Method and system for dispatching active power distribution network through grid connection of electric vehicles

Abstract

The invention relates to a method and a system for dispatching an active power distribution network through grid connection of an electric vehicle. The method comprises the following steps: acquiring operation data of the electric vehicle; constructing an electric vehicle cluster schedulable model and a multi-subject optimization scheduling model, wherein the multi-subject optimization scheduling model comprises a first stage and a second stage; the first stage is divided into an upper-layer power distribution network operator pre-scheduling model and a lower-layer electric vehicle aggregator charging and discharging decision model; in the second stage, a wind and light output uncertainty compensation model is constructed; and solving the multi-agent optimization scheduling model by using a KKT condition and a dual theorem to obtain a scheduling scheme of the active power distribution network by electric vehicle grid connection.

Description

Technical Field

[0001] This application relates to the field of electric vehicle scheduling, and mainly relates to a scheduling method and system for the grid connection of electric vehicles to the active distribution network. Background Art

[0002] With the increasingly serious global energy shortage and environmental pollution problems, electric vehicles, as a means of transportation with the advantages of clean environmental protection and low-carbon energy-saving travel, are gradually expanding their grid connection scale. The large-scale connected electric vehicles can participate in the optimal scheduling of the active distribution network as demand-side flexibility resources, and have great potential in improving the economic operation of the system and reducing the peak-valley difference of the load.

[0003] The existing research on the economic scheduling of the grid connection of electric vehicles to the active distribution network mostly focuses on the optimization of the interests of single subjects, ignoring the interest demands of subjects such as electric vehicle users, distributed energy owners, and electricity retailers, resulting in conflicts between global optimization and local interests; traditional stochastic or robust optimization methods are difficult to balance economy and system security, and when large-scale electric vehicles are connected, centralized optimization faces the problems of dimensionality disaster and communication delay, while distributed control lacks global coordination ability and does not fully consider the correlation of multi-source uncertainties such as electric vehicle travel patterns and renewable energy output; in addition, new devices such as distributed energy, energy storage, and electric vehicles lack unified communication protocols and control standards, resulting in difficulties in cross-platform collaboration; the aggregation and decoupling mechanisms of virtual power plants are imperfect, affecting the efficient integration of multi-subject resources.

[0004] As disclosed in the Chinese invention patent with the publication number "CN114462854A", a "Hierarchical Scheduling Method and System for Integrating New Energy and Electric Vehicles into the Grid" is disclosed, specifically including the following steps: Step 1, based on the characteristics of the new energy and electric vehicles integrated into the grid, establish a new energy prediction error model and an electric vehicle response error model; Step 2, establish an upper-layer model with the minimum total system load and the maximum total new energy consumption as the objective function, and establish a lower-layer model with the minimum cost of the electric vehicle agent, the minimum deviation between the agent's output and the scheduling plan, and the maximum electric vehicle response priority as the objective function. Substitute the models in Step 1 into the two-layer model in Step 2; Step 3, use an improved genetic algorithm to solve the two-layer model and obtain the scheduling result. However, the two-layer model of this method only includes the grid operator and the electric vehicle agent, and does not explicitly incorporate key entities such as distributed energy owners and electricity retailers, resulting in insufficient coordinated optimization of new energy consumption and load regulation and making it difficult to achieve global optimality. In addition, the improved genetic algorithm of this method belongs to a heuristic method, with the defects of high computational complexity and slow convergence speed, making it difficult to meet the real-time scheduling requirements and prone to falling into local optimality, affecting the quality of the optimization results. At the same time, this method lacks a dynamic adjustment and closed-loop feedback mechanism and cannot dynamically correct the prediction error based on actual operation data, weakening the system's ability to cope with uncertainties. Summary of the Invention

[0005] To solve the above problems existing in the prior art, the present application provides a scheduling method and system for an active distribution network with electric vehicles integrated into the grid.

[0006] The technical solution of the present application is as follows:

[0007] On the one hand, the present invention proposes a scheduling method for an active distribution network with electric vehicles integrated into the grid, the method including:

[0008] Obtain the operation data of electric vehicles; construct a dispatchable model for an electric vehicle cluster based on the operation data; construct a multi-agent optimal dispatch model based on the dispatchable model for the electric vehicle cluster, the multi-agent optimal dispatch model including a first stage and a second stage;

[0009] The first stage is divided into an upper-layer pre-dispatch model for the distribution network operator and a lower-layer charging and discharging decision model for the electric vehicle aggregator; wherein the upper-layer pre-dispatch model for the distribution network operator includes an objective function with the minimum total distribution network operation cost as the objective and constraint conditions with branch power flow constraints and safe operation constraints as the constraints; the lower-layer charging and discharging decision model for the electric vehicle aggregator includes an objective function with the maximum operating profit of the electric vehicle aggregator as the objective and constraint conditions with the operating constraints of the electric vehicle aggregator as the constraints; the second stage constructs a compensation model for the uncertainty of the wind and light output.

[0010] Solve the multi-agent optimal scheduling model by using the KKT conditions and the duality theorem to obtain the scheduling plan of the electric vehicle grid connection to the active distribution network.

