Projectile Impact Point Prediction Based on Genetic Algorithm BP Neural Network

Abstract

Aiming at the problems of long calculating time and cumulative error in traditional projectile impact point prediction methods, a prediction method based on BP neural network is proposed in this paper. In order to avoid the local minimum and slow convergence speed, genetic algorithm is used to optimize the initial weights and thresholds of the BP neural network. The model of genetic algorithm BP neural network for predicting the impact point of projectile is obtained by training the model with a large number of flight state parameters and impact point information, and the simulation test is carried out. The simulation results show that the method can predict the impact point with high accuracy, and it is superior to the numerical integration method in calculating time. Therefore, it is reasonable and feasible to use this method to predict the impact point of projectile, which provides a reference for the practical application of projectile impact point prediction.