A Multi-Source Cooperative Interaction Optimization Method and System Based on Capacity Domain

Abstract

This invention provides a multi-source collaborative interaction optimization method and system based on capacity domain, comprising: constructing a capacity demand domain for renewable energy considering the total output deviation of renewable energy over continuous periods; transforming the adjustment capabilities of each flexible resource within the system into corresponding constraints to construct a capacity adjustable domain reflecting the total adjustment capability of flexible resources over multiple periods; constructing a grid-source cluster interactive optimization potential game model and solving it using a distributed optimization algorithm and parallel optimization process; taking the sum of the capacity adjustable domains of all flexible resources in the system being no less than the total capacity demand domain of all renewable energy as a global constraint, adding it to the potential game model in the form of a penalty function to make the decision result feasible in all scenarios; improving the multi-source collaborative game process into a parallel process, enabling all game subjects to participate in optimization synchronously, and achieving collaborative interaction among various game subjects in the power grid through the combination of adjustable power domain and improved potential game, thus unifying individual optimization and global optimization.

Description

Technical Field

[0001] This invention relates to the field of multi-agent optimization scheduling technology, and in particular to a multi-source collaborative interaction optimization method and system based on capacity domain. Background Technology

[0002] With the gradual depletion of traditional energy sources and the booming development of renewable energy, developing and utilizing renewable energy to build a safe, efficient, clean, low-carbon, flexible, and intelligently integrated new power system has become an important way for power grids in various countries to achieve clean and green transformation. In 2022, the newly installed capacity and power generation of renewable energy in China reached 152 million kilowatts (accounting for 76.2% of the newly installed power generation capacity) and 2.7 trillion kilowatt-hours (accounting for 81% of the newly generated power generation), respectively. Under the trend of large-scale integration of renewable energy, appropriate operation optimization methods to ensure the safe, efficient, and economical operation of the power grid are a key link in the development and construction of new power systems.

[0003] However, in scenarios with a high proportion of renewable energy integration, the inherent randomness of renewable energy will pose challenges to the stable and reliable operation of the power system. Furthermore, with the development of distributed renewable energy and the electricity market, the entities responsible for the construction and management of renewable energy are no longer limited to the power grid. Entities with the necessary conditions can construct and manage renewable energy through application and approval processes, and participate as key players in power optimization scheduling and operation. Existing research considering the uncertainty of renewable energy mainly includes robust optimization, distributed robust optimization, and stochastic optimization. These methods effectively improve the economic efficiency of power grid operation under extreme scenarios or probabilistic conditions, but they cannot guarantee that the decision results can handle any realization value of random variables within the uncertain set of renewable energy, i.e., they cannot guarantee the feasibility of the decision results across all scenarios. Existing distributed optimization methods mainly include the ADMM algorithm and master-slave game theory, but existing methods either cannot avoid complex convergence proofs or require the design of a sequential optimization order, and they rarely consider the impact of uncertainty in distributed optimization algorithms. Therefore, they suffer from insufficient adaptability when facing complex power grid operation optimization scenarios with multiple homogeneous entities. Therefore, how to adopt appropriate optimization algorithms to achieve the combination of 1) the feasibility of system operation across all scenarios within the range of random fluctuations of renewable energy and 2) multi-agent distributed optimization to meet the needs of individual optimization and overall system optimization remains a topic that needs further exploration and development.

[0004] A search of existing technologies reveals a distributed scheduling method for multi-energy microgrids based on potential game theory (publication number CN112202206A). This method uses potential game theory to model the scheduling problem between different energy networks as a game problem between subnetworks, thereby achieving distributed optimal scheduling of multi-energy network systems with multi-agent characteristics. However, this invention does not consider the impact of renewable energy and its uncertainties on the elements and process of optimal scheduling. Furthermore, it does not consider the potential for uneven distribution of benefits among homogeneous agents that may result from using traditional potential game theory-based sequential optimization. Therefore, the performance of the proposed method needs further improvement. Summary of the Invention

[0005] To overcome the shortcomings of existing technologies, this invention proposes a multi-source collaborative interaction optimization method and system based on capacity domain. It constructs a capacity demand domain reflecting the total output deviation over multiple consecutive time periods based on historical scenarios of renewable energy, thus mapping the random fluctuation range of renewable energy. Conversely, from the perspective of flexible resources, it constructs a capacity adjustable domain for flexible resources under a certain output condition based on output constraints, thus mapping the adjustment capability range of flexible resources. The full coverage of the system's capacity demand domain by the system's capacity adjustable domain is used as a global constraint, ensuring that the optimization results meet the feasibility requirements of all scenarios within the random fluctuation range of renewable energy, guaranteeing the reliability and feasibility of system operation. Furthermore, a parallel potential game optimization method is proposed, modifying and improving the original potential game model and optimization process into a parallel game process to achieve fairness in the multi-party game.

