Multi-dimensional balanced scheduling shared energy storage optimization method, medium and program product

Abstract

The present invention discloses a multi-dimensional balanced scheduling shared energy storage optimization method, medium and program product, wherein the multi-dimensional balanced scheduling shared energy storage optimization method includes generating multiple optimization objective functions according to preset multi-dimensional optimization requirements; generating a real-time scheduling strategy based on multiple optimization objective functions in combination with dynamic priority rules and multi-objective optimization algorithms; in the scheduling process, obtaining the output power, grid load demand and energy storage status of renewable energy in real time, and dynamically adjusting the real-time scheduling strategy based on resource conflict decoupling mechanism and collaborative optimization strategy, so that energy storage can respond to grid load changes and complete the coordinated control of energy storage and renewable energy. The present invention improves the resource utilization efficiency of the shared energy storage system, reduces the operating cost of the power system, and enhances the system's adaptability and reliability to fluctuations in power load demand.

Description

Technical Field

[0001] The present invention belongs to the field of optimized scheduling of shared energy storage systems, and specifically relates to a multi-dimensional balanced scheduling shared energy storage optimization method, medium and program product. Background Art

[0002] In the current power grid environment, with the rapid development of renewable energy, especially the large-scale access of intermittent energy such as photovoltaic and wind power, traditional energy storage optimization methods mostly focus on a single dimension (such as cost, peak-valley balance, charging and discharging efficiency, etc.), which makes it difficult to cope with multi-dimensional and multi-objective scheduling needs and fully release the potential of energy storage systems.

[0003] The rise of shared energy storage systems has provided a new direction for resource sharing and optimized scheduling in power systems. Shared energy storage can not only improve the utilization efficiency of energy storage resources, but also reduce the overall investment and operating costs of the system by integrating the energy storage needs of different users. In the application of shared energy storage, how to reasonably dispatch energy storage resources to resolve charging and discharging conflicts, give priority to local consumption of renewable energy, and balance multi-dimensional needs (such as cost, system stability, and renewable energy acceptance rate) is still a complex and challenging task. Summary of the invention

[0004] In response to the above problems, the present invention proposes a multi-dimensional balanced scheduling shared energy storage optimization method, medium and program product, which improves the resource utilization efficiency of the shared energy storage system, reduces the operating cost of the power system, and enhances the system's adaptability and reliability to fluctuations in power load demand.

[0005] In order to achieve the above technical objectives and the above technical effects, the present invention is implemented through the following technical solutions:

[0006] In a first aspect, the present invention provides a multi-dimensional balanced scheduling shared energy storage optimization method, comprising:

[0007] Generate multiple optimization objective functions according to the preset multi-dimensional optimization requirements;

[0008] Generate real-time scheduling strategies based on multiple optimization objective functions, combined with dynamic priority rules and multi-objective optimization algorithms;

[0009] During the scheduling process, the output power of renewable energy, grid load demand and energy storage status are obtained in real time, and the real-time scheduling strategy is dynamically adjusted based on the resource conflict decoupling mechanism and collaborative optimization strategy, so that the energy storage can respond to changes in grid load and complete the coordinated control of energy storage and renewable energy.

[0010] In combination with the first aspect, optionally, the multi-dimensional optimization requirements include: cost dimension optimization requirements, power supply and demand balance optimization requirements, energy utilization efficiency optimization requirements and local consumption optimization requirements; the multiple optimization objective functions include: cost model objective function, power supply and demand balance model objective function, energy utilization efficiency model objective function and local consumption model objective function;

[0011] The objective function of the cost model is: , The mathematical expression is:

[0012] ;

[0013] in:

[0014] ;

[0015] ;

[0016] ;

[0017] ;

[0018] In the formula, For the moment The total cost, For the moment The charging cost, For the moment The discharge cost, For the moment Equipment maintenance costs; and Separately for the moment The charging and discharging prices; and Separately for the moment The charging energy and discharging energy of is the maintenance cost coefficient per unit energy;

