| Electric vehicle is powered by electricity and has very good clean and environmental protection benefits.While the proliferation of electric vehicles may improve the sustainability of transportation energy sources,it also poses some new challenges: On the one hand,the expansion of the scale of electric vehicles will cause a huge gap in charging facilities.On the other hand,random access of large-scale electric vehicles to the power grid will have a negative impact on the security and economy of the power system.Aiming at the above two problems,in order to realize the integrated development of electric vehicles and smart grid,this dissertation focused on electric vehicles and conducted an in-depth study on optimal location of charging infrastructures,orderly charging strategy and vehicle to grid(V2G)strategy based on the electric vehicles driving data collected by the self-built intelligent integrated operation management system platform of the project.The specific work is as follows:In order to describe the charging characteristics of electric vehicle users more accurately,this dissertation analyzed the travel and charging behavior characteristics of users through electric vehicles driving data,and revealed the distribution rule of electric vehicles charging load in time and space.The corresponding prediction models of three different charging load prediction methods,namely,historical average load prediction method,fleet characteristic prediction method and personal characteristic prediction method,were established.And the prediction effects of the three methods were compared through a section of driving data.The results show that the root-mean-square error of the personal characteristic prediction method is 2.5511 k W,which is far lower than other two methods and has the best effect.The prediction of charging load layed a foundation for optimal location of charging infrastructures and charging/discharging strategy of electric vehicles.In order to select the location of charging infrastructures more objectively and practically,this dissertation proposed an optimal location method of charging infrastructures based on the real-world driving data of electric vehicles.On the basis of analyzing the spatial distribution of electric vehicles charging demand,the selection method of charging infrastructure candidate locations was determined and the influence of data size was studied.The data-driven optimization model(DDOM)of charging infrastructure locations was established and solved by genetic algorithm.The optimized charging points not only reduced the over-discharge rate of electric vehicles from 68.1%to 39.6%,but also increased the daily utilization time of charging facilities by 63%.In addition,the relationship among over discharge rate and the number of charging points,the budget cost of equipment and the rated range were discussed,and the connection of the number increase of charging points and retention rate were analyzed.At last,the feasibility and effectiveness of this method were verified by the real-world data of charging stations.In order to reduce the negative impacts of large-scale electric vehicles charging,for grid to vehicle(G2V)one-way power flow mode,this dissertation proposed a centralized data-driven orderly charging strategy considering the user’s charging selection.The charging method considering the user’s charging choice was introduced,and the charging scheduling model of electric vehicle in time dimension was established.By comparing the performances of three charging strategies(data-driven charging strategy,model-driven charging strategy and random charging strategy)with the real-world driving data of electric vehicles,the results show that the data-driven orderly charging strategy reduces the peak-valley difference and equivalent load fluctuation of power grid load by 22.2% and 22.7% respectively,and greatly reduces the charging cost of users,which is better than the other two charging strategies.In addition,the influence of the user’s charging choice was analyzed,and the orderly charging strategy considering the user’s charging choice can effectively reduce the scheduling deviation caused by the user’s charging mode choice.For V2 G two-way power flow mode,this dissertation proposed a V2 G strategy driven by the real-world driving data of electric vehicles.The types of electric power service suitable for V2 G were discussed,the advantages of V2 G participating in peak shaving service were defined,and calculation methods of charging and discharging priority were determined respectively.In order to calculate the overall benefit of electric vehicle users in V2 G service more accurately,the loss cost of battery was analyzed in detail.On this basis,a user-vehicle-grid V2 G scheduling model was established,and the real-world driving data of electric vehicles were used for simulation analysis.The results show that this V2 G strategy can not only reduce the peak-valley difference and equivalent load fluctuation of power grid load by 32.2% and39.2% respectively,but also reduce the electricity cost for electric vehicle users,which verifies the feasibility and effectiveness of this V2 G strategy.The research results of this dissertation can provide theoretical guidance for the effectiveness and rationality of charging infrastructures construction,have important reference value for charging facility planning and future V2 G commercial operation mode selection,and have important significance for promoting the popularization of electric vehicles. |
PDF Full Text Request