Day-before-day multi-time scale electric vehicle cluster division strategy based on different charging completion times of electric vehicles

Abstract

The invention discloses a day-ahead-day multi-time scale electric vehicle cluster division strategy based on different charging completion times of electric vehicles. According to the method, modeling is carried out through historical travel statistical data, and the moment when the electric vehicles are connected to the network and the moment when the electric vehicles are disconnected from the network are determined; aiming at each scheduling period, considering constraints of a power grid, a vehicle owner, an electric vehicle battery and the like, solving a charging and discharging plan by two layers of strategies, and determining the total charging and discharging power of a cluster in a current time period by taking the minimum load change curve of a power distribution network in a scheduling interval as a target in the upper strategy; the lower-layer strategy takes the lowest cost of an electric vehicle owner as a target, finds a charging and discharging plan of a single electric vehicle, and introduces a scheduling penalty factor to adjust the difference with a cluster charging and discharging plan; according to the method, the load fluctuation can be effectively smoothed, the optimal real-time scheduling difficulty of the large-scale electric vehicle is remarkably reduced, and a better optimization effect and a higher solving speed are achieved for processing the access problem of the large-scale electric vehicle.

Description

Technical Field

[0001] The invention relates to the technical field of power grid dispatching, and in particular to a day-ahead and intra-day multi-time scale electric vehicle cluster division strategy based on different charging completion times of electric vehicles. Background Art

[0002] As a new type of controllable load, electric vehicles can be used as mobile energy storage devices after being connected to the grid. They can not only obtain electricity from the grid, but also provide energy to the grid through price incentive mechanisms to realize the Internet of Vehicles. However, with the increase in the number of electric vehicles, the disorderly charging of a large number of electric vehicles will have an impact on the grid, resulting in an increase in grid load and voltage fluctuations, affecting the normal operation of the grid. Proper guidance of the charging of large electric vehicles after they are connected to the grid can not only reduce the negative impact of random charging, solve the problems of aggravated peak-valley differences, reduced reliability and economy of distribution networks; it can also provide auxiliary services for grid optimization, achieve peak-to-valley shifting, stabilize grid load, reduce network losses, and bring certain economic benefits to the grid and car owners.

[0003] In the complex field of electric vehicle (EV) scheduling, there have been many relevant studies worth learning from, especially in the context of vehicle-to-grid (V2G) response. Through in-depth research on the orderly charging and discharging strategy of EVs through day-ahead optimal scheduling and real-time optimal scheduling, the driving parameters of EVs are carefully modeled and analyzed using vehicle historical data to predict the time distribution of charging and discharging demand for the next day. However, the shortcomings are also obvious: (1) Most of them subjectively define the optimal scheduling time of EV clusters. Blindly formulating scheduling strategies may have a negative impact on the power grid and increase power supply pressure; (2) EVs are different from traditional loads. Their access to the power grid is highly random. It is not feasible to use a single time scale to formulate scheduling plans, and the globality and accuracy of the scheduling process cannot be considered.

[0004] Therefore, it is necessary to design a multi-time scale electric vehicle cluster division strategy based on the day-ahead and intra-day charging completion times of electric vehicles to solve the above problems. Summary of the invention

[0005] The technical problem to be solved by the present invention is to provide a multi-time scale electric vehicle cluster division strategy based on the day-ahead and intraday different charging completion times of electric vehicles. The method constructs a charging model for the day-ahead stage based on historical data, and considers the charging willingness and travel plans of electric vehicles in the intraday stage in real time, so as to more accurately predict and formulate charging plans for electric vehicles.