[0019] The objective function of the power supply and demand balance model is: , The mathematical expression is:

[0020] ;

[0021] In the formula, Indicates time The power load demand, Indicates time The total output power of renewable energy and other power generation resources, where other power generation resources refer to power generation resources other than renewable energy in the system; and Separately for the moment The charging power and discharging power of the energy storage system; when the power supply is insufficient, increase the discharge power of the energy storage On the contrary, when there is excess power supply, the charging power of the energy storage is increased. ;

[0022] The objective function of the energy utilization efficiency model is: , The mathematical expression is:

[0023] ;

[0024] in:

[0025] ;

[0026] in, For the moment The total energy efficiency, and are the charging efficiency factor and the discharging efficiency factor, respectively. , ; and Separately for the moment The charging and discharging energy, is the time period;

[0027] The objective function of the local consumption model is: , The mathematical expression is:

[0028] ;

[0029] in:

[0030] ;

[0031] ;

[0032]

[0033] in, For the moment The transmission loss, Reverse path The distance Reverse path The loss coefficient, For the moment The reverse power, is the transmission efficiency factor, For the moment The local electricity load demand, For the moment The output power of renewable energy sources, For the moment excess power.

[0034] In combination with the first aspect, optionally, the method for generating the real-time scheduling strategy includes:

[0035] According to the preset dynamic priority rules, calculate the priority weights of each optimization objective function , priority weight The calculation formula is:

[0036] ,

[0037] in, For the moment The power load demand, Indicates time The state of energy storage, Indicates time The supply and demand balance is poor, Indicates time of renewable energy output; represents a weighted linear model or a nonlinear model; The values ​​of are cost, balance, efficiency, consumption;

[0038] Based on multiple optimization objective functions and the priority weights of each optimization objective function , generate the overall objective function, the expression of the overall objective function is:

[0039] ,

[0040] Solve the overall objective function and generate a real-time scheduling strategy.

[0041] In combination with the first aspect, optionally, the method for adjusting the real-time scheduling strategy includes:

[0042] Generate agents corresponding to each optimization objective function, and set corresponding constraints for each agent, and the objective function of each agent is the corresponding optimization objective function;

[0043] Based on the objective function of each agent, a global objective function is generated, and the expression of the global objective function is: ;

[0044] Based on the resource conflict decoupling mechanism and the collaborative optimization strategy, the global objective function is solved to obtain a new scheduling strategy to adjust the real-time scheduling strategy;

[0045] The resource conflict decoupling mechanism refers to introducing a reconciliation function to adjust the priority weight of the objective function of each agent if a conflict is identified between the agents;

[0046] The collaborative optimization strategy refers to the interactive decision-making of each agent based on game theory and collaborative evolution algorithm to achieve collaborative optimization in the form of Nash equilibrium.

[0047] In combination with the first aspect, optionally, the constraint conditions corresponding to the settings for each proxy include:

[0048] The agent corresponding to the objective function of the cost model is defined as the cost agent. The constraints of the cost agent include:

[0049] Charging constraints:

[0050] , , ;

[0051] Discharge constraints:

[0052] , , ;

[0053] Charging power constraints:

[0054] ;

[0055] Discharge power constraint:

[0056] ;

[0057] In the formula, For the moment Charging electricity price; For the moment The discharge price of For the moment Charging power; For the moment The discharge power; is the maximum charging power; is the maximum discharge power;

[0058] The agent corresponding to the objective function of the energy utilization efficiency model is defined as the efficiency agent. The constraints of the efficiency agent include:

[0059] ;

[0060] in,

[0061] ;

[0062] ;

[0063] In the formula, and Separately for the moment The amount of electricity lost during charging and the amount of electricity lost during discharging;

[0064] The agent corresponding to the objective function of the power supply and demand balance model is defined as the supply and demand balance agent, and the constraints of the supply and demand balance agent include:

[0065] ;

[0066] In the formula, For the moment The power load demand; For the moment of renewable energy output; For the moment The output power of other power generation resources in the system, wherein the other power generation resources refer to power generation resources other than renewable energy in the system;

[0067] The agent corresponding to the objective function of the local consumption model is defined as the local consumption agent, and the constraints of the local consumption agent include:

[0068]

[0069] in, is the fluctuation threshold.

