A Remaining Useful Life Prediction Model for Complex Electromechanical Equipments Based on Stochastic Process and Condition Data

Abstract

Aiming at the problem of residual life prediction of complex mechanical and electrical equipment, an improved Wiener process residual life prediction model is proposed. Firstly, aiming at the nonlinearity of degradation process of complex mechanical and electrical equipment, the general Wiener process prediction modeling method is improved, and a residual life prediction model based on random coefficient Wiener process is proposed. Secondly, aiming at the parameter estimation of degradation process, Based on the health assessment of the equipment, the distribution parameters of the process parameters are estimated by using the historical state data of similar equipment, and a construction method of the prior distribution of the degradation process parameters based on the state data is proposed. Finally, the information fusion is carried out by using Bayes statistical inference to construct the posterior distribution and Bayes estimation of the degradation process parameters, and the residual life of the equipment is estimated Life prediction. The case study shows that the residual life prediction model proposed in this paper has higher prediction accuracy and lower prediction uncertainty.