Power plant output control method and system based on multi-agent game

Abstract

The invention discloses a power plant output control method and system based on a multi-agent game. The method comprises the following steps: collecting historical data information of a game agent; establishing a multi-subject game model according to the historical data information; game playing is carried out by taking the lowest loss and the multi-main-body capacity balance as targets; and under the constraint of each main body, generating a control strategy of each main body. And the overall loss of the power system is effectively reduced. And the operation efficiency and the stability of the power system are improved. When the operation condition of the power plant changes, the control strategy can be quickly updated, and the response speed and flexibility of the system are enhanced. On the whole, the economical efficiency and the reliability of the power system are remarkably improved.

Description

Technical Field

[0001] The present invention relates to the technical field of power plant output control, and in particular to a power plant output control method and system based on multi-agent game. Background Art

[0002] In modern power systems, the balance between power production and consumption is the key to ensuring the stable operation of the power system. As the main power supply entity of the power system, power plants need to dynamically adjust their power generation (output) according to the needs of the power grid. With the introduction of renewable energy, the increasing volatility of load demand, and the complexity of the power market mechanism, traditional power plant output control methods face increasing challenges.

[0003] Traditional power plant output control methods are usually based on the optimization of a single power plant, lacking consideration of the synergy of multiple power plants in the entire power system. This approach may cause some power plants to operate at overload while other power plants are underutilized, leading to increased overall losses. In addition, the output adjustment of each power plant in the power system often relies on the instructions of the dispatching center rather than autonomous decision-making, which is slow to respond and lacks flexibility when dealing with emergencies (such as power plant failures or sudden load increases). Summary of the invention

[0004] In view of the above-mentioned problems, the present invention is proposed.

[0005] Therefore, the technical problem solved by the present invention is that the existing power plant output control method of multi-agent game has the problems of large power consumption and high cost.

[0006] In order to solve the above technical problems, the present invention provides the following technical solutions: a power plant output control method based on multi-agent game, comprising:

[0007] Collect historical data information of game entities;

[0008] According to the historical data information, a multi-agent game model is established;

[0009] The game is played with the goal of achieving minimum loss and balanced production capacity of multiple entities;

[0010] Under the constraints of each subject, the control strategy of each subject is generated.

[0011] As a preferred solution of the power plant output control method based on multi-agent game described in the present invention, wherein: the game subject includes each power plant under the regulation of the operator;

[0012] The historical data information includes power generation cost, power selling price, output, and loss during power distribution.

[0013] As a preferred solution of the power plant output control method based on multi-agent game described in the present invention, the multi-agent game model includes constructing a fitting model of output and loss in each power plant; and allocating output demand in each power plant according to the fitted model;

[0014] The fitting model includes fitting using multiple regression. After one fitting, a fitting curve is obtained, and the difference between the actual loss and the fitting loss of each data point is calculated:

[0015] ∈ i (P i )=|L i (P i )-L' i (P i )|

[0016] Among them, ∈ i (P i ) indicates that the output of the i-th power plant is P i The loss deviation, L i (P i ) indicates that the output of the i-th power plant is P i The actual loss, L' i (P i ) indicates that the output of the i-th power plant is P i The prediction loss, P i Indicates the input output data;

[0017] Set the deviation threshold ∈ threshold To judge as abnormal data, remove abnormal data points whose deviation exceeds the threshold:

[0018]

[0019] in, represents the data set of the ith power plant after removing abnormal data, represents the historical data set of the i-th power plant;

[0020] Using the removed data Refitting, repeating the process of eliminating abnormal data and refitting until no new abnormal data points are generated; after terminating the loop, output the fitting model L of the i-th power plant i , and integrate all the excluded data into Ai;

[0021] Among them, Ai represents the set of all excluded data in the i-th power plant;

[0022] Count the probability P of abnormal data in the i-th power plant and the mode Z in Ai;

[0023] For the fitted model L i Add interference terms to get the final fitting model L i '=L i +τ;

[0024] Among them, τ represents the interference term, the probability of P is Z, and the probability of 1-P is 0.

[0025] As a preferred solution of the power plant output control method based on multi-agent game described in the present invention, wherein: the minimum loss and the balance of production capacity of multiple agents are taken as the goals, and the goal of minimizing the sum of the losses is expressed as:

[0026]

[0027] Among them, M represents the number of game entities;

[0028] Consider the output P of each power plant i and expected output Define the equilibrium error, and the goal of multi-agent capacity equilibrium is expressed as:

[0029]

[0030] The combined optimization objective of the two objectives is expressed as:

[0031]

[0032] Among them, λ 1 and λ 2 is the weight coefficient, which sums to 1; represents the objective function, Minimize represents the optimization target as the minimum value; P M Represents the output of the Mth power plant.