[0006] In order to solve the above technical problems, the technical solution adopted by the present invention is: A multi-time scale electric vehicle cluster division strategy based on the day-ahead and intra-day charging completion times of electric vehicles includes: S1, detect the status of each charging device through the electric vehicle group management system EVGMS; S2, when the electric vehicle is connected to the charging facility, the EVGMS system establishes a day-ahead and intraday real-time multi-scale control strategy based on the day-ahead load model; S2, the day-ahead-intraday real-time multi-scale control strategy is divided into upper and lower strategies: the upper strategy solicits the charging and discharging power of the electric vehicle cluster, with the goal of minimizing the load change curve of the distribution network within the dispatching interval; the lower strategy optimizes the charging and discharging plan of a single electric vehicle, with the goal of minimizing the cost of the electric vehicle owner; S4, the EVGMS system updates the daily load situation and provides data for the next stage of scheduling.

[0007] Preferably, in step S2, the day-ahead load model includes a probability density function of the time when the electric vehicle returns home or leaves home, and the specific method is: Based on the NHTS database, the travel characteristics of electric vehicles are replaced with the travel characteristics of traditional energy-saving vehicles, and the statistical data are normalized to obtain the probability density function of the time when electric vehicles return home or leave home: ; ; In the formula, x represents any time in the cycle; and are the expected value and variance of the expected time for the i-th electric vehicle to be connected to the grid, respectively; and are the expected value and variance of the expected time for the i-th electric vehicle to be disconnected from the grid; represents the probability density function of the electric car returning home, represents the probability density function of an electric car leaving home.

[0008] Preferably, in step S2, before the electric vehicle is connected to the grid, the vehicle submits a travel plan and a charging and discharging schedule to the EVGMS, including whether to participate in V2G scheduling, off-grid time, and and off-grid expected charging status ; EVGMS obtains the grid connection time of electric vehicles from the battery management system BMS , State of Charge and battery capacity , EVGMS, as the center of information exchange and command transmission, reports the collected feedback to the regional distribution network and power grid company; the regional distribution dispatching department issues dispatching commands and formulates the charging and discharging plan of the entire cluster according to the current load demand of the power grid and the information reported by the EVGMS system; the V2G responsiveness k of electric vehicles is related to their grid connection / off-grid time and the time from maximum power Shortest charging time to full charge Related: ; ; In the formula, represents the parking time of electric vehicles; when k>1, the electric vehicle has V2G responsiveness, and the size of the k value is positively correlated with the flexibility and dispatchability of the electric vehicle.

[0009] Preferably, in step S3, the upper layer strategy minimizes the time from the start of unified scheduling. After unified scheduling The variance of the total load curve of the distribution network is obtained to obtain the total charging and discharging power of cluster j in time period t: ; ; In the formula, is the total load variance of the distribution network; J is the total number of clusters is the cluster j at time t The charge and discharge power on It is the basic load of the distribution network excluding the electric vehicle load; is the average load of the distribution network during the total period; The node power balance constraint of the distribution network is: ; ; Where K is the total number of distribution network nodes; and is the active power and reactive power of node m; and They are the active power and reactive power of electric vehicle charging and discharging respectively; and Remove base loads other than electric vehicles from the distribution network; and is the voltage amplitude; and represents the real and imaginary parts of the node conduction matrix.

[0010] The SOC constraints of the cluster are: ; ; ; In the formula, Indicates the battery capacity of an electric vehicle; is the cluster charge / discharge efficiency; is the total charging / discharging demand of cluster j during the optimal scheduling period.

[0011] Preferably, in step S3, the lower-level strategy formulates a charging and discharging strategy for a single electric vehicle with the goal of minimizing the cost of the owner of the i-th electric vehicle in group J under the premise of satisfying the needs of the vehicle owner: ; ; ; ; In the formula, V2G response costs for EV owners, including EV charging costs , Battery loss cost and discharge gain ; and Respectively i Electric vehicles in the time period t The charging and discharging power inside; I is the number of electric vehicles in the cluster; for t The electricity price at the time, is the battery loss coefficient, is the scheduling deviation penalty factor; The SOC constraint of an electric vehicle is: ; ; In the formula, and They represent the minimum and maximum SOC states of the battery of the i-th electric vehicle, and They represent the charging efficiency and discharging efficiency of the cluster respectively. The beneficial effects of the present invention are as follows: 1. When formulating the orderly charging and discharging strategy, the present invention introduces the idea of ​​real-time optimization into cluster scheduling based on the actual constraints and different characteristics of a single electric vehicle, comprehensively considers the constraints such as the power grid, the battery status of the electric vehicle and the charging and discharging needs of the owner, and establishes a real-time response cluster scheduling model for orderly charging and discharging of electric vehicles in the V2G mode.

