Font Size: a A A

Research Application Of Fault Diagnosis Expert System For Gun Control System Based On Rough Set And Rete Algorithm

Posted on:2019-06-18Degree:MasterType:Thesis
Country:ChinaCandidate:X X LiuFull Text:PDF
GTID:2382330551961077Subject:Control engineering
Abstract/Summary:PDF Full Text Request
Tank gun control system is the core control system of tank weapon part,which directly affects the tank's ability to fight on the battlefield and the survivability of the battlefield.However,the complex and harsh operating environment caused the frequent failure of the tank parts.Once only by experienced experts or soldiers of artificial diagnosis has been unable to meet demand.In the information age,the fault diagnosis expert system of tank gun control system arises at the historic moment.In this paper,several methods of knowledge acquisition in expert system are discussed firstly,and the knowledge acquisition methods of rough set and fault tree are selected.The steps of knowledge acquisition in rough set are analyzed in detail,and the flow chart of each step is given.Based on related data and data accumulated in the project,a fault tree of gun control system is constructed,and knowledge is acquired from fault tree by finding the minimum cut set.After discussing the knowledge representation and reasoning method applicable to this system,the most widely used RETE reasoning algorithm in the expert system is selected,and an improved method for the algorithm is put forward,that is,before the RETE network is established,the rule base is pretreated first to reduce the memory occupation of the algorithm.Experimental results show that this method can effectively reduce the memory footprint of RETE network.At last,the organization structure and function of the fault diagnosis expert system of the gun control system are studied.The expert system is written in C#programming language,and the SQL Server database are used to store the rules and data of expert system.The user management,rule management and other expert system functions are implemented.Through comparison and verification with the historical data,it is proved that the system can make a accurate judgement on the fault of the gun control system and give the maintenance advice,which has a high practical value.
Keywords/Search Tags:tank gun control system, fault diagnosis expert system, fault tree, rough set, rete
PDF Full Text Request
Related items