310 / 2018-05-03 15:47:15

The Feature Extraction and Identification of Micro-seismic Signals based on VMD and Singular Spectrum Entropy

singular spectrum entropy,rock fracture micro-seismic signals;,blasting vibrations,VMD

矿山动力灾害 > 岩爆冲击

The accurate identifications of effective micro-seismic events have great significance in the monitoring and early warnings of rock-burst hazards. However, the conventional identification methods have displayed difficulties in achieving satisfactory results. Therefore, in order to address these problems, a micro-seismic event recognition method based on variational mode decomposition (VMD) and multi-resolution singular spectrum entropy was proposed in this study. First, VMD was used to decompose the original signal into a series of variational modal components by frequency from high to low. Second, the characteristic pattern matrix was constructed according to the mode signals, and then the singular spectrum entropy was obtained. Finally, an identification model for micro-seismic events based on a support vector machine was established. This study’s results showed that the differences in the multi-resolution singular spectrum entropy feature vectors of the rock fracture micro-seismic and blasting vibration signals were distinct. In addition, the differences in the intrinsic characteristics of the two types of vibration signals were clearly identified. The method which was proposed in this study had the advantages of high classification accuracy and stability, and this study’s experimental results confirmed that it was an effective method for the identifications of rock fracture micro-seismic signals.