Research On SOC Estimation Of Lithium Ion Battery Based On Composite Model Method

With the continuous improvement of my country’s rail train technology,people generally choose relatively safe,fast and comfortable high-speed train as a vehicle for long-distance travel.Nowadays,the auxiliary power supply system of high-speed trains usually uses lithium-ion battery packs.However,the state of charge(SOC)of the battery is one of the key parameters to ensure the safe and comfortable operation of high-speed trains.Therefore,accurate estimation of lithium-ion batteries SOC technology has become a hot research topic in the current industry.In this thesis,the main research direction is to accurately calculate the SOC of on-board lithium-ion batteries,focusing on the construction of battery model and the estimation of the SOC of on-board lithium-ion batteries.First,the basic characteristic relation of the ternary lithium-ion battery pack was obtained.Then,according to the characteristics of high-speed train battery working environment and the basic characteristics of battery pack,a scheme to construct the improved second-order RC equivalent circuit model using MATLAB/SIMSCAPE components was developed,and the parameters of the battery model were identified based on polynomial fitting algorithm.Secondly,according to the actual operating conditions of the high-speed train and the actual performance of high-speed train battery,a solution for SOC estimation of on-board lithium batteries was proposed.It introduces a weight factor k to assign the proportion of the SOC results estimated by the improved Ampere-Hour integration method(AH)and the Adaptive Unscented Kalman Filter method(AUKF)in different discharge stages,so that the two can be combined with each other and make up for the single AUKF filtering method in low battery power moment and Ampere-hour integration method in the period of battery current fluctuation estimate the defects of inaccuracy,finally obtained a more accurate SOC estimation value of the on-board lithium battery.Finally,an experimental platform was built to test and verify the accuracy and convergence of the AH-AUKF composite model method for estimating the SOC of on-board lithium-ion batteries.The results show that:(1)Under different discharge conditions,the error of estimating the SOC of lithium-ion batteries by the AH-AUKF composite model method is always kept within 1.95%,which has a relatively high estimation accuracy.Under different initial battery power values,the AH-AUKF composite model method can rapidly converge the estimated value to the neighborhood of the reference value when estimating the battery SOC,which has good robustness and convergence.(2)Put the battery in the experimental environment for several consecutive days of testing,and the test results showed the battery SOC value estimated using the AH-AUKF composite model method in the actual environment,and the estimation error has always remained at about 2.5%,indicating that it also has a good estimation accuracy in the actual environment.To sum up,the estimation scheme proposed in this thesis solves the problem that the SOC estimation effect is not ideal when the battery is in low power and the current fluctuates greatly,and compensates the estimation error to the maximum extent.It improves the use efficiency and working time of the on-board lithium-ion battery.