基于劣化数据的综合传动装置剩余寿命预测

Remaining useful life (RUL) prediction of power-shift steering transmission(PSST) was presented based on multidimensional degradation monitoring data under uncertain measurement. The state space model and principle component analysis (PCA) was used to establish the degradation degree index. The RUL of PSST was defined based on the concept of first hit time (FHT) of stochastic process, and a PSST's degradation model was established based on Wiener process, considering the stochastic degradation and uncertain measurement. And then the maximum likelihood method was utilized to estimate the model parameter. The Kalman filtering technique was used to estimate and update the degradation state, and the RUL distribution was derived. Test results show that the proposed method can objectively describe the degradation law of the PSST, which is superior to the method without considering uncertain measurement, and can improve the accuracy of RUL prediction, which is helpful to the condition based maintenance.