Fault diagnosis of FOG SINS based on neural network

Based on nonlinear mapping relationship between fault symptom and fault type in subsystems of FOG SINS (fiber-optic gyroscope strapdown inertial system), BP (back propagation) and Elman neural network approaches were presented for fault diagnosis. Fault mechanism and failure behavior of FOG SINS was analyzed, then featured fault types were extracted from FOG SINS faults and the extracted features were regarded as fault symptom eigenvector. The process of fault diagnosis principal, fault diagnosis model and fault diagnosis algorithm were given using BP and Elman neural network with enough fault feature information. Trained BP and Elman were used for fault vector recognition and diagnosis to verify the proposed fault diagnosis model effectiveness and rationality. Training and test results of two neural networks were compared The conclusion was made.