Research On State Of Charge Estimation For Li-ion Battery Of Electric Vehicle

Li-ion battery is paid more and more attention due to its higher power density andenergy density, which is expected to be an ideal battery as an energy source for electricvehicle (EV). And people always pay great attention to safe, efficient, and proper applicationof Li-ion battery. as the basis of the EV’s energy management, State of Charge (SOC) ofLi-ion battery has a great influence on the performance of the EV.For the research and development on State of Charge estimation for Li-ion battery of EV,taking the LiFePO4battery as the research object. For solve the accuracy problem of the SOCestimation mainly from the battery model and estimation algorithm.Based on the equivalent circuit Thevenin model and identify the model parametersthrough the completely HPPC experiments at different temperature. Established the Theveninmodel based on temperature and SOC point2-d of charting method. Base on the dynamicrunning conditions of battery (UDDS condition and ECE condition), Analyzed and validatedthe precision of the model. The experimental result indicated the Thevenin model has highprecision, can accurate simulate the dynamic effects of battery, suitable for the high accuracyof battery model occasion.Based on the battery model, calculated the SOC of battery by use the extended kalmanfiltering (EKF) method combined with the Ah counting method, and imported theexperimental date to the simulation model. The experiment and simulation results show thatthe EKF method combine with Ah counting method has a very good anti-jamming andconvergence. Compensated for the error accumulation effect of the general Ah countingmethod which caused by the outside interference and the initial error of SOC.Designed the hardware and software of the SOC estimation system. The SOC estimatemodule based on the EKF algorithm is easy to realized. This method is suitable for EV’s SOCestimation.