An Improved And Optimized Positioning Method Of Microseismic Source In Mining

Automatically positioning a microseismic source and picking the first arrival are challenging and a focus in microseismic technology research. In order to reduce error caused by artificial picking arrival and speeds in different mediums, a new automatic-picking method is proposed in this paper to process microseismic data with MATLAB. There are four steps in the method: (1) de-noising and filtering microseismic signals using Empirical Mode Decomposition (EMD); (2) picking the first arrival using the traditional STA/LTA method with an improvement in the time-window and threshold value; (3) after finding the shortest first arrival, source positioning is evaluated with the "four-four" method; (4) after removing abnormal values from the sources positions using cluster analysis with specific dispersion threshold, optimized source positions are obtained by averaging. This method has been validated in the case study of a coal mine in China. Compared to other methods, the microseismic sources using this method can be positioned much more efficiently with high precision.