SOP locating and sizing planning method considering schedulable potential of electric vehicle
By evaluating the charging and discharging potential of electric vehicles and combining the active and reactive balance constraints of the distribution network, the SOP site selection and capacity setting are optimized, and the problem of insufficient flexibility of the distribution network is solved, and the efficient operation and cost reduction of the power grid is achieved. .
Patent Information
- Application Number
- CN202411123331.4
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-08-15
- Publication Date
- 2024-12-06
AI Technical Summary
The existing distribution network SOP site selection and capacity planning method provides insufficient flexibility on the time scale, resulting in limited improvement in the operation flexibility of the power grid, and the investment costs of energy storage systems and SOPs are high, which limits the supply and demand balance of the distribution network. ability.
A SOP site selection and capacity planning method that considers the dispatchable potential of electric vehicles is adopted. By detecting the distributed power supply and load power consumption data of new energy in the distribution network, combined with Monte Carlo sampling and simulated vehicle charging rules, the charging and discharging of electric vehicles is evaluated. Power and capacity changes, SOP operation constraints and active reactive balance constraints are built, planning costs are calculated and the node location and capacity of SOP are optimized, and the grid flexibility is improved.
It realizes the evaluation and optimization of the precise dispatchable potential of electric vehicles, improves the operating efficiency and reliability of the distribution network, reduces the investment costs of energy storage systems and SOPs, provides a more flexible and efficient grid planning method, and promotes Win-win development of power grid and electric vehicles.
Smart Images

Figure CN119093443A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of power distribution systems, in particular to the technical field of SOP site selection and capacity planning for distribution networks, and in particular to a SOP site selection and capacity planning method that takes into account the dispatchable potential of electric vehicles. Background Art
[0002] In response to the national carbon peak and carbon neutrality goals, a new power system with a gradually increasing proportion of new energy is built to promote the large-scale access of clean power resources. The strong uncertainty and volatility of DG lead to problems such as node voltage exceeding the limit and branch power overload in the distribution network, which poses new challenges to the operation of the distribution network. Electric vehicles are usually regarded as conventional loads that are plug-and-play, which wastes their considerable storage capacity. Studies have shown that the parking time of electric vehicles without fast charging requirements is much longer than their charging time. Therefore, the charging and discharging power of electric vehicles can be managed by charging stations to provide backup capacity for the power grid. At the same time, SOP, as a typical flexible interconnected power electronic device, can realize active power transmission and real-time control of reactive power. It can provide power flow distribution capabilities on a spatial scale, but the energy storage capacity is small and the flexibility adjustment capability on a time scale is limited. Although the energy storage system can effectively solve this problem, it is limited by its own high cost and is difficult to popularize on a large scale. Therefore, considering the dispatchable potential of electric vehicle clusters is of great significance for the site selection and capacity planning of SOP and improving the flexibility of the distribution system.
[0003] In existing research, when most scholars plan the site selection and capacity of the distribution network SOP, they usually plan it separately or coordinate it with energy storage devices. For example, the paper "Smart Soft Switch-Energy Storage Coordination Planning for Improving the Flexibility of New Distribution Systems" comprehensively considers the uncertainty of distributed power generation in different scenarios to coordinate the energy storage system and SOP. This method has the following disadvantages: First, the planning cost is high. Both the energy storage system and the energy storage devices in the SOP require additional investment costs, resulting in a high investment cost of equal annual value; second, the SOP planned in normal times does not improve the flexibility of the distribution network enough. Due to the limitation of energy storage investment costs, the improvement of supply and demand balance flexibility is limited; Summary of the invention
[0004] The purpose of the present invention is to solve the technical problem of insufficient flexibility in time scale when planning SOP in the prior art, and to provide a SOP site selection and sizing planning method that takes into account the dispatchable potential of electric vehicles.
[0005] In order to solve the above technical problems, the technical solution adopted by the present invention is:
[0006] A SOP site selection and sizing planning method considering the dispatchable potential of electric vehicles includes the following steps:
[0007] Step 1: Obtain data on power generation and load power consumption of new energy distributed power sources in the distribution network through detection technology;
[0008] Step 2: Simulate the charging rules of different types of vehicles through Monte Carlo sampling; evaluate the dispatchable potential of electric vehicles, and then calculate the charging and discharging power changes and the upper and lower limits of the capacity changes of the charging station based on the actual detection data obtained in Step 1;
[0009] Step 3, construct SOP operation constraints; then include the SOP operation constraints, the data measured in Step 1, and the upper and lower limits of the charging and discharging power changes and capacity changes of the charging station calculated in Step 2 into the active and reactive balance constraints of the distribution network nodes;
[0010] Step 4: Under the constraints of Step 2 and Step 3, calculate the planning cost and find the SOP planning scheme that minimizes the cost.
[0011] Step 5. Calculate branch load margin I BF and supply-demand balance flexibility index I FSD , and the operation flexibility of the distribution network is measured by calculating the difference between the two.
[0012] In the above Step 2, the charging and discharging power change and the upper and lower limits of the capacity change of the charging station are calculated; the specific steps are as follows:
[0013] Step 2.1. Use Monte Carlo sampling to simulate the charging patterns of different types of vehicles, including a number of private cars charging at night, a number of online ride-hailing vehicles charging at night, and a number of office workers charging during the day. Then assign them to charging stations and simulate the N-day charging and discharging history data of electric vehicles through N samplings. The N+1th sampling is the actual data.
[0014] Step 2.2: Aggregate electric vehicles to charging stations. By managing the charging and discharging of electric vehicles in the stations, electric vehicles can participate in the planning of distribution networks as flexible loads:
[0015]
[0016] Where: and are the total charging and discharging power of charging station j in period t respectively; and are the charging and discharging dispatching powers of electric vehicle n in period t respectively; is the set of electric vehicles in charging station j; T is the scheduling time set.
