The Study Of New Energy Vehicles Data Clustering Analysis And Applications

New energy vehicles with advanced power control and driving technologies have the characteristics of advanced principle,advanced technology and novel architectures.In the practical operation and applications,new energy vehicles produce a lot of data in various form,such as voltage,current,temperature of battery and driving motor,etc.Thus,how to analyze the new energy vehicles data by data mining for machine error retrieval and other tasks has become one difficult issue in industry.Data mining technology can extract or mine the potential valuable information from the vast amount of raw data,which has been widely concerned by computer scientist and data scientist.Clustering analysis is one of the most important data analysis techniques in data mining,which can cluster data samples by discovering the similarities between samples in an unsupervised way.After classification by clustering,data in the same cluster are as similar as possible,and inter-cluster data are as different as possible.In this thesis,clustering technology is used to analyze the data of new energy vehicles for knowledge mining.The main research contents are as follows:(1)In this thesis,we first introduce the concept and technology of data preprocessing and clustering analysis in data mining.According to the complex characteristics of new energy vehicles data,such as data noise,data missing and data incomplete,etc.This thesis will investigate how to obtain clean new energy vehicle data with unified formats by data integration and normalization,for the subsequent clustering analysis.(2)The classical k-means and hierarchical clustering algorithms are employed for clustering and mining new energy vehicle data,which mainly obtain valuable information of the energy consumption of new energy vehicles,fault analysis,driving habits and other issues to provide technical support for safety operation,safety production and rational use.(3)Extensive theoretical analysis and experimental results show that the clustering analysis can indeed capture potential useful information and knowledge from the new energy vehicle data accurately.The clustering analysis results can be accurate to determine the relationship between the new energy automotive data,and it is helpful for the operators and producers to discover the problems in energy consumption,failure analysis and driving behavior of new energy vehicles.