Research On Helicopter Rotor Fault Diagnosis Based On Grey Theory

Fault diagnosis of helicopter is the main method to promote its Reliability, Maintainability and Supportability (RMS), so it has significant meaning both in theory and practice. Rotor system is all-important for realize helicopter's function and safety flight as a lifting surface and controlling surface. The harm of rotor fault is serious, and the research to rotor system is still weakness in current most Helicopter Health and Usage Monitoring System (HUMS). Because of the real data of rotor faults is obtained difficultly, most research methods on rotor fault adopt qualitative analysis combine with test research. The dissertation used Grey theory as a rotor faults recognition method based on the data of helicopter rotor dynamic unbalance test. The main work of the dissertation as follows:(1) The basic method of Grey theory and the affect of association coefficient to relational degree were analyzed for fault diagnosis.(2) Hilbert-Huang transform was used for data processing and extracted marginal spectrum of different faults as fault feature and it worked resultfully. The dissertation used GM model to bate the infection of end effects caused by EMD, and it was proved more effective by simulation signal.(3) Theory of relational degree in GM was used to diagnose faults. Different feature extract methods of power spectrum by AR mode and energy spectrum by CWD were Compared,showed HHT method fits for Grey theory better. Equal weight theory of Grey theory was improved by calculate the normalization association coefficient of the related standard sequence, and successfully used in rotor unbalance fault diagnosis, especially to tab imbalance witch the precision reached 100%.(4) Relational degree obtained by unequal weight relational degree method was used as faults and composite faults degree recognition method. Due to the interrelationship of fault signal, a certain extent error happened during composite faults degree recognition. But 81.15% global precision was still got and supplied good information of the fault.