[0012] 2. Under the premise of ensuring the safe operation of the distribution network and the economic benefits of the car owners, the present invention divides the day-ahead-day real-time multi-scale control strategy into two layers of strategies. The upper layer coordinates the charging and discharging power of the electric vehicle cluster in each period from the perspective of the electric vehicle cluster, so that the variance of the distribution network load curve in the scheduling time period is minimized; the lower layer solves the economic cost problem of each electric vehicle owner in the cluster from the perspective of the optimal allocation of the charging and discharging power of a single electric vehicle. During the entire optimization process, each layer uses the optimization results of the previous layer as known conditions, and promotes information exchange between the two layers through multiple iterations; through the collaborative work between the upper and lower strategies, the effective management of the electric vehicle cluster and the load scheduling of the power grid are realized, the charging cost is reduced, and more flexibility and choices are provided for electric vehicle users and the power grid side. BRIEF DESCRIPTION OF THE DRAWINGS

[0013] Figure 1 This is a flow chart of the electric vehicle V2G real-time response scheduling of the present invention; Figure 2 This is a structural diagram of electric vehicle grouping rules in an embodiment of the present invention; Figure 3 This is the IEEE33-node distribution network diagram in the second embodiment of the present invention. DETAILED DESCRIPTION

[0014] Embodiment 1: like Figure 1 As shown in FIG. 1 , a multi-time scale electric vehicle cluster division strategy based on the day-ahead and intra-day charging completion times of electric vehicles includes: S1, detect the status of each charging device through the electric vehicle group management system EVGMS; S2, when the electric vehicle is connected to the charging facility, the EVGMS system establishes a day-ahead and intraday real-time multi-scale control strategy based on the day-ahead load model; S2, the day-ahead-intraday real-time multi-scale control strategy is divided into upper and lower strategies: the upper strategy solicits the charging and discharging power of the electric vehicle cluster, with the goal of minimizing the load change curve of the distribution network within the dispatching interval; the lower strategy optimizes the charging and discharging plan of a single electric vehicle, with the goal of minimizing the cost of the electric vehicle owner; S4, the EVGMS system updates the daily load situation and provides data for the next stage of scheduling.

[0015] Preferably, in step S2, the day-ahead load model includes a probability density function of the time when the electric vehicle returns home or leaves home, and the specific method is: Based on the NHTS database, the travel characteristics of electric vehicles are replaced with the travel characteristics of traditional energy-saving vehicles, and the statistical data are normalized to obtain the probability density function of the time when electric vehicles return home or leave home: ; ; In the formula, x represents any time in the cycle; and are the expected value and variance of the expected time for the i-th electric vehicle to be connected to the grid, respectively; and are the expected value and variance of the expected time for the i-th electric vehicle to be disconnected from the grid; represents the probability density function of the electric car returning home, represents the probability density function of an electric car leaving home.

[0016] Furthermore, the calculation in this embodiment is based on the results of the National Household Travel Survey (NHTS) of the Ministry of Transportation in 2022 and 2023.

