Variation of seismicity using reinforced seismic data for coal burst risk assessment in underground mines

Seismic data analysis methods have been widely used to assess the variation of seismicity in burst-prone mines for managing coal burst hazards. However, their performance can be significantly limited by the poor quality of seismic data. Due to the complex mining and geological conditions, the recorded seismic events may present high location errors, and an unfavourable layout of geophones in tabular longwalls also results in an incomplete seismic data catalogue. Therefore, this paper developed a seismic data reprocessing workflow to enhance the quality of seismic data, aiming to achieve better performance of seismic data analysis for coal burst risk assessment. As a first step, based on a modified seismic clustering method that can reduce the impacts from location errors, the potential damage zone in longwall roadways was identified where seismic events are related to possible dynamic failure. Within the potential damage zones, the seismic data were reinforced to address the inherent issues of location errors and seismic data integrity. Then, the reinforced seismic data were used to analyse spatial distributions of event counts and seismic energy in the longwall. The method was tested in a burst-prone longwall panel, and the reinforced seismic data presented higher correlation with the impending burst damage than the raw seismic data.