[0011] Preferably, the method further includes data cleaning of the operation data, and the data cleaning includes processing missing values, outliers, and data format standardization.

[0012] Preferably, the schedulable model of the electric vehicle cluster is expressed by the formula:

[0013]

[0014] In the formula, represents the charging power of the kth charging and discharging station at time t; represents the discharging power of the kth charging and discharging station at time t; u n,t represents the grid connection and disconnection state of the nth electric vehicle at time t; represents the charging power of the nth electric vehicle at time t; represents the discharging power of the nth electric vehicle at time t; represents the upper limit of the power of the kth charging and discharging station at time t; represents the lower limit of the power of the kth charging and discharging station at time t; represents the upper limit of the power of the nth electric vehicle; represents the lower limit of the power of the nth electric vehicle; represents the set of electric vehicles at the kth charging and discharging station; n represents the index value of the nth electric vehicle; k represents the index value of the kth charging and discharging station; t represents the time.

[0015] Preferably, the pre-scheduling model of the upper-level distribution network operator is specifically:

[0016] The objective function with the minimum total operation cost of the distribution network is expressed by the formula:

[0017] minC DS = C DG + C M + C loss - C L ;

[0018]

[0019]

[0020] In the formula, C DS represents the total operation cost of the distribution network; C DG represents the operation cost of the wind-solar unit; C M represents the interactive power purchase cost of the active distribution network; C lossRepresents the loss cost of the active distribution network; C L Represents the revenue from load electricity consumption; p WT Represents the cost per unit output of the wind turbine; P t WT Represents the per-unit output of the wind turbine at time t; p PV Represents the cost per unit output of the photovoltaic; P t PV Represents the per-unit output of the photovoltaic at time t; c WT Represents the penalty unit price for curtailed wind power; c PV Represents the penalty unit price for curtailed photovoltaics; ΔP t PV Represents the per-unit output of the photovoltaic at time t; ΔP t WT Represents the per-unit output of the wind turbine at time t; c md Represents the real-time electricity price of the active distribution network; P t buy Represents the power purchase of the active distribution network at time t; P t sell Represents the power sale of the active distribution network at time t; i represents the index value of the i-th node; j represents the index value of the j-th node; E represents the set of nodes; c loss Represents the unit cost of losses in the active distribution network; r ij Represents the resistance between the i-th node and the j-th node; I ij,t Represents the current between the i-th node and the j-th node at time t; c ds Represents the real-time electricity price of the active distribution network load; Represents the load power of the i-th node at time t; Represents the charging price set by the electric vehicle aggregator at time t; Represents the discharging price set by the electric vehicle aggregator at time t; T represents the time period;

[0021] The power flow constraint is expressed by the formula:

[0022]

[0023] In the formula, Represents the square of the voltage of the j-th node at time t; Represents the square of the voltage of the i-th node at time t; x ij Represents the reactance between the i-th node and the j-th node; P ij,t Represents the active power between the i-th node and the j-th node at time t; Q ij,t Represents the reactive power between the i-th node and the j-th node at time t; Represents the square of the current between the i-th node and the j-th node at time t; p j,tThe active power between the j-th node at time t; δ represents a preset positive number; → represents a mapping relationship; k:j→k represents the path from the j-th node to the k-th charging and discharging station; q j,t The reactive power between the j-th node at time t;

[0024] The safe operation constraints are expressed by the formula:

[0025]

[0026] In the formula, U i,min Represents the lower voltage limit of the i-th node; U i,max Represents the upper voltage limit of the i-th node; I ij,min Represents the lower current limit between the i-th node and the j-th node; I ij,max Represents the upper current limit between the i-th node and the j-th node.

[0027] Preferably, the charging and discharging decision-making model of the lower-layer electric vehicle aggregator is specifically:

[0028] The objective function with the maximum operating profit of the electric vehicle aggregator as the goal is expressed by the formula:

[0029]

[0030] In the formula, C EV Represents the operating profit of the electric vehicle aggregator;

[0031] The operating constraints of the electric vehicle aggregator are expressed by the formula:

[0032]

[0033] In the formula, Represents the upper limit of the equivalent charging power of the electric vehicle cluster at the k-th charging and discharging station; Represents the upper limit of the equivalent discharging power of the electric vehicle cluster at the k-th charging and discharging station; Represents the on-grid and off-grid state of the k-th charging and discharging station at time t; Represents the lower limit of the equivalent battery capacity of the electric vehicle cluster at the k-th charging and discharging station; Represents the upper limit of the equivalent battery capacity of the electric vehicle cluster at the k-th charging station; Represents the equivalent battery capacity of the electric vehicle cluster at the k-th charging and discharging station at time t; η c Represents the charging coefficient of the electric vehicle; η d Represents the discharging coefficient of the electric vehicle.