[0017] At the same time, according to Step 2.1, the historical parking data of electric vehicles is obtained, including the arrival time, departure time, battery power at arrival, battery power at departure, battery power boundary of electric vehicles and maximum charge and discharge power to evaluate the historical dispatchable potential of charging piles; among them, in order to realize the extension of the definition domain of individual charging stations, the state variable X is first introduced n,t Used to characterize the grid-connected status of electric vehicles, X n,t It can be directly calculated from the parking time of the electric vehicle:
[0018]
[0019] Where: X n,t represents the state of electric vehicle n at time t, X n,t =1 means that electric vehicle n is connected to the grid during period t, X n,t =0 means that electric vehicle n is off-grid during period t. represents the time period when electric vehicle n arrives at the charging station, Indicates the time period when electric vehicle n leaves the charging station.
[0020] Step 2.3: Based on the above two steps, electric vehicles are aggregated into charging stations and the state variable X is introduced. n,t Afterwards, the dispatch potential of the electric vehicle cluster is evaluated:
[0021]
[0022] Where: and They represent the maximum charging and discharging power of the generalized energy storage device at charging station j in period t; X n,t-1 represents the state of electric vehicle n at time t-1; and They represent the upper limits of charging and discharging power of electric vehicle n respectively; and represents the battery power safety boundary of electric vehicle n; and represents the power boundary of the generalized energy storage device at charging station j in period t; s n,arrival represents the initial charge of electric vehicle n; s n,leave represents the power grid of electric vehicle n when it leaves the charging station, also known as its expected power; △S j,t It represents the change in power of the generalized energy storage device at charging station j due to the change in the grid-connected state of electric vehicles in period t. φ is the dispatchable potential of the electric vehicle cluster in the dataset.
[0023] Step 2.4: The historical electric vehicle cluster data is compressed into historical dispatchable potential, which can be used as training data, and a dispatchable potential prediction model is further established based on back propagation (BP) neural network, time series and other methods.
[0024] Step 2.5, calculate the real-time dispatchable potential by time period, including the capacity change and power upper and lower limits of generalized energy storage. The charging station calculates the dispatchable potential based on the electric vehicles located in the charging station during the current time period. The calculation method is shown in formula (3). At the same time, as time goes by, the charging station continuously updates the dispatchable potential, but only executes a rolling time window plan during the dispatch process. Therefore, the real-time dispatchable potential is also called dynamic dispatchable potential.
[0025] In the above Step 3, the SOP operation constraints are first constructed; then the SOP operation constraints, the data measured in Step 1, and the upper and lower limits of the charging and discharging power changes and capacity changes of the charging station calculated in Step 2 are included in the active and reactive balance constraints of the distribution network nodes; the steps are as follows:
[0026] Step 3.1, construct SOP operation constraints. SOP will have certain losses during operation. It can also perform certain charging and discharging when necessary. As a soft switch, its power inflow and outflow should be balanced, with power injection as negative and outflow as positive. There are the following constraints:
[0027]
[0028]
[0029] Where: and are the active and reactive powers injected by SOP converters a and b at time t, respectively; is the charging / discharging power of the energy storage system in the SOP at time t; is the SOP transmission loss at time t; A SOP,a and A SOP,b are the loss coefficients of converter a and b respectively; and are the access capacities of converter a and converter b at t, which are usually considered equal and are recorded as
[0030] At the same time, when planning SOP, the upper limit of the number of planned SOP is 5, and the number of planned SOPs can be changed at will. If the number is different, the capacity and position of each SOP will also change accordingly; the unit optimization capacity of SOP is 10kV·A.
[0031] Step 3.2, include the SOP operation constraints, the data measured in Step 1, and the upper and lower limits of the charging and discharging power changes and capacity changes of the charging station calculated in Step 2 into the active and reactive balance constraints of the distribution network nodes:
[0032]
[0033] Where: P t MG and is the active and reactive power purchased from the upper main grid during period t; and are the active and reactive power generated by the nth DG in period t respectively; is the network loss of branch n in period t; and are respectively the active and reactive power consumed by the nth load in period t; Ω EV ,Ω DG ,Ω SOP and Ω b They are charging stations, distributed power sources, SOP installation nodes and all branches of the distribution network; is the active power detection value output by the nth DG at time t; P DG,n,t , Q DG,n,t are the active power and reactive power output by DG at node n during period t; β DG,n,t is the power factor angle of DG at node n during period t; is the net active power generated by the jth charging station at time t; and are the active and reactive power generated by the nth SOP at time t; β SOP,n,t is the power factor angle of the nth SOP at time t.
[0034] In the above Step 4, the planning cost is calculated under the constraints of Step 2 and Step 3, and the SOP planning scheme with the minimum cost is found; the steps are as follows:
[0035] Step 4.1: The planning cost is modeled with the minimum annual comprehensive operating cost of the distribution network as the objective function. The model is as follows:
[0036] f med =C SI +C OMS +C PC +C NLC (13)
[0037]
[0038] Where: F medis the annual comprehensive operating cost; C SI is the annual investment cost of SOP; C OMS is the annual operation and maintenance cost of SOP; C PC The cost of purchasing electricity from the upper grid; C NLC is the network loss cost; λ is the discount rate; Ω SOP Install node collection for SOP;y SOP is the useful life of the SOP; c s is the unit capacity investment cost of SOP; is the SOP capacity to be installed between nodes i and j; η SOP is the annual operation and maintenance cost of SOP; c g The unit price of electricity purchased from the upper power grid; c loss is the unit power network loss cost; P t MG is the active power injected by the upper main grid into the distribution network at time t; is the network loss power between nodes i and j at time t.
[0039] Step 4.2: Use the solver to solve the above model, and obtain the node position and capacity size of the SOP that minimizes the objective function value, which is the planning result.
[0040] The branch load margin I is calculated in Step 5 above. BF and supply-demand balance flexibility index I FSD The operational flexibility of the distribution network is measured by calculating the difference between the two. The specific steps are as follows:
[0041] Step 5.1. Calculate the branch load margin flexibility index and supply-demand balance flexibility index respectively:
[0042] The branch load margin flexibility index is used to evaluate the transmission capacity of network flexibility resources. It is mainly represented by the ratio of the current in the branch to the maximum current that the branch can withstand. The smaller the value, the greater the load margin of each branch, and the stronger the ability to cope with the uncertainty changes in flexibility demand:
[0043]
[0044] Where: I BF is the average load rate of the network branch, Ω B is the branch set of the distribution network; L zm,t is the load rate of branch z,m at time t; I zm,t is the current flowing through branch z,m at time t; I zm is the maximum value of the current allowed to flow through branch z,m.
