Maximum correlation Pearson correlation coefficient deconvolution bearing fault diagnosis algorithm based on generalized spherical coordinate transformation
YANG Gang
CHENG Lei
XU Wuyi
DENG Qin
KANG Haoming
Abstract：[Objective]The maximum correlation Pearson correlation coefficient deconvolution(MCPCCD)algorithm can enhance the amplitude-frequency characteristics of bearing fault impulses under strong background noise while preserving their phase-frequency characteristics.However,the algorithm is prone to failure under the interference of random impacts and harmonic components,and its performance is limited by the accuracy of the prior deconvolution period.To address these issues,an improved MCPCCD method was proposed.[Methods]Firstly,the coefficients of the deconvolution filter were projected onto an L-1 dimensional sphere with a radius of 1 via generalized spherical coordinate transformation to narrow the search space.Secondly,the sparrow search algorithm was adopted to optimize the filter coefficients for signal filtering.Finally,simulated signals and measured bearing signals of high-speed train traction motors were employed to verify the proposed algorithm.[Results]The results show that the proposed algorithm solves the failure problem of the original MCPCCD algorithm under strong interference and maintains satisfactory performance even when the deconvolution period deviates.The results of the comparative studies demonstrate that the proposed algorithm possesses better anti-interference capability than the maximum correlated kurtosis deconvolution algorithm and the empirical mode decomposition algorithm.
Keywords：Maximum correlation Pearson correlation coefficient deconvolutionFault diagnosisFault characteristic intensity indexGeneralized spherical coordinate transformationSparrow search algorithm
Publication Date：2026-06-30
Online Publishing Date：2026-09-12(First online date of this platform, not the publication date of the document)
Pages：10( 52-61 )