[0070] In combination with the first aspect, optionally, the collaborative optimization strategy includes the following steps:

[0071] Step (1): At the initial stage of the game, each agent selects a preliminary charging and discharging strategy based on its own optimization goal, and calculates the corresponding performance index based on its objective function and constraints;

[0072] Step (2): Each agent sends its own charging and discharging strategy and corresponding performance indicators to other agents;

[0073] Step (3): Each agent generates an optimal charging and discharging strategy based on the charging and discharging strategies and corresponding performance indicators sent by other agents, according to the game theory algorithm, and its own objective function and constraints;

[0074] (4) Repeat steps (2) and (3) until the charging and discharging strategies of each agent are stable and coordinated optimization among the agents is achieved.

[0075] In combination with the first aspect, optionally, conflicts between agents It is calculated using the following formula:

[0076] ,

[0077] In the formula, Indicates time No. The objective function of an agent With The objective function of an agent The covariance of , the larger the covariance, the smaller the conflict; Indicates time No. The objective function of an agent The standard deviation of Indicates time No. The objective function of an agent The standard deviation of

[0078] Use conflicts to calculate the priority weight of each agent , the formula for priority weight is:

[0079] ;

[0080] The expression of the harmonic function is:

[0081] ;

[0082] In the formula, For the moment The harmonic function is the sum of the ratios of all agents’ priority weights and the objective function; It's time No. The objective function of an agent is represents the total number of agents;

[0083] Each agent uses the following formula to calculate the optimal charging and discharging strategy:

[0084]

[0085] In the formula, Indicates The optimal charging and discharging strategy of each agent; Indicates The current charging and discharging strategy of each agent; Indicates that except The charging and discharging strategies of other agents other than the first agent, Each agent generates its own optimal charging and discharging strategy given the strategies of other agents.

[0086] In combination with the first aspect, optionally, the multi-dimensional balanced scheduling shared energy storage optimization method further includes:

[0087] Predict future electricity load demand and renewable energy output based on pre-built forecast models;

[0088] According to the preset dynamic priority rules, calculate the priority weights of each optimization objective function , priority weight The calculation formula is:

[0089] ,

[0090] in, For the moment The power load demand, Indicates time The state of energy storage, Indicates time The supply and demand balance is poor, represents the output power of renewable energy; represents a weighted linear model or a nonlinear model; Represents different optimization objectives, and its values ​​are cost, balance, efficiency, and consumption;

[0091] Based on multiple optimization objective functions and the priority weights of each optimization objective function , generate the overall objective function, the expression of the overall objective function is:

[0092] ,

[0093] Solving the overall objective function and generating a prediction scheduling strategy;

[0094] When the difference between the real-time power load demand and the output power of renewable energy and the predicted power load demand and the output power of renewable energy is less than a preset error range, the prediction scheduling strategy is executed;

[0095] When the difference between the real-time power load demand and the output power of renewable energy and the predicted power load demand and the output power of renewable energy is greater than or equal to a preset error range, a real-time scheduling strategy is generated according to the real-time power load demand and the output power of renewable energy, and the real-time scheduling strategy is executed;

[0096] If the prediction and scheduling strategy is executed, the output power of renewable energy, grid load demand and energy storage status are obtained in real time, and the prediction and scheduling strategy is dynamically adjusted based on the resource conflict decoupling mechanism and collaborative optimization strategy, so that the energy storage can respond to changes in grid load and complete the coordinated control of energy storage and renewable energy.

