Joint Estimation Of SOC And SOH For Lithium-ion Battery Packs Considering Temperature And Capacity Correctio

With the decreasing and non-renewable oil resources and the increasing number of traditional fuel vehicles,environmental problems and energy crisis are becoming increasingly serious.The state vigorously develops new energy vehicles to reduce a series of problems caused by fuel vehicles.As a kind of new energy vehicle,electric vehicle takes battery as the power source,and has been widely concerned by people.Battery Management System(BMS)is an important part of electric vehicles.The accurate estimation of battery state of charge(SOC)and battery state of health(SOH)has always been the focus and difficulty of BMS technology research.In this paper,the following studies have been carried out around the joint estimation of SOC and SOH of lithium-ion batteries:Firstly,the working principle of lithium-ion battery is described,and the experiment of lithium-ion battery is carried out.Understand the relationship between its external characteristics and some influencing factors.The change rule and working characteristics of lithium-ion battery parameters at different charge-discharge current ratio,temperature and temperature are also investigated.The results show that the initial discharge voltage of the battery is slightly different at different discharge rates.The discharge capacity of the battery is also different.The discharge capacity of the battery decreases with the increase of discharge rate.Temperature also has a certain impact on the battery capacity.When the working temperature of the battery decreases with the ambient temperature,the energy released by the battery and the actual capacity will decrease when the battery is discharged from the full charge state to the cut-off voltage.The influence of temperature and discharge rate on the capacity was analyzed,and the actual capacity of the battery was corrected by adding an influence factor.Then the commonly used battery models are analyzed,and the second-order Thevenin equivalent circuit model is selected to establish the mathematical model considering the model accuracy and computational complexity.The offline and online parameter identification of the model is carried out based on the battery discharge data.The off-line identification method is to identify different internal resistance and capacitance values at different SOC points by fitting the zero input response voltage curve.And the recursive least square method with variable forgetting factor is used for online parameter identification to obtain the change curve of model parameters with SOC.Under pulse discharge condition and FUDS discharge condition,the current data is imported into the model.By comparing the output value of the model with the experimental voltage,the accuracy of the model parameters is verified.Then,based on the basic principle of extended Kalman filter algorithm,the battery capacity is selected as the characteristic quantity of SOH.The improved Sage-Husa adaptive algorithm is introduced based on the double extended Kalman filter algorithm.Realize the real-time update of the system covariance matrix.In order to reduce the system computation,further add the multi-time scale theory for optimization.The SOC estimation adopts micro scale,and the SOH estimation adopts macro scale,which greatly reduces the calculation amount of the algorithm.The algorithm is verified at different temperatures and different working conditions.The SOC estimation error is kept within 2.4%,and the capacity estimation error is kept within 1.2%.The performance of the algorithm is analyzed when the initial state is biased,and the robustness of the algorithm is verified.Finally,the battery management system experimental platform was built.The system hardware circuit is mainly composed of main control unit,battery status information acquisition unit and power supply unit.The algorithm in this paper is written into BMS to test SOC and SOH estimation.The experimental results show that the SOC and SOH estimation algorithms designed in this paper have been fully verified in the experimental test platform.During the experiment,the estimation error of the two is controlled within4%,which conforms to the technical requirements of the battery management system for electric vehicles in China.It shows that the algorithm proposed in this paper is accurate and reliable,and has certain reference value in practical engineering.