An Enhanced Variational Mode Decomposition and Permutation Entropy Algorithm for Signal Denoising Applications

Abstract

Partial discharge (PD) detection is critical for assessing the insulation performance of power cables, as PD signals are often contaminated by periodic narrowband interference, white noise, and random pulse interference. Among these, narrowband interference poses significant challenges due to its large amplitude and high energy, which obscure the PD signals. Variational Mode Decomposition (VMD) has shown promise in denoising PD signals, but its performance heavily depends on the optimal selection of the number of modes (K). To address this issue, this paper proposes a novel method for determining the optimal K value based on the average orthogonal value, enabling accurate and intuitive selection to improve VMD decomposition performance. Additionally, permutation entropy is employed to reconstruct effective components and suppress periodic narrowband interference. Simulation and experimental results demonstrate that the proposed method effectively removes interference, minimizes PD signal waveform distortion, and achieves a high noise suppression ratio compared to conventional VMD and ensemble empirical mode decomposition methods. This approach provides a robust and reliable technique for PD signal denoising in power cable insulation assessment.