Exploration of intelligent prediction methods for damage effect based on parameters of liner

Predicting the armor piercing power of different structural parameters of liner has always been a focus of attention for scholars. Conventional numerical simulation methods require a lot of repetitive work and are time-consuming to obtain armor piercing power. Therefore, in order to achieve the effect of quickly predicting the armor piercing power of flat top shaped liner with different structural parameters, four structural parameters were selected, boundary conditions were given, and an orthogonal experimental scheme was designed for numerical simulation to obtain a dataset. Four intelligent prediction methods were used to establish a armor piercing power prediction model for flat top shaped liner with different structural parameters, and comparative experiments were conducted. The results show that the four intelligent prediction algorithms can accurately predict different structural parameters of liner. Support vector regression algorithm has the best fitting effect and shortest computation time among the four machine learning algorithms; The XGboosting algorithm has the best prediction accuracy among four machine learning algorithms; while the prediction model established by random forest algorithm has the worst performance.