Research And Analysis Of Vehicle Battery Data Based On Data Mining Technology

With the increasing environmental pollution and the shortage of non-renewable energy,the market demands more and more new energy vehicles.As the main power of new energy vehicles,vehicle-mounted battery technology has attracted extensive attention from all walks of life.At the same time,with the promotion of the concept of big data,the information contained in the data began to be valued by manufacturers and created huge economic benefits,and data mining technology was gradually applied in various stages of industrial production.In order to further improve and improve the new energy vehicle battery industry technology,mining new energy vehicles in the process of driving battery data generated potential information value.This paper studies and analyzes the data of vehicle battery based on data mining technology.Firstly,an effective data mining method is developed for the existing vehicle-mounted battery data set after analyzing the manufacturer's demand.According to the scheme,the data in the battery cell information table and the extreme value information table were mined,and the respective algorithms were determined from the extreme value information of temperature abnormal data detection,the correlation analysis of each feature and the clustering analysis of total current extreme value.Finally,the algorithm was implemented by python.Among them,aiming at the problem that k-means algorithm is not accurate enough in the result of total current clustering analysis,it is improved and the initial clustering center is selected according to a certain probability distribution to avoid the reliability problem caused by random sampling.Finally,an example is given to show that the improved k-means++ algorithm has better clustering accuracy and computing speed.In this paper,the total current is clustered according to the current difference of vehicle battery in different driving behaviors.According to the clustering results,the proportion of different driving behaviors is obtained,and the road condition information in the driving process is inferred.In the correlation analysis,this paper found that the battery temperature in the battery box was weakly correlated with the battery current,and the abnormal temperature distribution of the battery in the battery pack was determined through the detection of temperature anomalies.The results of two data analyses show that the temperature of the battery in the battery box is affected by its position in the battery box.For this reason,the manufacturers put forward Suggestions to improve the battery in different positions of the battery section,that is,to optimize the performance of the battery in the position of high abnormal temperature,enhancethe stability and other measures,to increase the stability of the whole battery pack.