[0045] The supply-demand balance flexibility index is used to measure whether the flexibility supply can meet the flexibility demand. The larger the value, the more the flexibility resources can meet the flexibility demand, and the better the overall flexibility of the distribution network:
[0046]
[0047]
[0048] Where: I FSD It is an indicator of supply and demand balance flexibility; UFSD ,I DFSD The sum of the ratios of the dispatchable margin of flexible resources to the change in net load when the net load increases or decreases within a dispatch cycle; are the upward and downward flexibility adjustment capabilities provided by the j-th charging pile to the distribution network at time t; They are the upward and downward flexibility adjustment capabilities provided by the upper main network at time t; are the total upward and downward flexibility adjustment capabilities of the distribution network at time t; T is the total number of time periods in the dispatching cycle. The dispatching cycle in this paper is 24 hours, and the dispatching time interval is 1 hour; wt is the state variable of the net load power change at time t; P t nl , are the net load power at time t and t+1 respectively.
[0049] Step 5.2: The two distribution network flexibility evaluation indicators calculated in the first two steps are coupled, influenced and interacted with each other. In the planning process, node flexibility and branch flexibility are combined to jointly determine the quality of system flexibility:
[0050] F I =I FSD -I BF (20)
[0051] Where: FI represents system flexibility. The larger its value is, the better the system flexibility is.
[0052] Compared with the prior art, the present invention has the following technical effects:
[0053] 1) Based on the aggregation of electric vehicles into energy storage devices, the present invention realizes the accurate evaluation and optimal utilization of the dispatchable potential of electric vehicles through advanced algorithms and models. From the charging state and battery capacity of electric vehicles to their response speed and dispatch strategy in the power grid, the present invention can comprehensively simulate and optimize the full physical process of electric vehicles as energy storage devices.
[0054] 2) In the present invention, the dispatching potential of electric vehicles is combined with the time-series variation characteristics of distributed generation output, and dynamic response and optimization of power grid operation are achieved through logic design. Regardless of how the distributed generation output changes, the present invention can ensure that the dispatching potential of electric vehicles is fully utilized through intelligent dispatching algorithms, while avoiding the risks of power grid operation caused by improper dispatching.
[0055] 3) In the distribution network operation constraint model, the present invention not only considers the dispatchable potential of electric vehicles, but also the operational safety and economy of the power grid. By establishing a comprehensive model, the present invention can effectively solve the matching problem between the dispatching potential of electric vehicles and the operation demand of the power grid, and realize the optimal configuration of electric vehicles in the site selection and capacity planning of the power supply business area.
[0056] 4) Compared with the traditional SOP site selection and capacity planning method, the present invention provides a more flexible and efficient power grid planning method by introducing the dispatchable potential of electric vehicles and the time-series variation characteristics of distributed power generation output. The present invention can not only improve the operating efficiency and reliability of the power grid, but also provide electric vehicle users with more opportunities to participate in power grid scheduling, achieving a win-win development of the power grid and electric vehicles. BRIEF DESCRIPTION OF THE DRAWINGS
[0057] Figure 1 The overall framework flow chart of the SOP site selection and sizing planning method considering the dispatchable potential of electric vehicles;
[0058] Figure 2 Improved wiring diagram for 33 nodes of the distribution network;
[0059] Figure 3 To obtain data waveforms of new energy distributed power sources and loads in distribution networks through detection technology;
[0060] Figure 4 The upper and lower bounds of SOC of the day-ahead and real-time are shown in Figure 3, taking charging station 3 as an example;
[0061] Figure 5 The upper and lower bounds of the day-ahead and real-time charging and discharging power are shown in Figure 3, taking charging station 3 as an example;
[0062] Figure 6 Node voltage diagram for traditional planning;
[0063] Figure 7 The node voltage diagram considering EV planning;
[0064] Figure 8 The real-time charging and discharging power diagram of charging station 1 (positive means charging, negative means discharging; the same below);
[0065] Fig. 9 Real-time charging and discharging power diagram for charging station 2;
[0066] Fig.10 Real-time charging and discharging power diagram of charging station 3;
[0067] Fig.11 A real-time charging and discharging power diagram for charging station 4; DETAILED DESCRIPTION
[0068] like Figure 1 As shown in , a SOP site selection and sizing planning method considering the dispatchable potential of electric vehicles includes the following steps:
[0069] Step 1: Obtain data on power generation and load power consumption of new energy distributed power sources in the distribution network through detection technology;
[0070] Step 2: Simulate the charging rules of different types of vehicles through Monte Carlo sampling; evaluate the dispatchable potential of electric vehicles, and then calculate the charging and discharging power changes and the upper and lower limits of the capacity changes of the charging station based on the actual detection data obtained in Step 1;
[0071] Step 3, construct SOP operation constraints; then include the SOP operation constraints, the data measured in Step 1, and the upper and lower limits of the charging and discharging power changes and capacity changes of the charging station calculated in Step 2 into the active and reactive balance constraints of the distribution network nodes;
[0072] Step 4: Under the constraints of Step 2 and Step 3, calculate the planning cost and find the SOP planning scheme that minimizes the cost.
[0073] Step 5. Calculate branch load margin I BF and supply-demand balance flexibility index I FSD , and the operation flexibility of the distribution network is measured by calculating the difference between the two.
[0074] In Step 2, the charging and discharging power change and the upper and lower limits of the capacity change of the charging station are calculated; the specific steps are as follows:
[0075] Step 2.1. Use Monte Carlo sampling to simulate the charging patterns of different types of vehicles, including 800 private cars charging at night, 400 online ride-hailing vehicles charging at night, and 800 office workers charging during the day. The vehicles are assigned to charging stations and sampled 1,000 times to simulate the 1,000-day charging and discharging history of electric vehicles. The 1,001th sampling is the actual data.
