Research Of Battery Sorting And Performance Evaluation Based On Fuzzy Concept Used In Electric Vehicles

With the increasing popularity of electric vehicles and the wide application of power batteries in automotive power systems,the life span and performance of power batteries used as vehicle energy source affect the comprehensive performance of electric vehicles directly.The inconsistency of single battery will not only reduce the performance of the battery group,but also cause safety problems.Because the fuzzy set theory has good flexibility and intelligence,it is very suitable for the complex system such as batteries.To solve the problem of the inconsistency of single battery,this paper combines fuzzy clustering and fuzzy comprehensive evaluation method,and proposes a method for assessing and sorting property of power battery based on fuzzy theory and it has been studied in depth.Firstly,the development of electric vehicles and power applications and the research results of assessing and sorting batteries and the inconsistency of power batteries performance are introduced in this paper.After introducing and analyzing some common methods of clustering,charging and discharging batteries experiments are designed to acquire original clustering sample,and a kind of sorting method is designed which is based on clustering method.All of them are based on the distance function which is a characterization of different single battery.Through analyzing the results of theory,definite sorting sample results are obtained using this method,but there are some problems,like the distortion of sorting results.Secondly,common Fuzzy C-means are overviewed,and common Fuzzy Cmeans are studied in this paper.The paper analyzed the voltage platform on the curve of battery discharging,and then designed a Fuzzy C-mean of sorting batteries using discharging voltage platform.In this method,original sample data are obtained in the experiment of charging and discharging the samples,and then three features points are selected from the voltage platform,and they are standardized as clustering objects.Then this paper designed program to analyze different results in different categories,and selected suitable classifying number via the validity indicators.The paper paralleled the results into groups through the judging criteria of the batteries' state of charge,and did the cycle life test experiment which is compared with reference experiment.It can be proved that it is necessary to the use of sorting batteries into groups,and the method is effective from the experiment results.In the end,to solve the problems of performance degradation and group consistency,which are caused by low performance of single battery,the paper proposed a method of assessing single battery performance in groups combined with fuzzy comprehensive evaluation.To be exact,general steps of fuzzy comprehensive evaluation are given,and the evaluation object and influence factor of battery performance are defined.The weights of electrical performance,safety and other norms are defined through Delphi and coefficient of variation,following by selecting different membership to establish functions evaluation matrix through the indicator characteristics and obtaining fuzzy comprehensive evaluation set.At the same time,the paper designed batteries holding and HPPC experiments to get evaluation data sample,and introduced detailed theory procedure.At last,the validation of evaluation results is analyzed.