Tacholess order tracking method for gearbox based on time-varying filter and energy centrobaric correction method

Abstract

Under the premise that the instantaneous speed can be accurately measured or precise estimated, order tracking is considered to be a classical and effective technique for non-stationary vibration analysis of rotating machinery. The meshing frequency components of the vibration signal measured from complex gearbox system interfere with each other mutually. The energy centrobaric correction method exhibits exact instantaneous frequency estimation (IFE) ability under the interference, but is susceptible to strong noise. Combining with time-varying filter and energy centrobaric correction method, a tacholess order tracking technique for gear fault diagnosis under the rotating frequency multi-linear fluctuation is proposed. The time-varying filter is designed to filter the strong noise signal, and then IFE could be accurately calculated from the filtered signal by energy centrobaric correction method. The instantaneous phase of reference shaft based on the Vold-Kalman filter (VKF) is used for angular resampling of the original vibration signal. The results show that the shaft rotating speed could be accurately identified from extracted gear meshing components. The order spectrum obtained by the proposed method with tacholess is substantially the same as the order spectrum obtained by a tachometer. Simulation analysis and experimental results verify the advantages of proposed tacholess order tracking technology for monitoring and fault diagnosis of gearbox vibration under the condition of strong noise and large speed variation.