[0076] Step 2.2: Aggregate electric vehicles to charging stations. By managing the charging and discharging of electric vehicles in the stations, electric vehicles can participate in the planning of distribution networks as flexible loads:
[0077]
[0078] Where: and are the total charging and discharging power of charging station j in period t respectively; and are the charging and discharging dispatching powers of electric vehicle n in period t respectively; is the set of electric vehicles in charging station j; T is the scheduling time set.
[0079] At the same time, according to Step 2.1, the historical parking data of electric vehicles is obtained, including the arrival time, departure time, battery power at arrival, battery power at departure, battery power boundary of electric vehicles and maximum charge and discharge power to evaluate the historical dispatchable potential of charging piles; among them, in order to realize the extension of the definition domain of individual charging stations, the state variable X is first introduced n,t Used to characterize the grid-connected status of electric vehicles, X n,t It can be directly calculated from the parking time of the electric vehicle:
[0080]
[0081] Where: X n,t represents the state of electric vehicle n at time t, X n,t =1 means that electric vehicle n is connected to the grid during period t, X n,t =0 means that electric vehicle n is off-grid during period t. represents the time period when electric vehicle n arrives at the charging station, Indicates the time period when electric vehicle n leaves the charging station.
[0082] Step 2.3: Based on the above two steps, electric vehicles are aggregated into charging stations and the state variable X is introduced. n,t Afterwards, the dispatch potential of the electric vehicle cluster is evaluated:
[0083]
[0084] Where: and They represent the maximum charging and discharging power of the generalized energy storage device at charging station j in period t; X n,t-1 represents the state of electric vehicle n at time t-1; and They represent the upper limits of charging and discharging power of electric vehicle n respectively; and represents the battery power safety boundary of electric vehicle n; and represents the power boundary of the generalized energy storage device at charging station j in period t; s n,arrivalrepresents the initial charge of electric vehicle n; s n,leave represents the power grid of electric vehicle n when it leaves the charging station, also known as its expected power; △S j,t It represents the change in power of the generalized energy storage device at charging station j due to the change in the grid-connected state of electric vehicles in period t. φ is the dispatchable potential of the electric vehicle cluster in the dataset.
[0085] Step 2.4: The historical electric vehicle cluster data is compressed into historical dispatchable potential, which can be used as training data, and a dispatchable potential prediction model is further established based on back propagation (BP) neural network, time series and other methods.
[0086] Step 2.5, calculate the real-time dispatchable potential by time period, including the capacity change and power upper and lower limits of generalized energy storage. The charging station calculates the dispatchable potential based on the electric vehicles located in the charging station during the current time period. The calculation method is shown in formula (3). At the same time, as time goes by, the charging station continuously updates the dispatchable potential, but only executes a rolling time window plan during the dispatch process. Therefore, the real-time dispatchable potential is also called dynamic dispatchable potential.
[0087] In Step 3, the SOP operation constraints are first constructed; then the SOP operation constraints, the data measured in Step 1, and the upper and lower limits of the charging and discharging power changes and capacity changes of the charging station calculated in Step 2 are included in the active and reactive balance constraints of the distribution network nodes; the steps are as follows:
[0088] Step 3.1, construct SOP operation constraints. SOP will have certain losses during operation. It can also perform certain charging and discharging when necessary. As a soft switch, its power inflow and outflow should be balanced, with power injection as negative and outflow as positive. There are the following constraints:
[0089]
[0090] Where: and are the active and reactive powers injected by SOP converters a and b at time t, respectively; is the charging / discharging power of the energy storage system in the SOP at time t; is the SOP transmission loss at time t; A SOP,a and A SOP,b are the loss coefficients of converter a and b respectively; and are the access capacities of converter a and converter b at t, which are usually considered equal and are recorded as
[0091] At the same time, when planning SOP, the upper limit of the number of planned SOP is 5, and the number of planned SOPs can be changed at will. If the number is different, the capacity and position of each SOP will also change accordingly; the unit optimization capacity of SOP is 10kV·A.
[0092] Step 3.2, include the SOP operation constraints, the data measured in Step 1, and the upper and lower limits of the charging and discharging power changes and capacity changes of the charging station calculated in Step 2 into the active and reactive balance constraints of the distribution network nodes:
[0093]
[0094]
[0095] Where: P t MG and is the active and reactive power purchased from the upper main grid during period t; and are the active and reactive power generated by the nth DG in period t, respectively; is the network loss of branch n in period t; and are respectively the active and reactive power consumed by the nth load in period t; Ω EV ,Ω DG ,Ω SOP and Ω b They are charging stations, distributed power sources, SOP installation nodes and all branches of the distribution network; is the active power detection value output by the nth DG at time t; P DG,n,t , Q DG,n,t are the active power and reactive power output by DG at node n during period t; β DG,n,t is the power factor angle of DG at node n during period t; is the net active power generated by the jth charging station at time t; and are the active and reactive power generated by the nth SOP at time t; β SOP,n,t is the power factor angle of the nth SOP at time t.
[0096] In Step 4, under the constraints of Step 2 and Step 3, the planning cost is calculated and the SOP planning scheme with the minimum cost is found; the steps are as follows:
[0097] Step 4.1: The planning cost is modeled with the minimum annual comprehensive operating cost of the distribution network as the objective function. The model is as follows:
[0098] f med=C SI +C OMS +C PC +C NLC (13)
[0099]
[0100] Where: F med is the annual comprehensive operating cost; C SI is the annual investment cost of SOP; C OMS is the annual operation and maintenance cost of SOP; C PC The cost of purchasing electricity from the upper grid; C NLC is the network loss cost; λ is the discount rate; Ω SOP Install node collection for SOP;y SOP is the useful life of the SOP; c s is the unit capacity investment cost of SOP; is the SOP capacity to be installed between nodes i and j; η SOP is the annual operation and maintenance cost of SOP; c g The unit price of electricity purchased from the upper power grid; c loss is the unit power network loss cost; P t MG is the active power injected by the upper main grid into the distribution network at time t; is the network loss power between nodes i and j at time t.