[0033] As a preferred solution of the power plant output control method based on multi-agent game described in the present invention, the constraints of each subject include ensuring that the total output of all power plants meets the needs of the power system:

[0034]

[0035] Where D is the total demand of the system;

[0036] The output of each power plant cannot exceed its maximum capacity and cannot fall below its minimum capacity:

[0037]

[0038] in, represents the maximum capacity of power plant i, represents the minimum capacity of power plant i.

[0039] As a preferred solution of the power plant output control method based on multi-agent game described in the present invention, wherein: by taking the minimum loss and the production capacity balance of multiple subjects as the game objectives under the constraint conditions of each subject, the output of each subject is optimized;

[0040] First, under the constraints of each subject, the output is randomly assigned to each subject;

[0041] Predicting the loss of each subject using the fitted model;

[0042] The output of each subject is adjusted according to the optimization target until the optimization target is achieved.

[0043] As a preferred solution of the power plant output control method based on multi-agent game described in the present invention, the control strategy includes taking the game result as the target output value of the power plant, and controlling the output of the generator set in actual operation to achieve the target output value;

[0044] If an accident occurs in power plant i, resulting in a change in the capacity limit, the maximum capacity is reduced according to the uploaded reduction ratio based on the accident situation, and the constraints on power plant i are updated and the control strategy is regenerated.

[0045] A power plant output control system based on multi-agent game using the method of the present invention is characterized by:

[0046] A collection unit collects historical data information of game entities;

[0047] An analysis unit, which establishes a multi-agent game model based on the historical data information;

[0048] The game unit plays the game with the goal of minimum loss and balanced production capacity of multiple entities;

[0049] The control unit generates control strategies for each subject under the constraints of each subject.

[0050] A computer device comprises: a memory and a processor; the memory stores a computer program, wherein the processor implements the steps of any one of the methods of the present invention when executing the computer program.

[0051] A computer-readable storage medium stores a computer program, wherein the computer program, when executed by a processor, implements the steps of any one of the methods of the present invention.

[0052] Beneficial effects of the present invention: The power plant output control method based on multi-agent game provided by the present invention accurately describes the complex relationship between power plant output and loss based on the multivariate regression fitting model of historical data, thereby effectively reducing the overall loss of the power system. Secondly, by taking the minimum loss and production capacity balance as the game goals, each power plant can automatically adjust the output without relying on central dispatch, thereby improving the operating efficiency and stability of the power system. In addition, the present invention also has dynamic adjustment capabilities, and can quickly update the control strategy when the operating conditions of the power plant change, thereby enhancing the response speed and flexibility of the system. Overall, the present invention significantly improves the economy and reliability of the power system and has broad application prospects. BRIEF DESCRIPTION OF THE DRAWINGS

[0053] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other accompanying drawings can be obtained based on these accompanying drawings without paying creative work.

[0054] Figure 1 An overall flow chart of a power plant output control method based on multi-agent game provided for the first embodiment of the present invention. DETAILED DESCRIPTION

[0055] In order to make the above-mentioned purposes, features and advantages of the present invention more obvious and easy to understand, the specific implementation methods of the present invention are described in detail below in conjunction with the drawings of the specification. Obviously, the described embodiments are part of the embodiments of the present invention, but not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary persons in the art without creative work should fall within the scope of protection of the present invention.

[0056] Example 1, reference Figure 1 , which is an embodiment of the present invention, provides a power plant output control method based on multi-agent game, comprising:

[0057] S1: Collect historical data information of game entities.

[0058] The game subjects include various power plants under the regulation of operators; the historical data information includes power generation costs, electricity sales prices, output conditions, and losses during power distribution.

[0059] S2: Establish a multi-agent game model based on the historical data information.

[0060] The multi-agent game model includes constructing a fitting model of output and loss in each power plant; and allocating output demand in each power plant according to the fitted model.

[0061] The fitting model includes fitting using multiple regression. After one fitting, a fitting curve is obtained, and the difference between the actual loss and the fitting loss of each data point is calculated:

[0062] ∈ i (P i )=|L i (P i )-L' i (P i )|

[0063] Among them, ∈ i (P i ) indicates that the output of the i-th power plant is P i The loss deviation, L i (P i ) indicates that the output of the i-th power plant is P i The actual loss, L' i (P i ) indicates that the output of the i-th power plant is P i The prediction loss, P i Indicates the input output data.

[0064] Set the deviation threshold ∈ threshold To judge as abnormal data, remove abnormal data points whose deviation exceeds the threshold:

[0065]

[0066] in, represents the data set of the ith power plant after removing abnormal data, Represents the historical data set of the i-th power plant.

