Fault diagnosis of avionics system based on artificial neural network

Prognostics and health management can be used for monitoring and life prediction of integrated modular avionics systems. As one of the methods of prognostics and health management, the BP neural network has good self-learning, self-adaptation and generalization capabilities, but it is easy to fall into a local minimum during the operation. Therefore, this paper chooses to use genetic algorithm to optimize the initial weights and thresholds of the BP neural network training, which can greatly improve the accuracy of the BP neural network for fault diagnosis; at the same time, choose to use unsupervised learning self-organizing feature map for fault diagnosis. The simulation results show that the fault diagnosis accuracy rate of the BP neural network optimized by the genetic algorithm is greatly improved, and self-organizing feature map can reach a higher fault sample classification accuracy rate after 200 training.