[0006] The present invention is achieved by at least one of the following technical solutions.

[0007] A multi-source collaborative interaction optimization method based on capacity domain includes the following steps:

[0008] Based on the historical predicted output scenarios of renewable energy and the corresponding historical actual output scenarios, a capacity demand domain reflecting the output deviation of renewable energy over continuous periods is constructed.

[0009] Based on the constraint space of capacity-flexible resources, construct the capacity-adjustable domain of capacity-flexible resources;

[0010] Based on the constraint space of power-type flexible resources, a capacity adjustable domain for power-type flexible resources is constructed.

[0011] A multi-source collaborative interaction optimization model based on potential game theory is constructed, and the model is solved using a parallel optimization process.

[0012] Furthermore, a capacity demand domain reflecting the output deviation of renewable energy over continuous periods is constructed, specifically including:

[0013] For each renewable energy entity, a scenario pair is constructed based on historical predicted output scenarios and corresponding historical actual output scenarios. Each scenario pair is represented as follows:

[0014]

[0015]

[0016] In the formula, M represents the number of renewable energy entities, N represents the number of scenarios with concentrated application scenarios, and T represents the total number of optimization periods. For the m-th RES and the n-th historical predicted power scene dataset, For the m-th RES and the n-th historical real-world scenario dataset, It contributes to the prediction of the m-th RES and the n-th scene for the t-th time period. For the m-th RES and n-th scenario pair, the actual output at time t is given by the deviation rate of the actual output relative to the predicted output for each scenario pair:

[0017]

[0018] In the formula, Let the power deviation rate of the m-th RES and the n-th scenario be at time t. Considering all possible consecutive time periods within a day, calculate the total predicted power deviation to form the power deviation matrix for this scenario pair, which is expressed as:

[0019]

[0020] In the formula, Let m be the power deviation matrix of the m-th renewable energy subject and the n-th scenario pair. Let the total deviation of the predicted power output of the m-th renewable energy subject in the n-th scenario be expressed as:

[0021]

[0022] In the formula, Let be the planned output power of the m-th renewable energy source in time period b, Δt be the optimization time period length, i be the calculation start time period, and j be the calculation end time period. Considering the maximum value of the total power deviation, the capacity demand domain of renewable energy is constructed, expressed as:

[0023]

[0024] In the formula, This is the capacity demand matrix for the m-th renewable energy source. Let be the reduced capacity demand matrix for the m-th renewable energy source, and together with , they constitute the capacity demand domain for that renewable energy source. and Let $\frac{m}{j}$ represent the upward and downward capacity demands of the $m$-th renewable energy source from time period $i$ to time period $j$. The calculation process is as follows:

[0025]

[0026]

[0027] When the scenarios are sufficiently typical and the number is large enough, it is considered and It is the boundary of the predicted output deviation of the m-th renewable energy subject. By constructing the capacity demand domain, the boundary characteristics of the power deviation of each renewable energy subject in any continuous period are obtained.

[0028] Furthermore, the capacity-adjustable domain of capacity-based flexible resources is constructed, specifically including:

[0029] The charging and discharging results of capacity-type flexible resources over multiple time periods have a cumulative effect, which will be reflected in the remaining capacity and affect the regulation capability in subsequent time periods. Capacity-type flexible resources include energy storage, and the adjustable capacity domain of capacity-type flexible resources is expressed as:

[0030]

[0031] In the formula, ω cn,up With ω cn,dn These are the capacity capability matrices for increasing and decreasing capacity flexibility resources, respectively. Together, they constitute the adjustable capacity domain of capacity flexibility resources. and These represent the total upward and downward capacity available for capacity-type flexibility resources from time period i to time period j. The calculation process for each element in the matrix is ​​as follows:

[0032]

[0033]

[0034] In the formula, λ represents the charging and discharging efficiency of capacity-type flexible resources, and S max S min Let S(j) represent the maximum and minimum remaining capacity of the capacity-type flexible resource, and the remaining capacity of the capacity-type flexible resource at time j, respectively. For capacity-type flexible resources, the maximum charge / discharge power is given by P, where Δt is the optimization time period length. cn (b) represents the charging and discharging power of capacity-type flexibility resources during time period b, where discharging is positive and charging is negative.

