Fatigue Fracture Of Computer Knowledge Expression And Reliability

Knowledge acquisition and representation of failure analysis as well as thereliability of conclusion are studied in this paper.The way of knowledge acquisitionand representation are given through the comparison of various knowledgeacquisition algorithms and representation methods.Fatigue failure,one of the mostdangerous and frequently occurred failure modes,is applied in this thesis forknowledge acquisition and representation.After summarizing the characteristics ofvarious uncertainty reasoning methods,the algorithm of uncertainty reasoning andreliability of aided failure analysis are proposed.A basic test of the system is alsocarried on through a failure case. The document-based processing algorithm is mainly used to acquire domainknowledge in accordance with the train of thought of Computer Aided FailureAnalysis Expert System for Mechanical Components.This algorithm can improvethe accuracy and reliability of acquired knowledge and obey to the human ways ofpreserving failure analysis knowledge. The knowledge of failure analysis can be classified as the basic knowledge andcase knowledge. Rules aided by procedurals and object-oriented scheme areappropriate for representing the basic knowledge,and frame method for the failurecase knowledge. The content of rules expresses the experts' experience;proceduralsstores the difficult operations and fixed algorithms;and object oriented schemerepresents the hierarchy of knowledge.Frames address all of the attributes of a case. The classical probability-based method is suitable for describing the strength ofbackward chaining rules,and certainty factor for describing the strength of forwardchaining rules. As for the failure case knowledge, case adaptation implies itsuncertainty reasoning.The reliability of conclusion is obtained through uncertaintyreasoning. The results of system test make clear that the system is feasible.However, inorder to complete uncertainty reasoning,further efforts should to be made.