Scholarlink

<p>The accurate detection and analysis of partial discharge (PD) signals play a critical role in engineering practices for assessing insulation conditions and providing early warnings of potential faults in power equipment. To mitigate interference and enhance the acquisition of discernible and accurate partial discharge signals, this paper proposes a joint denoising method that combines adaptive singular value decomposition (ASVD) with the discrete wavelet transform (DWT). By adaptively selecting singular values and denoising thresholds, the proposed method extracts the principal components of the signal, effectively suppressing random noise. Leveraging the multi-resolution analysis capability of the DWT, its decomposition and reconstruction processes are further utilized to eliminate high-frequency noise components. Both simulated PD signals and experimentally acquired PD signals are employed as processing targets in this study, enabling comprehensive evaluation and comparative analyses of various algorithms (Zhong et al. in IEEE Transact Instrument Measure 69(11):8866–8873, 2020); (Jha et al. in IRBM 42(1):65–72, 2021). Experimental results demonstrate that the proposed method consistently outperforms traditional denoising techniques in terms of improving the signal-to-noise ratio while preserving important signal details. Specifically, the method achieves a signal-to-noise ratio (SNR) of 7.66 dB and a normalized correlation coefficient (NCC) of 0.912537 for the simulated PD signal. For the real PD signal, the noise reduction ratio (NRR) reaches up to 7.88 dB, indicating its high practical value and broad application potential.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Denoising of partial discharge signals based on adaptive singular value decomposition and discrete wavelet transform

  • 期刊论文

  • 发表日期：

    2026年2月3日

查看全文
  • Yun Zhao,
  • Hong Liu,
  • Rui Tang,
  • Heng Niu,
  • Dong Sun,
  • Hongyi Lin

摘要

The accurate detection and analysis of partial discharge (PD) signals play a critical role in engineering practices for assessing insulation conditions and providing early warnings of potential faults in power equipment. To mitigate interference and enhance the acquisition of discernible and accurate partial discharge signals, this paper proposes a joint denoising method that combines adaptive singular value decomposition (ASVD) with the discrete wavelet transform (DWT). By adaptively selecting singular values and denoising thresholds, the proposed method extracts the principal components of the signal, effectively suppressing random noise. Leveraging the multi-resolution analysis capability of the DWT, its decomposition and reconstruction processes are further utilized to eliminate high-frequency noise components. Both simulated PD signals and experimentally acquired PD signals are employed as processing targets in this study, enabling comprehensive evaluation and comparative analyses of various algorithms (Zhong et al. in IEEE Transact Instrument Measure 69(11):8866–8873, 2020); (Jha et al. in IRBM 42(1):65–72, 2021). Experimental results demonstrate that the proposed method consistently outperforms traditional denoising techniques in terms of improving the signal-to-noise ratio while preserving important signal details. Specifically, the method achieves a signal-to-noise ratio (SNR) of 7.66 dB and a normalized correlation coefficient (NCC) of 0.912537 for the simulated PD signal. For the real PD signal, the noise reduction ratio (NRR) reaches up to 7.88 dB, indicating its high practical value and broad application potential.