[0017] Preferably, in step S2, before the electric vehicle is connected to the grid, the vehicle submits a travel plan and a charging and discharging schedule to the EVGMS, including whether to participate in V2G scheduling, off-grid time, and and off-grid expected charging status ; EVGMS obtains the grid connection time of electric vehicles from the battery management system BMS , State of Charge and battery capacity , EVGMS, as the center of information exchange and command transmission, reports the collected feedback to the regional distribution network and power grid company; the regional distribution dispatching department issues dispatching commands and formulates the charging and discharging plan of the entire cluster according to the current load demand of the power grid and the information reported by the EVGMS system; the V2G responsiveness k of electric vehicles is related to their grid connection / off-grid time and the time from maximum power Shortest charging time to full charge Related: ; ; In the formula, represents the parking time of electric vehicles; when k>1, electric vehicles have V2G responsiveness, and the size of the k value is positively correlated with the flexibility and dispatchability of electric vehicles.

[0018] Preferably, in step S3, the upper layer strategy minimizes the time from the start of unified scheduling. After unified scheduling The variance of the total load curve of the distribution network is obtained to obtain the total charging and discharging power of cluster j in time period t: ; ; In the formula, is the total load variance of the distribution network; J is the total number of clusters is the cluster j at time t The charge and discharge power on It is the basic load of the distribution network excluding the electric vehicle load; is the average load of the distribution network during the total period; The node power balance constraint of the distribution network is: ; ; Where K is the total number of distribution network nodes; and is the active power and reactive power of node m; and They are the active power and reactive power of electric vehicle charging and discharging respectively; and Remove base loads other than electric vehicles from the distribution network; and is the voltage amplitude; and represents the real and imaginary parts of the node conduction matrix.

[0019] The SOC constraints of the cluster are: ; ; ; In the formula, Indicates the battery capacity of an electric vehicle; is the cluster charge / discharge efficiency; is the total charging / discharging demand of cluster j during the optimal scheduling period.

[0020] Preferably, in step S3, the lower-level strategy formulates a charging and discharging strategy for a single electric vehicle with the goal of minimizing the cost of the owner of the i-th electric vehicle in group J under the premise of satisfying the needs of the vehicle owner: ; ; ; ; In the formula, V2G response costs for EV owners, including EV charging costs , Battery loss cost and discharge gain ; and Respectively i Electric vehicles in the time period t The charging and discharging power inside; I is the number of electric vehicles in the cluster; for t The electricity price at the time, is the battery loss coefficient, is the scheduling deviation penalty factor; The SOC constraint of an electric vehicle is: ; ; In the formula, and They represent the minimum and maximum SOC states of the battery of the i-th electric vehicle, and They represent the charging efficiency and discharging efficiency of the cluster respectively.

[0021] Embodiment 2: In this embodiment, a structure diagram of the electric vehicle grouping rules described in this strategy is proposed; Figure 2 In the structure diagram shown, the EVGMS system first collects the status information of electric vehicles connected to charging, including battery capacity, battery SOC status, grid-connected and grid-off time; then it screens whether the electric vehicle owners choose V2G and calculates the responsiveness of V2G vehicles; finally, it formulates a detailed cluster division strategy for electric vehicles that meet the requirements.

[0022] Figure 3 The IEEE 33 node distribution network structure diagram used in this strategy is given. Figure 3 It can be seen that the entire distribution network structure diagram is divided into three areas.

Claims

1. A multi-time scale electric vehicle cluster division strategy based on the day-ahead and day-intraday charging completion times of electric vehicles, characterized in that: include: S1, detect the status of each charging device through the electric vehicle group management system EVGMS; S2, when the electric vehicle is connected to the charging facility, the EVGMS system establishes a day-ahead and intraday real-time multi-scale control strategy based on the day-ahead load model; S2, the day-ahead-intraday real-time multi-scale control strategy is divided into upper and lower strategies: the upper strategy solicits the charging and discharging power of the electric vehicle cluster, with the goal of minimizing the load change curve of the distribution network within the dispatching interval; the lower strategy optimizes the charging and discharging plan of a single electric vehicle, with the goal of minimizing the cost of the electric vehicle owner; S4, the EVGMS system updates the daily load situation and provides data for the next stage of scheduling.

