Operating Parameters Optimization Based on BPNN and NSGA-III in Rotating Mode for Pneumatic Directional Drilling

Abstract

Drilling operations face the problems of low drilling efficiency and more difficult slag discharge in coal seams. A multi-objective optimization method is proposed to solve these problems. Rate of penetration (ROP), air pressure, and pull-out pressure are determined as the optimization objectives by characterization of the rotating mode. The maximum information coefficient (MIC) method is used to select the decision variables related to the optimization objectives, which are feed pressure, air volume, borehole depth, coal seam hardness, and rotary pressure. Then, a multi-objective optimization model is established using back propagation neural network (BPNN). The Non-Dominated Sorting Genetic Algorithm-III (NSGA-III) is used to solve the optimal operating parameters when the ROP is maximum and the air pressure and the pull-out pressure are minimum. Comparative experiments show that the method proposed in this study is effective. The results of this study can provide a new solution to improving drilling efficiency and resolving slag discharge difficulties in coal mines.

Overall scheme of operating parameters optimization