An adaptive narrowband demodulation methodology for planetary gear in RV reducers using modulation mechanisms

Abstract To address the challenges of the substantial computational burden associated with manual sideband selection, and the mapping modeling between planetary gear faults and modulation signals, this paper proposes an adaptive narrowband demodulation methodology for rotary vector (RV) reducers using modulation mechanisms. The fault vibration signal model is derived based on system dynamics modeling and Fourier series theory. The proposed adaptive fault demodulation method intelligently filters and preserves the frequency band containing rich fault information through the adaptive median absolute deviation (AMAD) threshold method, reduces asynchronous interference components by corner domain synchronous averaging technology, and extracts fault characteristic signals through narrowband demodulation. The effectiveness and practicality of the proposed fault demodulation method are verified through fault vibration experiments of the planetary gear in an RV reducer. The spectrum results obtained by Hilbert transformation and inverse transformation show that the error between the theoretical model and the actual model is close to 0, and the skewness and kurtosis are close to the ideal state, which proves the reliability of the fault vibration signal modeling. Compared with the fault demodulation methods such as angular synchronous average and forward difference, the AMAD narrow-band demodulation method achieves the highest signal-to-noise ratio in the simulation signal experiment and fault vibration experiment of RV reducer planetary gear under different working conditions, the former error is small, and the latter reaches more than 40, which proves that the fault demodulation method proposed in this paper has good demodulation effect.