Flaw detection in high speed train's rail based on EMD and PSD

This paper proposes a new flaw detection method based on EMD and PSD with application to high speed train's rail. The signals in different scenarios are generated by utilizing the finite element model. In this paper, the empirical mode decomposition method is used to decompose the investigated signals into intrinsic mode functions and the flaw features are extracted by utilizing power spectral density. By comparing the relationship between signals with and without flaw in power spectral density, the decision-making criterion for the existence and position of the flaw is founded. Simulation results demonstrate the effectiveness of the proposed method.