[0097] In a second aspect, the present invention provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the multi-dimensional balanced scheduling shared energy storage optimization method described in any one of the first aspects.

[0098] In a third aspect, the present invention provides a computer program product, including a computer program / instruction, which, when executed by a processor, implements the multi-dimensional balanced scheduling shared energy storage optimization method described in any one of the first aspects.

[0099] Compared with the prior art, the present invention has the following beneficial effects:

[0100] The present invention realizes intelligent control and coordinated optimization of energy storage by comprehensively considering multi-dimensional factors such as energy storage cost, power supply and demand balance, energy utilization efficiency and local consumption, thereby improving the resource utilization efficiency of shared energy storage, reducing the operating cost of the power system, and enhancing the adaptability and reliability to fluctuations in power load demand.

[0101] Furthermore, the present invention proposes to predict future electricity load demand and output power of renewable energy, and generate a predictive scheduling strategy based on the predicted electricity load demand and output power of renewable energy, and dynamically adjust the predictive scheduling strategy according to real-time data feedback during actual execution to maintain the stability of shared energy storage under dynamic demand. BRIEF DESCRIPTION OF THE DRAWINGS

[0102] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments are briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative labor, among which:

[0103] Figure 1 A schematic diagram of a flow chart of a multi-dimensional balanced scheduling shared energy storage optimization method according to an embodiment of the present invention;

[0104] Figure 2 The figure is a schematic diagram of the generation and execution flow of a prediction optimization strategy according to an embodiment of the present invention. DETAILED DESCRIPTION

[0105] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0106] In addition, if there are descriptions involving "first", "second", etc. in the embodiments of the present invention, the descriptions of "first", "second", etc. are only used for descriptive purposes and cannot be understood as indicating or suggesting their relative importance or implicitly indicating the number of the indicated technical features. Therefore, the features defined as "first" and "second" may explicitly or implicitly include at least one of the features. In addition, the technical solutions between the various embodiments can be combined with each other, but they must be based on the ability of ordinary technicians in the field to implement them. When the combination of technical solutions is contradictory or cannot be implemented, it should be deemed that such a combination of technical solutions does not exist and is not within the scope of protection required by the present invention.

[0107] Example 1

[0108] The embodiment of the present invention provides a multi-dimensional balanced scheduling shared energy storage optimization method, such as Figure 1 As shown, the following steps are included:

[0109] (1) Generate multiple optimization objective functions based on the preset multi-dimensional optimization requirements;

[0110] (2) Generate a real-time scheduling strategy based on multiple optimization objective functions, combined with dynamic priority rules and multi-objective optimization algorithms;

[0111] (3) During the scheduling process, the output power of renewable energy, grid load demand and energy storage status are obtained in real time, and the real-time scheduling strategy is dynamically adjusted based on the resource conflict decoupling mechanism and collaborative optimization strategy, so that the energy storage can respond to changes in the grid load and complete the coordinated control of energy storage and renewable energy.

[0112] In a specific implementation of the embodiment of the present invention, the multi-dimensional optimization requirements include: cost dimension, supply and demand balance dimension, efficiency dimension and local consumption dimension; the multiple optimization objective functions include: cost model objective function, power supply and demand balance model objective function, energy utilization efficiency model objective function and local consumption model objective function;

[0113] The objective function of the cost model is: , its optimization goal is to minimize the total cost. The mathematical expression is:

[0114] ;

[0115] in:

[0116] ;

[0117] ;

[0118] ;

[0119] ;

[0120] In the formula, For the moment The total cost, For the moment The charging cost, For the moment The discharge cost, For the moment Equipment maintenance costs; and Separately for the moment The charging and discharging prices; and Separately for the moment The charging energy and discharging energy of is the maintenance cost coefficient per unit energy; the cost model objective function is used to evaluate the total cost in real time in different dispatching strategies, and dynamically adjust the dispatching strategy to minimize the overall cost of the shared energy storage system;

