基于VMD-SSA的埋地输气管道泄漏声波信号降噪研究

Noise Reduction of Leakage Acoustic Signal of Buried Gas Pipeline Based on VMD-SSA

针对埋地输气管道泄漏声波信号沿土壤介质传播过程中易受噪声干扰的问题,提出基于变分模态分解结合奇异谱分析的联合降噪方法。先对含噪信号进行VMD分解,并根据能量熵分布图重构去除高频噪声。再对余下含中低频噪声的信号分量进行SSA分析,结合奇异值差分谱和能量贡献选取最优奇异值重构得到最终降噪信号。实验结果表明,相较于小波阈值、变分模态分解等降噪方法,改进方法降噪前后信噪比均增量提高27.1%,均方根误差均减量提高13.8%,是一种更有效的降噪方法。

In order to solve the problem that the acoustic signal of buried gas pipeline leakage is easily disturbed by noise when it propagates along soil medium,a joint noise reduction method based on variational mode decomposition(VMD)and singular spectrum analysis(SSA)is proposed.Firstly,the noisy signal is decomposed by VMD,and the high-frequency noise is removed by reconstruction according to the energy entropy distribution map;the remaining signal components with medium and low-frequency noise are analyzed by SSA,and the optimal singular value is selected by combining the singular value difference spectrum and energy contribution to reconstruct the final denoised signal.The experimental results show that compared with wavelet threshold,variational mode decomposition and other denoising methods,the signal-to-noise ratio increment and root mean square error reduction of the improved method increase by 27.1%and 13.8%respectively,which is a more effective denoising method.