[0035] Furthermore, the capacity adjustable domain of power-type flexible resources is constructed, specifically including:

[0036] Power-type flexibility resources do not have a cumulative effect across multiple time periods and will not affect the regulation capability in subsequent time periods. Capacity-type flexibility resources include gas turbines. The adjustable power output of power-type flexibility resources in each time period should meet the following requirements:

[0037]

[0038]

[0039] R≥[P G (t)+ΔP G,up (t)]-[P G (t-1)-ΔP G,dn [(t-1)]≥-R (14)

[0040] R≥(P G (t)-ΔP G,dn (t))-(P G (t-1)+ΔP G,up (t-1))≥-R (15)

[0041] ΔP G,up (t)≥0, ΔP G,dn (t)≥0 (16)

[0042] Equations (12) and (13) represent the output constraints of the adjustable power, and equations (14) and (15) represent the ramp rate constraints of the adjustable power. In these equations, P... G (t) represents the output power of the power-type flexible resource in time period t, ΔP G,up (t) and ΔP G,dn (t) represents the adjustable power and adjustable power of power-type flexibility resources during time period t, respectively. and These are the upper and lower limits of the output of the power-type flexible resource, respectively, and R is the ramp rate of the power-type flexible resource;

[0043] Based on the adjustable power range of power-type flexible resources at different times, the adjustable capacity domain of power-type flexible resources is constructed:

[0044]

[0045] In the formula, ω G,up With ω G,dn These are the up-adjustment and down-adjustment capacity capability matrices for power-type flexibility resources, which together constitute the adjustable capacity domain of power-type flexibility resources. and These represent the total upward and downward capacity that power-type flexibility resources can provide from time period i to time period j. The calculation process for each element in the matrix is ​​as follows:

[0046]

[0047]

[0048] Where Δt is the length of the optimization period.

[0049] Furthermore, the multi-source collaborative interaction optimization model includes three elements: players, payoff function, and strategy space.

[0050] Furthermore, the types of stakeholders include: renewable energy operators, capacity-type flexible resource operators, and power-type flexible resource operators.

[0051] Furthermore, the payoff function for the players is:

[0052] Power-type flexible resource operators as insiders:

[0053] The revenue function for power-type flexible resources includes operating costs and electricity sales revenue, namely:

[0054]

[0055] In the formula, F G(k) Let c be the revenue function for the k-th iteration of power-type flexibility resources. sell (t) represents the electricity price during time period t, a, b, and c are the generation cost coefficients of power-type flexible resources, Δt is the optimization period length, α is the penalty factor, and the superscript k is the value of the variable or parameter in the k-th iteration; θ G(k) (t) represents the power deviation that needs to be addressed in the k-th iteration and time period t of the power-type flexibility resource, as shown in Equation (21); and These are the upward and downward adjustable capacity deficits that need to be addressed in the k-th iteration of the power-type flexibility resource, specifically as shown in equations (22) and (23):

[0056] θ G(k) (t)=G G θ (k-1) (t)-P G(k-1) (t)+P G(k) (t) (21)

[0057]

[0058]

[0059] In the formula, For the capacity of the d-th renewable energy source, G cn For capacity-type flexible resource capacity, G G For power-type flexible resource capacity, θ (k) (t), and These represent the system power deviation, upward adjustable capacity deficit, and downward adjustable capacity deficit that need to be addressed per unit capacity during the k-th iteration, as shown in equations (24)-(26):

[0060]

[0061]

[0062]

[0063] In the formula, L(t) represents the load power during time period t, and is the responsibility of capacity-based flexible resource operators.

[0064] The revenue function of capacity-based flexibility resources includes operating costs and electricity trading costs, namely:

[0065]

[0066] In the formula, F cn(k) Let c be the revenue function of capacity-type flexibility resources in the k-th iteration. cn θ is the operating cost coefficient for capacity-type flexible resources. cn(k) (t) represents the power deviation that needs to be addressed in the k-th iteration and time period t of the capacity-type flexibility resource, as shown in Equation (28); and These represent the upward and downward adjustable capacity deficits that need to be addressed in the k-th iteration of the capacity-type flexibility resource, from time period i to time period j, as shown in equations (29) and (30):

[0067] θ cn(k) (t)=G cn θ (k-1) (t)-P cn(k-1) (t)+P cn(k) (t) (28)

[0068]

[0069]

[0070] Renewable Energy Operator Insider:

[0071] Renewable energy operators are self-interested individuals who generate revenue by selling electricity to the grid.

[0072]

[0073] In the formula, Let m be the revenue function of the m-th renewable energy operator in the k-th iteration. Let be the power deviation that the m-th renewable energy operator needs to resolve in the k-th iteration during the t-th time period, as shown in Equation (32); and These represent the upward and downward adjustable capacity deficits that the m-th renewable energy operator needs to address in the k-th iteration, specifically as shown in equations (33) and (34):

[0074]

[0075]

[0076]

[0077] Furthermore, the strategy space of the players includes the strategy space of power-type flexible resource operators, the strategy space of capacity-type flexible resource operators, and the strategy space of renewable energy operators.

