Research On Product Remaining Useful Life Prediction Method Based On Degradation Data

Prognostics and health management technology (PHM)is a new type of equipment comprehensive security technology, both to improve equipment safety and reduce maintenance costs. Prognostics and health management technology include two parts: Prognostics and management. Among them, Prognostics is based on the current or historical performance of the system to predict the future health status of components, for example, to determine the remaining useful life of the equipment. So remaining useful life(RUL) estimation is a key issue in Prognostics and health management technology.Accurately predicting the remaining useful life of stochastic degradation devices is the the basis of health management, also the key difficult issue. For the complex equipment,the stochastic and versatile degradation rules are difficult to describe by mechanism modeling,and the traditional methods based on lifetime data are not suitable for expensive systems with scare failure data.In this paper, based on the degradation data of production, the remaining life estimation method is studied in two aspects:1. For linear model, based on the first passage time, we propose a Wiener process degradation model that takes into account both random coefficient uncertainty and measurement uncertainty, and deduced the analytic solution of the probability density function of the remaining useful life of the wiener degradation equipment with parametric noise and easurement error. At the same time, the on-line residual life prediction is achieved. The validity of the proposed method is verified by Monte Carlo simulation. And finally the experimental results show that the proposed method can significantly improve the accuracy of the prediction by the laser data.2. For nonlinear model, based on cycle life degradation data of lithium-ion battery, lifedegradation process of the lithium-ion battery is analyzed and the empirical degradation model is adopted.We proposed a method for predicting remaining useful life of lithium-ion battery based on Extended Kalman Filter/Kalman Filter algorithm. Extended Kalman Filter algorithm is used to estimate the parameters of historical data, and then use Kalman Filter algorithm to estimate the remaining useful life of lithium-ion battery. Using the University of Maryland's lithium-ion battery data to validate the effectiveness of the algorithm, and the algorithm was evaluated with MAE index.