Study On Lithium-ion Battery Modeling And SOC Estimation Algorithm Based On AIEKF Algorithm

The energy crisis and environmental crisis have gradually developed into two huge problems that plague the development of the world and human survival.As a transportation in the future,electric vehicles have become a hot spot of concern and development at home and abroad with their advantages of energy saving,environmental protection and high efficiency.As a power source for electric vehicles,accurate estimation of their state of charge(SOC)is great significance for lithium-ion batteries.High-precision SOC estimation can not only effectively prevent overcharging and overdischarging of lithium batteries,improve battery life,but also can improve the battery capacity utilization efficiency,maximizing the cruising range of electric vehicles under the premise of limited capacity.In this paper,the current mainstream high specific energy density ternary lithium battery is selected as the research object,and then an improved second-order RC equivalent circuit model was established.The model parameters are identified based on experimental data.Based on the Extend Kalman filter(EKF)algorithm,the noise covariance update rules and iterative theory are introduced to form an adaptive iterative extended Kalman filter(AIEKF)to estimate the SOC of lithium-ion battery.The experimental results show the effectiveness of the AIEKF algorithm.Here are the specific work done in this article:Firstly,this paper studies and analyzes the current development status and technical route of lithium-ion batteries,and also great insight to the current research status of SOC estimation algorithms research actuality.Then it introduces the working principle of lithium-ion batteries and analyzes the implementation methods and existing problems of current SOC estimation algorithm.Proposing an improved SOC estimation algorithm to effectively overcome these problems.The battery test platform is built and the ternary lithium-ion battery is charged and discharged under different temperatures and different charge and discharge rates,including pulse charge and discharge test and capacity test,working condition test and so on.Based on these experimental data,the model parameters were identified and the characteristics of the lithium battery were analyzed.An improved second-order RC model is proposed and the model is built in Matlab.Finally,the parameters of the identification are brought in model,and the accuracy of the model parameters is verified by the measured voltage data.Then the principle of extended Kalman filter is introduced briefly.On this basis,the improved method is introduced as a key point.The algorithm model based on Matlab/Simulink is established,and the improved SOC estimation algorithm is verified under the dynamic test condition(DST)and the urban road circulation condition(UDDS).The results show that the improved algorithm AIEKF has higher estimation accuracy and faster convergence speed compared with the classical extended Kalman(EKF)and adaptive extended Kalman filter algorithm(AEKF).The hardware circuit of battery management system related to SOC estimation is designed,and choose the mainstream Freescale chip for automotive.The designed hardware circuit mainly includes the current sampling module,temperature sampling module,voltage sampling module and CAN communication module and so on.The underlying driver is implemented in C language.The SOC algorithm uses simulink model to build and implement embedded code generation,and finally realizes the integration of the underlying code and functional layer programs.Experiments show that the adaptive iterative Kalman filter algorithm has higher estimation accuracy and faster convergence speed in SOC estimation under complex conditions,and also provides a new idea for SOC estimation method.