used. Artificial Neural Networks is used to enhance the efficiency in the identification of these cracks. A dataset is created by performing modal analysis using simulation tools like ANSYS. This dataset would be input to the neural network as a training set. Hence a neural network would be trained which will give crack location as the output.