Abstract To further improve the safety management efficiency of coal mines, this study optimizes the prediction of heat transfer in the surrounding rock of coal mine tunnels using numerical simulation methods. It establishes a temperature field model based on Fourier's law and the unsteady heat conduction equation and enhances computational efficiency by combining dynamic grid adaptive technology. Meanwhile, this study adopts Bayesian optimization and Gaussian process regression to invert the thermal physical parameters of the surrounding rock. The results show that when the ventilation rate increases from 0.5 meters per second (m/s) to 1.5 m/s, the inlet temperature of the tunnel decreases from 21.0°C to 19.8°C. At the same time, the terminal temperature drops from 26.8°C to 23.0°C, and the heat flux density increases from 14.3 W/m² to 28.5 W/m². In addition, compared with traditional models, the proposed method's accuracy is improved by 15%. This study is committed to constructing a more accurate theoretical framework and technical scheme for thermal environment management and safety monitoring in coal mine production processes. Moreover, it is expected that the proposed research results can produce practical application benefits in fields such as mine thermal hazard prevention, energy consumption optimization, and safe production.
Zhi et al. (Fri,) studied this question.