Damage Identification in Thin Steel Beams Containing a Horizontal Crack Using the Artificial Neural Networks

This investigation presents damage identification in thin steel beams containing a horizontal crack using artificial neural networks. In this way, finite element modeling of the cracked beam is developed to generate natural frequencies corresponding to various horizontal cracks scenarios. Then, the artificial neural network is used to create a predictor model for localizing horizontal cracks in steel beams. Results of the current paper show that The proposed technique is an effective method for detecting horizontal crack damage in steel beams. The regression index obtained in this study is equal to 0.979.