Recursive Model-Based Order Tracking Filters for Solving Overlapping and Interference Problem in Rotating Machines

Abstract

Order tracking is a crucial method for time-frequency analysis of vibration signals which can provide important information for fault diagnosis in rotating machines. Several existing techniques such as the short-time Fourier transform (STFT) and the Vold-Kalman filter (VKF) can be used for time-frequency analysis under both stationary and non-stationary operating conditions. However, the problems of frequency components overlapping and interference from other frequencies are still not solved well. In this paper, we propose two alternative order tracking filters to handle these two problems. The results of numerical studies illustrate the effectiveness of the proposed methods.