A multi-agent collaborative planning method and medium for distributed power sources

Abstract

This invention provides a distributed generation multi-agent collaborative planning method and medium, including: (1) constructing a DG multi-agent collaborative planning model: the distribution network structure, branch data information and node data information of a certain actual power grid node; (2) based on the distribution network model in step S1, establishing optimization objective functions with the goal of maximizing the revenue of each market entity; (3) setting optimization constraints for market entities DG operators, power users and DNOs; (4) combining the objective functions and constraints of steps S2 and S3 to construct a distribution network DG multi-agent game collaborative optimization planning model; (5) using a solver to solve the distribution network DG multi-agent game collaborative optimization planning problem constructed in step S4: obtaining the optimal DG planning scheme.

Description

Technical Field

[0001] This invention relates to the field of power systems, and specifically to a multi-entity collaborative planning method and medium for distributed power sources. Background Technology

[0002] The integration of clean, environmentally friendly, and flexible distributed generation (DG) power sources into the main power grid has become a popular trend. While the addition of new electricity retailers and the implementation of various demand response projects have provided abundant power allocation resources, they have also brought about problems such as reduced system reliability, insufficient power absorption, and overvoltage. In the current electricity market environment, DG operators, distribution network operators, and users have formed an orderly but non-cooperative game relationship. If the conflicts between the generation and supply sides, and between the supply and user sides can be reasonably coordinated, the resources of both "source" and "load" can be fully utilized to improve the sustainable development of the power grid. Therefore, researching multi-stakeholder collaborative planning methods for DG applied to distribution networks, rationally planning the location and capacity of DG, deeply considering the actual needs of electricity users, and balancing and optimizing the interests of various market players is of great significance. Summary of the Invention

[0003] To address the shortcomings of existing technologies, this invention proposes a multi-agent collaborative planning method and medium for distributed power sources.

[0004] The objective of this invention can be achieved through the following technical solutions:

[0005] A multi-agent collaborative planning method for distributed power sources includes the following steps:

[0006] Step S1: Construct a multi-agent collaborative planning model for distributed power sources;

[0007] Step S2: Establish optimization objective functions for distributed power generation operators, power users, and distribution network operators, respectively, with the goal of maximizing the revenue of each market participant.

[0008] Step S3: Set optimization constraints for market participants, including distributed power generation operators, power users, and distribution network operators;

[0009] Step S4: Based on the objective function and optimization constraints, construct a multi-agent game-theoretic collaborative optimization planning model for distributed power sources in the distribution network;

[0010] Step S5: Solve the multi-agent game-theoretic collaborative optimization planning model of distributed generation in the distribution network to obtain the optimal planning scheme for distributed generation.

[0011] Furthermore, in step S1, a multi-agent collaborative planning model for distributed power sources is constructed:

[0012] The actual distribution network structure includes the connection relationships between nodes of the distribution network; the branch data information includes the impedance of the network lines; the node data information includes the load size of each node of the distribution network, the capacity of the distributed power supply equipment, and the installation location.

[0013] Furthermore, the objective function of the distributed power operator includes: distributed power sales revenue, initial investment in distributed power construction, and operation and maintenance expenses of distributed power, as follows:

[0014]

[0015] q es =f eb2.0 -μ1·P i.DG (2)

[0016] In the formula, Ω is the set of distributed power sources to be connected to; q es It is the unit electricity price of distributed power sources; P i . DG It is the actual active power provided by the distributed power source at bus i, and P i.DG With q es Negative correlation; μ1 is a fixed parameter value; f eb2.0 It refers to the feed-in tariff for distributed power generation; q om q in These are the cost coefficients for distributed power source operation, maintenance, and construction investment; P ZJ.DG It is the installed capacity of distributed power sources at bus i.

[0017] Furthermore, based on the differentiated load demands, loads are divided into friendly loads that meet basic user requirements and peak loads that meet the maximum user demands, i.e., unfriendly loads. Load data from the distribution network over a year is statistically analyzed for forecasting, and the month with the middle load volume is selected as the reference value P. f0 Calculate the monthly load and P f0 The deviation; if the detected load deviation value ΔP fi Greater than If it is a non-friendly load, then it is a friendly load; otherwise, it is a friendly load.

