Fatigue crack damage detection based on improved energy-power entropy under dynamic temperature

High-speed train structures, operating under time-varying conditions, present a significant challenge in the field of Structural Health Monitoring (SHM) due to uncertainties introduced during the extraction of damage indexes from signals. In this paper, a novel method for quantifying fatigue crack based on improved energy-power entropy under dynamic temperature environments is introduced. Two new types of information entropy, namely energy singular spectral entropy (ES) and power singular spectral entropy (PS) are proposed. A baseline Gaussian Mixture Model (GMM) is constructed using the information entropy acquired under dynamic temperature when the structure is in a healthy state. Subsequently, an online GMM is constructed via an online update mechanism utilizing a moving feature sample set. The minimum matching Kullback–Leibler (KL) distance of the probability component is employed to quantitatively depict the cumulative migration trend of the GMM under damage conditions, thereby aiding in damage detection. The experimental results verify the reliable performance of the GMM-KL method in detecting fatigue cracks under dynamic temperature.