Uncertainty of microseismic sources identification and probabilistic location in underground excavation

Abstract

Microseismic (MS) source location is an integral component of MS technology and essential to understanding the rock failure mechanism and avoiding potential geological hazards in underground rock excavation. However, accurate location remains challenging owing to the complex geological conditions and unknown rock failure mechanisms. In this study, a novel location framework was developed to locate the MS source positions and their uncertainties based on probabilistic programming. Probabilistic programming was utilized to determine the coordinates of the MS source and its variation using the Markov Chain Monte Carlo (MCMC) method based on the waveform equation. A classical benchmark problem was utilized to verify and illustrate the developed framework. The developed framework can not only locate the position of the MS source but also determine its variation due to the uncertainty during the monitoring and excavation. The located MS source is in agreement with the actual positions. The results show that the developed framework is a scientific, accurate, reasonable, and promising tool for the location of MS sources. Then, the developed framework was applied to locate the position of the blasting in a practical mine. This further proved that the developed framework could locate the MS source, providing an excellent uncertainty analysis tool for underground rock excavation.