[0018] Furthermore, the power user objective function W US This includes electricity cost savings after users participate in demand-side response, interruption load compensation costs, and user satisfaction (W). C.US The objective function is as follows:

[0019]

[0020]

[0021]

[0022] In the formula, γ eb It is the price that users pay to the power distribution network operator; t d is the actual electricity consumption of the user at time t; t Let α represent the user's electricity demand at time t; t = 1, ..., N represents the time period; parameter α t The value is negative, and it is related to the elasticity of electricity prices; parameter β t A positive value indicates a different load type; γ bc It is the user interruption load compensation cost coefficient; the user's planned interruption load amount and P ZJ.DG Positive correlation; μ2 is a fixed parameter value.

[0023] Furthermore, the objective function of the distribution network operator includes: electricity sales revenue, line loss and fault repair expenses, electricity purchase expenses from the upper-level grid, electricity purchase expenses from distributed power generation operators, contractual obligation deposits for contracts with distributed power generation operators, and user satisfaction W. C.US The objective function is as follows:

[0024]

[0025] q bz =μ3·P ZJ.DG (7)

[0026] In the formula, γ cb It is the line loss and fault repair expenditure coefficient for distribution network operators; f eb1 Electricity purchase price from the main grid; P GM.DG It refers to the electricity purchased from distributed power sources under contracts between distribution network operators and distributed power generation operators; q bz It is the economic loss coefficient suffered by the distribution network operator due to the failure of the distributed power generation operator to fulfill the power generation commitment stipulated in the contract; μ3 is a fixed parameter value.

[0027] A computer-readable storage medium storing computer instructions that, when executed, enable the above-described planning method.

[0028] The beneficial effects of this invention are:

[0029] Based on the theory of non-cooperative complete information game theory, a multi-agent collaborative planning method for distributed generation is proposed, enabling the formulation of location and capacity schemes for distributed generation. Distributed generation operators, distribution network operators, and users continuously optimize their decisions and interact in the game process, formulating the optimal location and capacity scheme. This improves their own revenue while balancing the revenue of all market participants. While meeting the economic benefits for users, it also comprehensively considers the issues of power comfort and satisfaction, promoting the healthy development of the distribution network, stimulating the activity of the electricity market, and meeting the requirements of healthy market competition and economic efficiency. Attached Figure Description

[0030] The invention will now be further described with reference to the accompanying drawings.

[0031] Figure 1 This is a flowchart illustrating the present invention.

[0032] Figure 2 This is the power distribution network diagram used in the implementation examples of this invention, where PV is the photovoltaic installation point and WT is the wind turbine installation point. Detailed Implementation

[0033] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0034] In some examples of this invention, a multi-agent cooperative planning method for distributed power sources is disclosed, which may include the following steps:

[0035] Step S1: Construct a multi-agent collaborative planning model for distributed power sources

[0036] A real distribution network structure includes the connection relationships between the nodes of the distribution network; branch data information specifically refers to the impedance of the network lines; node data information specifically includes the load size of each node of the distribution network, the capacity of the DG equipment, and the installation location.

[0037] Step S2: Establish the objective function for each market participant

[0038] The traditional monopolistic power distribution and sales model of distribution network operators (DNOs) is gradually transitioning to a new power distribution and sales model with the participation of new market players. DNOs purchase electricity from the main grid and distributed energy operators, and then supply power to loads through the distribution network.

[0039] Optimization objective functions are established for distributed power generation operators, power users, and distribution network operators, respectively. Step S2 comprises three steps, as follows:

[0040] Step S21: The objective function of the planning model for distributed power operators mainly includes: distributed power sales revenue, initial construction investment in distributed power, and operation and maintenance expenses of distributed power. The objective function is as follows:

[0041]

[0042] q es =f eb2.0 -μ1·P i.DG (2)

[0043] In the formula, Ω is the set of distributed power sources to be connected to; q es It is the unit electricity price of distributed power sources; P i.DG It is the actual active power provided by the distributed power source at bus i, and P i.DG With q es Negative correlation; μ1 is a fixed parameter value; f eb2.0 It refers to the feed-in tariff for distributed power generation; q om q in These are the cost coefficients for distributed power source operation, maintenance, and construction investment; P ZJ.DG It is the installed capacity of distributed power sources at bus i.

