Monitoring And Analysis System Of Electric Vehicle Battery Based On Cloud Service

With the development of new energy industry,electric vehicles has become the future direction of automobile industry,and lithium-ion battery has been the vast majority of electric vehicle energy sources.Due to the lithium-ion battery's unique electrochemical properties,compared with the traditional fossil fuels,lithium-ion battery exist larger defectsin the use of life,energy reserving and other problems.Battery management system(BMS)has become one of the key technologies of electric vehicles.The main functions of BMS include the ability to accurately estimate the state of battery charge,the ability to monitor the Real-time state of the battery,and the ability to help the driver to get a better driving experience,by fixing their driving behaviors.In the process of driving electric vehicles,with the cause of real-time traffic,user'sdriving behavior,objective cell aging condition,the accuracy of electric vehicle battery management system will be lower and lower.In order to improve the overall performance of the battery,extend the service life of the battery,a new battery management system are in urgent need.It is necessary to establish a long-term use of the precise battery management system and these objective factors into consideration,through the analysis of a large number of historical data.In this paper,According to the cloud server,and vehicle intelligent terminal,a new battery management system of electric vehicle based on cloud services is set up,realizing the data interaction between the vehicle,the cloud,and the users.In addition to the basic functions of battery management system,we made a deeper research on SoC estimation function.At first,an improved Thevenin model for lithium-ion battery is established.After the establishment of battery equivalent circuit model,parameters of the model are identified through linear least square method,and thenbattery's SoCis predicted by using the open-circuit-voltage method.In this paper,battery's SoC-OCV curve is verified through a battery of dynamic characteristic test,and then the OCV is estimated by Kalman filtering method,and then the battery's SoC is calculatedaccording to the battery'sSoC-OCV curve.Meanwhile,in this paper,BMS also build some models,and do some data analysis in the light of vehicle information,driving behavior analysis model and traffic information analysis model is set up.Through this two models travel informations are transferred to be moreintuitive.BP neural network is used to predict the future current curvein the process ofdriving.Thereby through the Ampere-Hour integral method,the future SOC to be consumed will be calculated.And real-time correction will fix the error strongly,through the closed loop system.Finally,as the experimental results shown,the method can be more effective on the remaining SoC estimates,also canmodify the SoC of the arrival destination in real time.