[0101] Step 4.2: Use the solver to solve the above model, and obtain the node position and capacity size of the SOP that minimizes the objective function value, which is the planning result.
[0102] Calculate the branch load margin I in Step 5 BF and supply-demand balance flexibility index I FSD The operational flexibility of the distribution network is measured by calculating the difference between the two. The specific steps are as follows:
[0103] Step 5.1. Calculate the branch load margin flexibility index and supply-demand balance flexibility index respectively:
[0104] The branch load margin flexibility index is used to evaluate the transmission capacity of network flexibility resources. It is mainly represented by the ratio of the current in the branch to the maximum current that the branch can withstand. The smaller the value, the greater the load margin of each branch, and the stronger the ability to cope with the uncertainty changes in flexibility demand:
[0105]
[0106] Where: I BF is the average load rate of the network branch, Ω Bis the branch set of the distribution network; L zm,t is the load rate of branch z,m at time t; I zm,t is the current flowing through branch z,m at time t; I zm is the maximum value of the current allowed to flow through branch z,m.
[0107] The supply-demand balance flexibility index is used to measure whether the flexibility supply can meet the flexibility demand. The larger the value, the more the flexibility resources can meet the flexibility demand, and the better the overall flexibility of the distribution network:
[0108]
[0109] Where: I FSD It is an indicator of supply and demand balance flexibility; UFSD ,I DFSD The sum of the ratios of the dispatchable margin of flexible resources to the change in net load when the net load increases or decreases within a dispatch cycle; are the upward and downward flexibility adjustment capabilities provided by the j-th charging pile to the distribution network at time t; They are the upward and downward flexibility adjustment capabilities provided by the upper main network at time t; are the total upward and downward flexibility adjustment capabilities of the distribution network at time t; T is the total number of time periods in the dispatching cycle. The dispatching cycle in this paper is 24 hours, and the dispatching time interval is 1 hour; wt is the state variable of the net load power change at time t; P t nl , are the net load power at time t and t+1 respectively.
[0110] Step 5.2: The two distribution network flexibility evaluation indicators calculated in the first two steps are coupled, influenced and interacted with each other. In the planning process, node flexibility and branch flexibility are combined to jointly determine the quality of system flexibility:
[0111] F I =I FSD -I BF (20)
[0112] Where: FI represents system flexibility. The larger its value is, the better the system flexibility is.
[0113] Example:
[0114] The present invention is explained in detail with reference to the data set and attached figures of distributed power sources (wind power, photovoltaic power) and loads in the distribution network. The specific implementation steps are as follows:
[0115] S1: The data set of distributed power sources and loads selected in this embodiment is derived from the literature "Day-ahead Optimal Dispatch of AC / DC Hybrid Distribution Networks with SOP" by Chu Zhuang. Four photovoltaic systems are connected at nodes 7, 15, 22, and 26, with access capacities of 800kVA, 800kVA, 500kVA, and 400kVA, respectively; three wind turbines are connected at nodes 9, 23, and 30, with access capacities of 800kVA, 500kVA, and 300kVA, respectively; it contains one-day operating data of distributed power sources (wind power, photovoltaic power) and loads, with a measurement interval of 1 hour; the data set is shown in Table 1 below:
[0116]
[0117]
[0118] Table 1 Distributed power supply and load operation data set
[0119] S2: Simulate the charging rules of different types of vehicles through Monte Carlo sampling; evaluate the dispatchable potential of electric vehicles, and then calculate the upper and lower limits of the charging and discharging power changes and capacity changes of the charging station based on the actual detection data obtained in S1; taking charging station 3 as an example, the plot is obtained Figure 4 and Figure 5 The specific steps are as follows:
[0120] Step 2.1. Use Monte Carlo sampling to simulate the charging patterns of different types of vehicles, including 800 private cars charging at night, 400 online ride-hailing vehicles charging at night, and 800 office workers charging during the day. The vehicles are assigned to charging stations and sampled 1,000 times to simulate the 1,000-day charging and discharging history of electric vehicles. The 1,001th sampling is the actual data.
[0121] Step 2.2: Aggregate electric vehicles to charging stations. By managing the charging and discharging of electric vehicles in the stations, electric vehicles can participate in the planning of distribution networks as flexible loads:
[0122]
[0123] Where: and are the total charging and discharging power of charging station j in period t respectively; and are the charging and discharging dispatching powers of electric vehicle n in period t respectively; is the set of electric vehicles in charging station j; T is the scheduling time set.
[0124]
[0125] Table 2 Electric vehicle sampling parameters
[0126] N(μ,σ 2 ): Normal distribution with mathematical expectation μ and standard deviation σ;
[0127] U(a,b): uniform distribution on the interval [a,b];
[0128] S B : Rated capacity of electric vehicles. The present invention assumes that electric vehicles have uniform specifications.
[0129]
[0130]
[0131] Table 3 Charging station sampling parameters
[0132] At the same time, according to Step 2.1, the historical parking data of electric vehicles is obtained, including the arrival time, departure time, battery power at arrival, battery power at departure, battery power boundary of electric vehicles and maximum charge and discharge power to evaluate the historical dispatchable potential of charging piles; among them, in order to realize the extension of the definition domain of individual charging stations, the state variable X is first introduced n,t Used to characterize the grid-connected status of electric vehicles, X n,t It can be directly calculated from the parking time of the electric vehicle:
[0133]
[0134] Where: X n,t represents the state of electric vehicle n at time t, X n,t =1 means that electric vehicle n is connected to the grid during period t, X n,t =0 means that electric vehicle n is off-grid during period t. represents the time period when electric vehicle n arrives at the charging station, Indicates the time period when electric vehicle n leaves the charging station.