[0044] Step S22: For electricity users, the goal is to reduce electricity costs by adjusting their electricity usage plans. Based on differentiated load demands, loads are categorized into "friendly" loads (those meeting basic user requirements) and "peak" loads (those meeting maximum user demands), i.e., "unfriendly" loads. Based on distribution network experience data, load data for the entire year is statistically analyzed for forecasting, and the month with the highest load is selected as the reference value P. f0 Calculate the monthly load and P f0 The deviation. If the detected load deviation value ΔP fi Greater than If it is a non-friendly load, then it is a friendly load; otherwise, it is a friendly load.

[0045] The objective function of the user benefit model is W. US This includes electricity cost savings after users participate in demand-side response, interruption load compensation costs, and user satisfaction (W). C.US The objective function is as follows:

[0046]

[0047]

[0048]

[0049] In the formula, γ eb It is the price that users pay to the power distribution network operator; t d is the actual electricity consumption of the user at time t; t Let α represent the user's electricity demand at time t; t = 1, ..., N represents the time period; parameter α t The value is negative, and it is related to the elasticity of electricity prices; parameter β t A positive value for β indicates a different load type (if the load is a friendly load, β...). t Take 200; if the load is an unfriendly type of load, β t Take 100); γ bc It is the user interruption load compensation cost coefficient; the user's planned interruption load amount and P ZJ.DG Positive correlation; μ2 is a fixed parameter value.

[0050] Step S23: Compared to distribution network operators, the goal is to reduce expenditures on transmission line losses, initial construction, and purchasing electricity from upstream suppliers, while increasing electricity sales revenue to maximize profits.

[0051] The objective function W of the distribution network operator planning model DN It mainly consists of the following parts: electricity sales revenue, line loss and fault repair expenses, electricity purchase expenses from the upper-level power grid, electricity purchase expenses from distributed power operators, contractual guarantee deposits for distributed power contracts, and user satisfaction W. C.US The objective function is as follows:

[0052]

[0053] q bz =μ3 · P ZJ.DG (7)

[0054] In the formula, γ cb It is the coefficient for line loss and fault repair expenditure of the distribution network operator; f eb1 Electricity purchase price from the main grid; P GM.DG It refers to the electricity purchased from distributed power sources under contracts between distribution network operators and distributed power generation operators; q bz It is the economic loss coefficient suffered by the distribution network operator due to the failure of the distributed power generation operator to fulfill the power generation commitment stipulated in the contract; μ3 is a fixed parameter value.

[0055] Step S3: Set optimization constraints for each market participant

[0056] Optimization constraints are set for distributed power generation operators, power users, and distribution network operators, respectively. Step S3 includes three steps, as follows:

[0057] Step S31: The investment constraints for distributed power operators are shown in the following formula:

[0058] 0≤P ZJ.DG ≤P ZJ.max.DG (8)

[0059] In the formula, P ZJ.max.DG It represents the maximum capacity of devices that can be connected to the distributed power source at node i.

[0060] Step S32: User constraints are as follows:

[0061] The user's minimum load demand during time period t must be met, i.e., the actual electricity consumption l t It should be greater than the uninterruptible load. t.un Less than the upper limit of power generation g t.max With the maximum load demand l t.max As shown below:

[0062] l t.un ≤l t ≤min(l t.max ,g t.max (9)

[0063] The actual interruptible load of the user must be less than the maximum interruptible load, that is:

[0064]

[0065] In the formula, δ is the maximum value of the interruptible load proportionality coefficient.

[0066] Step S33: The constraints of the distribution network operator's planning model are: power flow constraints of the distribution network, active power constraints of each line, and node voltage limits.

