Application of BP neural network in fault diagnosis of FOG SINS

Taking FOG SINS (fiber-optic gyroscope strapdown inertial system) as an object, a new fault diagnostic scheme based on BP (back-propagation) neural network is proposed. Being capable of training and simulating data off-line, neural networks provide a solution to overcome some drawbacks of the quantitative fault diagnosis. The fault tree of FOG SINS is analyzed, which is the basis of the study of neural network fault diagnosis technology. The structure and inferential mechanism of BP network used for elementary fault diagnosis are discussed in detail. Training simulation results of the neural network are given and an improved effect with real data is obtained, which show the feasibility of the proposed scheme. Finally the design steps of fault detection system based on neural network for FOG SINS are summarized.