Research on the Method of Gun Power Failure Prediction Based on Improved Grey Neural Network

The artillery power system is the power supply equipment of the artillery weapon system, and its reliability is related to the combat efficiency of the artillery. In order to realize the fault diagnosis and prediction of the artillery power system and improve the reliability of the power system, this paper takes the artillery power system as the object and proposes to combine the gray model with the BP neural network to predict the artillery power failure. The fault prediction effects of traditional GM (1, 1) gray model, BP neural network, traditional gray neural network and improved GM-BP gray neural network are compared and analyzed by simulation. The simulation results show that the improved GM-BP gray neural network improves the prediction accuracy and speeds up the convergence speed, verifies the effectiveness and correctness of the method, and provides an effective engineering application basis for the fault prediction of the artillery power system.