[0078] Furthermore, the strategy space of power-type flexible resource operators includes power constraints and capacity adjustable domain constraints; the strategy space of capacity-type flexible resource operators includes capacity constraints, charge / discharge power constraints, state of charge constraints, and capacity adjustable domain constraints; the strategy space of renewable energy operators includes power constraints. The capacity domain constraints of power-type flexible resources are shown in equations (12)-(19), and the capacity domain constraints of capacity-type flexible resources are shown in equations (9)-(11).

[0079] Furthermore, the optimization process of the multi-source collaborative interaction optimization model is as follows:

[0080] Step 1: Input the load forecast data for the next 24 hours into the coordination center;

[0081] Step 2: Set k = 0. Take the lower limit of power output for power-type flexible resources, and do not take action for capacity-type flexible resources. Take all matrix elements in the capacity adjustable domain of all flexible resources to zero. Renewable energy outputs power according to its maximum predicted output power, and calculate the capacity demand domain under the maximum output power condition. Send the maximum predicted output power data of renewable energy, the capacity adjustable domain data and the capacity demand domain data to the coordination center.

[0082] Step 3: The coordination center determines whether the result meets the iteration termination condition. The game termination condition is set as the sum of the absolute values ​​of the power deviation and the adjustable capacity deficit in each time period, ε. rDoes it meet the given precision, i.e.:

[0083]

[0084] In the formula, ε is the accuracy parameter, M is the number of renewable energy sources, T is the total number of optimization periods, and the superscript k is the value of the variable or parameter in the k-th iteration. Let P be the planned power output of the m-th renewable energy source during time period t. cn (t) represents the charging and discharging power of capacity-type flexibility resources during time period t, P G (t) represents the output power of the power-type flexible resource during time period t. and These represent the total upward and downward capacity that power-type flexibility resources can provide from time period i to time period j, respectively. and These represent the total upward and downward capacity available for capacity-type flexibility resources from time period i to time period j, respectively. and Let k be the upward and downward capacity demand of the m-th renewable energy entity from time period i to time period j. If the demand is met, proceed to step 6; otherwise, let k = k + 1 and proceed to step 4.

[0085] Step 4: The coordination center calculates the unresolved power deviation and adjustable capacity deficit of the personnel according to formulas (24)-(26), and sends the results to the personnel.

[0086] Step 5: Flexible resource operators use the argmax decision rule to make decisions with the goal of maximizing the revenue function, update the planned output and adjustable capacity domain, and send the results to the coordination center; Renewable energy operators use the argmax decision rule to make decisions with the goal of maximizing the revenue function, update the planned output, calculate the capacity demand domain under the new planned output, send the results to the coordination center, and proceed to step 3.

[0087] Step 6: The game ends, and each player adjusts their effort based on the game results.

[0088] The system implementing the capacity-domain-based multi-source collaborative interaction optimization method includes several distributed player systems, each player system comprising:

[0089] Communication module: Used to send its own decision results and receive decision results from other players or system power deficits;

[0090] Intelligent computing module: responsible for updating its own decision-making results.

[0091] Compared with existing technologies, the beneficial effects of the present invention are as follows:

[0092] (1) Construct a capacity demand domain for renewable energy to reflect the total random fluctuation of renewable energy over multiple consecutive periods; construct a capacity adjustable domain for flexible resources to reflect the total adjustment capacity of flexible resources over multiple consecutive periods. The sum of the capacity adjustable domains of all flexible resources in the system is not less than the total capacity demand domain of all renewable energy as a global constraint, and is added to the potential game model in the form of a penalty function so that the decision result has the feasibility of the entire system operation scenario.

[0093] (2) By introducing the potential game into the optimization system, the problem is decomposed into multiple sub-problems and distributed solutions are obtained by constructing multiple potential game models between flexible resources and renewable energy. This achieves the combination of individual optimization and global optimization under the self-interest and autonomy of individuals.

[0094] (3) Improve the serial game process of traditional potential game to a parallel game process, so that all players are in an equal position in the game and achieve fairness in the participation of homogeneous individuals in the game. Attached Figure Description

[0095] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort, wherein:

[0096] Figure 1 A schematic diagram illustrating the construction of capacity demand domains based on three historical output scenarios of renewable energy.

