The State Estimation Study On Power Lithium-ion Battery Of Electric Vehicles Based On Date-driven

In recent decade,the ownership of Electric Vehicles(EVs)has been increasing in major regions of the world year by year.On the one hand,owing to the world’s tight energy reserves,on the other hand,because of the serious impact of global environmental problems on people’s lives,it is urgent to solve environmental problems,which have accelerated the development of the EVs industry.In the paper,the health state estimation of lithium-ion battery refers to the state of charge(SOC)and state of remaining capacity(also known as State of Health(SOH))of the lithium-ion battery.The definitions of other critical states of lithium-ion batteries are given in the current state of research.The main reason for studying these two states of lithium-ion batteries in this paper is that the accuracy of SOC estimation determines the accuracy of BMS estimation of other critical states,which is expressed as the range of EVs;while the accuracy of SOH estimation determines remain useful life,which is related to the degradation state of lithium-ion batteries,it is related to the normal safe and reliable use of lithium-ion batteries,and also to the precise maintenance of lithium-ion batteries.In summary,the following studies were conducted on the reliability estimation of lithium-ion health state SOC and SOH:(1)Firstly,the electrochemical reaction characteristics and basic principles of Lithium-ion battery were summarized and analyzed,and then based in Urban Dynamometer Driving Schedule(UDDS),operating condition discharge test platform was built;(2)Based on the previous research,the 2-RC equivalent circuit model of lithiumion battery is constructed from two dimensions,i.e.,2-RC equivalent circuit model(2-RC means the circuit contains two RC branch structures),and the recursive least squares(RLS)and adaptive genetic algorithm(AGA)are introduced for the two equivalent circuit models by using the current and voltage data collected from the working condition discharge test.The errors of the two parameter identification algorithms are compared and analyzed to support the reliability estimation of the next lithium-ion battery health state;(3)Based on KF Algorithm and PF Algorithm,developed two health state estimation algorithms for battery in the paper,which are DEKF and EKPF-EKF algorithm.Where DEKF estimated the SOC of battery based on the 2-RC equivalent circuit integer-order model,then verified the accuracy of the proposed algorithm based on UDDS,the mean error is 1.2%.Similarly,EKPF-EKF performs the joint estimation of SOC,SOH for Lithium-ion battery based on the 2-RC Li-ion battery equivalent circuit fractional-order model which ultimately analyzes and compares the FOEKPF,IOEKPF,and FOPF algorithms,and concludes that the average error of SOC estimated by FOEKPF-EKF is 0.48%.Finally,the reliability of the developed lithium-ion battery health state estimation algorithm was verified from two dimensions.What is lacking,however,is the complexity of actual road driving conditions and the lack of actual road driving tests,which need to be further explored in future research.