Research on intelligent denoising method of weak electromagnetic signal induced by coal rock fracture

Electromagnetic radiation (EMR) can reflect coal rock stress, rupture and gas seepage outflow, and is an important means of monitoring coal and rock dynamic disasters such as rock burst, coal and gas outburst. However, there are also problems such as electromagnetic radiation is susceptible to interference and high signal noise, which affect the accuracy of early warning indicators and risk early warning. At present, there is still a lack of effective EMR de-noising means, so it is necessary to find the optimal de-noising method of EMR signal. In view of the above problems, this paper carried out the coal and rock uniaxial compression EMR monitoring experiment, collected the EMR signals of different noise levels in the whole process of coal and rock loading. time-varying filtering empirical mode decomposition (TVF-EMD), variational mode decomposition (VMD) and successive variational mode decomposition (SVMD) algorithms optimized by gray wolf optimization (GWO) algorithm are applied to the de-noising process of EMR signal of coal rock fracture. Compared with the conventional empirical mode decomposition (EMD) and wavelet packet transform (WPT) algorithms, the signal to noise ratio (SNR) after denoising is analyzed. The experimental results show that GWO-TVF-EMD algorithm can adapt to the de-noising of EMR signals with different noise levels, and the average increase in SNR after de-noising is 19.74 compared with the original signal, which is significantly better than other algorithms. The research results are of great significance for the improvement of EMR monitoring and early warning technology.