[0097] Figure 2 The optimization structure diagram of the multi-source collaborative interaction optimization method and system based on capacity domain provided by this invention;

[0098] Figure 3 The optimization flowchart of the multi-source collaborative interaction optimization method and system based on capacity domain provided by the present invention is shown. Detailed Implementation

[0099] This invention proposes a multi-source collaborative interaction optimization method and system based on capacity domain, which is used to solve the feasibility of grid optimization operation in all scenarios and the problem of multi-entity optimization collaboration under the background of large-scale distributed energy grid connection.

[0100] To enable those skilled in the art to better understand the present invention, the invention will be further described in detail below with reference to the accompanying drawings and specific embodiments. Obviously, the described embodiments are only a part of the embodiments of the present invention, and not all of them. 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.

[0101] Example 1

[0102] The purpose of this embodiment is to provide a method for constructing the renewable energy capacity demand domain in a multi-source collaborative interaction optimization method and system based on capacity domain, specifically including:

[0103] like Figure 1 As shown, taking three time periods as an example, the construction and implementation of the renewable energy capacity demand domain is as follows:

[0104] Assuming three predicted historical power output scenarios for renewable energy are [1,1,3], [3,3,1], and [2,2,2], and their corresponding actual historical power output scenarios are [2,2,2], [2,2,2], and [2,2,2], respectively, and the planned power output for the day is [3,3,3], then under these three historical scenarios, the total power deviation between the actual historical power output and the predicted historical power output over a continuous period is calculated, yielding the corresponding power deviation matrix:

[0105]

[0106] Calculate the maximum and minimum values ​​of the corresponding matrix elements under the three power deviation matrices, and combine equations (7) and (8) to construct the renewable energy capacity demand domain as follows:

[0107]

[0108] Example 2

[0109] The purpose of this embodiment is to provide a multi-source collaborative interaction optimization method and system based on capacity domain, including:

[0110] like Figure 2 and Figure 3 As shown, a multi-source collaborative interaction optimization method and system based on capacity domain includes the following steps when M renewable energy operators, one capacity-type flexibility resource operator (such as energy storage), and one power-type flexibility resource (such as a gas turbine) participate in operation optimization:

[0111] Step 1: Input the load forecast data for the next 24 hours into the coordination center;

[0112] Step 2: Set k=1. For power-type flexible resources, set the lower limit of power output. For capacity-type flexible resources, leave them inactive. Set all matrix elements within the capacity adjustable domain of all flexible resources to zero. Calculate the capacity demand domain under the maximum predicted output power condition for renewable energy. Send the maximum predicted output power data, capacity adjustable domain data, and capacity demand domain data of renewable energy to the coordination center.

[0113] Step 3: The coordination center determines whether the results meet the iteration termination condition. The game termination condition is set as whether the sum of the absolute values ​​of the power deviation and the adjustable capacity deficit in each time period meets the given precision, i.e.:

[0114]

[0115] In the formula, ε is the precision parameter. If satisfied, proceed to step 6; otherwise, let k = k + 1 and proceed to step 4.

[0116] Step 4: The coordination center calculates the unresolved power deviation and adjustable capacity deficit of the personnel according to formulas (24)-(26), and sends the results to the personnel.

[0117] Step 5: Flexible resource operators adopt the argmax decision rule to make decisions with the goal of maximizing the revenue function, and update the planned output and adjustable capacity domain. The revenue function for power-type flexible resources is shown in Equation (20), and the planned output and adjustable capacity domain are updated simultaneously; the revenue function for capacity-type flexible resources is shown in Equation (27), and the planned output and adjustable capacity domain are updated simultaneously. Meanwhile, renewable energy operators adopt the argmax decision rule to make decisions with the goal of maximizing the revenue function, and update the planned output. The revenue function for renewable energy operators is shown in Equation (31). After renewable energy completes its decision, it recalculates the capacity demand domain according to the process described in Example 1. After all stakeholders complete their decisions, they send the results of the planned output and adjustable capacity domain or capacity demand domain to the coordination center. Proceed to Step 3;

[0118] Step 6: The game ends, and each player adjusts their effort based on the game results.

[0119] Example 3

[0120] The purpose of this embodiment is to provide a multi-source collaborative interaction optimization method and system based on capacity domain, including:

[0121] like Figure 2 and Figure 3 As shown, a multi-source collaborative interaction optimization method and system based on capacity domain is proposed. When M renewable energy operators, one capacity-type flexibility resource operator (such as energy storage), and one power-type flexibility resource (such as gas turbine) participate in operation optimization, if the random fluctuations of renewable energy are not considered, the original model degenerates into a deterministic parallel potential game that does not consider the capacity adjustable domain and the capacity demand domain. The method includes the following steps:

[0122] Step 1: Input the load forecast data for the next 24 hours into the coordination center;

[0123] Step 2: Set k = 0. For power-type flexibility resources, set the lower limit of output power. For capacity-type flexibility resources, do not activate. Set all matrix elements within the capacity adjustable domain of all flexibility resources to zero. Renewable energy outputs power based on its maximum predicted output power. Send the maximum predicted output power data of renewable energy to the coordination center.