[0067] Using the removed data Refitting, repeating the process of eliminating abnormal data and refitting until no new abnormal data points are generated; after terminating the loop, output the fitting model L of the i-th power plant i , and all the excluded data are combined into Ai. Where Ai represents the set of all excluded data in the i-th power plant.

[0068] Count the probability P of abnormal data in the i-th power plant and the mode Z in Ai.

[0069] For the fitted model L i Add interference terms to get the final fitting model L i '=L i+τ. τ represents the interference term, the probability of P is Z, and the probability of 1-P is 0.

[0070] It should be noted that by eliminating abnormal data and reducing the fitting error caused by noise or abnormal points, the final fitting model can better reflect the true relationship between power and loss. By statistically calculating the probability of abnormal data and introducing interference terms in the fitting model, the model can not only cope with normal operation, but also consider possible emergencies in actual operation (such as equipment failure or data noise), thereby improving the performance of the model in uncertain environments. Through more accurate loss prediction, the output distribution of each power plant is optimized, loss minimization and capacity balance are achieved, thereby improving the operating efficiency and economic benefits of the power system.

[0071] S3: Take the minimum loss and the balance of production capacity of multiple agents as the goal to conduct a game. Under the constraints of each agent, generate the control strategy of each agent.

[0072] Furthermore, minimum loss means minimizing the energy loss in the power generation, transmission and distribution process in the entire power system. This is the key goal to maximize the economy and efficiency of the power system. By minimizing the objective function of the total loss:

[0073]

[0074] Among them, M represents the number of game entities.

[0075] Capacity balancing means that under the premise of meeting electricity demand, the output of each power plant should be as close to its expected output as possible to avoid overloading or no-load operation of some power plants. Consider the output P of each power plant i and expected output Define the equilibrium error, and the goal of multi-agent capacity equilibrium is expressed as:

[0076]

[0077] The combined optimization objective of the two objectives is expressed as:

[0078]

[0079] Among them, λ 1 and λ 2 is the weight coefficient, which sums to 1; represents the objective function, Minimize represents the optimization target as the minimum value; P M Represents the output of the Mth power plant.

[0080] It should also be noted that the constraints of each subject include ensuring that the total output of all power plants meets the needs of the power system:

[0081]

[0082] Where D is the total demand of the system.

[0083] The output of each power plant cannot exceed its maximum capacity and cannot fall below its minimum capacity:

[0084]

[0085] in, represents the maximum capacity of power plant i, represents the minimum capacity of power plant i.

[0086] Furthermore, by taking the minimum loss and the balance of production capacity of multiple entities as the game objectives under the constraints of each entity, the output of each entity is optimized.

[0087] First, under the constraints of each subject, the output is randomly assigned to each subject. The loss of each subject is predicted through the fitting model. The output of each subject is adjusted according to the optimization target until the optimization target is achieved.

[0088] The control strategy includes taking the game result as the target output value of the power plant and controlling the output of the generator set in actual operation to achieve the target output value. If an accident occurs in power plant i, resulting in a change in the capacity limit, the maximum capacity is reduced according to the uploaded reduction ratio according to the accident situation, and the constraints on power plant i are updated, and the control strategy is regenerated.

[0089] On the other hand, this embodiment also provides a power plant output control system based on multi-agent game, which includes:

[0090] The collection unit collects historical data information of the game subject.

[0091] The analysis unit establishes a multi-agent game model based on the historical data information.

[0092] The game unit plays the game with the goal of minimizing losses and balancing the production capacity of multiple entities.

[0093] The control unit generates control strategies for each subject under the constraints of each subject.

[0094] If the above functions are implemented in the form of software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art or the part of the technical solution, can be embodied in the form of a software product. The computer software product is stored in a storage medium, including several instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk and other media that can store program codes.

[0095] The logic and / or steps represented in the flowchart or otherwise described herein, for example, can be considered as an ordered list of executable instructions for implementing logical functions, and can be embodied in any computer-readable medium for use by an instruction execution system, device or apparatus (such as a computer-based system, a system including a processor, or other system that can fetch instructions from an instruction execution system, device or apparatus and execute instructions), or in conjunction with such instruction execution systems, devices or apparatuses. For the purposes of this specification, "computer-readable medium" can be any device that can contain, store, communicate, propagate or transmit a program for use by an instruction execution system, device or apparatus, or in conjunction with such instruction execution systems, devices or apparatuses.

