Answers are grounded in this paper and cite their sources.
The shaped charge jet is widely used in oil exploration. However, the high-dimensional parameters and complex nonlinear relationships that influence the jet performance make it a challenge to precisely optimize parameters to enhance jet performance. This study proposes a combined numerical and machine learning prediction method, which leverages a data-driven approach to capture the relationship between structural parameters and jet performance. The validity is confirmed with theoretical models, simulations and experiments. Based on similarity theory, the study further predicts jet performance for different diameters and derives the dimensionless parameters for the enhanced structure. The results show that the Gaussian Process Regression model is selected for its superior ability of prediction in small datasets, with 0.9966 of R 2 , 26.12 of MAE and 839.13 of MSE. The average error of the prediction results is only 1.6%. The data-driven learning outcomes align with general theories, while delineating the dimensionless ranges that influence jet performance. Finally, the enhanced shaped charge demonstrates a significant improvement, maintaining a high jet head velocity while achieving a smooth velocity gradient. Additionally, compared to orthogonal opitimized methods, the depth of penetration was increased by 37.9%. This study provides a promising solution for optimizing the performance of shaped charges.
No takes yet. Share an insight, caveat, or question.
Cao et al. (2026) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: