Prediction Method of Lithium Battery's State of Charge Based on No Trace of Calman Filter

摘要

The Thevenin equivalent circuit model is established for single lithium battery, current and voltage data to identify the parameters of the equivalent circuit is obtained by the discharge experiment, and the open circuit voltage and charge state relationship curve was obtained by curve fitting. On this basis, design the extended Kalman filter algorithm and unscented Kalman filter algorithm on the lithium battery state of charge, then use Matlab/Simulik simulation, the results of the state prediction of the two different algorithms are compared. The analysis results show that two kinds of algorithm are effective for single lithium battery state of charge estimation, and no trace of Calman filter algorithm can effectively solve the the problem of accuracy is not high of the extended Calman filter, which due to the linear approximation.