Neural Networks Fault Diagnosis And Application In The Navigation System

With the rapid development of science and technology,single navigation system has been unable to meet the accuracy requirements currently.Therefore,as the computer technology and the modern control theory advance,the integrated navigation system is widely used on various types of carriers.However,the complexity of the system makes the fault occurred more frequently,fault diagnosis technology has become an important method to improve the accuracy and reliability of integrated navigation system.This paper overviews the development and the basic principles of neural network,gives a brief introduction to its classification and learning methods.As well as the basic principle of particle swarm optimization algorithm.Due to the disadvantage of the particle swarm algorithm is slow in late convergence and easily to fall into the local optimal solution based on the standard particle swarm optimization(PSO)algorithm,an improved PSO algorithm with acceleration factor and inertia weight adjustable is proposed.On the basis of INS / GPS combined navigation loosely coupled model,the error model of the integrated navigation system is established.Under this premise,a fault diagnosis scheme based on improved PSO-BP network is designed,taking gyroscope and integrated navigation fault diagnosis as two examples,verifies the improved PSO-BP network fault diagnosis scheme by simulation,which verified the feasibility and accuracy of the scheme.