Damage Localization in Rail Section Using Single AE Sensor Data: An Experimental Investigation with Deep Learning Approach

Abstract

ABSTRACT Railways serve as a vital link for global trade and transportation in any country, but the rail sections are susceptible to damage due to factors such as traffic, extreme environment, and other unavoidable conditions. Monitoring such damages in real time is crucial to prevent casualties and economic losses. Non-destructive testing (NDT) techniques have been used for damage localization, and the acoustic emission (AE) technique has gained attention for real-time monitoring. However, conventional AE approaches are complex, time-consuming, and require multiple sensors. An alternative method is needed for easy and effective implementation of damage localization in rail sections using AE signals. In this study presents, a deep learning approach deploying Artificial Neural Network (ANN) and Support Vector Machine (SVM) models under AI is illustrated experimentally for easy and effective implementation of the damage localization process in the rail section. The novelty in this approach is the application of single AE sensor data which makes the damage localization process economical and less time-consuming. These findings have significant implications for the scientific community and rail transportation industries, ensuring safe and efficient operations..