[0135] Step 2.3: Based on the above two steps, electric vehicles are aggregated into charging stations and the state variable X is introduced. n,t Afterwards, the dispatch potential of the electric vehicle cluster is evaluated:
[0136]
[0137] Where: and They represent the maximum charging and discharging power of the generalized energy storage device at charging station j in period t; X n,t-1 represents the state of electric vehicle n at time t-1; and They represent the upper limits of charging and discharging power of electric vehicle n respectively; and represents the battery power safety boundary of electric vehicle n; and represents the power boundary of the generalized energy storage device at charging station j in period t; s n,arrival represents the initial charge of electric vehicle n; s n,leave represents the power grid of electric vehicle n when it leaves the charging station, also known as its expected power; △S j,t It represents the change in power of the generalized energy storage device at charging station j due to the change in the grid-connected state of electric vehicles in period t. φ is the dispatchable potential of the electric vehicle cluster in the dataset.
[0138] Step 2.4: The historical electric vehicle cluster data is compressed into historical dispatchable potential, which can be used as training data, and a dispatchable potential prediction model is further established based on back propagation (BP) neural network, time series and other methods.
[0139] Step 2.5, calculate the real-time dispatchable potential by time period, including the capacity change and power upper and lower limits of the generalized energy storage. The charging station calculates the dispatchable potential based on the electric vehicles located in the charging station during the current time period. The calculation method is shown in formula (3). At the same time, as time goes by, the charging station continuously updates the dispatchable potential, but only executes a rolling time window plan during the dispatch process. Therefore, the real-time dispatchable potential is also called dynamic dispatchable potential. The calculated charging and discharging power upper and lower limits and SOC limits of the charging station are taken as an example in the attached figure. Figure 3 shown.
[0140] S3: First, construct the SOP operation constraints; then, include the SOP operation constraints, the data measured in Step 1, and the upper and lower limits of the charging and discharging power changes and capacity changes of the charging station calculated in Step 2 into the active and reactive balance constraints of the distribution network nodes; the steps are as follows:
[0141] Step 3.1, construct SOP operation constraints. SOP will have certain losses during operation. It can also perform certain charging and discharging when necessary. As a soft switch, its power inflow and outflow should be balanced, with power injection as negative and outflow as positive. There are the following constraints:
[0142]
[0143] Where: and are the active and reactive powers injected by SOP converters a and b at time t, respectively; is the charging / discharging power of the energy storage system in the SOP at time t; is the SOP transmission loss at time t; A SOP,a and A SOP,bare the loss coefficients of converter a and b respectively; and are the access capacities of converter a and converter b at t, which are usually considered equal and are recorded as
[0144] At the same time, when planning SOP, the upper limit of the number of planned SOPs is 5, and the number of planned SOPs can be changed at will. The capacity and position of each SOP will change accordingly if the number is different; the unit optimization capacity of SOP is 10kV·A. The planning parameters are shown in Table 4 below:
[0145]
[0146]
[0147] Table 4 SOP planning parameters
[0148] Step 3.2, include the SOP operation constraints, the data measured in Step 1, and the upper and lower limits of the charging and discharging power changes and capacity changes of the charging station calculated in Step 2 into the active and reactive balance constraints of the distribution network nodes:
[0149]
[0150] Where: P t MG and is the active and reactive power purchased from the upper main grid during period t; and are the active and reactive power generated by the nth DG in period t respectively; is the network loss of branch n in period t; and are respectively the active and reactive power consumed by the nth load in period t; Ω EV ,Ω DG ,Ω SOP and Ω b They are charging stations, distributed power sources, SOP installation nodes and all branches of the distribution network; is the active power detection value output by the nth DG at time t; P DG,n,t , Q DG,n,t are the active power and reactive power output by DG at node n during period t; β DG,n,t is the power factor angle of DG at node n during period t; is the net active power generated by the jth charging station at time t; and are the active and reactive power generated by the nth SOP at time t; β SOP,n,tis the power factor angle of the nth SOP at time t.
[0151] S4: Under the constraints of Step 2 and Step 3, calculate the planning cost and find the SOP planning scheme that minimizes the cost; the steps are as follows:
[0152] f med =C SI +C OMS +C PC +C NLC (13)
[0153]
[0154] Where: F med is the annual comprehensive operating cost; C SI is the annual investment cost of SOP; C OMS is the annual operation and maintenance cost of SOP; C PC The cost of purchasing electricity from the upper grid; C NLC is the network loss cost; λ is the discount rate; Ω SOP Install node collection for SOP;y SOP is the useful life of the SOP; c s is the unit capacity investment cost of SOP; is the SOP capacity to be installed between nodes i and j; η SOP is the annual operation and maintenance cost of SOP; c g The unit price of electricity purchased from the upper power grid; c loss is the unit power network loss cost; P t MG is the active power injected by the upper main grid into the distribution network at time t; is the network loss power between nodes i and j at time t.
[0155] Step 4.2: Use the solver to solve the above model, and obtain the node position and capacity size of the SOP that minimizes the objective function value, which is the planning result. The planning results are shown in Table 5 below:
[0156] Table 5 Comparison of economic performance of SOP planning
[0157]
[0158] S5: Calculate branch load margin I BF and supply-demand balance flexibility index I FSD The operational flexibility of the distribution network is measured by calculating the difference between the two. The specific steps are as follows:
[0159] Step 5.1. Calculate the branch load margin flexibility index and supply-demand balance flexibility index respectively:
[0160] The branch load margin flexibility index is used to evaluate the transmission capacity of network flexibility resources. It is mainly represented by the ratio of the current in the branch to the maximum current that the branch can withstand. The smaller the value, the greater the load margin of each branch, and the stronger the ability to cope with the uncertainty changes in flexibility demand:
[0161]
[0162] Where: I BF is the average load rate of the network branch, Ω B is the branch set of the distribution network; L zm,t is the load rate of branch z,m at time t; I zm,t is the current flowing through branch z,m at time t; I zm is the maximum value of the current allowed to flow through branch z,m.
