Prediction of penetration depth of earth penetrator based on neural network

Abstract An artificial intelligence neural network model is established in this essay to seek a more general method for predicting penetration depth of earth penetrator, to comprehensively analyze the effect of various parameters on penetration depth as well as to predict the penetration depth of earth penetrator.This paper, by means of numerical simulation, and determined the ordnance penetrator warhead curvature radius, the length of the projectile, the density of the projectile,the density of the target protective layer, the elastic modulus of the target protective layer and the hit velocity of the earth penetrator.This six key parameters as the input data of neural network model, and by using numerical simulation to obtain the data needed for training the neural network model samples. According to the characteristics of six input data and one output data of the neural network model, the possible structure of the neural network model is set, and the optimal model structure is selected through training. We built neural network model to forecast the ordnance penetrator penetration depth, analyzes the six key parameter’s influence on the depth of penetration, the results show that reducing the warhead curvature radius, increasing the length and density of the projectile, properly increasing the impact velocity of the projectile can improve the penetration ability of the earth penetrating projectile, and increasing the density and elastic modulus of the target protective layer can improve the anti-penetration ability of the protective layer.