Precursor characteristics of rock bursts in chain pillars based on microseismic monitoring

To address the technical challenges of frequent rock bursts in high-stress chain pillars of deep mines with thick and hard roofs, poor adaptability of on-site rockburst precursor identification, and difficulty in selecting effective early warning parameters, a systematic study on rockburst precursor identification was carried out, taking the 3 − 1 103 working face of a mine in the Xinjie mining area as the research object, and combining laboratory cyclic loading–unloading uniaxial compression acoustic emission (AE) tests with in-situ real-time microseismic monitoring. The unified precursor law of “fluctuating increase” for both laboratory AE and field microseismic signals during the failure and instability of high-stress wide chain pillars was revealed. The shift cumulative energy and energy deviation value, which exhibit strong anti-interference capability, were selected as the core early warning parameters. Based on the Mann–Kendall non-parametric trend test, a precursor identification model for rock burst in high-stress wide chain pillars was constructed. Compared with existing rockburst precursor information identification methods, a warning scheme consisting of “mechanism research–parameter optimization–model construction–field verification” was established for the scenario of deep thick and hard roofs with high-stress wide chain pillars. Field practice demonstrates that the model can provide an early warning of 8–32 h, offering a theoretical basis and engineering reference for rock burst prevention and control in mines with similar geological and mining conditions.