Electric Vehicle Charging And Discharging Bi-level Scheduling Strategies Considering User’s Future Mobility

In the process of achieving the peak of carbon dioxide emissions and carbon neutrality in China,accelerating the formation of green and low-carbon transport mode has placed a demand on the development of transportation electrification.While the popularity of electric vehicles(EVs)saves energy and reduces pollutant emissions,the randomness of their charging behavior puts enormous pressure on grid operation.With the development of Vehicle-to-Grid(V2G)technology,the mobile energy storage characteristics of EVs provides a way to improve the power grid operation.The charging demands of EV users during working and entertainment hours overlaps with the original load peaks of the power grid,tends to cause "peak on peak" phenomenon of the power grid load.Therefore,this paper conducts the research from two aspects: modeling the future mobility of users and scheduling the charging and discharging of EVs.The main research contents are as follows:1)Modeling users’ future mobility.Based on the 2017 National Household Travel Survey(NHTS)published by the US Department of Transportation and the State of Charge(SOC)data of EV charging start and end time in a local distribution network area,this paper analyzes users’ travel behaviors and charging behaviors,further introduces road network speed data on the basis of a static road network model,and establishes dynamic traffic model with dynamic changes in road section speed.The paper realizes the simulation of users’ future mobility based on dynamic traffic constraints,and further proposes a road network space-time travel chain based on space-time trajectory space to finely model the changes in time,space and EV speed during users’ future mobility.2)Energy consumption calculation and path recommendation for users’ future mobility.Based on the driving data of a hybrid vehicle in pure electric mode,this paper conducts correlation analysis and principal component analysis to determine the driving conditions characteristics considering the influence on energy consumption.In order to solve the problem that it is difficult to calculate the air condition operating power for the users’ future mobility,establish a probabilistic model to determine the users’ air condition operating power under different ambient temperatures.Further,considering the influence of driving conditions on energy consumption,establish a neural network energy consumption prediction model to calculate the energy consumption of feasible paths in the user’s road network space-time travel chain.Finally,establish a path evaluation system considering the users’ path preference to recommend the optimal path for the user,and the feasibility of the proposed scheme is verified through examples.3)EV charging and discharging bi-level scheduling strategy considering users’ future mobility.This paper analyses the impact of battery depth of discharge and starting SOC of discharging on battery life and establishes an economic model of battery loss based on cycle range.This paper further establishes a bi-level EV charging and discharging scheduling model,where the upper layer model develops the scheduling plan for each time period with the objective of minimizing the peak-to-valley difference of the power grid and minimizing the scheduling deviation between the upper layer and lower layer,the lower layer model aims at tracking the upper level scheduling plan which considers the users’ charging demand and battery cycle range to develop charging and discharging strategies for EVs.Finally,this paper verifies the effectiveness of the proposed strategy by examples.