[0067]

[0068]

[0069] In the formula, P i With Q i These represent the active and reactive power injections at node i, respectively; U i with U j These are the voltage magnitudes at nodes i and j, respectively; G ij With B ij These are the conductance and susceptance of branch ij, respectively; O ij U represents the voltage phase angle difference between nodes i and j; j∈i represents the node adjacent to node i; i.min with U i.max These are the minimum and maximum values ​​of the voltage amplitude at node i, respectively; P ij.t With P ij.maxThese represent the power flowing through the line between nodes i and j, and their maximum values.

[0070] Step S4: Construct a multi-agent game-theoretic collaborative optimization planning model for distributed power sources in the distribution network.

[0071] First, the distribution network operator formulates a power purchase strategy based on the distributed generation planning decision results, and calculates the maximum value of the user-side objective function, letting the power user objective function W... US The first derivative is 0, as shown below:

[0072]

[0073] Find:

[0074]

[0075] Taking the second derivative of both sides of the equation, we get...

[0076]

[0077] Because of parameter α t The value is negative, parameter β t If the value is positive, then the Hessien matrix is ​​a negative definite matrix, and the user objective function W... US In l t * That is, take the maximum value, i.e., l t * This represents the optimal electricity consumption decision for power users under a known distributed power generation planning decision.

[0078] Then based on the obtained l t * Seeking the optimal power purchase decision for distribution network operators, such that the objective function W of the distribution network operators is satisfied. DN To obtain the maximum value, substitute equation (14) into W. DN have to:

[0079]

[0080] Step S5: Solve the multi-agent game-theoretic collaborative optimization planning problem of distribution network DG using a solver.

[0081] Combining the objective function and constraints established in steps S2 and S3, and for the multi-agent game-theoretic collaborative optimization planning model of distributed power sources in the distribution network constructed in step S4, the CPLEX solver is called to obtain the optimal planning scheme for distributed power sources.

[0082] In a specific embodiment of this application, a simulation analysis is performed using a real power distribution network as an example, as follows: Figure 2As shown, the planned operating period is 10 years, the discount rate is 0.1, the annual load growth rate of the system is 10%, and the simulated operating scenario is based on 24-hour forecast data of typical days in the four seasons. The relevant parameters of the distributed power source are shown in Table 1 below.

[0083] The parameters for wind, solar, and energy storage are shown in Table 1-3 below:

[0084] Table 1 Distributed Power Supply Parameters

[0085]

[0086]

[0087] Based on the load characteristics within the planning area, the load is divided into commercial users and general users. Commercial users account for 20% of the total users, and the distribution is shown in Table 2 below.

[0088] Table 2 User Distribution

[0089] Business users 3，4，9 Regular users 2，5，6，7，8，10，11，12

[0090] The randomness and volatility of distributed power generation output, along with the peak-valley difference in electricity demand, not only waste power generation resources but also increase power supply costs. To address the differentiated needs of users, time-of-use pricing is implemented to encourage users to develop reasonable electricity consumption plans. The actual time-of-use pricing situation in the region is shown in Table 3 below.

[0091] Table 3 Time-of-use electricity pricing information for commercial and general users

[0092]

[0093] Based on the load demand of each node, and considering the interests of distributed power sources, distribution network operators, and users, the installation location and capacity configuration are determined to ensure the safe and economical operation of the distribution network. The specific results are shown in Table 4 below.

[0094] Table 4 Game Theory Optimization Scheme

[0095]

[0096] In some examples of this invention, a computer-readable storage medium storing instructions is also involved. When executed, the instructions enable the distributed power multi-agent collaborative planning method in any of the above examples. More specifically, the instructions can be a computer-readable language. The computer described above can be a general-purpose computer device or a special-purpose computer device. In specific implementations, the computer can be a desktop computer, a portable computer, a network server, a PDA (Personal Digital Assistant), a mobile phone, a tablet computer, a wireless terminal device, a communication device, or an embedded device. The storage medium can be any available medium accessible to the computer or a data storage device such as a server or data center that integrates one or more available media. For example, the storage medium is, but is not limited to, magnetic media (e.g., floppy disks, hard disks, magnetic tapes), optical media (e.g., Digital Versatile Discs (DVDs)), or semiconductor media (e.g., solid-state drives (SSDs)).

[0097] In the description of this specification, references to terms such as "an embodiment," "example," "specific example," etc., indicate that a specific feature, structure, material, or characteristic described in connection with that embodiment or example is included in at least one embodiment or example of the invention. In this specification, illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples.

