Quality Assessment of Microseismic P-Phase Arrival Picks and Its Application of Source Location in Coal Mining

Abstract

Abstract Correctly identifying abnormal and false P-phase arrival picks (P-pick) in underground coal mining is essential to microseismic source location. Manual judgement and identification are time-consuming with the increasingly growing monitoring data. To eliminate the effects of false P-picks, a novel microseismic source location with weighted P-picks was proposed, and ten waveform parameters were selected to characterize the difference between two types of signals with usable and unusable P-picks. The discriminant analysis experiment has revealed that the prediction rate of unusable P-pick set increases dramatically with the sample size when the sample size is less than 2,000 and the prediction rates of unusable P-pick set are around 88% when the sample size is greater than 2,000, while the prediction rates of usable P-pick set are around 80%, which is little affected by the sample size. Considering the prediction rates of usable and unusable P-pick populations, the discrimination function with a sample size of 3,000 was selected to identify the usable and unusable P-picks. The identification rates of usable and unusable P-pick populations are up to 83.24% and 88.99%, respectively. The application of P-pick discriminant analysis model in source location was discussed. The location case and long-term result show that the P-pick discriminant model and its application in source location perform well.