2. The multi-time scale electric vehicle cluster division strategy based on the day-ahead and day-intraday electric vehicle charging completion time according to claim 1 is characterized in that: In step S2, the day-ahead load model includes the probability density function of the time when the electric vehicle returns home or leaves home, and the specific method is as follows: Based on the NHTS database, the travel characteristics of electric vehicles are replaced with the travel characteristics of traditional energy-saving vehicles, and the statistical data are normalized to obtain the probability density function of the time when electric vehicles return home or leave home: ； ； In the formula, Indicates any moment in the cycle; and are the expected value and variance of the expected time for the i-th electric vehicle to be connected to the grid, respectively; and are the expected value and variance of the expected time for the i-th electric vehicle to be disconnected from the grid; represents the probability density function of the electric car returning home, represents the probability density function of electric vehicles leaving home; exp represents the normalized analysis of the data.

3. The multi-time scale electric vehicle cluster division strategy based on the day-ahead and day-intraday electric vehicle charging completion time according to claim 2 is characterized in that: In step S2, before the electric vehicle is connected to the grid, the vehicle submits a travel plan and a charging and discharging schedule to EVGMS, including whether to participate in V2G scheduling and the time to leave the grid. and off-grid expected charging status ; EVGMS obtains the grid connection time of electric vehicles from the battery management system BMS , State of Charge and battery capacity , EVGMS, as the center of information exchange and command transmission, reports the collected feedback to the regional distribution network and power grid company; the regional distribution dispatching department issues dispatching commands and formulates the charging and discharging plan of the entire cluster according to the current load demand of the power grid and the information reported by the EVGMS system; the V2G responsiveness k of electric vehicles is related to their grid connection / off-grid time and the time from maximum power Shortest charging time to full charge Related: ； ； In the formula, represents the parking time of electric vehicles; when k > 1, the electric vehicle has V2G responsiveness, and the value of k is positively correlated with the flexibility and dispatchability of the electric vehicle.

4. The multi-time scale electric vehicle cluster division strategy based on the day-ahead and day-intraday electric vehicle charging completion time according to claim 3 is characterized in that: In step S3, the upper layer strategy minimizes the time from the beginning of unified scheduling. After unified scheduling The variance of the total load curve of the distribution network is obtained to obtain the total charging and discharging power of cluster j in time period t: ； ； In the formula, is the total load variance of the distribution network; J is the total number of clusters is the cluster j at time t The charge and discharge power on It is the basic load of the distribution network excluding the electric vehicle load; is the average load of the distribution network during the total period; The node power balance constraint of the distribution network is: ； ； Where K is the total number of distribution network nodes; and is the active power and reactive power of node m; and They are the active power and reactive power of electric vehicle charging and discharging respectively; and Remove base loads other than electric vehicles from the distribution network; and is the voltage amplitude; and represents the real and imaginary parts of the node conduction matrix; The SOC constraints of the cluster are: ； ； ； In the formula, Indicates the battery capacity of an electric vehicle; is the cluster charge / discharge efficiency; is the total charging / discharging demand of cluster j during the optimal scheduling period.

5. The multi-time scale electric vehicle cluster division strategy based on the day-ahead and day-intraday electric vehicle charging completion time according to claim 4 is characterized in that: In step S3, the lower-level strategy formulates a charging and discharging strategy for a single electric vehicle with the goal of minimizing the cost of the owner of the i-th electric vehicle in group J, while satisfying the needs of the owners: ； ； ； ； In the formula, V2G response costs for EV owners, including EV charging costs , Battery loss cost and discharge gain ; and Respectively i Electric vehicles in the time period t Internal charging and discharging power; I is the number of electric vehicles in the cluster; for t The electricity price at the time, is the battery loss coefficient, is the scheduling deviation penalty factor; The SOC constraint of an electric vehicle is: ； ； In the formula, and They represent the minimum and maximum SOC states of the battery of the i-th electric vehicle, and They represent the charging efficiency and discharging efficiency of the cluster respectively.

Images