A Method of FaultIdentification Using BP Neural Network for Flight Control Systems

The reliable flight control systems are necessary to autonomous flight of unmanned vehicle(UAV).Fault patterns classification algorithm is an effective method to improve the reliability in complex control systems.In order to deal with high-dimension,nonlinear,and uncertainty in UAV,a fault identification method based improved BP neural network is introduced to identify flight control system failures.Firstly,neural network parameter is optimized by efficient conjugate gradient optimization algorithm.Then the failures of actuators,sensors and system were diagnosed by using proposed algorithm.The scheme is illustrated through simulations,applying the longitudinal flight control system of an UAV.The simulation results show that the proposed algorithm is reliable and simple.