Inversion of rock failure mechanisms based on waveform principal component extraction and hybrid moment tensor and its application

The moment tensor (MT) inversion method, developed based on the theory of elastic wave propagation, has now become an important tool for studying rock mass failure mechanisms in the mining industry. However, the complex mining environment, filled with various noise sources, often results in poor microseismic (MS) waveform quality, and the propagation medium is complex and variable. These factors make the inversion of rock failure mechanisms using MS data often unreliable, which is highly detrimental to the interpretation of the processes and mechanisms underlying rock mass dynamic disasters. In light of this, this paper proposes a joint MT inversion method based on multichannel singular spectrum analysis (MSSA) combined with a hybrid MT inversion method, and conducts effect verification and field application: First, the primary components of the MS waveforms are extracted using the MSSA method, achieving the separation of rock failure signals from noise signals. Then, it is combined with the hybrid MT inversion method to reduce the impact of noise and propagation medium on the MT inversion, and the accuracy of the proposed method is validated using laboratory acoustic emission data. Finally, the method is applied in engineering practice to interpret and analyze the formation mechanism of seepage channels in the research area of the Shirengou Iron Mine.