Case study on pre-warning and protective measures against rockbursts utilizing the microseismic method in deep underground mining

Abstract

Predicting rockbursts and implementing effective protection measures are of paramount importance for safe mining in deep underground mines. A more advanced approach to recognizing rockburst hazards in underground mining involves the integration of various techniques, including microseismic (MS) parameters, localization of high-magnitude MS events, seismic tomography, deep learning models, and on-site surveys. In this study, tomography results and the distribution of MS events were utilized to identify the mining activity region and velocity anomaly in the Shaanxi Zhenao mine. Subsequently, multiple MS parameters, such as microseismic moment (MSM), microseismic energy (MSE), apparent stress (AS), b-value, S value, and comprehensive microseismicity intensity (CMSI), were examined at the 842 level. Based on these analyses, a potential rockburst area was accurately determined, and appropriate protective measures were implemented in the short term at the goaf of 842 level by integrating the results from in-site surveys, deep learning predictions, and the microseismic method. It's noted that a rockburst incident occurred in the goaf four days later; however, the adjacent area remained undamaged due to the effectiveness of the protective measures in place. This case study indicates that the utilization of MS information and deep learning models can serve as a valuable pre-warning method for assessing the risk of rockburst.