[0124] Step 3: The coordination center determines whether the result meets the iteration termination condition. The game termination condition is set as whether the sum of the absolute values ​​of the power deviations in each time period meets the given precision, i.e.:

[0125]

[0126] In the formula, ε is the precision parameter. If satisfied, proceed to step 6; otherwise, let k = k + 1 and proceed to step 4.

[0127] Step 4: The coordination center calculates the power deviation to be resolved for the players according to formulas (24)-(26) and sends the results to the players.

[0128] Step 5: The flexible resource operator uses the argmax decision rule to update the planned output, aiming to maximize the revenue function. The revenue function for power-type flexible resources is shown below:

[0129]

[0130] Simultaneously update the planned output and adjustable capacity domains.

[0131] The payoff function for capacity-based flexible resources is shown below:

[0132]

[0133] Simultaneously update the planned output and adjustable capacity domains.

[0134] Meanwhile, renewable energy operators adopt the argmax decision rule to make decisions and update planned output with the goal of maximizing the revenue function. The revenue function of renewable energy operators is shown in equation (42).

[0135]

[0136] After all stakeholders have made their decisions, they will send the planned contribution results to the coordination center. Proceed to step 3;

[0137] Step 6: The game ends, and each player adjusts their effort based on the game results.

[0138] Those skilled in the art will understand that the modules or steps described above can be implemented using general-purpose computer devices. Optionally, they can be implemented using computer-executable program code, thereby allowing them to be stored in a storage device for execution by a computer device, or they can be fabricated as separate integrated circuit modules, or multiple modules or steps can be fabricated as a single integrated circuit module. This disclosure is not limited to any particular combination of hardware and software.

[0139] The above embodiments are implementation methods of the present invention, but the implementation methods of the present invention are not limited to the above embodiments. Any modifications, alterations, substitutions, combinations, or simplifications made without departing from the spirit and principle of the present invention shall be considered equivalent substitutions and shall be included within the protection scope of the present invention.

Claims

1. A multi-source collaborative interaction optimization method based on capacity domain, characterized in that, Includes the following steps: Based on the historical predicted output scenarios of renewable energy and the corresponding historical actual output scenarios, a capacity demand domain reflecting the output deviation of renewable energy over continuous periods is constructed. Based on the constraint space of capacity-flexible resources, construct the capacity-adjustable domain of capacity-flexible resources; Based on the constraint space of power-type flexible resources, a capacity adjustable domain for power-type flexible resources is constructed. A multi-source collaborative interaction optimization model based on potential game theory is constructed, and the model is solved using a parallel optimization process. The optimization process of the multi-source collaborative interaction optimization model is as follows: Step 1: Input the load forecast data for the next 24 hours into the coordination center; Step 2: Set k = 0. Take the lower limit of power output for power-type flexible resources, and do not take action for capacity-type flexible resources. Take all matrix elements in the capacity adjustable domain of all flexible resources to zero. Renewable energy outputs power according to its maximum predicted output power, and calculate the capacity demand domain under the maximum output power condition. Send the maximum predicted output power data of renewable energy, the capacity adjustable domain data and the capacity demand domain data to the coordination center. Step 3: The coordination center determines whether the result meets the iteration termination condition. The game termination condition is set as the sum of the absolute values ​​of the power deviation and the adjustable capacity deficit in each time period, ε. r Does it meet the given precision, i.e.: In the formula, ε is the accuracy parameter, M is the number of renewable energy sources, T is the total number of optimization periods, and the superscript k is the value of the variable or parameter in the k-th iteration. Let P be the planned power output of the m-th renewable energy source during time period t. cn (t) represents the charging and discharging power of capacity-type flexibility resources during time period t, P G (t) represents the output power of the power-type flexible resource during time period t. and These represent the total upward and downward capacity that power-type flexibility resources can provide from time period i to time period j, respectively. and These represent the total upward and downward capacity available for capacity-type flexibility resources from time period i to time period j, respectively. and Let k be the upward and downward capacity demand of the m-th renewable energy entity from time period i to time period j. If the demand is met, proceed to step 6; otherwise, let k = k + 1 and proceed to step 4. Step 4: The coordination center calculates the unresolved power deviation and adjustable capacity deficit of the personnel according to formulas (24)-(26), and sends the results to the personnel. For the capacity of the d-th renewable energy source, G cn For capacity-type flexible resource capacity, G G For power-type flexible resource capacity, θ (k) (t), and These represent the system power deviation, upward adjustable capacity deficit, and downward adjustable capacity deficit that need to be addressed per unit capacity during the k-th iteration, as shown in equations (24)-(26): In the formula, L(t) represents the load power during time period t; Step 5: Flexible resource operators use the argmax decision rule to make decisions with the goal of maximizing the revenue function, update the planned output and adjustable capacity domain, and send the results to the coordination center; Renewable energy operators use the argmax decision rule to make decisions with the goal of maximizing the revenue function, update the planned output, calculate the capacity demand domain under the new planned output, send the results to the coordination center, and proceed to step 3. Step 6: The game ends, and each player adjusts their effort based on the game results.

