Partial discharge denoising method and system based on adaptive principal component analysis

By using adaptive principal component analysis combined with a noise feature database and a Gaussian mixture model, efficient noise reduction and accurate identification of partial discharge signals are achieved, solving the problem of insufficient noise suppression capability in existing technologies and improving the accuracy and efficiency of signal detection.

CN122132825APending Publication Date: 2026-06-02STATE GRID ZHEJIANG ELECTRIC POWER CO LTD JIAXING POWER SUPPLY CO +1
View PDF 0 Cites 0 Cited by

Patent Information

Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
STATE GRID ZHEJIANG ELECTRIC POWER CO LTD JIAXING POWER SUPPLY CO
Filing Date
2026-03-03
Publication Date
2026-06-02

AI Technical Summary

Technical Problem

Existing technologies have insufficient noise suppression capabilities in partial discharge detection, low signal recognition accuracy and efficiency, and rely on manual parameter adjustment, making them unable to effectively handle partial discharge signals in complex noise environments.

Method used

An adaptive principal component analysis method is adopted. By constructing a noise feature database, effective components are screened using successive variational mode decomposition and cross-correlation coefficients. Signal separation is performed by combining Gaussian mixture model and residual network. Deep noise suppression is performed by combining finite length and infinite impulse response filters. Finally, a closed-loop Bayesian calibration mechanism is used to optimize signal reconstruction.

Benefits of technology

Achieving high-fidelity partial discharge signal reconstruction in complex noise environments improves noise suppression capability and signal recognition accuracy, reduces computational complexity and power consumption, adapts to various noise signals, and enhances detection efficiency.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN122132825A_ABST
    Figure CN122132825A_ABST
Patent Text Reader

Abstract

The present application relates to the technical field of power cable state monitoring and defect positioning, and discloses a partial discharge noise reduction method and system based on adaptive principal component analysis, comprising: constructing a noise feature database, using SVMD to adaptively decompose the partial discharge signal to be denoised, extracting intrinsic mode components through a constraint minimization model, and using a cross-correlation coefficient to identify and screen effective components; constructing a Gaussian mixture model and counting a priori, separating noise and partial discharge signals through a residual network and variational inference; combining a FIR filter and an IIR filter to perform deep noise suppression and signal reconstruction on the separated partial discharge signal; and using a closed-loop Bayesian calibration mechanism to dynamically optimize the reconstructed partial discharge signal, so as to obtain a denoised partial discharge signal. The present application can adaptively denoise the cable partial discharge signal, and improve noise suppression capability, signal identification accuracy and efficiency.
Need to check novelty before this filing date? Find Prior Art