[0096] More specific examples of computer-readable media (a non-exhaustive list) include the following: an electrical connection with one or more wires (electronic device), a portable computer disk case (magnetic device), a random access memory (RAM), a read-only memory (ROM), an erasable and programmable read-only memory (EPROM or flash memory), an optical fiber device, and a portable compact disk read-only memory (CDROM). In addition, the computer-readable medium may even be a paper or other suitable medium on which the program is printed, since the program may be obtained electronically, for example, by optically scanning the paper or other medium, followed by editing, deciphering or, if necessary, processing in another suitable manner, and then stored in a computer memory.

[0097] It should be understood that the various parts of the present invention can be implemented by hardware, software, firmware or a combination thereof. In the above-mentioned embodiments, a plurality of steps or methods can be implemented by software or firmware stored in a memory and executed by a suitable instruction execution system. For example, if implemented by hardware, as in another embodiment, it can be implemented by any one of the following technologies known in the art or their combination: a discrete logic circuit having a logic gate circuit for implementing a logic function for a data signal, a dedicated integrated circuit having a suitable combination of logic gate circuits, a programmable gate array (PGA), a field programmable gate array (FPGA), etc.

[0098] Example 2 is an embodiment of the present invention, which provides a power plant output control method based on multi-agent game. In order to verify the beneficial effects of the present invention, scientific demonstration is carried out through economic benefit calculation and simulation experiments.

[0099] The experiment selected five power plants of different sizes as game entities, namely Power Plant A, Power Plant B, Power Plant C, Power Plant D and Power Plant E. These power plants operate in different geographical locations and power grid environments, and have their own power generation characteristics and cost structures. The experimental data includes power generation costs, electricity prices, historical output conditions and distribution losses. The specific process of the experiment is as follows:

[0100] First, the historical operating data of the five power plants in the past year were collected, including the output (in megawatts), losses (in megawatts), power generation costs (in yuan / MWh), and electricity prices (in yuan / MWh) of each power plant. The data was cleaned and preprocessed to remove obvious abnormal data points and normalized to ensure data consistency and comparability.

[0101] The relationship between the output and loss of each power plant was fitted using the multivariate regression method. After the initial fitting, the deviation between the actual loss and the predicted loss was calculated for each data point. Abnormal data points with deviations exceeding the threshold were removed, and the model was refitted after removal until the fitting results were stable and the deviations were within an acceptable range.

[0102] The experiment sets two optimization goals, namely, minimum loss and capacity balance. The minimum loss goal aims to minimize the total loss of the five power plants, while the capacity balance goal ensures that the actual output of each power plant is close to its expected output. Through multi-objective optimization, these two goals are combined and the weight coefficient λ is set 1 and λ 2 They are 0.6 and 0.4 respectively, to balance loss minimization and production capacity balance.

[0103] Under the condition of meeting the total output demand and the maximum / minimum capacity constraints of each power plant, the optimal output allocation strategy of each power plant is calculated by randomly assigning the initial output and using the game theory model. Each power plant is adjusted according to the loss value predicted by the fitting model until the optimization objective function reaches the minimum value. The final generated control strategy is used to guide the actual operation of the power plant.

[0104] During the experiment, the actual output, losses, power generation costs, electricity sales revenue and other parameters of each power plant were recorded. The experiment was repeated many times, simulating different power demands and market environments each time to ensure the reliability and universality of the results.

[0105] Table 1 Data record table

[0106]

[0107] It can be seen from the experimental data in the table that the operating efficiency of the five power plants has been significantly improved by implementing the power plant output control method based on multi-agent game. Specifically, the output of each power plant after optimization is higher than the initial allocated output. This is because the optimization model fully considers the power generation cost and electricity selling price of each power plant. On the premise of meeting the total demand of the system, it improves the output level of efficient power plants and reduces the over-generation of inefficient power plants. After optimization, the actual losses of each power plant were significantly reduced, and the loss reduction ratio ranged from 12.5% ​​to 25%. These results show that through reasonable output allocation, the power loss can be effectively reduced, thereby improving the overall economic benefits of the power system.

[0108] In addition, the optimized output adjustment also demonstrates a good capacity balancing effect. Experiments show that the output of each power plant is closer to its expected output value, thus avoiding overload or no-load operation of individual power plants. Compared with the traditional output allocation method, this invention not only takes into account the economic benefits of the power plant, but also achieves the dual goals of minimizing losses and balancing capacity through multi-objective optimization.

[0109] In the process of comparing existing technologies, traditional methods usually optimize based on a single goal (such as cost minimization), often ignoring the overall system loss and the rationality of capacity allocation. The present invention, by combining a multi-agent game model, not only fully considers the individual characteristics and system requirements of each power plant, but also introduces a multivariate regression fitting model based on historical data to ensure the accuracy and applicability of the model. At the same time, the dynamic adjustment mechanism in the experiment enables the method to update the output allocation strategy in a timely manner when the operating conditions of the power plant change, thereby enhancing the robustness and flexibility of the system.

