Research on evolution characteristics and correlation relationship of microseismic multidimensional information in mining process

Microseismic monitoring technology plays a critical role in predicting rockbursts in coal mines, yet the deeper potential embedded in microseismic data remains insufficiently explored. This study investigates the evolution characteristics of microseismic signals across four key dimensions—time, energy, space, and frequency—and quantitatively analyzes their interrelationships. The results show that both the multifractal parameter Δα and spatial correlation length exhibit a clear upward trend prior to roof weighting and gradually decline after pressure release. In contrast, the main frequency and b -value decrease as roof weighting approaches and recover afterward. These four indicators demonstrate a precursor period of approximately 2–4 days. Correlation analysis further reveals that all four parameters are not only significantly correlated with working face pressure but also exhibit long-range dependence, with the multifractal parameter Δα showing the strongest correlation. These findings contribute to improving the predictive effectiveness of microseismic monitoring in dynamic disaster early warning.