Research And Application Of Fault Diagnosis Expert System For Gun Control System Based On Fuzzy Reasoning

There are still many deficiencies in the field of fault diagnosis for gun control system in China.On the one hand,the form of fault detection is single and inefficient.On the other hand,there are not many products that can be used for actual diagnosis,and most of them can only detect circuit-board-level faults.To deal with the problem,a fault diagnosis expert system for gun control system which can diagnose component fault is designed in this paper.In addition,in order to solve the problem of interference signal affecting normal reasoning,fuzzy inference is introduced into fault diagnosis expert system of gun control system,and the related research is carried out.In order to obtain more complete knowledge for accurate reasoning,the fault diagnosis rules of component level are obtained by fault tree analysis and circuit principle analysis,and the result is that the accuracy level of the fault diagnosis is improved.At the same time,the fault grade is introduced,which simplifies the rules and improves the efficiency of reasoningThe T-S fuzzy inference system is built by using neural network,and the selection of fuzzy inference input for fault diagnosis of gun control system is analyzed.In order to obtain the fault diagnosis rules of complex circuits,a subtractive clustering algorithm for gun control system is proposed.The improved subtractive clustering can effectively realize the rules reduction and remove redundant fault diagnosis rules of gun control system.Finally,a fault diagnosis expert system for gun control system which can diagnose fault of component level is formed by combining precise reasoning with fuzzy reasoning.The expert system completes the reasoning according to the input data,and gives the fault diagnosis,fault location,reasoning explanation and maintenance suggestion.