Research On Online Velocity Optimization To Reduce Smart Electric Vehicles Energy Consumption Under Slope Conditions

Automotive manufacturers have launched electric vehicles to meet the increasingly strict emission or energy consumption standards for automotive industry and the approaching time of the ban on internal combustion engines.Limited driving range is still one drawback of electric vehicles,so extending their driving range is important to the development of the electric vehicle industry.Meanwhile,smart vehicles are being popular,and the vehicles are expected to drive autonomously without the driver's operation in the future.In these two aspects,the energy consumption and driving range of electric vehicles are closely related its driving speed which should be autonomously decided by autonomous vehicles.This research belongs to an enterprise-university cooperation program between BMW-Brilliance Automotive and Dalian University of Technology.The research topic is proposed by BMW according to the latest research situation in automotive industry:to realize that a smart electric vehicle with autonomous driving function can independently decide the optimal driving speed during downhill driving to extend its driving range.During our research,it was found that the use case can be extended to all slope road driving situation,and this paper focuses on the research of multiple dynamic programming algorithms to optimize the driving speed for smart electric vehicles to improve their energy recovery and reduce their driving energy consumption,which could reduce the user's cost and pollution.The main work of this paper includes:(1)An optimum trategy is proposed to achieve efficient conversion among potential energy,kinetic energy and electric energy in smart electric vehicles by optimizing vehicle driving speed.Therefore,during downhill driving,the reduced potential energy of a smart electric vehicle can be converted into electric energy and stored in the battery to the maximum extent,or the smart electric vehicle energy consumption can be minimized to extend its driving range.The reasons why energy transfer efficiency is improved are also studied.(2)For the problem that discrete-distance DP(dynamic programming)under fixed boundary conditions cannot calculate the optimum velocity in real time,divided mesh discrete-distance DP are proposed to reduce discrete-distance DP calculation time while achieving the same results,and real-time DP optimum velocity calculation is realized.Based on variable boundary conditions,discrete-time bivariate DP is proposed,which allows time-dependent inputs.Therefore,discrete-time bivariate DP is able to calculate the optimum velocity under complex traffic conditions.(3)To solve the problem that discrete-time bivariate DP cannot calculate the optimum velocity in real time,a vector data net solver is proposed to improve discrete-time bivariate DP calculation efficiency without losing accuracy.Therefore rolling optimization is introduced and discrete-time bivariate DP online real-time calculation based on instant traffic events which affect the optimum velocity profiles can be realized.(4)As the most navigation maps do not contain slope information that supports the optimum velocity calculation,a method that combines data fitting and data mining to set up a road slope database is proposed.In this method,data fitting is applied to restore the accurate 3D driving trail for individual vehicle.Next,the 3D driving trajectories from all online vehicles are uploaded to a server to create a road surface point cloud database,and K-means is used as the data mining strategy to find the most representative points as the road surface coordinates and corresponding slope values to set up the road slope database,which supports the DP optimum velocity calculation methods.(5)Simulation results are verified by real-car test.Test results indicate that the optimal vehicle speed calculation strategies proposed in this paper can calculate the optimal vehicle speed in real time,to achieve efficient conversion among potential energy,kinetic energy and electric energy of smart electric vehicles with autonomous driving functions to reduce their energy consumption with better riding comfort.