Font Size: a A A

Research Of Case-Reasoning On Aero Engine Endoscopes Fault Diagnosis

Posted on:2012-09-22Degree:MasterType:Thesis
Country:ChinaCandidate:Y L ZhangFull Text:PDF
GTID:2248330395962595Subject:Pattern Recognition and Intelligent Systems
Abstract/Summary:PDF Full Text Request
With the development of the aviation technique, the requirements of civil equipmentmaintenance have become higher and higher. So it promotes a new challenge that how tocarry out the fault diagnosis and maintenance for the plane on the base of assuring the safety,lowing the cost and improving the reliability. With the development of artificial intelligence,the diagnosis techniques have entered into an intelligent diagnosis phase. Using the variousmethods in the fault diagnosis, this can improve the levels and efficiency of fault diagnosisgreatly.Theory and application of intelligent fault diagnose has been studied in this paper, Tomake analysis of endoscope damage pictures, image processing methods based oncharacteristic parameters have been included, Meanwhile case base has been establishedbased on it. In the applying of case base, the combination of Artificial neural network(ANN)and Case-based reasoning(CBR)are used in fault diagnosis, in this big complex system, themethod of fault diagnosis based on ANN is lack of clarity, which can’t have a good result, andthe one based on CBR run slowly besides of its imperfect diagnosis precision. According tothese, they have been combined in this paper, and raised the project. In this project, ANN istreated as a pre-classifier because of its function in pattern matching, and then cases in CBRare been indexed by the ANN results, which also guide the retrieve of cases. Then the methodof ANN-CBR hybrid reasoning has been raised.ANN-CBR hybrid reasoning method mainly uses the BP neural network model to trainengine damage image, the outputs as the condition of case base index, finally put forward toachieve the retrieval function through CBR method.The endoscope damage pictures are checked and simulated by the artificial neuralnetwork, case-based reasoning and ANN-CBR hybrid reasoning, through comparing the faultdiagnosis results, which show that the performance of the hybrid reasoning is better than thesituation of individual evidently, which proved this method to be practical as well as valid.
Keywords/Search Tags:damage image, fault diagnosis, artificial neural network, case-based reasoning, ANN-CBR hybrid reasoning
PDF Full Text Request
Related items