The Fault Diagnosis Expert System Of Satellite Based On Neural Network And Its Application

The fault diagnosis technique is a key technology to ensure longevous service and the high reliability of satellite. Based on the spaceflight technology innovation program of"Research on the Testing Method of Reliability----Research on the Fault Diagnostic Expert System and Neural Network"and the Pre-Research Defense Projects of"Research on longevous service and high reliability of Satellites"of the National tenth-five-year plan, taking attitude control system of a satellite as object, the theories and the applications of the neural networks and expert system in the fault diagnosis are studied in this dissertation. This paper includes following parts:At first, the basic theory of neural networks and expert system was analyzed. Then, by researching the similarities and differences of the neural network and expert system from knowledge acquisition, representation and reasoning, their own advantages and disadvantages are compared, accordingly, a starting point that the neural network combines with the expert system was fount out. Later, following studying some domestic and international combination ways, a model of expert system based on neural network was designed, and its function was simply described.The number of the hidden layers of mul-tilayer perceptrons(MLPs) is analyzed, and three-layer perceptrons neural network is adopted; By analyzing the mechanism of the neural cells in hidden layer ,a method for combining Genetic Algorithm and BP Algorithm to optimize the design of the neural networks is presented, and it solves the defects of getting into infinitesimal locally and low convergence efficiently, it can also solve the problem that it can usually obtain nearly global optimization solution within shorter time through using Genetic Algorithm method lonely; Several examples validate that this algorithm can simplify the neural networks effectively, and it makes the neural networks solve the practical problem of fault diagnosis more effectively.The course of building expert system based on neural network is studied. The course of building the necessary module of the diagnostic system, such as knowledge acquisition, knowledge base, reasoning method, explanation system,...