Study On Fault Diagnosis Expert System For Gunner's Scope

The gunner's scope is an important part of the tank sighting system.It is installed with architecture structure in the tank structure with the function of observing,searching and aiming at the inside of the tank.The precision is high and the structure is complex.It is difficult to ensure the accuracy and maintenance efficiency by relying on the maintenance experience of professionals.It is necessary to develop an intelligent detection system to quickly and accurately detect the performance index of the length of the gun.Therefore,a gunner's sight diagnosis expert system is designed in this thesis.The main research contents are stated as following:A knowledge acquisition method of the gunner's scope diagnosis expert system is proposed.The subtraction clustering algorithm and fault tree analysis method are used to acquire the knowledge of the components of the gun sight and the control signal fault.In the subtractive clustering algorithm,the Spearman correlation coefficient is introduced to eliminate the interference of irrelevant input,which can avoid the redundancy of knowledge.According to the composition and working principle of the gunner's scope,the fault tree of the gunner sight is constructed and the minimum cut set is obtained.The logical relationship between the faults is clarified and the specific cause of the fault is determined to complete the knowledge acquisition.A knowledge based on the fault diagnosis expert system for gunner's scope was constructed.The knowledge expression method of production rules is clarified.The SQL Server 2012 database is used as a construction tool.According to the reasoning method of the inference engine,the knowledge base is divided into four parts: condition table,rule table,conclusion table and explanation table for construction and management.The inference engine of the gunner's sight diagnosis expert system is designed.The improvement of the KMP character matching algorithm improves the matching efficiency,and the fault level is set to solve the problem of rule matching conflict.Based on the idea of forward reasoning method,the improved KMP character matching algorithm is used to match the facts with the rules to realize the design of the inference engine.The software development of the gunner's sight diagnosis expert system is realized.Thestructure of the expert system is divided into four modules of software for development.The process of the system software was is designed.A system software is developed by using Visual Studio 2013 programming tool and SQL Server 2012 database.The example tests show that the fault diagnosis function of the gunner sight can be realized and reasonable maintenance suggestions are given.