PATTERN RECOGNITION OF MINE MICROSEISMIC AND BLASTING EVENTS BASED ON WAVE FRACTAL FEATURES
Xuelong Li,Zhonghui Li,Enyuan Wang,Yunpei Liang,Baolin Li,Peng Chen,Yongjie Liu
DOI: https://doi.org/10.1142/s0218348x18500299
2018-01-01
Fractals
Abstract:A microseismic (MS) monitoring system in a mine can monitor the MS signals generated by coal rock rupture and blasting waves and can distinguish the two types of waves more clearly to monitor and analyze the rupture and evolution process of coal rock. According to the nonlinearity characteristics of the waveform, the fractal characteristics of a mine’s MS and blasting waves are analyzed by simple fractal and multifractal theory, and the simple fractal dimension [Formula: see text] and multifractal parameters are obtained, respectively. Results show that the simple fractal dimension [Formula: see text] reflects the complexity and frequency structure of the wave. The simple fractal dimension [Formula: see text] of a blasting wave is larger than that of a mine MS wave, which indicates that the blasting wave is relatively complex with higher frequency, while the mine MS wave is relatively simple with lower frequency. However, the simple fractal dimension [Formula: see text] can only describe the wave integrity features, not the local features. The multifractal parameters can describe the local characteristics of the wave more finely, and the multifractal spectrum describes the probability information of the singularity exponent [Formula: see text]. The singularity exponential range and multifractal spectral width [Formula: see text] of the blasting wave are smaller than those of the mine MS wave, which indicates that the probability measure of distribution unevenness and the degree of partial parameter fluctuation of the blasting wave are more severe than those of the mine MS wave. Wave signal analysis based on simple fractal and multifractal methods can not only obtain the characteristics of the wave strength and spectral structure but also other important information, such as local singularity. Therefore, it is possible to more clearly and conspicuously identify mine MS and blasting waves, so that coal rock rupture can be monitored more accurately.