In-situ Fatigue Damage Detection for Railway Structures Using Nonlinear Ultrasonic Guided Waves

Abstract

This paper presents a damage detection strategy for rail tracks by taking advantage of the nonlinear interactions between guided waves and fatigue cracks. A Local Interaction Simulation Approach (LISA) was developed to achieve efficient simulation of ultrasonic guided wave propagation in railway structures. The stick-slip contact dynamics at the fatigue cracks was integrated into the LISA model using the penalty method and Coulomb friction law. The LISA procedure was coded with the Compute Unified Device Architecture (CUDA), which enables the highly parallelized supercomputing on powerful Graphics Processing Units (GPUs). Guided waves generated at the transmission location propagate along the rail track, interact with the fatigue crack, carry the Contact Acoustic Nonlinearity (CAN) with them, and are finally picked up at the sensing location. Various fatigue crack size cases were simulated to investigate the change of nonlinear spectrum characteristics from the growing damage severity. A damage index was developed based on the nonlinear ultrasonic energy proportion in the sensing signals to monitor the existence and severity of fatigue cracks. This study shows that the nonlinear ultrasonic techniques possess promising potential for the in-situ health monitoring of railway structures.