Research On EVs Charging Facilities Deployment And Charging Scheduling Strategy Based On Time And Space Characteristics

The implementation of electric vehicle(EV)industrial policy and advancements in battery technology have been instrumental in promoting energy conservation and emissions reduction in low-carbon societies.However,with the rapid growth of EV ownership,large-scale charging activities will have negative impacts on the grid and EV users,such as exacerbating load peak-valley differences and prolonging queuing time.To address these challenges and ensure the safe operation of the power grid while enhancing user charging satisfaction,it is crucial to strategically plan charging facilities and develop effective charging scheduling strategies.Thus,the thesis conducts research on the key issues involved in the deployment of charging facilities,and charging scheduling management.It starts with how to predict the spatial and temporal distribution of charging demand,and studies the layout of charging facilities and their capacity.On this basis,the thesis further studies the joint dispatch management of EVs.The main contribution of the thesis is as follows:(1)Aiming at the problem of uncertainty in the distribution of charging demand in time and space,a charging demand prediction method driven by behavioral characteristic data is proposed in the thesis.Firstly,the method analyzes the travel characteristics of EVs,and constructs the daily travel chain of EVs,including the destination chain,departure time chain,and arrival time chain.Secondly,the thesis analyzes the charging characteristics of EV users,including charging frequency,charging location and charging mode.By combining travel and charging characteristics,a comprehensive charging demand forecasting process is established.Thirdly,the thesis conducts simulation analysis and discussion on EV charging demand in various regions,time periods,and spaces based on real-world scenarios.The simulation results demonstrate that the spatial and temporal distribution of charging demand correlates with users’ daily commuting patterns and the density of workplaces and residences.Regarding time,there is a certain delay compared to the commuting period.Regarding space,higher building densities result in greater charging demand.The spatial and temporal distribution prediction of charging demand provides a practical basis for researching the placement and capacity of charging facilities,as well as managing charging scheduling.(2)To address the crucial challenge,the deployment of charging facilities in the EV ecosystem,an elastic charging demand-oriented charging facility planning method is proposed in the thesis.Firstly,according to the obtained travel chain data,the thesis screens out the places where the residence time exceeds the threshold by setting the residence time threshold.The process identifies charging demand areas and potential charging facility candidate points through clustering techniques.Secondly,in order to plan the deployment of charging facilities,the thesis establishes an elastic demand model for EVs constrained by driving distance and waiting time.In this model,each demand area generates demand components belonging to each candidate point based on the attractiveness of different candidate points.Thirdly,with the goal of maximizing the operator’s revenue,the thesis establishes the optimal deployment model of charging facilities and transforms it into a nonlinear integer programming problem.Finally,the thesis employs a quantum genetic algorithm(QGA)to solve the location layout of charging facilities and the number of charging piles.According to the simulation results,for an area with a total charging demand of 5000,the distance between charging stations is set to 2km,and the maximum capacity of charging piles is set to 20-25,the operator can obtain a profit of 3.5×107(?) and the proportion of users who abandon charging is relatively low,approximately 13.5%.The simulation results show that the matching of charging facilities and their capacity deployment with charging demand is very important in increasing charging station revenue and reducing the proportion of users abandoning charging.(3)Limited by the randomness and unpredictability of EV user behavior,a complete dynamic joint charging scheduling strategy is proposed in the thesis.The strategy comprises two key components:the real-time selection of charging stations(CSs)and the charging demand decision of EVs.Firstly,to maintain load balance among CSs,a CS’s matching control mechanism is designed in the thesis,which takes into account both load balance and EV driving distance constraints.Secondly,in order to maximize the charging utility of EVs,the thesis constructs a game framework,where the charging decision of EV users is formulated as a game problem with charging price constraints.Simulation results demonstrate the effectiveness of the proposed strategy.For a scenario with 500 EVs,the DRCS strategy introduced in this thesis outperforms the NNCS and CDNSA strategies.Specifically,the DRCS strategy increases user charging utility by approximately 40%and 17%compared to NNCS and CDNSA respectively.It also reduces waiting time by about 31%and 18%,respectively.Moreover,it improves charging operator revenue by around 14%and 7%,respectively.Furthermore,the resource utilization increases by about 22%and 10%,respectively.The load balancing factor is reduced to 0.17.