An adaptive extraction method for rail crack acoustic emission signal under strong wheel-rail rolling noise of high-speed railway

Abstract

Aiming to detect the weak rail crack signal under strong Wheel-rail Rolling Noise (WRRN) in high-speed railway by Acoustic Emission (AE) technology, a Hurst exponent-improved Adaptive Line Enhancer (ALE) is put forward. The Hurst exponent is adopted to describe the irregularity and fractality property, and is introduced into the cost function of ALE through its power-law relation with the structure function of the fractional Brownian motion so that the optimal objective of the adaptive filter is improved to adapt to the rail crack signal. Compared with the ALE and the Shannon entropy-improved ALE, the proposed method has the best performance. The Hurst exponent-improved ALE not only suppresses the strong WRRN in high-speed condition, but also enhances the rail crack signal and the signal to noise ratio to a large extent. Moreover, with the entire frequency components of the desired signal reserved, the method has the benefits of low computational cost and simple implementation. The research offers a method for extraction of the weak rail crack signal, solves the WRRN problem, and makes AE technology applicable in high-speed rail defect detection.