The Research On Prediction Method Of Remaining Available Energy Of Lithium Ion Power Battery

The accurate estimation of the energy state and remaining available energy of lithium-ion battery is still a challenging problem for electric vehicles,which is the basis of electric vehicle energy management and is crucial for the estimation of residual driving range.Therefore,special research is necessary.In this paper,lithium-ion battery is taken as the research object,and the energy aspect of the battery is researched.An approach for predicting remaining available energy based on State-of-Energy and energy-conversion-efficiency map is proposed.Firstly,based on the theory of fractional calculus and electrochemical impedance spectroscopy,the structure of the battery model is established,and the recursive least squares algorithm is used to realize the online identification of the battery model parameters,which provides a model basis for establishing the battery energy state estimator.Secondly,to overcoming the drawback of conventional State-of-Energy definition,a novel definition of State-of-Energy based on the first law of thermodynamics is presented considering the conversion between chemical energy,electric energy and thermal energy of the battery during charging and discharging.Thirdly,there are inevitable uncertainties about system noise and measurement noise in the applied stochastic dynamic battery system,and the problems of estimator's non-convergence in state estimation.An adaptive State-of-Energy estimator is established via the battery model and Square-Root Unscented Kalman Filter.Finally,through the experimental data,the energy conversion rate MAP based on the battery operating condition and energy state is established,and the prediction method of the remaining energy available for the battery is established based on the prediction of the operating conditions of the battery.Then,he reliability and accuracy of the proposed approach are verified by the experiments under various representative dynamic cycles and charging condition.Experimental results show that the errors of State-of-Energy estimation and remaining available energy prediction remain within 3% and 4% respectively.