[0034] Preferably, the wind and light output uncertainty compensation model is specifically:

[0035] Construct an objective function for coping with the worst-case scenario, which is expressed by the formula as follows:

[0036] maxminC DS =ΔC DG +ΔC M +ΔC loss -ΔC L ;

[0037] In the formula, ΔC DG represents the operating cost of the wind-solar power units under the worst-case scenario; ΔC M represents the interactive power purchase cost of the active distribution network under the worst-case scenario; ΔC loss represents the loss cost of the active distribution network under the worst-case scenario; ΔC L represents the load electricity consumption revenue under the worst-case scenario;

[0038] The constraint conditions are expressed by the formula as follows:

[0039]

[0040] In the formula, represents the PV grid-connected power of the i-th node at time t; represents the predicted PV output of the i-th node at time t; represents the maximum predicted error of the PV output of the i-th node at time t; represents the wind turbine grid-connected power of the i-th node at time t; represents the predicted wind turbine output of the i-th node at time t; represents the maximum predicted error of the wind turbine output of the i-th node at time t; represents the positive-direction error auxiliary variable of the PV output of the i-th node at time t; represents the negative-direction error auxiliary variable of the PV output of the i-th node at time t; represents the positive-direction error auxiliary variable of the wind turbine output of the i-th node at time t; represents the negative-direction error auxiliary variable of the wind turbine output of the i-th node at time t; Γ PV represents the adjustable error parameter of the PV output; Γ WT represents the adjustable error parameter of the wind turbine output.

[0041] Preferably, the KKT conditions and the duality theorem are used to solve the multi-agent optimal scheduling model, specifically as follows:

[0042] The KKT conditions transform the lower-layer electric vehicle aggregator's charging and discharging decision model into the constraints of the upper-layer distribution network operator's pre-scheduling model, and transform the objective functions of the first stage and the second stage into a mixed-integer linear programming problem. Using the duality theorem, the first stage is regarded as the main problem and the second stage is regarded as the sub-problem, specifically as follows:

[0043] The main problem is expressed by the formula:

[0044]

[0045] where λ represents the auxiliary variable for coping with uncertainty in the second stage; M represents a preset constant; l represents the index value of the l-th iteration. When l = 1, λ = 0 indicates no uncertainty scenario constraint. When l > 1, represents the newly added uncertainty scenario constraint in the l-th iteration, s represents the time period corresponding to the worst scenario found by the sub-problem;

[0046] The sub-problem is expressed by the formula:

[0047]

[0048]

[0049] Use a mixed-integer linear programming solver to iteratively solve the main problem and the sub-problem to obtain the candidate solution x l of the main problem and the candidate solution λ l of the sub-problem, where x l represents the candidate solution of the l-th iteration, and λ l represents the candidate solution of the sub-problem of the l-th iteration;

[0050] The iterative solution process is specifically based on the current candidate solution x l of the main problem to find the wind-solar power output error scenario λ l that maximizes the cost of the sub-problem, that is, the candidate worst scenario found by the sub-problem; until the maximum number of iterations is reached or the sub-problem converges. The sub-problem converges specifically as follows: if λ l > 0, then add the initial uncertainty scenario constraint λ l ≥∑ t∈s (c PV (ΔP t PV ) l + c WT (ΔP t WT ) l ) to the constraints of the main problem and re-solve the main problem, where (ΔP t PV ) l represents the photovoltaic unit output power at time t corresponding to the l-th iteration, and (ΔP t WT ) l represents the wind turbine unit output power at time t corresponding to the l-th iteration; if λ l≤0, stop the iteration, the sub - problem converges to obtain the worst - case scenario found by the sub - problem, that is, the main problem converges to obtain the optimal solution of the main problem, namely, obtain the scheduling scheme of the electric vehicle grid connection to the active distribution network.

[0051] On the other hand, the present invention also proposes an economic scheduling system for the electric vehicle grid connection to the active distribution network. The system includes a data acquisition module, a model construction module, a solution module, and a result output module, where:

[0052] The data acquisition module is used to acquire the operation data of the electric vehicle; and transmit the operation data to the model construction module;

[0053] The model construction module is used to construct a schedulable model of the electric vehicle cluster based on the operation data; and construct a multi - agent optimal scheduling model based on the schedulable model of the electric vehicle cluster. The multi - agent optimal scheduling model includes a first stage and a second stage;

[0054] The first stage is divided into an upper - layer distribution network operator pre - scheduling model and a lower - layer electric vehicle aggregator charging and discharging decision - making model; wherein the upper - layer distribution network operator pre - scheduling model includes an objective function with the minimum total operation cost of the distribution network as the goal, and constraint conditions with branch power flow constraints and safe operation constraints as the constraints; the lower - layer electric vehicle aggregator charging and discharging decision - making model includes an objective function with the maximum operation profit of the electric vehicle aggregator as the goal, and constraint conditions with the operation constraints of the electric vehicle aggregator as the constraints; the second stage constructs a wind - solar power output uncertainty compensation model;

[0055] The solution module is used to solve the multi - agent optimal scheduling model by using the KKT condition and the duality theorem, and obtain the scheduling scheme of the electric vehicle grid connection to the active distribution network;

[0056] The result output module is used to display the economic scheduling scheme of the electric vehicle grid connection to the active distribution network.

[0057] On yet another aspect, the present invention also proposes an electronic device, including a memory, a processor, and a computer program stored on the memory and executable on the processor. When the processor executes the program, it implements a scheduling method for the electric vehicle grid connection to the active distribution network as described in any embodiment of the present invention.

