A Denoising Method for Ship-Radiated Noise Based on Optimized Variational Mode Decomposition with Snake Optimization and Dual-Threshold Criteria of Correlation Coefficient

Abstract

The ship-radiated noise (SN) is easily affected by other hydroacoustic objects or complex ocean noise when it spreads through water. In order to reduce the impact from the environment, a denoising method for SN based on optimized variational mode decomposition with snake optimization (SO-VMD) and dual-threshold criteria of correlation coefficient (CC) is proposed in this paper. The first step is to optimize the parameter combination, that is, decomposition number K and penalty factor α, of variational mode decomposition (VMD) by snake optimization (SO) with envelope entropy (EE). Then, the input signal using the optimized results is decomposed and the intrinsic mode functions (IMFs) are obtained. After that, the IMFs are classified into three classes with the dual-threshold criteria of CC, including signal components, signal-noise components, and noise components. Finally, all the signal components and the processed signal-noise components denoised by wavelet threshold (WT) are reconstructed together. Simulations performed in this paper demonstrate that SO is the more appropriate optimization for VMD and the proposed method has the more outstanding performance in denoising different kinds of test signals. In addition, experiments on measured SNs show that the proposed method is effective and accurate in denoising.