| With the wide application of new energy vehicles,the safety and reliability of lithium-ion power batteries have attracted much attention.Efficient and reliable battery management system is an important foundation to ensure the safe and efficient operation of lithium-ion battery system.Because lithium-ion battery is a complex electrochemical thermal coupling system,establishing a simple and efficient battery model that can fully reflect the dynamic characteristics of lithium-ion battery is of great significance for the state estimation and management of lithium-ion battery.Firstly,the single particle model(ESP1 model)directly adding electrolyte phase voltage in the electrochemical model is improved,an improved single particle model(ESP2 model)considering the non-uniformity of electrode open circuit potential in the electrode thickness direction is established and an improved single particle equivalent circuit model(ECM-ESP2 model)based on ESP2 model is established by approximation method and error correction method.It completely uses the same parameters as the traditional equivalent circuit model,such as resistance and capacitance,and its complexity and number of parameters are lower than the ESP2 model in the form of non equivalent circuit.Secondly,the temperature dependent ECM-ESP2 model is established by using Arrhenius equation which describes the relationship between electrochemical parameters and temperature of lithium-ion battery.Then,a parameter identification method is designed for the developed temperature dependent ECM-ESP2 model.Finally,the state of charge(SOC)and state of power(SOP)of lithium-ion battery are estimated by using the temperature dependent ECM-ESP2 model.The main works is as follows.1)Based on the analysis of the internal potential distribution of the pseudo two-dimensional model(P2D model)and the single particle model with directly adding the electrolyte phase voltage,an improved single particle model considering the non-uniformity of the open circuit potential in the electrode thickness direction is proposed.The accuracy of the improved single particle model is higher than that of the existing single particle model without adding electrolyte phase voltage and directly adding electrolyte phase voltage.2)This paper presents a method to approximate the transfer function of the lithium ion concentration difference between the surface and the average concentration of solid particles by using the integer order transfer function composed of multiple first-order systems.The parameter value of the approximate transfer function is obtained by minimizing the square sum of the amplitude error in frequency domain between the approximate transfer function and the original transfer function.Then,in order to simplify the calculation of lithium ion concentration in electrolyte phase in the improved single particle model,the transfer function for calculating lithium ion concentration distribution in electrolyte phase is obtained by Laplace transform of polynomial approximation method.3)This paper systematically analyzes the linearization form of Butler-Volmer equation and the law and characteristics of linearization error.On this basis,a method of using conversion coefficient to correct the linearization error of Butler-Volmer equation is proposed.In this paper,the analytical solutions of the potential distribution of solid phase and electrolyte phase in the improved single particle model are derived.On this basis,the transfer function of electrolyte phase voltage is obtained.Based on the potential distribution of electrolyte,the modified value of terminal voltage of improved single particle model is deduced.The equivalent circuit model of the improved single particle model(ECM-ESP2model)is obtained by integrating the transfer function of lithium ion concentration on the surface of solid particles and electrolyte phase and the transfer function of the rest of the improved single particle model.The accuracy of the model is verified by several conditions.The electrochemical impedance spectrum of ECM-ESP2 model is also analyzed and compared with that of P2 D model.The results show that the ECM-ESP2 model proposed in this paper also has high accuracy in frequency domain.4)A temperature dependent ECM-ESP2 model with temperature as one of the model inputs is established by using Arrhenius equation.The computational complexity and the number of parameters are lower than the commonly used multi RC equivalent circuit model.In this paper,the state space equation for calculating the average temperature of cylindrical batteries is derived,so that the battery temperature can be calculated by a simpler state space equation.Combined with the temperature dependent ECM-ESP2 model and the state space equation for calculating the battery temperature,an electrothermal coupling model based on ECM-ESP2 model is established in this paper.The comparison with the electrothermal coupling model based on P2 D model shows that the electrothermal coupling model based on ECM-ESP2 model has high accuracy.5)Based on the OCV-SOC test data of the full battery,this paper designs a method to identify the capacity of lithium-ion battery and the electrode SOC range of positive and negative electrodes,and then obtain the mapping relationship between the open circuit potential of electrodes and the normalized electrode SOC.The temperature coefficient of the open circuit potential of the battery is obtained through the variable temperature shelving experiment.Then,this paper proposes a method to identify the interface reaction resistance of the battery,the ohmic DC internal resistance without interface reaction resistance and their temperature characteristics by combining the electrochemical impedance spectroscopy(EIS)curve of the battery and the HPPC test and using the nonlinear characteristics of the interface reaction resistance.In this paper,taking the test data of multiple SOC test points at multiple temperatures as the excitation data,the parameter values related to lithium ion diffusion in ECM-ESP2 model and the activation energy parameter values controlling its temperature characteristics are identified by genetic algorithm.Through simulation and experiment,the modeling accuracy of ECM-ESP2 model and traditional multi RC equivalent circuit model under different temperatures is compared.The results show that the modeling accuracy of ECM-ESP2 model,especially in the low SOC section of battery,is significantly higher than that of traditional multi RC equivalent circuit model.6)The SOC estimation methods based on ECM-ESP2 model and traditional multi RC equivalent circuit model are compared.Simulation and experiments show that due to the improvement of model accuracy,the SOC estimation accuracy based on ECM-ESP2 model is higher than that based on traditional multi RC equivalent circuit model.Aiming at the problem that the inaccurate battery temperature leads to the reduction of model accuracy and SOC estimation accuracy,an SOC estimation method based on interactive multi model is designed.7)This paper studies the phenomenon and reason of large error of lower cut-off voltage constraint current estimated by traditional linear approximate terminal voltage prediction method in SOP estimation.To solve this problem,a calculation method of lower cut-off voltage constrained current based on non approximate terminal voltage prediction method based on ECM-ESP2 model is proposed in this paper.The simulation and experimental results show that this method can significantly improve the SOP estimation accuracy in the low SOC section of the battery.In general,this study establishes an equivalent circuit model based on the improved single particle model with low model complexity.This new model can significantly improve the accuracy of terminal voltage prediction and state estimation of lithium-ion battery.This study has practical significance and improvement for the engineering application and theory development of lithium-ion battery management system. |
PDF Full Text Request