[0058] On yet another aspect, the present invention also proposes a computer - readable storage medium, on which a computer program is stored. When the program is executed by a processor, it implements a scheduling method for the electric vehicle grid connection to the active distribution network as described in any embodiment of the present invention.

[0059] Compared with the prior art, the beneficial effects of the present invention are:

[0060] 1) The present invention provides a dispatching method and system for an electric vehicle integrated grid connection to an active distribution network. By explicitly constructing a multi-agent model including a distribution network operator, an electric vehicle aggregator, and a wind-solar power generation entity, the balance ability between global economy and local interests is enhanced;

[0061] 2) The present invention provides a dispatching method and system for an electric vehicle integrated grid connection to an active distribution network. In the first stage, branch power flow constraints and safe operation constraints are incorporated. In the second stage, the wind-solar power output uncertainty compensation model improves the adaptability of the dispatching scheme to real-time data deviation and enhances the operation safety of the distribution network through probability constraints and compensation mechanisms; by constructing a multi-agent optimal dispatching model, the closed-loop feedback ability to uncertainty is enhanced, and the flexibility to cope with real-time operation state changes is improved;

[0062] 3) The present invention provides a dispatching method and system for an electric vehicle integrated grid connection to an active distribution network. Based on the exact solution method of KKT conditions and duality theorem, the calculation efficiency of large-scale optimization problems is improved, the defect that heuristic algorithms are prone to fall into local optima is avoided, the convergence of the dispatching scheme is enhanced, the load smoothing and peak shaving potential is improved, and the aggregation efficiency of electric vehicle resources is enhanced. BRIEF DESCRIPTION OF THE DRAWINGS

[0063] Figure 1 is the flowchart of the method in the embodiment of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0064] The following describes the specific embodiments of the present invention to facilitate those skilled in the art of the present technology to understand the present invention. However, it should be clear that the present invention is not limited to the scope of the specific embodiments. For those of ordinary skill in the art of the present technology, as long as various changes are within the spirit and scope of the present invention defined and determined by the appended claims, these changes are obvious, and all inventions and creations using the concept of the present invention are within the scope of protection.

[0065] The present invention provides the following technical solution: a dispatching method and system for an electric vehicle integrated grid connection to an active distribution network.

[0066] Embodiment 1

[0067] Specifically refer to Figure 1 , this embodiment provides a dispatching method for an electric vehicle integrated grid connection to an active distribution network. The specific steps include:

[0068] S1. Obtain the operation data of the electric vehicle; construct a dispatchable model of the electric vehicle cluster based on the operation data;

[0069] The method further includes data cleaning of the operation data, and the data cleaning includes processing missing values, outliers, and data format standardization;

[0070] S11. The schedulable model of the electric vehicle cluster is expressed by the formula:

[0071]

[0072] In the formula, represents the charging power of the k-th charging and discharging station at time t; represents the discharging power of the k-th charging and discharging station at time t; u n,t represents the grid-connected and off-grid state of the n-th electric vehicle at time t; represents the charging power of the n-th electric vehicle at time t; represents the discharging power of the n-th electric vehicle at time t; represents the upper limit of the power of the k-th charging and discharging station at time t; represents the lower limit of the power of the k-th charging and discharging station at time t; represents the upper limit of the power of the n-th electric vehicle; represents the lower limit of the power of the n-th electric vehicle; represents the set of electric vehicles at the k-th charging and discharging station; n represents the index value of the n-th electric vehicle; k represents the index value of the k-th charging and discharging station; t represents the time;

[0073] S2. Based on the schedulable model of the electric vehicle cluster, a multi-agent optimal scheduling model is constructed. The multi-agent optimal scheduling model includes a first stage and a second stage;

[0074] S21. The first stage is divided into an upper-layer distribution network operator pre-scheduling model and a lower-layer electric vehicle aggregator charging and discharging decision-making model;

[0075] S211. The upper-layer distribution network operator pre-scheduling model includes an objective function with the minimum total operation cost of the distribution network as the goal, which is expressed by the formula:

[0076] minC DS = C DG + C M + C loss - C L ;

[0077]

[0078]

[0079] In the formula, C DS represents the total operation cost of the distribution network; C DG represents the operation cost of the wind-solar unit; C M represents the interactive power purchase cost of the active distribution network; C loss represents the loss cost of the active distribution network; C LRepresents the load electricity revenue; p WT Represents the cost per unit output of the wind turbine; P t WT Represents the unit output of the wind turbine at time t; p PV Represents the cost per unit output of the photovoltaic; P t PV Represents the unit output of the photovoltaic at time t; c WT Represents the penalty unit price for abandoned wind turbines; c PV Represents the penalty unit price for abandoned photovoltaic; ΔP t PV Represents the unit output of the photovoltaic at time t; ΔP t WT Represents the unit output of the wind turbine at time t; c md Represents the real-time electricity price of the active distribution network; P t buy Represents the purchased power of the active distribution network at time t; P t sell Represents the sold power of the active distribution network at time t; i represents the index value of the i-th node; j represents the index value of the j-th node; E represents the set of nodes; c loss Represents the unit cost of losses in the active distribution network; r ij Represents the resistance between the i-th node and the j-th node; I ij,t Represents the current between the i-th node and the j-th node at time t; c ds Represents the real-time electricity price of the active distribution network load; Represents the load power of the i-th node at time t; Represents the charging price set by the electric vehicle aggregator at time t; Represents the discharging price set by the electric vehicle aggregator at time t; T represents the time period;