2. The multi-source collaborative interaction optimization method based on capacity domain according to claim 1, characterized in that, Constructing a capacity demand domain that reflects the output deviation of renewable energy over continuous periods, specifically including: For each renewable energy entity, a scenario pair is constructed based on historical predicted output scenarios and corresponding historical actual output scenarios. Each scenario pair is represented as follows: In the formula, M represents the number of renewable energy entities, N represents the number of scenarios with concentrated application scenarios, and T represents the total number of optimization periods. For the m-th RES and the n-th historical predicted power scene dataset, For the m-th RES and the n-th historical real-world scenario dataset, It contributes to the prediction of the m-th RES and the n-th scene for the t-th time period. For the m-th RES and n-th scenario pair, the actual output at time t is given by the deviation rate of the actual output relative to the predicted output for each scenario pair: In the formula, Let the power deviation rate of the m-th RES and the n-th scenario be at time t. Considering all possible consecutive time periods within a day, calculate the total predicted power deviation to form the power deviation matrix for this scenario pair, which is expressed as: In the formula, Let m be the power deviation matrix of the m-th renewable energy subject and the n-th scenario pair. Let the total deviation of the predicted power output of the m-th renewable energy subject in the n-th scenario be expressed as: In the formula, Let be the planned output power of the m-th renewable energy source in time period b, Δt be the optimization time period length, i be the calculation start time period, and j be the calculation end time period. Considering the maximum value of the total power deviation, the capacity demand domain of renewable energy is constructed, expressed as: In the formula, This is the capacity demand matrix for the m-th renewable energy source. Let be the reduced capacity demand matrix for the m-th renewable energy source, and together with , they constitute the capacity demand domain for that renewable energy source. and i,j=1,2,...,T represent the upward and downward capacity demands of the m-th renewable energy source from time period i to time period j, respectively. The calculation process is as follows: When the scenarios are sufficiently typical and the number is large enough, it is considered and It is the boundary of the predicted output deviation of the m-th renewable energy subject. By constructing the capacity demand domain, the boundary characteristics of the power deviation of each renewable energy subject in any continuous period are obtained.

3. The multi-source collaborative interaction optimization method based on capacity domain according to claim 1, characterized in that, Constructing a capacity-adjustable domain for capacity-flexible resources specifically includes: The charging and discharging results of capacity-type flexible resources over multiple time periods have a cumulative effect, which will be reflected in the remaining capacity and affect the regulation capability in subsequent time periods. Capacity-type flexible resources include energy storage, and the adjustable capacity domain of capacity-type flexible resources is expressed as: In the formula, ω cn,up With ω cn,dn These are the capacity capability matrices for increasing and decreasing capacity flexibility resources, respectively. Together, they constitute the adjustable capacity domain of capacity flexibility resources. and These represent the total upward and downward capacity available for capacity-type flexibility resources from time period i to time period j. The calculation process for each element in the matrix is ​​as follows: In the formula, λ represents the charging and discharging efficiency of capacity-type flexible resources, and S max S min Let S(j) represent the maximum and minimum remaining capacity of the capacity-type flexible resource, and the remaining capacity of the capacity-type flexible resource at time j, respectively. For capacity-type flexible resources, the maximum charge / discharge power is given by P, where Δt is the optimization time period length. cn (b) represents the charging and discharging power of capacity-type flexibility resources during time period b, where discharging is positive and charging is negative.

