A bearing fault signal denoising method based on MCEEMD and cane mouse optimization algorithm

By optimizing the MCEEMD parameters using the sugarcane mouse optimization algorithm and combining cross-correlation coefficient screening and wavelet threshold denoising, the problems of suppressing and preserving bearing fault signals in complex noise environments are solved, thereby improving the reliability and accuracy of fault identification.

CN122196365APending Publication Date: 2026-06-12FUZHOU UNIV
View PDF 2 Cites 0 Cited by

Patent Information

Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
FUZHOU UNIV
Filing Date
2026-03-10
Publication Date
2026-06-12

AI Technical Summary

Technical Problem

Existing technologies struggle to effectively suppress noise interference and retain bearing fault characteristic information in complex noise environments, affecting the accuracy and reliability of fault identification.

Method used

The MCEEMD method based on the cane rat optimization algorithm is adopted. The parameters of MCEEMD are optimized by local minimum envelope entropy. Combined with cross-correlation coefficient screening and improved wavelet threshold denoising, the adaptive decomposition and denoising of the signal are realized.

Benefits of technology

It significantly suppresses noise interference, effectively preserves fault characteristic information, improves the reliability and accuracy of fault identification, has strong adaptability, and reduces computational complexity.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN122196365A_ABST
    Figure CN122196365A_ABST
Patent Text Reader

Abstract

The application provides a bearing fault signal denoising method based on a MCEEMD algorithm of a cane mouse optimization algorithm, takes a local minimum envelope entropy as an adaptive function, and adaptively searches for key parameters of MCEEMD through the cane mouse optimization algorithm; and based on the optimized parameters, the original fault signal is decomposed through MCEEMD to obtain a plurality of IMF components; the correlation coefficient method is used to screen relevant components, the improved wavelet threshold denoising method is used to process the relevant components and discard irrelevant components; the denoised components are reconstructed to realize the denoising of the signal; the application can effectively retain the characteristic information related to the fault state in the signal while significantly suppressing the noise interference, thereby improving the reliability and accuracy of fault identification.
Need to check novelty before this filing date? Find Prior Art