[0080] The constraint conditions with branch power flow constraints and safe operation constraints as constraints;

[0081] The said power flow constraint, expressed by the formula as:

[0082]

[0083] In the formula, Represents the square of the voltage of the j-th node at time t; Represents the square of the voltage of the i-th node at time t; x ij Represents the reactance between the i-th node and the j-th node; P ij,t Represents the active power between the i-th node and the j-th node at time t; Q ij,t Represents the reactive power between the i-th node and the j-th node at time t; represents the square of the current between the i-th node and the j-th node at time t; p j,t represents the active power between the j-th nodes at time t; δ represents a preset positive number; → represents a mapping relationship; k:j→k represents the path from the j-th node to the k-th charging and discharging station; q j,t represents the reactive power between the j-th nodes at time t;

[0084] The safe operation constraint is expressed by the formula:

[0085]

[0086] In the formula, U i,min represents the lower voltage limit of the i-th node; U i,max represents the upper voltage limit of the i-th node; I ij,min represents the lower current limit between the i-th node and the j-th node; I ij,max represents the upper current limit between the i-th node and the j-th node;

[0087] S212. The charging and discharging decision-making model of the lower-layer electric vehicle aggregator; The objective function aiming at the maximum operating profit of the electric vehicle aggregator is expressed by the formula:

[0088]

[0089] In the formula, C EV represents the operating profit of the electric vehicle aggregator;

[0090] The constraint conditions with the operating constraints of the electric vehicle aggregator as the constraints are expressed by the formula:

[0091]

[0092] In the formula, represents the upper limit of the equivalent charging power of the electric vehicle cluster at the k-th charging and discharging station; represents the upper limit of the equivalent discharging power of the electric vehicle cluster at the k-th charging and discharging station; represents the on-grid / off-grid state of the k-th charging and discharging station at time t; represents the lower limit of the equivalent battery capacity of the electric vehicle cluster at the k-th charging and discharging station; represents the upper limit of the equivalent battery capacity of the electric vehicle cluster at the k-th charging station; represents the equivalent battery capacity of the electric vehicle cluster at the k-th charging and discharging station at time t; η c represents the charging coefficient of the electric vehicle; η d represents the discharging coefficient of the electric vehicle;

[0093] S22. The construction of the wind and light output uncertainty compensation model in the second stage;

[0094] Construct an objective function for coping with the worst-case scenario, which is expressed by the formula as follows:

[0095] maxminC DS = ΔC DG + ΔC M + ΔC loss - ΔC L ;

[0096] In the formula, ΔC DG represents the operating cost of the wind-solar power units under the worst-case scenario; ΔC M represents the interactive power purchase cost of the active distribution network under the worst-case scenario; ΔC loss represents the loss cost of the active distribution network under the worst-case scenario; ΔC L represents the load electricity consumption benefit under the worst-case scenario;

[0097] The constraint conditions are expressed by the formula as follows:

[0098]

[0099] In the formula, represents the PV grid-connected power of the i-th node at time t; represents the PV predicted output of the i-th node at time t; represents the maximum PV prediction error of the i-th node at time t; represents the wind turbine grid-connected power of the i-th node at time t; represents the wind turbine predicted output of the i-th node at time t; represents the maximum wind turbine prediction error of the i-th node at time t; represents the positive-direction error auxiliary variable of the PV output of the i-th node at time t; represents the negative-direction error auxiliary variable of the PV output of the i-th node at time t; represents the positive-direction error auxiliary variable of the wind turbine output of the i-th node at time t; represents the negative-direction error auxiliary variable of the wind turbine output of the i-th node at time t; Γ PV represents the adjustable error parameter of the PV output; Γ WT represents the adjustable error parameter of the wind turbine output;

[0100] S3. Use the KKT conditions and the duality theorem to solve the multi-agent optimal scheduling model. The KKT conditions transform the charging and discharging decision model of the lower-layer electric vehicle aggregator into the constraints of the upper-layer distribution network operator's pre-scheduling model, and transform the objective functions of the first stage and the second stage into mixed-integer linear programming problems. Use the duality theorem to regard the first stage as the main problem and the second stage as the sub-problem. Specifically:

[0101] The main problem is expressed by the formula as follows:

[0102]

[0103] In the formula, λ represents the auxiliary variable for coping with uncertainty in the second stage; M represents a preset constant; l represents the index value of the l-th iteration. When l = 1, λ = 0 indicates no uncertainty scenario constraint. When l > 1, represents the newly added uncertainty scenario constraint in the l-th iteration, s represents the time period corresponding to the worst scenario found by the sub-problem;

[0104] The sub-problem is expressed by the formula as follows:

[0105]

[0106]

