Research On SOC Estimation Method In Ev Battery

With the increasingly serious global environmental pollution as well as the world’s growing shortage of non-renewable energy, the current environmental problems to the poor countries have to a lot of trouble, which led the world for electric vehicle research boom. The most important components of electric vehicles is the battery, the battery core technology is the battery state of charge(State of Charge, SOC) estimation;Battery SOC estimation accuracy is higher, the battery remaining state of the driver can have a better understanding, but also can improve their performance and battery life of electric vehicles is of great significance and research value in the future. Therefore, this paper to estimate the battery SOC for the study, the main work is as follows:Firstly, the current situation and development of electric vehicles, battery electric vehicles leads to the heart, while the battery SOC estimation is the core power battery technology, brief battery SOC estimation method and the status quo.Secondly, the analysis and comparison of several commonly used on the market power battery, lithium battery finally selected as the research object of this paper describes some of the basic characteristics of the lithium battery and works,It also introduces several factors affect lithium battery SOC, such as temperature, aging, charge-discharge rate and so on;Then it introduces several common battery SOC estimation method, when Ah-integration, open circuit voltage, Kalman filtering, and ultimately choose unscented Kalman filter method for estimating the SOC method such as a battery.Again, introduces several common battery model, such as Rint, Thevenin, PNGV peer model, analyze the advantages and disadvantages of these types of models, and finally choose to Thevenin basis, a series RC circuit second order RC model, confirmed by experiments correspondence between the open-circuit voltage and the SOC, while the improved model parameter identification and verification to ensure that the improved model can accurately simulate the actual battery status.Finally, the theoretical principles of Kalman Filter Method and unscented Kalman filter, unscented Kalman filter algorithm is mainly suitable for nonlinear systems, battery power just in the actual use of the process is non-linear;Its main principle is to provide pre-estimate of the cycle and then continuously update and select the best estimates, and the error correction has effect, closer to the theoretical value; The experimental data import unscented Kalman filter model built by the MATLAB, the battery SOC simulation and analysis of simulation results with theoretical values for comparison, the estimated unscented Kalman filter algorithm has high power battery accuracy.