Application Research On Data Mining In New Energy Vehicle Accident Analysis

In recent years,the production and sales of china’s new energy vehicle have been increasing year by year and been in the world’s leading position for six years.While new energy vehicles are popularized,the safety has also attracted people’s attention,and the occasional fire accidents have aroused widespread concern in society.New energy vehicle’s fire accidents are affected by many factors,How to find potential and common development rules from numerous accidents is the primary problem faced in the analysis of new energy vehicle accidents.Data mining is a kind of data analysis method,aiming at digging out implicitly unknown concepts,rules,regulations and modes with potential value to decision making.This paper,based on the features of data collection of new energy vehicle’s accident and the key problems in data analysis application,takes the historical data of new energy vehicle power battery as the data mining object,uses data mining related theories,feature extraction,data dimensionality reduction,cluster analysis,constructs the accident analysis model based on data mining,and conducts accident analysis research on new energy vehicles.This paper achieves the following results:(1)Aiming at the problems of completeness,objectivity and standardization in the process of data collection and transmission,this paper uses data preprocessing technology to process the data uniformly and build a data set for model verification.(2)Aiming at the limitations of traditional methods in the analysis of new energy vehicle accidents,this paper uses wavelet transform,singular value decomposition,k-means clustering,correlation analysis and other methods,constructes two accident analysis models based on data mining technology,uses the data of normal vehicles and accident vehicles to verify the effectiveness of the model.(3)Aiming at the process and standardize of new energy vehicle accident analysis,this paper designs a analysis management system of new energy vehicles accident data,introduces the multi-source data preprocessing system,accident data analysis system and visualization module,then analyzes the abnormal data patterns what excavated by the system.The result can provide a reference for the application of data mining technology in new energy vehicle accident analysis,and it has important practical significance in improving the depth of new energy vehicle accident investigation and the efficiency of accident analysis,real-time monitoring and early warning of new energy vehicle operation,maintaining the safety of consumers’ lives and property.