Research On The Health Intelligent Evaluation Method Of The Lithium Ion Battery

Lithium ion battery in civil and military fields has been widely used. To build efficient and intelligent management system of lithium ion battery is the key to ensuring the safe use of lithium ion battery. The intelligent health evaluation method of lithium ion battery is the core technology of the management system of lithium ion battery. Therefore, the research of health evaluation method of lithium ion battery with rapidity and high precision has great significance.This paper mainly studies the estimation methods of battery state of charge and prediction methods of remaining useful life which are among the intelligent health evaluation methods of lithium ion battery. The main contents are as following:1) The working principle of lithium ion battery is introduced. The basic concept of lithium ion battery performance and state parameters is elucidated. Charging methods of lithium ion battery are summarized. The environment temperature, discharge current and the battery aging influence rules to the battery actual capacity are summarized through the experimental study.2) The estimation method of battery state of charge using random forest is proposed. The estimation accuracy of the ampere hour integral method relies heavily on the accuracy of the initial value of the battery state of charge, the estimation method of empirical model and extended kalman filter can correct the error of the initial value of the battery state of charge, but its estimation error is too large. In order to solve the problems mentioned above, the estimation method of battery state of charge using random forest is proposed. The steady and dynamic discharge experimental results of single lithium ion battery show that its estimation accuracy is not affected by initial value of the battery state of charge and is higher than that of the estimation method of empirical model and extended kalman filter. It also has the ability to automatically recognize discharge conditions.3) The prediction method of remaining useful life of lithium ion battery using GA and ARIMA is studied. The traditional ARIMA model order determination method is too complicated. The order determination process has the problem of subjective selection. In order to solve the problems mentioned above, the prediction method of remaining useful life of lithium ion battery using GA and ARIMA is studied. The experimental results show that it can effectively simplify the order determination process and improve the precision of prediction.