[0107] Use a mixed-integer linear programming solver to iteratively solve the main problem and the sub-problem to obtain the candidate solution x l of the main problem and the candidate solution λ l of the sub-problem, where x l represents the candidate solution of the l-th iteration, and λ l represents the candidate solution of the sub-problem in the l-th iteration;

[0108] S4. The iterative solution process is specifically based on the current candidate solution x l of the main problem to find the wind-solar power output error scenario λ l that maximizes the cost of the sub-problem, that is, the candidate worst scenario found by the sub-problem; until the maximum number of iterations is reached or the sub-problem converges. The sub-problem convergence specifically means that if λ l > 0, then add the initial uncertainty scenario constraint λ l ≥∑ t∈s (c PV (ΔP t PV ) l +c WT (ΔP t WT ) l ) to the constraints of the main problem and re-solve the main problem, where (ΔP t PV ) l represents the photovoltaic unit output power at time t corresponding to the l-th iteration, and (ΔP t WT ) l represents the wind turbine unit output power at time t corresponding to the l-th iteration; if λ l≤0, stop the iteration. The sub - problem converges to obtain the worst - case scenario found by the sub - problem, that is, the master - problem converges to obtain the optimal solution of the master - problem.

[0109] Since the sub - agent problem is specifically a function of the auxiliary variable λ for coping with uncertainty in the second stage, and the master - agent problem is specifically a function only containing the auxiliary variable λ for coping with uncertainty in the second stage.

[0110] By obtaining the optimal solution of the sub - agent problem, the auxiliary variable λ for coping with uncertainty in the second stage can be obtained, that is, by solving the optimal solution of the master - agent problem, the scheduling scheme of the electric vehicle grid connection to the active distribution network is obtained.

[0111] Embodiment 2

[0112] This embodiment provides an economic scheduling system for electric vehicle grid connection to an active distribution network. The system includes a data acquisition module, a model construction module, a solution module, and a result output module, where:

[0113] The data acquisition module is used to acquire the operation data of electric vehicles and transmit the operation data to the model construction module.

[0114] The model construction module is used to construct a schedulable model of an electric vehicle cluster based on the operation data, and construct a multi - agent optimal scheduling model based on the schedulable model of the electric vehicle cluster. The multi - agent optimal scheduling model includes a first stage and a second stage.

[0115] The first stage is divided into an upper - layer distribution network operator pre - scheduling model and a lower - layer electric vehicle aggregator charging and discharging decision - making model. The upper - layer distribution network operator pre - scheduling model includes an objective function with the minimum total operation cost of the distribution network as the goal and constraint conditions with branch power flow constraints and safe operation constraints as the constraints. The lower - layer electric vehicle aggregator charging and discharging decision - making model includes an objective function with the maximum operation profit of the electric vehicle aggregator as the goal and constraint conditions with the operation constraints of the electric vehicle aggregator as the constraints. The second stage constructs a wind - solar power output uncertainty compensation model.

[0116] The solution module is used to solve the multi - agent optimal scheduling model by using the KKT conditions and the duality theorem to obtain the scheduling scheme of the electric vehicle grid connection to the active distribution network.

[0117] The result output module is used to display the economic scheduling scheme of the electric vehicle grid connection to the active distribution network.

[0118] Embodiment 3

[0119] This embodiment provides an electronic device, including a memory, a processor, and a computer program stored on the memory and executable on the processor. When the processor executes the program, it implements a scheduling method for an electric vehicle to connect to the grid and dispatch the active distribution network as described in any embodiment of the present invention.

[0120] Embodiment 4

[0121] This embodiment provides a computer-readable storage medium, on which a computer program is stored. When the program is executed by a processor, it implements a scheduling method for an electric vehicle to connect to the grid and dispatch the active distribution network as described in any embodiment of the present invention.

[0122] It should be noted that the systems, electronic devices, and computer-readable storage media described in the present invention are all based on the same principle as the method described in Embodiment 1, and will not be elaborated herein.

[0123] The above are only the embodiments of the present invention, and do not limit the patent scope of the present invention. Any equivalent structure or equivalent process transformation made by using the content of the specification and drawings of the present invention, or directly or indirectly applied in other related technical fields, shall be equally included in the patent protection scope of the present invention.

Claims

1. A dispatching method for an electric vehicle connected to the grid for an active distribution network, characterized in that, The method includes: Obtaining the operation data of the electric vehicle; constructing a dispatchable model for the electric vehicle cluster based on the operation data; Constructing a multi-agent optimal dispatch model based on the dispatchable model of the electric vehicle cluster, where the multi-agent optimal dispatch model includes a first stage and a second stage; The first stage is divided into an upper-layer distribution network operator pre-dispatch model and a lower-layer electric vehicle aggregator charging and discharging decision model; among them, the upper-layer distribution network operator pre-dispatch model includes an objective function with the minimum total operation cost of the distribution network as the goal, and constraint conditions with branch power flow constraints and safe operation constraints as the constraints; the lower-layer electric vehicle aggregator charging and discharging decision model includes an objective function with the maximum operation profit of the electric vehicle aggregator as the goal, and constraint conditions with the operation constraints of the electric vehicle aggregator as the constraints; the second stage constructs a wind-solar power output uncertainty compensation model; Using the KKT conditions and the duality theorem to solve the multi-agent optimal dispatch model to obtain the dispatch plan of the electric vehicle grid connection to the active distribution network.

