The application of order tracking for vibration analysis of a varying speed rotor with a propagating transverse crack

Propagating cracked rotor vibrations may contain substantial information on the rotor crack conditions. Capturing the characteristic vibrations due to the non-stationary response of rotors of which the speed change, is useful for condition monitoring of such systems. From the literature it is clear that vibrations at harmonics of the rotational speed, as well as transient responses, may be considered key indicators of cracks in rotors. Since faults other than cracks, such as misalignment and imbalance also generate harmonic vibrations, the non-order related vibrations therefore become important characteristics in detecting cracked rotor problems. But these signals may present with small amplitude and may easily be missed among the non-stationary harmonic vibrations. Therefore, clear identification of harmonic as well as non-order related (transient) vibrations are both important for detecting cracks in rotor systems. In this paper, a finite element model is used to calculate the response of a rotor with a propagating transverse crack under varying rotational speed conditions. various order tracking techniques, i.e. computed order tracking, Vold-Kalman filter order tracking, Gabor order tracking, are implemented to remove the varying speed harmonic vibrations so that the non-order related vibrations in which information about the cracks are contained, are emphasized. And some other signal processing methods that may achieve similar effects, i.e. double re-sampling method and intrinsic mode function from empirical mode decomposition, are also discussed for comparisons to these order tracking techniques. The paper demonstrates the advantages of order tracking methods in rotor crack detection.

► This paper presents order tracking analysis for enhancement of crack detection.
► Various signal processing techniques to order tracking methods are discussed.
► Results show that order tracking methods are better solution to crack detection.