Optimization and practice of deep hole loosening presplitting blasting parameters of protective layer working face based on genetic algorithm

In order to solve the problem of hard rock faults encountered in the back mining of the protective layer working face of the 13 coal mine of Pingmei, it is proposed to optimise the deep hole loose presplitting blasting parameters of the protective layer to improve the advancing speed of the working face. The optimization model of deep hole loose presplitting blasting parameters is set up using genetic algorithm, the model takes hole depth, hole diameter, hole spacing and charge as independent variables, and the advancement speed of the protective layer working face as the dependent variable, and takes the improvement of the advancement speed of the protective layer working face as the core of the optimization, and applies 100 groups of blasting parameters on site to optimize the deep hole loose presplitting blasting parameters through computer programming in MATLAB. The experimental results show that after 25, 50, 75 and 100 repetitions of optimization, the propulsion speed of the protective layer working face increased by 7.4%, 17.6%, 20.2% and 23.1% compared with the original propulsion speed, respectively; after 100 repetitions of optimization, the propulsion speed of the protective layer working face gradually stabilizes and tends to be optimal, and finally 10 optimized blasting parameter combinations corresponding to the optimal propulsion speed can be obtained. Combined with the site construction conditions, the final determination of the protective layer deep hole loose presplitting blasting hole depth of 40 m, hole diameter of 75 mm, hole spacing of 2 m, and filling amount of 82.5 kg is finally obtained. Industrial tests show that using the optimised blasting parameters, the protective layer fault hard rock is sufficiently crushed, the coal mine machine blocking loss is reduced by 31.4%, and the propulsion speed is increased by 22.2%, achieving the goals of speed improvement and consumption reduction. Using genetic algorithm to optimise the parameters of deep hole loose presplitting blasting can not only improve the blasting effect and reduce human interference, but also has certain reference significance for deep hole loose presplitting blasting of protective layer.