Expert System For Diagnosis And War Damage Of Certain Military Engine

Human exploration and research on artificial intelligence has never stopped.With the development of science and technology,artificial intelligence technology has stood at the forefront of The Times.Expert system is one of the fastest and earliest fields of artificial intelligence development.Along with the progress of Internet technology and intelligent theory,it has also ushered in the opportunity of technological leap.In the expert system of engine fault diagnosis,the common analysis method is fault tree analysis.Common reasoning strategies include:forward reasoning,blindly searching,using algorithms to deduce engine failures by operation parameters or only logical reasoning.The military engine fault diagnosis expert system only contains the vehicle maintenance,ignoring the war damage repair.Combined with the characteristics of the current expert system,the main contents of this paper are as follows:(1)Establish the fault tree with fault phenomenon as top event and intermediate event and fault cause as bottom event.The fault tree is expressed by frame representation and production representation,and expert system knowledge base is established.(2)Maintenance expert fault diagnosis was researched,hybrid reasoning method is more useful,realizing fault diagnosis of the forward reasoning and reverse validation;By repeatedly comparing the advantages and disadvantages between logical reasoning and parameter algorithm diagnosis,the logical reasoning expert system is diagnostic completeness and low efficiency,the parameter algorithm diagnosis is efficient immediacy and limitations,the expert system of logical reasoning that takes parameter algorithm as heuristic information is established,and the advantages of both is realized.Because the width priority search between fault phenomena is inefficiency and the depth of the fault tree is limited,the system adopts the depth priority heuristic search(3)For the fault tree with abnormal noise of the engine,the expert experience method is used to directly establish the membership degree between the fault phenomenon and the fault cause,and a fuzzy relation matrix is established.The vector of known fault phenomena is established by the input fault phenomena of maintenance personnel.For the multi-stage fault tree,the data association between engine parameters and fault causes and fault parameters is firstly established.According to the input parameters,Mamdani fuzzy reasoning method is used to combine logical reasoning to achieve the design of the reasoning machine.The design of the reasoning machine has application prospect.(4)Combined with the battlefield environment,established the concept of battlefield repairment,created the life tree analysis,defined the concept of the tree of life,built the method of depiction and analysis method.Study is applied on application significance,qualitative analysis and quantitative analysis methods.