Rotating Equipment Defect Detection Using the Algorithmof Mode Isolation

Noise;Crack;Rotor;Experimental modal analysis;EMA;Bearing;Residue;Modal

Wagner, Benjamin ; Mechanical Engineering

University:Georgia Institute of Technology

Department:Mechanical Engineering

关键词: Noise;    Crack;    Rotor;    Experimental modal analysis;    EMA;    Bearing;    Residue;    Modal;   

Others  :  https://smartech.gatech.edu/bitstream/1853/16230/1/wagner_benjamin_b_200708_phd.pdf

美国|英语

来源: SMARTech Repository

【 摘 要 】

Findings from a project involving rotating equipment defect detection using the Algorithmof Mode Isolation (AMI) are presented. The prototypical system evaluated is arotating shaft, supported by hydrodynamic bearings at both ends, with one disk mountedto the shaft. Shaft cracks and bearing wear are the two equipment defects considered.An existing model of the prototypical system from the literature, termed thesimplifiedmodel.is modified to simulate the presence of a transverse shaft crack at mid-span. Thismodified model is termed thestandard model.Ritz series analysis, in conjunction with apreviously published description of the compliance related to the presence of a transverseshaft crack, is used to describe the decrease in shaft stiffness associated with the crack.The directional frequency response function (dFRF) is shown in the literature to providebenefits over the standard frequency response function (FRF) in both system identificationand shaft crack detection for rotating equipment. The existing version of AMI is modifiedto process dFRFs and termed Two-Sided AMI. The performance of Two-Sided AMI isverified through system identification work using both the simplified model and a rigidrotor model from the literature. The results confirm the benefits of using the dFRF forsystem identification of isotropic systems. AMI and Two-Sided AMI are experimental modalanalysis (EMA) routines, which estimate modal properties based on a frequency domainexpression of system response. Eigenvalues and associated modal residues are the modalproperties considered in the present work.Three defect detection studies are fully described. In the first, the simplified model isused to investigate bearing wear detection. Various damage metrics related to the eigenvalueand the residue are evaluated. The results show that residue-based metrics are sensitiveto bearing wear. Next, the standard model is used in an in-depth investigation of shaftcrack detection. When a shaft crack is present, the standard model is time-varying in boththe fixed and moving coordinate systems. Therefore, this analysis is also used to evaluateperforming EMA on non-modal data. In addition to continuing the evaluation of variousxiv damage metrics, the shaft crack study also investigates the effects of noise and coordinatesystem choice (fixed or moving) on shaft crack detection. Crack detection through EMAprocessing of noisy, non-modal data is found to be feasible. The eigenvalue-based damagemetrics show promise. Finally, the standard model is used in a dual-defect study. Thesystem is configured with both a shaft crack and a worn bearing. One defect is heldconstant while the magnitude of the other is increased. The results suggest that AMI isusable for defect detection of rotating machinery in the presence of multiple system defects,even though the response data is not that of a time-invariant system. The relative meritsof both input data types, the FRF and the dFRF, are evaluated in each study.