[0121] The objective function of the power supply and demand balance model is: , its optimization goal is to minimize the supply and demand balance error, The mathematical expression is:

[0122] ;

[0123] In the formula, Indicates time The power load demand, Indicates time The total output power of renewable energy and other power generation resources, where other power generation resources refer to power generation resources other than renewable energy in the system; and Separately for the moment The charging power and discharging power of the energy storage system. When the power supply is insufficient, increase the discharging power of the energy storage system. On the contrary, when there is excess power supply, the charging power of the energy storage system is increased. The objective function of the power supply and demand balance model is used to ensure dynamic balance under the conditions of supply and demand changes, and avoid overload or shortage of the power grid.

[0124] The objective function of the energy utilization efficiency model is: , its optimization goal is to maximize the energy utilization efficiency, The mathematical expression is:

[0125] ;

[0126] in:

[0127] ;

[0128] in, For the moment The total energy efficiency, and are the charging efficiency factor and the discharging efficiency factor, respectively. , ; and Separately for the moment The charging and discharging energy, is the time period; the energy utilization efficiency model objective function is mainly used to quantify the energy loss and system efficiency in the charging and discharging process. By analyzing the impact of different charging and discharging rates, frequencies and depths on the energy storage capacity utilization efficiency, the charging and discharging strategy is dynamically adjusted to maximize the energy utilization efficiency and reduce energy loss while extending the service life of the equipment.

[0129] The objective function of the local consumption model is: , its optimization goal is to minimize the transmission loss (i.e., minimize the reverse power loss). The mathematical expression is:

[0130] ;

[0131] in:

[0132] ;

[0133] ;

[0134] ;

[0135] in, For the moment The transmission loss, Reverse path The distance Reverse path The loss coefficient, For the moment Reverse power (when local consumption is insufficient, electricity needs to be reversed to the grid or other demand points, thus generating reverse power), is the transmission efficiency factor, For the moment The local electricity load demand, For the moment The output power (i.e. electricity supply) of renewable energy, For the moment The excess power, when Less than When Carry out local consumption or transmission. The objective function of the local consumption model is used to achieve: give priority to local consumption of renewable energy, reduce grid reverse transmission losses, and improve grid system stability. When the power load demand is low, give priority to local consumption to reduce the loss of photovoltaic reverse transmission to the grid. Optimize the reverse transmission path. When the local load is insufficient, calculate the optimal path for reverse transmission to the grid or surrounding loads, reduce energy loss during the reverse transmission process, and improve power utilization efficiency. By minimizing transmission losses , improve the absorption rate of renewable energy and reduce electricity waste.

[0136] In a specific implementation of the embodiment of the present invention, the method for generating the real-time scheduling strategy includes:

[0137] According to the preset dynamic priority rules, calculate the priority weights of each optimization objective function , used to prioritize the most important optimization goals in different situations, priority weights The calculation formula is:

[0138] ,

[0139] in, For the moment The power load demand, Indicates time The state of energy storage, Indicates time The supply and demand balance is poor, Indicates time of renewable energy output; Represents a weighted linear model or nonlinear model, where the weight coefficients are set based on historical data or expert experience, thereby dynamically adjusting the priority of the optimization objectives to ensure that the system can respond flexibly under changing load conditions; The values ​​are cost, balance, efficiency, and consumption. The dynamic priority rule is based on the power load demand characteristics and resource availability in different time periods. It gives priority to reducing costs during peak power load demand periods and gives priority to improving energy utilization efficiency and local consumption efficiency of renewable energy during low demand periods. Through dynamic adjustment of priorities, the system can respond flexibly under changing load conditions.

