Research On Lithium-Ion Battery Life Test System And RCL Indirect Prediction

As an advanced green battery,lithium-ion battery has been widely studied and applied in various fields,due to its advantages of high energy density,light weight,long life and environmental protection.However,lithium-ion battery has a decline in performance.If the above situation isn’t properly handled,the normal operation of the electrical system will be affected,and disastrous result may happen.Therefore,the performance parameters of lithium-ion battery are accurately measured.And the remaining cycle life(RCL)is accurately predicted.They are foundations to guarantee the reliability and safety of lithium-ion battery.Meanwhile,it’s also an important part of the key technology of lithium-ion battery health management system.In this paper,the life test system of lithium-ion battery and the indirect prediction method of RCL are deeply studied.Firstly,based on the research of the life test method of lithium ion battery,the lithium-ion battery life test system for modeling life testing as a key function is designed.The charge and discharge ratio,temperature and depth of discharge as factors are considered for routine life test.Meanwhile,some evaluation life tests can also be carried out.The test accuracy and stability of the life test system is verified by experiments on 18650 lithium-ion batteries produced by a company.Moreover,the measured data can be displayed in the human-computer interaction interface and saved to the upper computer.Furthermore,the measured data also can provide data support for the study of the RCL indirect prediction of lithium-ion battery in the rear.Secondly,though the degradation process of the lithium-ion battery is analyzed,the decrease of the actual capacity is the direct expression of the degradation process of the lithium-ion battery.Moreover,based on the measured data of 18650 lithium-ion battery and the public data of NASA lithium-ion battery,the data of voltage is processed to obtain time interval of equal discharge voltage as the indirect life characteristic parameter of lithium-ion battery.Meanwhile,correlation coefficient method is used to prove strong correlation between actual capacity and time interval of equal discharge voltage.The strong correlation indicates the time interval of equal discharge voltage as the indirect life characteristic parameter is feasible and reasonable.Furthermore,the RCL indirect prediction experiment of lithium-ion battery is carried out based on the Relate Vector Machine algorithm.The experimental results show that the multi-step prediction ability of the Relate Vector Machine algorithm is weak,but the regression ability is strong.Finally,due to the results of the RCL indirect prediction of lithium-ion battery based on Relate Vector Machine algorithm,a fusion algorithm of the RCL indirect prediction is proposed.The Particle Filter algorithm and the Autoregressive AR model are introduced respectively in this paper.Meanwhile,the advantages of the two above algorithms are combined with the Relate Vector Machine algorithm mentioned earlier,in this paper.Then,the fusion algorithm is obtained.Moreover the RCL indirect prediction experiment of lithium-ion battery is carried out based on the fusion algorithm.The experimental results show that the fusion algorithm can improve the prediction accuracy.Furthermore,the validity and universality of this fusion algorithm also are proved.