[0098] The foregoing has shown and described the basic principles, main features, and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited to the above embodiments. The embodiments and descriptions in the specification are merely illustrative of the principles of the invention. Various changes and modifications can be made to the invention without departing from its spirit and scope, and all such changes and modifications fall within the scope of the claimed invention.

Claims

1. A multi-agent collaborative planning method for distributed power sources, characterized in that, Includes the following steps: Step S1: Construct a multi-agent collaborative planning model for distributed power sources; Step S2: Establish optimization objective functions for distributed power generation operators, power users, and distribution network operators, respectively, with the goal of maximizing the revenue of each market participant. Step S3: Set optimization constraints for market participants, including distributed power generation operators, power users, and distribution network operators; Step S4: Based on the objective function and optimization constraints, construct a multi-agent game-theoretic collaborative optimization planning model for distributed power sources in the distribution network; Step S5: Solve the multi-agent game-theoretic collaborative optimization planning model of distributed generation in the distribution network to obtain the optimal planning scheme for distributed generation; The objective function of the distributed power operator includes: distributed power sales revenue, initial investment in distributed power construction, and operation and maintenance expenses of distributed power. The objective function is as follows: (1) (2) In the formula, Ω is the set of distributed power supply nodes to be connected; q es It is the unit electricity price of distributed power sources; P i . DG It is a busbar i The actual active power provided by the distributed power source, and P i . DG and q es Negative correlation; μ 1 is a fixed parameter value; f eb2.0 It refers to the grid connection price of distributed power generation; q om , q in These are the cost coefficients for the operation, maintenance, and construction investment of distributed power sources; P ZJ . DG It is a busbar i Distributed power generation capacity; The objective function of the electricity user W US This includes electricity cost savings after users participate in demand-side response, interruption load compensation costs, and user satisfaction. W C.US The objective function is as follows: (3) (4) (5) In the formula, It is the price that users pay to the power distribution network operator; It is the first t The user's actual electricity consumption at any given time; For the first t Real-time user power consumption; t =1, ..., N Indicates a time period; parameters The value is negative, which is related to the elasticity of electricity prices; parameter A positive value indicates a different load type; It is the user interruption load compensation cost coefficient; the user's planned interruption load amount and P ZJ . DG Positive correlation; For fixed parameter values; The objective function of the distribution network operator includes: electricity sales revenue, line loss and fault repair expenses, electricity purchase expenses from the upper-level grid, electricity purchase expenses from distributed power generation operators, contractual obligation deposits with distributed power generation companies, and user satisfaction. W C.US The objective function is as follows: (6) (7) In the formula, It is the line loss and fault repair expenditure coefficient of the distribution network operator; The price of electricity purchased from the main grid; P GM . DG It refers to the amount of electricity purchased from distributed power sources under the contract between the distribution network operator and the distributed power source operator. It is the economic loss coefficient suffered by the distribution network operator due to the failure of the distributed power generation operator to fulfill the power generation commitments stipulated in the contract; It is a fixed parameter value.

2. The distributed power source multi-agent collaborative planning method according to claim 1, characterized in that, In step S1, a multi-agent collaborative planning model for distributed power sources is constructed: The actual distribution network structure includes the connection relationships between various nodes in the distribution network; branch data information includes the impedance of network lines; node data information includes the load size of each node in the distribution network, the capacity of distributed power generation equipment, and the installation location.

3. The distributed power source multi-agent collaborative planning method according to claim 1, characterized in that, Based on differentiated load demands, loads are categorized into friendly loads that meet basic user requirements and peak loads that meet the maximum user demand, i.e., unfriendly loads. Load data from the distribution network over a year is used for forecasting, with the month with the middle load volume selected as a reference value. Calculate the monthly load and The deviation; if the detected load deviation value Δ P fi Greater than If it is a non-friendly load, then it is a friendly load; otherwise, it is a friendly load.

4. A computer-readable storage medium storing computer instructions, characterized in that, When the instruction is executed, it can realize the planning method described in any one of claims 1 to 3.

Images