[0163] The supply-demand balance flexibility index is used to measure whether the flexibility supply can meet the flexibility demand. The larger the value, the more the flexibility resources can meet the flexibility demand, and the better the overall flexibility of the distribution network:
[0164]
[0165] Where: I FSD It is an indicator of supply and demand balance flexibility; UFSD ,I DFSD The sum of the ratios of the dispatchable margin of flexible resources to the change in net load when the net load increases or decreases within a dispatch cycle; are the upward and downward flexibility adjustment capabilities provided by the j-th charging pile to the distribution network at time t; They are the upward and downward flexibility adjustment capabilities provided by the upper main network at time t; are the total upward and downward flexibility adjustment capabilities of the distribution network at time t; T is the total number of time periods in the dispatching cycle. The dispatching cycle in this paper is 24 hours, and the dispatching time interval is 1 hour; wt is the state variable of the net load power change at time t; P t nl , are the net load power at time t and t+1 respectively.
[0166] Step 5.2: The two distribution network flexibility evaluation indicators calculated in the first two steps are coupled, influenced and interacted with each other. In the planning process, node flexibility and branch flexibility are combined to jointly determine the quality of system flexibility:
[0167] F I =I FSD -I BF(20)
[0168] Where: FI represents system flexibility. The larger its value is, the better the system flexibility is.
[0169] The flexibility calculation results are shown in Table 6 below:
[0170]
[0171]
[0172] Table 6 Comparison of SOP planning flexibility
[0173] In general, the present invention proposes a SOP site selection and sizing planning method considering the dispatchable potential of electric vehicles (EVs), which effectively improves the economy and flexibility of the distribution network by integrating the energy storage capacity of EVs and the fluctuation characteristics of distributed generation (DG). Since the existing methods have problems such as high cost and insufficient flexibility improvement in the planning process, the present invention first obtains the data of distributed power sources and loads in the distribution network through detection technology; then, the charging patterns of different types of vehicles are simulated through Monte Carlo sampling, and the dispatchable potential of electric vehicles is evaluated; the electric vehicles are aggregated to charging stations, their charging and discharging power and dispatchable potential are calculated, and a prediction model is established; then the upper and lower limits of the charging and discharging power and the upper and lower limits of the capacity of the charging stations and SOPs are defined, and these data are included in the active and reactive balance constraints of the distribution network nodes to construct the SOP operation constraints; on the premise of meeting the constraints, the planning cost is calculated, and the SOP planning scheme with the minimum cost is found; finally, the branch load margin and the supply and demand balance flexibility index are calculated, and the operating flexibility of the distribution network is measured by calculating the difference between the two; and the economic and flexibility results are compared; through this method, not only the energy storage capacity of electric vehicles can be effectively utilized, the investment cost of the energy storage system and SOP can be reduced, but also the flexibility and economy of the distribution network can be significantly improved. Therefore, the SOP site selection and capacity planning method proposed in the present invention fully considers the dispatchable potential of electric vehicles and their improvement on the flexibility of the distribution network, solves the shortcomings of traditional methods in cost and flexibility, and significantly improves the economy and reliability of the distribution network.
Claims
1. A SOP site selection and capacity planning method considering the dispatchable potential of electric vehicles, characterized in that: The following steps are involved: Step 1: Obtain data on power generation and load power consumption of new energy distributed power sources in the distribution network through detection technology; Step 2: Simulate the charging rules of different types of vehicles through Monte Carlo sampling; evaluate the dispatchable potential of electric vehicles, and then calculate the charging and discharging power changes and the upper and lower limits of the capacity changes of the charging station based on the actual detection data obtained in Step 1; Step 3: Construct SOP operation constraints; then include the SOP operation constraints, the data measured in Step 1, and the upper and lower limits of the charging and discharging power changes and capacity changes of the charging station calculated in Step 2 into the active and reactive balance constraints of the distribution network nodes; Step 4: Under the constraints of Step 2 and Step 3, calculate the planning cost and find the SOP planning scheme that minimizes the cost; Step 5: Calculate branch load margin I BF and supply-demand balance flexibility index I FSD , and the operation flexibility of the distribution network is measured by calculating the difference between the two.
2. The method according to claim 1, characterized in that In Step 2, the charging and discharging power change and the upper and lower limits of the capacity change of the charging station are calculated; the specific steps are as follows: Step 2.1: Use Monte Carlo sampling to simulate the charging patterns of different types of vehicles, including a number of private cars charging at night, a number of online ride-hailing vehicles charging at night, and a number of office workers charging during the day; and assign them to charging stations. Use N sampling times to simulate the N-day charging and discharging history data of electric vehicles, and the N+1th sampling is the actual data; Step 2.2: Aggregate electric vehicles to charging stations. By managing the charging and discharging of electric vehicles in the stations, electric vehicles can participate in the planning of distribution networks as flexible loads: Where: and are the total charging and discharging power of charging station j in period t respectively; and are the charging and discharging dispatching powers of electric vehicle n in period t respectively; is the set of electric vehicles in charging station j; T is the scheduling time set; At the same time, according to Step 2.1, the historical parking data of electric vehicles is obtained, including the arrival time, departure time, battery power at arrival, battery power at departure, battery power boundary of electric vehicles and maximum charge and discharge power to evaluate the historical dispatchable potential of charging piles; among them, in order to realize the extension of the definition domain of individual charging stations, the state variable X is first introduced n,t Used to characterize the grid-connected status of electric vehicles, X n,t It can be directly calculated from the parking time of the electric vehicle: Where: X n,t represents the state of electric vehicle n at time t, X n,t =1 means that electric vehicle n is connected to the grid during period t, X n,t =0 means that electric vehicle n is off-grid during period t. represents the time period when electric vehicle n arrives at the charging station, represents the time period when electric vehicle n leaves the charging station; Step 2.3: Based on the above two steps, electric vehicles are aggregated into charging stations and the state variable X is introduced. n,t Afterwards, the dispatch potential of the electric vehicle cluster is evaluated: Where: and They represent the maximum charging and discharging power of the generalized energy storage device at charging station j in period t; X n,t-1 represents the state of electric vehicle n at time t-1; and They represent the upper limits of charging and discharging power of electric vehicle n respectively; and represents the battery power safety boundary of electric vehicle n; and represents the power boundary of the generalized energy storage device at charging station j in period t; s n,arrival represents the initial charge of electric vehicle n; s n,leave represents the power grid of electric vehicle n when it leaves the charging station, or its expected power; ΔS j,t represents the change in power of the generalized energy storage device at charging station j due to the change in the grid-connected state of electric vehicles during period t, and φ is the dispatchable potential of the electric vehicle cluster in the dataset; Step 2.4: The historical electric vehicle cluster data is compressed into historical dispatchable potential, which can be used as training data, and further based on back propagation (BP neural network, time series and other methods to establish a dispatchable potential prediction model; Step 2.5: Calculate the real-time dispatchable potential by time period, including the capacity change and power upper and lower limits of the generalized energy storage; the charging station calculates the dispatchable potential based on the electric vehicles located in the charging station during the current time period, and the calculation method is shown in formula (3); at the same time, as time goes by, the charging station continuously updates the dispatchable potential, but only executes a rolling time window plan during the dispatch process; therefore, the real-time dispatchable potential is also called dynamic dispatchable potential.
