Acoustic emission-based intelligent identification and dual early warning for coal fatigue failure under multi-stage cyclic loading

Aiming at the problem of insufficient accuracy and timeliness in the early warning of coal instability under multi-stage cyclic loading, this paper proposes an intelligent recognition and dual early warning method for coal damage based on acoustic emission parameters. The multi-stage cyclic compression test of coal is carried out to obtain five characteristic parameters of acoustic emission, and the average event intensity ( ERR ) is used to characterize the sudden change characteristics of coal energy release. The Sparrow Search Algorithm (SSA) is adopted to optimize the hybrid model of Transformer and Gated Recurrent Unit (GRU), so as to realize the high-precision recognition of coal damage stages. On this basis, the early warning coefficient ( EW ) is calculated by using the Isolation Forest algorithm and the CRITIC-TODIM method, and a dual early warning system of “ preliminary prompt with ERR and comprehensive confirmation with EW ” is established. The results show that the recognition accuracy of the optimized model for the fourth damage stage is increased by 11.7% compared with the SSA-GRU model. The early warning lead time of ERR is between 60.5 s and 185.7 s, and that of EW is between 10.5 s and 101.5 s. The proposed method integrates physical mechanism and data-driven technology, which can effectively improve the accuracy and reliability of coal damage early warning under multi-stage cyclic loading, and can provide theoretical support and technical reference for the early warning of dynamic disasters in deep coal mines.