Research on structural parameter optimization of elliptical bipolar linear shaped charge based on machine learning

Numerical simulation based on SPH method, compared with laboratory experiments, using the grey correlation theory to analyze the correlation between the parameters of the elliptical bipolar linear shaped charge and the performance of the shaped charge jet. The structure of shaped charge is optimized by machine learning to obtain the optimal structural parameters, and it is compared with the rock crack development of shaped charge blasting in practical application. The results show that the structural parameters of the shaped charge have the same influence on the jet head velocity, and there are certain differences in the impact on the jet length. The fitted curve of the support vector machine (SVM) regression model based on the genetic algorithm (GA) is high prediction accurate. By comparing the optimization results with the actual engineering application of the shaped charge structure, the rock breaking effect has been significantly improved, which has important guiding significance for the actual engineering application. Machine learning; Shaped charge; Blasting; Parameter optimization.