2. The dispatching method of an electric vehicle connected to the grid for an active distribution network according to claim 1, characterized in that, The method further includes data cleaning of the operation data, and the data cleaning includes processing missing values, outliers, and data format unification.

3. A dispatching method for an electric vehicle to connect to the grid and dispatch an active distribution network according to claim 1, characterized in that, The dispatchable model of the electric vehicle cluster is expressed by the formula: In the formula, represents the charging power of the k-th charging and discharging station at time t; represents the discharging power of the k-th charging and discharging station at time t; u n,t represents the grid-connected and off-grid state of the n-th electric vehicle at time t; represents the charging power of the n-th electric vehicle at time t; represents the discharging power of the n-th electric vehicle at time t; represents the upper limit of the power of the k-th charging and discharging station at time t; represents the lower limit of the power of the k-th charging and discharging station at time t; represents the upper limit of the power of the n-th electric vehicle; represents the lower limit of the power of the n-th electric vehicle; represents the set of electric vehicles at the k-th charging and discharging station; n represents the index value of the n-th electric vehicle; k represents the index value of the k-th charging and discharging station; t represents the time.

4. A dispatching method for an electric vehicle to be connected to the grid for an active distribution network according to claim 1, characterized in that, The upper-layer distribution network operator pre-dispatch model is specifically: The objective function with the minimum total operation cost of the distribution network as the goal is expressed by the formula: minC DS = C DG + C M + C loss - C L ; Where, C DS represents the total operating cost of the distribution network; C DG represents the operating cost of the wind-solar units; C M represents the interactive power purchase cost of the active distribution network; C loss represents the loss cost of the active distribution network; C L represents the load electricity consumption revenue; p WT represents the unit output cost of the wind turbine; represents the unit output of the wind turbine at time t; p PV represents the unit output cost of the photovoltaic; represents the unit output of the photovoltaic at time t; c WT represents the penalty unit price for abandoned wind turbines; c PV represents the penalty unit price for abandoned photovoltaics; represents the unit output of the photovoltaic at time t; represents the unit output of the wind turbine at time t; c md represents the real-time electricity price of the active distribution network; represents the purchased power of the active distribution network at time t; represents the sold power of the active distribution network at time t; i represents the index value of the i-th node; j represents the index value of the j-th node; E represents the set of nodes; c loss represents the loss unit cost of the active distribution network; r ij represents the resistance between the i-th node and the j-th node; I ij,t represents the current between the i-th node and the j-th node at time t; c ds represents the real-time electricity price of the active distribution network load; represents the load power of the i-th node at time t; represents the charging price set by the electric vehicle aggregator at time t; represents the discharging price set by the electric vehicle aggregator at time t; T represents the time period; The power flow constraint is expressed by the formula: Wherein, represents the square of the voltage of the j-th node at time t; represents the square of the voltage of the i-th node at time t; x ij represents the reactance between the i-th node and the j-th node; P ij,t represents the active power between the i-th node and the j-th node at time t; Q ij,t represents the reactive power between the i-th node and the j-th node at time t; represents the square of the current between the i-th node and the j-th node at time t; p j,t represents the active power between the j-th nodes at time t; δ represents a preset positive number; → represents a mapping relationship; k:j→k represents the path from the j-th node to the k-th charging and discharging station; q j,t represents the reactive power between the j-th nodes at time t; The safe operation constraint is expressed by the formula: where U i,min represents the lower voltage limit of the i-th node; U i,max represents the upper voltage limit of the i-th node; I ij,min represents the lower limit of the current between the i-th node and the j-th node; I ij,max represents the upper limit of the current between the i-th node and the j-th node.

5. A dispatching method for an electric vehicle to be connected to the grid for an active distribution network according to claim 1, characterized in that, The lower-layer electric vehicle aggregator charging and discharging decision model is specifically: The objective function with the maximum operation profit of the electric vehicle aggregator as the goal is expressed by the formula: where C EV represents the operating profit of the electric vehicle aggregator; The operation constraint of the electric vehicle aggregator is expressed by the formula: In the formula, represents the upper limit of the equivalent charging power of the electric vehicle cluster at the k-th charging and discharging station; represents the upper limit of the equivalent discharging power of the electric vehicle cluster at the k-th charging and discharging station; represents the grid-connected and off-grid state of the k-th charging and discharging station at time t; represents the lower limit of the equivalent battery capacity of the electric vehicle cluster at the k-th charging and discharging station; represents the upper limit of the equivalent battery capacity of the electric vehicle cluster at the k-th charging station; represents the equivalent battery capacity of the electric vehicle cluster at the k-th charging and discharging station at time t; η c represents the charging coefficient of the electric vehicle; η d represents the discharging coefficient of the electric vehicle.

