Estimation Of Remaining Driving Range For Pure Electric Vehicle Based On Condition Recognition And Driving Style Recognition

The theme of today’s era is energy and environmental protection,and electric vehicles have the characteristics of pollution-free,low energy consumption,so all countries attach great importance to its development,but restricted by the factors of driving range,resulting in the popularization of electric vehicles encountered obstacles.The extended driving range of pure electric vehicles is generally short.Accurate estimation of the remaining extended driving range can guide users to travel,alleviate the problem of "range anxiety" of users,and be beneficial to the development of pure electric vehicles.The accurate estimation of the remaining driving range of pure electric vehicles is not only subject to the vehicle structure parameters,but also to the driving conditions and driving styles.The structural parameters of the vehicle have been determined at the time of vehicle manufacturing,and the structural parameters of the vehicle will not change during the driving process.However,the type of working condition and the driver’s driving style with random characteristics are not invariable in the process of vehicle driving.Therefore,based on the influence of different driving conditions and driving styles on the remaining driving miles,this paper proposes an estimation method of the remaining driving miles of pure electric vehicles considering driving conditions and driving styles based on the establishment of the working condition identification model and driving style identification model,and compared with other estimation methods,in order to reflect the advantages of this method.Specific research contents are as follows:Firstly,the factors affecting the remaining driving distance of pure electric vehicles are divided into fixed factors and random factors.The fixed factors are the structural parameters of the vehicle,and the random factors are the type of working condition and driving style.Combined with the previous research results,the influence rules of the random factors on the remaining driving distance are introduced.Secondly,based on the working condition data in the ADVISOR software,the system clustering algorithm and principal component analysis method combined with LVQ(Learning vector quantization)neural network were used to establish the working condition identification model.The driving style evaluation index is analyzed,and the driving style identification model is established by using k-means clustering algorithm and fuzzy reasoning system in combination with the driving style real car data of a company.Thirdly,based on the model of working condition identification and driving style identification,the method of determining the correction coefficient of driving style to the residual driving mileage under different working conditions is presented,and the estimation algorithm of residual driving mileage based on working condition identification and driving style identification is proposed.Finally,the residual driving distance estimation algorithms based on average power consumption,based on condition identification and based on condition identification and driving style identification are compared.The results show that the estimated value of the residual driving mileage based on the condition recognition and driving style recognition is the closest to the actual residual driving mileage,which indicates that the estimation method of the residual driving mileage proposed in this paper has the higher accuracy.