[0140] Based on multiple optimization objective functions and the priority weights of each optimization objective function , generate the overall objective function, the expression of the overall objective function is:

[0141] ,

[0142] Solve the overall objective function and generate a real-time scheduling strategy. In the solution process, an optimization algorithm based on Pareto frontier analysis can be used to determine the compromise solution between different optimization objectives through non-dominated sorting. The optimal solution of each optimization objective function will constitute the Pareto frontier. The optimal compromise solution is selected from multiple Pareto frontier solutions to ensure the balance of the system under multiple dimensional optimization objectives. In the specific implementation process, the scheduling strategy includes charging and discharging time, charging and discharging power, total cost, power supply and demand balance difference, energy utilization efficiency, and local consumption.

[0143] In a specific implementation of the embodiment of the present invention, the method for adjusting the real-time scheduling strategy includes:

[0144] Generate agents corresponding to each optimization objective function, and set corresponding constraints for each agent, and the objective function of each agent is the corresponding optimization objective function;

[0145] Based on the objective function of each agent, a global objective function is generated, and the expression of the global objective function is: ;

[0146] Based on the resource conflict decoupling mechanism and the collaborative optimization strategy, the global objective function is solved to obtain a new scheduling strategy to adjust the real-time scheduling strategy;

[0147] The resource conflict decoupling mechanism refers to introducing a reconciliation function to adjust the priority weight of the objective function of each agent if a conflict is identified between the agents;

[0148] The collaborative optimization strategy refers to the interactive decision-making of each agent based on game theory and collaborative evolution algorithm to achieve collaborative optimization in the form of Nash equilibrium.

[0149] In a specific implementation of the embodiment of the present invention, the setting of corresponding optimization rules for each agent includes:

[0150] The agent corresponding to the objective function of the cost model is defined as the cost agent. For the cost agent, the system is prioritized to meet the lowest cost dispatch during the peak period of power load demand, and the charging and discharging path that meets the economic priority is selected. The charging and discharging path includes the charging and discharging time and the charging and discharging power. Its constraints include:

[0151] Charging constraints:

[0152] , , ,

[0153] The charging constraint means charging when the electricity price is low;

[0154] Discharge constraints:

[0155] , , ,

[0156] The discharge constraint indicates that the electricity is discharged when the electricity price is high;

[0157] Charging power constraints:

[0158] ;

[0159] Discharge power constraint:

[0160] ;

[0161] In the formula, For the moment Charging electricity price; For the moment The discharge price of For the moment Charging power; For the moment The discharge power; is the maximum charging power; is the maximum discharge power;

[0162] The agent corresponding to the objective function of the energy utilization efficiency model is defined as the efficiency agent. For the efficiency agent, its optimization rule is: in the valley period, the efficiency optimization path is preferentially selected. The efficiency optimization path includes the charge and discharge amount and the energy storage loss amount. Its constraints include:

[0163] ;

[0164] in,

[0165] ;

[0166] ;

[0167] In the formula, and Separately for the moment The power loss during charging and discharging;

[0168] The agent corresponding to the objective function of the power supply and demand balance model is defined as the supply and demand balance agent. For the supply and demand balance agent, its optimization rules are: to maintain the dynamic balance of the power grid, the battery charging and discharging needs to meet the power load demand, when the power grid is insufficient, it needs to provide power through discharge, and when the power grid is oversupplied, it absorbs excess power through charging. Its constraints include:

[0169]

[0170] In the formula, For the moment The power load demand; For the moment of renewable energy output; For the moment The supply of other power generation resources in the system, where other power generation resources refer to power generation resources other than renewable energy in the system, such as hydropower generation resources, thermal power generation resources, etc.;

[0171] The agent corresponding to the objective function of the local consumption model is defined as the local consumption agent. For the local consumption agent, its optimization rule is: in the period when the fluctuation is greater than the set threshold, the local consumption of renewable energy is given priority. Its constraints include:

[0172]

[0173] in, is the fluctuation threshold.

