基于油液光谱数据的离合器剩余寿命随机过程预测

The residual life prediction of wet clutch based on oil spectrum data significantly impacts on the condition monitoring and reliability of integrated transmission device. Aiming at the problems of high randomness of oil spectral data and single performance index and large error of existing methods, the prediction of clutch remaining life is carried out using the advantages of real-time and accuracy of binary Wiener process. Firstly, combined with the wet clutch life test, the indicator elements Cu and Pb and the failure threshold of the remaining life prediction of the clutch are extracted through the oil supplement and change correction of the spectral data of the whole life cycle: Secondly, the correlation characteristics of indicator elements are analyzed by MATLAB copula function, and the correlation function of residual life is derived: Thirdly, according to the inverse Gaussian principle, the performance degradation mathematical models of the unary and binary Wiener processes of the above two indicator elements are established: Finally, the maximum likelihood estimation method is used to estimate the parameters, and the univariate and binary performance degradation mathematical models are used to predict the remaining life of the tested clutch. By comparing the predicted results with the experimental results, the deviation of residual life prediction of binary Wiener process is 6 % ~ 2 2 % in the range of 150~240 h: Compared with the univariate Wiener process, the accuracy of residual life prediction is improved by more than 9%. The results show that the binary Wiener process model and its prediction method have the advantages of real-time solid prediction and high prediction accuracy. At the same time, this method can be extended to related fields such as on-line monitoring of equipment status and residual life prediction.