4. The multi-source collaborative interaction optimization method based on capacity domain according to claim 1, characterized in that, Constructing a capacity-adjustable domain for power-type flexible resources specifically includes: Power-type flexibility resources do not have a cumulative effect across multiple time periods and will not affect the regulation capability in subsequent time periods. Capacity-type flexibility resources include gas turbines. The adjustable power output of power-type flexibility resources in each time period should meet the following requirements: R≥[P G (t)+ΔP G,up (t)]-[P G (t-1)-ΔP G,dn (t-1)]≥-R (14) R≥(P G (t)-ΔP G,dn (t))-(P G (t-1)+ΔP G,up (t-1))≥-R (15) ΔP G,up (t)≥0,ΔP G,dn (t)≥0 (16) Equations (12) and (13) represent the output constraints of the adjustable power, and equations (14) and (15) represent the ramp rate constraints of the adjustable power. In these equations, P... G (t) represents the output power of the power-type flexible resource in time period t, ΔP G,up (t) and ΔP G,dn (t) represents the adjustable power and adjustable power of power-type flexibility resources during time period t, respectively. and These are the upper and lower limits of the output of the power-type flexible resource, respectively, and R is the ramp rate of the power-type flexible resource; Based on the adjustable power range of power-type flexible resources at different times, the adjustable capacity domain of power-type flexible resources is constructed: In the formula, ω G,up With ω G,dn These are the up-adjustment and down-adjustment capacity capability matrices for power-type flexibility resources, which together constitute the adjustable capacity domain of power-type flexibility resources. and These represent the total upward and downward capacity that power-type flexibility resources can provide from time period i to time period j. The calculation process for each element in the matrix is ​​as follows: Where Δt is the length of the optimization period.

5. The multi-source collaborative interaction optimization method based on capacity domain according to claim 1, characterized in that, The multi-source collaborative interaction optimization model includes three elements: players, payoff function, and strategy space.

6. The multi-source collaborative interaction optimization method based on capacity domain according to claim 5, characterized in that, The types of stakeholders mentioned include: renewable energy operators, capacity-based flexible resource operators, and power-based flexible resource operators.

7. The multi-source collaborative interaction optimization method based on capacity domain according to claim 5, characterized in that, The payoff function for the player is: Power-type flexible resource operators as insiders: The revenue function for power-type flexible resources includes operating costs and electricity sales revenue, namely: In the formula, F G(k) Let c be the revenue function for the k-th iteration of power-type flexibility resources. sell (t) represents the electricity price during time period t, a, b, and c are the generation cost coefficients of power-type flexible resources, Δt is the optimization period length, α is the penalty factor, and the superscript k is the value of the variable or parameter in the k-th iteration; θ G(k) (t) represents the power deviation that needs to be addressed in the k-th iteration and time period t of the power-type flexibility resource, as shown in Equation (21); and These are the upward and downward adjustable capacity deficits that need to be addressed in the k-th iteration of the power-type flexibility resource, specifically as shown in equations (22) and (23): θ G(k) (t)=G G θ (k-1) (t)-P G(k-1) (t)+P G(k) (t) (21) Operators of capacity-flexible resources: The revenue function of capacity-based flexibility resources includes operating costs and electricity trading costs, namely: In the formula, F cn(k) Let c be the revenue function of capacity-type flexibility resources in the k-th iteration. cn θ represents the operating cost coefficient for capacity-type flexible resources. cn(k) (t) represents the power deviation that needs to be addressed in the k-th iteration and time period t of the capacity-type flexibility resource, as shown in Equation (28); and These represent the upward and downward adjustable capacity deficits that need to be addressed in the k-th iteration of the capacity-type flexibility resource, from time period i to time period j, as shown in equations (29) and (30): θ cn(k) (t)=G cn θ (k-1) (t)-P cn(k-1) (t)+P cn(k) (t) (28) Renewable Energy Operator Insider: Renewable energy operators are self-interested individuals who generate revenue by selling electricity to the grid. In the formula, Let m be the revenue function of the m-th renewable energy operator in the k-th iteration. Let be the power deviation that the m-th renewable energy operator needs to resolve in the k-th iteration during the t-th time period, as shown in Equation (32); and These represent the upward and downward adjustable capacity deficits that the m-th renewable energy operator needs to address in the k-th iteration, specifically as shown in equations (33) and (34):

8. The multi-source collaborative interaction optimization method and system based on capacity domain according to claim 5, characterized in that, The strategy space of the players includes the strategy space of power-type flexible resource operators, the strategy space of capacity-type flexible resource operators, and the strategy space of renewable energy operators.

9. The multi-source collaborative interaction optimization method and system based on capacity domain according to claim 8, characterized in that, The strategy space for power-type flexible resource operators includes power constraints and capacity adjustable domain constraints; the strategy space for capacity-type flexible resource operators includes capacity constraints, charge / discharge power constraints, state of charge constraints, and capacity adjustable domain constraints; the strategy space for renewable energy operators includes power constraints.

Images