6. A dispatching method for an electric vehicle to be connected to the grid for an active distribution network according to claim 1, characterized in that The wind-solar power output uncertainty compensation model is specifically: Constructing an objective function for coping with the worst-case scenario, which is expressed by the formula: maxminC DS = ΔC DG + ΔC M + ΔC loss - ΔC L ; where, ΔC DG represents the operating cost of the wind-solar power unit under the worst-case scenario; ΔC M represents the interactive power purchase cost of the active distribution network under the worst-case scenario; ΔC loss represents the loss cost of the active distribution network under the worst-case scenario; ΔC L represents the load electricity consumption revenue under the worst-case scenario; The constraint condition is expressed by the formula: Wherein, represents the photovoltaic grid-connected power of the i-th node at time t; represents the predicted photovoltaic output of the i-th node at time t; represents the maximum predicted error of the photovoltaic of the i-th node at time t; represents the wind turbine grid-connected power of the i-th node at time t; represents the predicted wind turbine output of the i-th node at time t; represents the maximum predicted error of the wind turbine of the i-th node at time t; represents the positive-direction error auxiliary variable of the photovoltaic output of the i-th node at time t; represents the negative-direction error auxiliary variable of the photovoltaic output of the i-th node at time t; represents the positive-direction error auxiliary variable of the wind turbine output of the i-th node at time t; represents the negative-direction error auxiliary variable of the wind turbine output of the i-th node at time t; Γ PV represents the adjustable error parameter of the photovoltaic output; Γ WT represents the adjustable error parameter of the wind turbine output.

7. A dispatching method for an electric vehicle to be connected to the grid for an active distribution network according to claim 1, characterized in that, Using the KKT conditions and the duality theorem to solve the multi-agent optimal dispatch model, specifically: The KKT conditions transform the lower-layer electric vehicle aggregator charging and discharging decision model into the constraints of the upper-layer distribution network operator pre-dispatch model, and transform the objective functions of the first stage and the second stage into a mixed-integer linear programming problem. Using the duality theorem, the first stage is regarded as the main problem and the second stage is regarded as the sub-problem, specifically: The main problem is expressed by the formula: Wherein, λ represents an auxiliary variable for coping with uncertainty in the second stage; M represents a preset constant; l represents the index value of the l-th iteration, where when l = 1, λ = 0 represents no uncertainty scenario constraint, and when l > 1, represents the uncertainty scenario constraint newly added in the l-th iteration, and s represents the time period corresponding to the worst scenario found by the sub-problem; The sub-problem is expressed by the formula: The master problem and the subproblem are iteratively solved using a mixed-integer linear programming solver to obtain a candidate solution x for the master problem l , and a candidate solution λ for the subproblem l , where x l represents the candidate solution at the l-th iteration, and λ l represents the candidate solution of the subproblem at the l-th iteration; The iterative solution process is specifically based on the current candidate solution x of the master problem l , and find the wind-solar output error scenario λ that maximizes the sub-problem cost l , that is, the candidate worst-case scenario found by the sub-problem; until the maximum number of iterations is reached or the sub-problem converges. The sub-problem converges specifically as follows: if λ l > 0, then add the initial uncertainty scenario constraint to the constraints of the master problem and re-solve the master problem, where represents the photovoltaic unit output power at time t corresponding to the l-th iteration, represents the wind turbine unit output power at time t corresponding to the i-th iteration; if λ l ≤ 0, stop the iteration, the sub-problem converges to obtain the worst-case scenario found by the sub-problem, that is, the master problem converges to obtain the optimal solution of the master problem, that is, obtain the scheduling plan of the electric vehicle grid connection to the active distribution network.

8. An economic dispatching system for an electric vehicle connected to the grid to the active distribution network, characterized in that, The system includes a data acquisition module, a model construction module, a solution module, and a result output module, where: The data acquisition module is used to obtain the operation data of the electric vehicle; and transmit the operation data to the model construction module; The model construction module is used to construct a dispatchable model for the electric vehicle cluster based on the operation data; construct a multi-agent optimal dispatch model based on the dispatchable model of the electric vehicle cluster, where the multi-agent optimal dispatch model includes a first stage and a second stage; The first stage is divided into an upper-layer distribution network operator pre-scheduling model and a lower-layer electric vehicle aggregator charging and discharging decision-making model; among them, the upper-layer distribution network operator pre-scheduling model includes an objective function with the minimum total operating cost of the distribution network as the goal, and constraint conditions with branch power flow constraints and safe operation constraints as the constraints; the lower-layer electric vehicle aggregator charging and discharging decision-making model includes an objective function with the maximum operating profit of the electric vehicle aggregator as the goal, and constraint conditions with the operating constraints of the electric vehicle aggregator as the constraints; the second stage constructs a wind-solar power output uncertainty compensation model; The solving module is used to solve the multi-agent optimal scheduling model by using the KKT conditions and the duality theorem, and obtain the scheduling plan of the electric vehicle grid connection to the active distribution network; The result output module is used to display the economic scheduling plan of the electric vehicle grid connection to the active distribution network.

9. An electronic device, comprising a memory, a processor, and a computer program stored on the memory and executable on the processor, characterized in that, When the processor executes the program, it implements a scheduling method for the grid connection of electric vehicles to the active distribution network according to any one of claims 1 to 7.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the program is executed by the processor, it implements a scheduling method for the grid connection of electric vehicles to the active distribution network according to any one of claims 1 to 7.

Images