Study On Closed-loop SOC Estimation Based On Variable Parameter Battery Model

Today,energy and environmental issues have become an important issue in the development of the world.As the representative of conserving energy and protecting environment,electric vehicles are favored by consumers because of their advantages such as low pollution,zero emissions and low cost.As the main source of the electric vehicles,the power lithium battery has become a hot research in recent years.State-of-Charge(SOC)directly reflects the remaining capacity of the battery and the mileage of the electric vehicles in some degree.The accurate estimation of SOC is of great significance to manage the battery power effectively,predict the battery endurance accurately and avoid the over-charge or over-discharge of the battery.Taking the power lithium battery of electric vehicles as the research object,this dissertation studies the estimation of battery SOC that is based on two main factors including the battery model and the filtering algorithm.Firstly,this dissertation chooses Thevenin model which is easy to carry out the identification of model parameters and has a clear physical meaning as the research object by analyzing the working principle as well as its advantages and disadvantages of common battery model.Aiming at the problem that the parameters of commonly used battery model are fixed and the scope of application is limited and combining the laws of change of battery resistances in the actual work process,the variable parameter Thevenin model affected by the temperature and the SOC is established.In order to make the battery model accurate and less complicated,the model parameters are identified by the Design of Experiment(DOE)method based on Central Composite Designs(CCD)and the least square method in this dissertation.The accuracy of the filtering algorithm plays an important role in ensuring the estimation accuracy of SOC.Then,by analyzing the applicated principles of the commonly used filtering algorithms,this dissertation has proposed an improved unscented Kalman particle filter(IUPF)algorithm to solve the problem that the estimated accuracy of the algorithm is affected when the system noise is larger.The system state noise and the measurement noise are simultaneously introduced into the sample point so that the noises are synmmetrically sampled and imported into the process of the algorithm calculation to ensure the accuracy of the algorithm.The IUPF algorithm which is adopted based on variable parameter Thevenin model can reduce the impacts of noises on the estimated accuracy of system while ensuring the scope of the model.This dissertation chooses the same type and batch of multiple 9A-h cylindrical lithium-iron phosphate batteries as the research object.Aimed at charging or discharging cabinet and high or low temperature box,the experiment takes multiple charge and discharge experiments under different conditions.Experimental and simulation results show that the SOC estimation method based on IUPF algorithm and variable parameter battery model can keep a higher estimation accuracy over a large temperature range,while solving the problem that the scope of application is limited as well as keeping the accuracy of the model.Especially when the system state noise and measurement noise impact seriously,the accuracy of the model is improved,and the method has better robustness to the disturbance caused by the model parameters.