HELICOPTER VIBRATION MONITORING USING UNIFIED FAULT-STRUCTURE-SENSOR ANALYSIS

VIBRATION MONITORING R&D HAS BEEN SIGNAL PROCESSING ORIENTED AND PROBLEMS ARISE BECAUSE TECHNIQUES DEVELOPED FOR ONE MACHINE ARE NOT TRANSFERABLE TO OTHERS. TIDG PROPOSES A UNIFIED APPROACH TO MONITORING R&D WHICH WILL USE MODELING AND DATA TO PREDICT FAULT SIGNATURES, USE CEPSTRAL AND OTHER FORMS OF PROCESSING TO REMOVE OR DEFINE THE MACHINE'S STRUCTURAL RESPONSE TO THE DEFECT, AND USE UNIQUE SENSOR AND NOISE REDUCTION CONCEPTS TO IMPROVE FAULT SIGNAL-TO-NOISE RATIO AT THE SENSOR. DETECTION ALGORITHMS, SIMILAR TO SONAR DETECTION PROCESSING, WILL THEN BE USED TO DETECT AND IDENTIFY FAULTS WHEN THEY OCCUR. PHASE I WILL INVESTIGATE THE FEASIBILITY OF THIS APPROACH BY PERFORMING AN ANALYSIS FOR A SPECIFIC GEARBOX. GEAR AND BEARING MODELS WILL BE USED TO PREDICT FAULT SIGNATURES. THE GEARBOX STRUCTURAL RESPONSE WILL BE MEASURED, AND THE APPLICABILITY OF RECENT MACHINERY STRUCTURAL RESPONSE RESEARCH WILL BE INVESTIGATED. METHODS FOR SIGNAL-TO-NOISE IMPROVEMENT WILL BE DEVISED AND INVESTIGATED. A SYSTEM INTEGRATION AND FEASIBILITY STUDY (BASED ON SONAR SYSTEM ANALYSIS TECHNIQUES) WILL THEN BE DONE. THE APPROACH HAS THE PROMISE OF HAVING WIDE APPLICATION TO VIBRATION MONITORING, SINCE THE RESULTS AND PRODCEDURES DERIVED FOR ONE GEARBOX WILL BE APPLICABLE TO MANY OTHERS; ONLY THE SPECIFIC GEARBOX PARAMETERS WILL CHANGE.