3. The method according to claim 1, characterized in that In the Step 3, the SOP operation constraints are first constructed; then the SOP operation constraints, the data measured in Step 1, and the upper and lower limits of the charging and discharging power change and capacity change of the charging station calculated in Step 2 are included in the active and reactive balance constraints of the distribution network nodes; The steps are as follows: Step 3.1: Construct SOP operation constraints. SOP will have certain losses during operation. It can also perform certain charging and discharging when necessary. As a soft switch, its power inflow and outflow should be balanced, with power injection as negative and outflow as positive. There are the following constraints: Where: and are the active and reactive powers injected by SOP converters a and b at time t, respectively; is the charging / discharging power of the energy storage system in the SOP at time t; is the SOP transmission loss at time t; A SOP,a and A SOP,b are the loss coefficients of converter a and b respectively; and are the access capacities of converter a and converter b at t, which are usually considered equal and are recorded as At the same time, when planning SOP, you can change the planned quantity by yourself. If the quantity is different, the capacity and location of each SOP obtained by planning will also change accordingly; Step 3.2: Include the SOP operation constraints, the data measured in Step 1, and the upper and lower limits of the charging and discharging power changes and capacity changes of the charging station calculated in Step 2 into the active and reactive balance constraints of the distribution network nodes: Where: P t MG and is the active and reactive power purchased from the upper main grid during period t; and are the active and reactive power generated by the nth DG in period t, respectively; is the network loss of branch n in period t; and are respectively the active and reactive power consumed by the nth load in period t; Ω EV ,Ω DG ,Ω SOP and Ω b They are charging stations, distributed power sources, SOP installation nodes and all branches of the distribution network; is the active power detection value output by the nth DG at time t; P DG,n,t , Q DG,n,t are the active power and reactive power output by DG at node n during period t; β DG,n,t is the power factor angle of DG at node n during period t; is the net active power generated by the jth charging station at time t; and are the active and reactive power generated by the nth SOP at time t; β SOP,n,t is the power factor angle of the nth SOP at time t.
4. The method according to claim 1, characterized in that: In Step 4, under the constraints of Step 2 and Step 3, the planning cost is calculated and the SOP planning scheme with the minimum cost is found. The specific steps are as follows: Step 4.1: The planning cost is modeled with the minimum annual comprehensive operating cost of the distribution network as the objective function. The model is as follows: min F med =C SI +C OMS +C PC +C NLC (13) Where: F med is the annual comprehensive operating cost; C SI is the annual investment cost of SOP; C OMS is the annual operation and maintenance cost of SOP; C PC The cost of purchasing electricity from the upper grid; C NLC is the network loss cost; λ is the discount rate; Ω SOP Install node set for SOP;y SOP is the useful life of the SOP; c s is the unit capacity investment cost of SOP; is the SOP capacity to be installed between nodes i and j; η SOP is the annual operation and maintenance cost of SOP; c g The unit price of electricity purchased from the upper power grid; c loss is the unit power network loss cost; P t MG is the active power injected by the upper main grid into the distribution network at time t; is the network loss power between nodes i and j at time t; Step 4.2: Use the solver to solve the above model, and obtain the node position and capacity size of the SOP that minimizes the objective function value, which is the planning result.
5. The method according to claim 1, characterized in that Calculate the branch load margin I in Step 5 BF and supply-demand balance flexibility index I FSD , and the difference between the two is calculated to measure the flexibility of distribution network operation; the specific steps are as follows: Step 5.
1. Calculate the branch load margin flexibility index and supply-demand balance flexibility index respectively: The branch load margin flexibility index is used to evaluate the transmission capacity of network flexibility resources. It is mainly represented by the ratio of the current in the branch to the maximum current that the branch can withstand. The smaller the value, the greater the load margin of each branch, and the stronger the ability to cope with the uncertainty changes in flexibility demand: Where: I BF is the average load rate of the network branch, Ω B is the branch set of the distribution network; L zm,t is the load rate of branch z,m at time t; I zm,t is the current flowing through branch z,m at time t; I zm is the maximum value of the current allowed to flow through branch z,m; The supply-demand balance flexibility index is used to measure whether the flexibility supply can meet the flexibility demand. The larger the value, the more the flexibility resources can meet the flexibility demand, and the better the overall flexibility of the distribution network: Where: I FSD It is an indicator of supply and demand balance flexibility; UFSD ,I DFSD The sum of the ratios of the dispatchable margin of flexible resources to the change in net load when the net load increases or decreases within a dispatch cycle; are the upward and downward flexibility adjustment capabilities provided by the j-th charging pile to the distribution network at time t; They are the upward and downward flexibility adjustment capabilities provided by the upper main network at time t; are the total upward and downward flexibility adjustment capabilities of the distribution network at time t; T is the total number of time periods in the dispatching cycle. The dispatching cycle in this paper is 24 hours, and the dispatching time interval is 1 hour; w t is the state variable of the net load power change at time t; P t nl , are the net load power at time t and t+1 respectively; Step 5.2: The two distribution network flexibility evaluation indicators calculated in the first two steps are coupled, mutually influential, and interact with each other; in the planning process, node flexibility and branch flexibility are combined to jointly determine the quality of system flexibility: F I =I FSD -I BF (20) Where: F I Indicates system flexibility. The larger the value, the better the system flexibility.
