Research On Modeling And State Estimation Method Of Power Lithium-ion Battery Under Wide Temperature Range

With the continuous reduction of fossil fuels and the increasingly prominent environmental problems,countries around the world are constantly increasing investment in the research and development of new energy vehicles,especially electric vehicles.Battery Management System(BMS)is the core component of electric vehicle lithium-ion Battery,which is an important basis to ensure the safe and reliable operation of electric vehicles.However,the time-varying nonlinear characteristics of power lithium batteries and their susceptibility to external factors,such as temperature,aging and dynamic working conditions,it is a great challenge to BMS for accurately estimating the states of power lithium battery.In this paper,research is carried out on SOC and energy state estimation of lithium battery for vehicle power,and the specific research work includes:(1)A test platform for power lithium batteries is built.Basic and aging experiments are carried out in a wide temperature range(0～50 ℃)for two types of lithium batteries with different materials and different structures,and experimental data of lithium battery characteristics are accumulated.Based on the experimental data,the track of the change of the Open Circuit Voltage(OCV)and the maximum available capacity(energy)with the working environment temperature and aging state of the lithium battery is analyzed.The experimental results show that the operating environment temperature has a significant effect on the open circuit voltage and the maximum available capacity(energy)of power lithium battery.The experimental database of lithium battery is established to provide data support for the subsequent state estimation research and the verification of the accuracy and reliability of the estimation algorithm.(2)For the influence of working environment temperature on the parameters of power lithium battery model and state of charge(SOC)estimation,a state of charge(SOC)estimation method of power lithium battery based on temperature compensation model is proposed.This method can effectively reduce the influence of temperature on the open circuit voltage and other parameters of lithium battery,so as to improve the accuracy and reliability of SOC estimation.Aiming at the shortcomings of traditional Kalman filter in dealing with time-varying nonlinear systems,an online estimation method of parameters and SOC of lithium battery based on particle filter is proposed,and the convergence speed and estimation accuracy of particle filter are improved by adaptive noise covariance.The simulation results show that the SOC estimation method based on temperature compensation model can achieve accurate estimation under complex conditions such as dynamic temperature,dynamic working conditions and aging.(3)The open circuit voltage(OCV)is the key factor of SOE estimation,a SOE estimation method based on adaptive parameter model is proposed.The OCV is incorporated into the parameter estimator for on-line estimation to improve the identification accuracy of the model.To reduce the dimension and complexity of the parameter estimator,a recursive least squares method with forgetting factor is proposed to identify the battery model parameters.In addition,considering the shortage of fixed forgetting factor,a recursive least square method with variable forgetting factor is proposed.Considering the deficiency of traditional Kalman filter which assumes that the statistical characteristic of noise is known,The H-infinity filter is proposed to estimate SOE.An adaptive SOE estimation method based on variable forgetting factor recursive least square method and H infinite filter is established.The accuracy and robustness of the proposed estimation method are verified by constant current and dynamic cycles.(4)Considering the issue of multi states joint estimation of power lithium battery,an estimation method based on multi time scale is proposed.The SOC and SOE of lithium battery are estimated by micro timescale,and the parameters of lithium battery model are identified by macro time scale.Aiming at the issue of time interval selection of multi timescale,the accuracy and stability of lithium battery model are studied theoretically.According to the conservative assumption of Kalman filter for the system noise characteristic,H infinite filter is used as the estimation method to estimate the parameters and multi states of lithium battery online.The accuracy and reliability of the method are verified by different temperature working conditions and different materials of lithium batteries.(5)Aiming at the application and verification of power lithium battery state estimation methods,methods and technologies such as lithium battery modeling and state estimation are integrated into the battery management system to realize the fine and efficient management of power lithium battery.Based on the development trend of battery management system,a distributed architecture is adopted to develop the battery management system.In order to verify the reliability of the state estimation method proposed in this paper,the reliability of the estimation method proposed in this paper is verified through the bench test of power lithium battery pack,which lays a solid foundation for the engineering application of battery management system.