Adaptive State Of Charge Estimation Based On The Improved Feedback Extended Kalman Filtering For High-power Lithiumion Batteries

The rechargeable battery is getting familiar in electric vehicles,hybrid electric vehicles,as well as many stationary devices which are very necessary for daily life due to high energy,stability,and longevity.Considering all the useful rechargeable batteries,lithium-ion is a more attractive and reasonable choice that can serve to prolong its service lifespan,energy-saving,emission reduction,and reduce internal losses.However,the internal mechanism and structure are comparatively complex so that the dynamic behavior and state of charge(SOC)is affected by various complex factors such as discharging current,self-discharge.On the other hand,the Parameters are crucial as it indicates the performance and longevity of the battery.Therefore,an accurate SOC estimation charging status is a symbolic rating indicator for energy control techniques used in those fields which can reflect its residual charging states.Therefore,it is a very necessary task to process lithium-ion batteries and accurately measures their state-ofcharge parameters.In this paper,the following research is carried out to achieve accurate SOC estimation of powerful lithium-ion batteries.1.Dealing with internal reaction,the dynamic high-order electrical model is hired2.Mathematical derivation is calculated according to model requirements3.Model parameters are identified by the state-space description4.SOC with Open circuit Voltage(OCV)appeared through the hybrid pulse experiment5.To verify the terminal voltage,a simulation model has designed and implemented6.To deal with non-linear behavior,improved feedback extended Kalman filter(FEKF) algorithm is introduced with its mathematical explanation7.The algorithm is written by S-function and it verifies through the high-order model To,identified the lithium-ion battery parameters and investigate the internal reaction,a dynamic high-order equivalent circuit model is hired which is expand by 2-RC networks in series with its ohmic internal resistance.Also,the RC values(Resistance and Capacity)were measured and two high-order circuit conditions were resolved.Furthermore,the hybrid pulse experiment helps to identify the parameters which are used to estimate the residual SOC.Then the calculation is performed by OCV and the improved FEKF algorithm is performed.The dynamic high-order model simulation results appeared to be less than 0.030 V which indicates the optimization strategy can improve the SOC estimation remarkably.Finally,the iterative prediction and correction strategies are introduced in the process,and the initial error is reduced by less than 0.0325 V convergence employing an improved FEKF.