Leakage Signal Denoising of Water Supply Pipelines Based on the Gray Wolf Optimization Algorithm

Abstract

Abstract To address the issue of suboptimal denoising performance in variational mode decomposition (VMD) due to the reliance on empirical parameter selection, this paper proposes a novel joint denoising method (GVMD) based on the gray wolf optimization (GWO) algorithm for the optimal selection of VMD parameters. First, the GWO algorithm is used to optimize the number of intrinsic mode functions and the penalty factor. Then, the signal is processed by combining the Pearson correlation coefficient (PCC) and dynamic time warping (DTW) to effectively distinguish between noise and valid signals. Finally, the signal is reconstructed to obtain the negative pressure wave (NPW) signal. The simulation results demonstrate that, compared to wavelet transform (WT), VMD, and the VMD optimization algorithm based on northern goshawk [Northern Goshawk optimization (NGO)-VMD], the proposed method achieves the highest signal-to-noise ratio (SNR) of 16.57 dB and the lowest root-mean-square error (RMSE) of 0.151. Additionally, a leakage simulation experimental platform was established to further validate the algorithm’s performance. Experimental results indicate that GVMD produces a smoother NPW signal, preserves the waveform characteristics of the signal, and is capable of identifying distinct NPW inflection points.