Study on the evolution law of acoustic emission time series characteristics of coal-rock assemblage under true triaxial conditions

Introduction This study aims to investigate the fractal characteristics of acoustic emission (AE) signals during coal-rock rupture and their predictive significance for impact dynamic hazards. Methods True triaxial loading/unloading tests were conducted on five coal-rock composites with varying coal-thickness ratios using the TRW-3000 system, with AE signals analyzed to correlate stress evolution and temporal-spatial fractal features. Results Key findings reveal that: 1) AE parameters exhibit stress-dependent behavior, growing slowly initially, stabilizing during stress-holding, and surging exponentially before peak stress, with ∼90% of cumulative energy released pre-destabilization; 2) Ring counts follow an Λ-type trend against absolute energy, while coal-thickness proportion induces a mirror-image N-type pattern in AE parameters; 3) High-energy/high-ringcount AE events cluster spatiotemporally before failure, with clear time sequence precursors—short-duration ground sounds and consecutive rises in energy/frequency serve as critical early-warning indicators, whereas isolated signals lack hazard relevance. Discussion These fractal patterns and precursor thresholds enhance real-time monitoring and risk assessment frameworks, offering actionable strategies for mitigating coal-rock dynamic disasters in underground mining.