[0110] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present invention, which should all be included in the scope of the claims of the present invention.

Claims

1. A power plant output control method based on multi-agent game, characterized in that: include: Collect historical data information of game entities; According to the historical data information, a multi-agent game model is established; The game is played with the goal of achieving minimum loss and balanced production capacity of multiple entities; Under the constraints of each subject, the control strategy of each subject is generated.

2. The power plant output control method based on multi-agent game as claimed in claim 1, characterized in that: The game subjects include various power plants under the regulation of the operator; The historical data information includes power generation cost, power selling price, output, and loss during power distribution.

3. The power plant output control method based on multi-agent game as claimed in claim 2, characterized in that: The multi-agent game model includes constructing a fitting model of output and loss in each power plant; and allocating output demand in each power plant according to the fitted model; The fitting model includes fitting using multiple regression. After one fitting, a fitting curve is obtained, and the difference between the actual loss and the fitting loss of each data point is calculated: ∈ i (P i )=|L i (P i )-L’ i (P i )| Among them, ∈ i (P i ) indicates that the output of the i-th power plant is P i The loss deviation, L i 9P i ) indicates that the output of the i-th power plant is P i The actual loss, L' i (P i ) indicates that the output of the i-th power plant is P i The prediction loss, P i Indicates the input output data; Set the deviation threshold ∈ threshold To judge as abnormal data, remove abnormal data points whose deviation exceeds the threshold: in, represents the data set of the ith power plant after removing abnormal data, represents the historical data set of the i-th power plant; Using the removed data Refitting, repeating the process of eliminating abnormal data and refitting until no new abnormal data points are generated; after terminating the loop, output the fitting model L of the i-th power plant i , and integrate all the excluded data into Ai; Among them, Ai represents the set of all excluded data in the i-th power plant; Count the probability P of abnormal data in the i-th power plant and the mode Z in Ai; For the fitted model L i Add interference terms to get the final fitting model L i '=L i +τ; Among them, τ represents the interference term, the probability of P is Z, and the probability of 1-P is 0.

4. The power plant output control method based on multi-agent game as claimed in claim 3 is characterized by: The goal of taking the minimum loss and the balance of production capacity of multiple agents as the objectives includes minimizing the sum of the losses as follows: Among them, M represents the number of game entities; Consider the output P of each power plant i and expected output Define the equilibrium error, and the goal of multi-agent capacity equilibrium is expressed as: The combined optimization objective of the two objectives is expressed as: Among them, λ1 and λ2 are weight coefficients, which add up to 1; represents the objective function, Minimize represents the optimization target as the minimum value; P M Represents the output of the Mth power plant.

5. The power plant output control method based on multi-agent game as claimed in claim 4, characterized in that: The constraints of each entity include ensuring that the total output of all power plants meets the needs of the power system: Where D is the total demand of the system; The output of each power plant cannot exceed its maximum capacity and cannot fall below its minimum capacity: in, represents the maximum capacity of power plant i, represents the minimum capacity of power plant i.

6. The power plant output control method based on multi-agent game as claimed in claim 5, characterized in that: By optimizing the output of each subject under the constraints of each subject, with the lowest loss and balanced production capacity of multiple subjects as the game objectives; First, under the constraints of each subject, the output is randomly assigned to each subject; Predicting the loss of each subject using the fitted model; The output of each subject is adjusted according to the optimization target until the optimization target is achieved.

7. The power plant output control method based on multi-agent game as claimed in claim 6, characterized in that: The control strategy includes taking the game result as the target output value of the power plant and controlling the output of the generator set in actual operation to achieve the target output value; If an accident occurs in power plant i, resulting in a change in the capacity limit, the maximum capacity is reduced according to the uploaded reduction ratio based on the accident situation, and the constraints on power plant i are updated and the control strategy is regenerated.

8. A power plant output control system based on multi-agent game using the method according to any one of claims 1 to 7, characterized in that: A collection unit collects historical data information of game entities; An analysis unit, which establishes a multi-agent game model based on the historical data information; The game unit plays the game with the goal of minimum loss and balanced production capacity of multiple entities; The control unit generates control strategies for each subject under the constraints of each subject.

9. A computer device comprising: Memory and processor; The memory stores a computer program, characterized in that: when the processor executes the computer program, the steps of the power plant output control method based on multi-agent game are implemented.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of a power plant output control method based on multi-agent game are implemented.

Images