To reduce the gas disaster of high gas coal seam and improve the efficiency of gas extraction by high drilling, the layout parameters of drilling holes in Pingdingshan coal mine are optimized. Based on the analysis and calculation of the "three zones" of the movement towards the overly strata of No.10 coal in Pingdingshan coal mine, the height of caving zone and fissure zone in 24,130 working face are 10.06–14.46 m and 38.75–49.95 m respectively. The elevation angle, azimuth angle and the length of high-level boreholes are studied and analyzed by COMSOL numerical simulation software. The simulation results show that the optimum layout parameters of high-level boreholes are as follows: The elevation angle of borehole should be controlled at 9°–12°, the azimuth angle should be 30°–45°, and the length of borehole should be 150 m. Then the optimum layout parameters of high-level boreholes are determined for engineering application of 24,130 working face. Borehole data onto actual mine show that the optimum layout parameters of high-level boreholes were elevation angle between 8°and 11°, azimuth angle between 30° and 42°, and length of boreholes between 145 and 155 m. The simulation results are basically consistent with the measured data. The maximum gas concentration in working face, upper corner and return air roadway is stably controlled below 1%. The safe mining of 24,130 working face is ensured, which provided a certain reference value of gas control in the goaf of Pingdingshan mine and adjacent mines.

To address the problem of fault branch recognition in mine ventilation systems, a one-class classification algorithm is introduced to construct the MC-OCSVM ventilation system fault diagnosis model, which is integrated with multiple OCSVMs. This model adopts uniform hyperparameters and transforms the ventilation system fault diagnosis problem into a maximum decision distance problem, to realize the effective use of mine monitoring wind speed data. The experiments on public KEEL datasets verify that the one-class classification integration model can solve the multiclassification problem and that the MC-OCSVM model has better generalizability than other one-class classification integration models do. The experiment is carried out in the Buertai coal mine, and the results show that the proposed algorithm can identify fault branches quickly and accurately, with an accuracy of 93.2% and a single fault diagnosis time is 1.2 s, highlighting its strong robustness.