[0174] In a specific implementation of the embodiment of the present invention, the collaborative optimization strategy includes the following steps:

[0175] Step (1): In the initial stage of the game, each agent selects a preliminary charging and discharging strategy according to its optimization goal, calculates the corresponding performance indicators based on its objective function and constraints, and provides a basis for the next decision;

[0176] Step (2): Each agent sends its own charging and discharging strategy and corresponding performance indicators to other agents to help other agents adjust their own charging and discharging strategies to ensure the overall coordination of the system;

[0177] Step (3): Each agent generates the optimal charging and discharging strategy based on the charging and discharging strategies and corresponding performance indicators sent by other agents, according to the game theory algorithm, its own objective function and constraints, and gradually adjusts it during the game process until a Nash equilibrium is reached;

[0178] (4) Repeat steps (2) and (3) until the charging and discharging strategies of each agent are stable and coordinated optimization among the agents is achieved. Specifically, the game process will continue until the charging and discharging strategies of all agents are stable. If in a certain round of iteration, the charging and discharging strategies of all agents change very little (i.e., reach equilibrium), the iteration process is stopped. When the game process converges, the charging and discharging strategies of each agent will no longer change significantly, and the entire system will achieve coordinated optimization.

[0179] In a specific implementation of the embodiment of the present invention, the conflicts between the agents It is calculated using the following formula:

[0180] ,

[0181] In the formula, Indicates time No. The objective function of an agent With The objective function of an agent The covariance of , the larger the covariance, the smaller the conflict; Indicates time No. The objective function of an agent The standard deviation of Indicates time No. The objective function of an agent The standard deviation of

[0182] Use conflicts to calculate the priority weight of each agent , the formula for priority weight is:

[0183] ;

[0184] The expression of the harmonic function is:

[0185] ;

[0186] In the formula, For the moment The harmonic function is the sum of the ratios of all agents’ priority weights and the objective function; It's time No. The objective function of an agent is represents the total number of agents; thus, the priority weight It will be adjusted according to the conflicts between agents so that the objective function of the agent with less conflict will receive more attention.

[0187] The reconciliation function is often used to balance the relationship between different objectives during the optimization process. It is designed to optimize the various optimization objectives of the system as a whole in a "coordinated" manner. During the scheduling process, each agent's decision is subject to some constraints, which are usually the maximum limit of the resources that the agent can use. By setting constraints, it can be ensured that the system remains within the feasible range during the optimization process. These constraints ensure that all agents do not exceed the available resources of the system in decisions such as charging, discharging, and storage.

[0188] In the multi-agent collaborative optimization process, Nash equilibrium is a core concept in game theory, which means that in a multi-party game, all agents cannot obtain better results by unilaterally adjusting their strategies, that is, each agent's strategy has reached the best given the strategies of other agents. To this end, each agent in the present invention uses the following formula to calculate the optimal charging and discharging strategy:

[0189]

[0190] In the formula, Indicates The optimal charging and discharging strategy of each agent; Indicates The current charging and discharging strategy of each agent; Indicates that except The charging and discharging strategies of other agents other than the first agent, Each agent generates its own optimal charging and discharging strategy given the strategies of other agents. The conditions for Nash equilibrium include: (1) Local optimality: Each agent selects its own optimal charging and discharging strategy under the premise that the charging and discharging strategies of other agents remain unchanged, so that the agent's objective function is minimized (or maximized), thereby achieving a local optimal solution. (2) Global optimal solution: The strategies of all agents jointly reach a global equilibrium state. Although each agent does not consider the goals of other agents when optimizing its own goals, the overall system reaches an equilibrium state in multiple rounds of interaction in the game.

[0191] In a specific implementation of the embodiment of the present invention, Figure 2 As shown, the multi-dimensional balanced scheduling shared energy storage optimization method also includes:

[0192] Predict future electricity load demand and renewable energy output based on pre-built forecasting models (i.e., deep learning load forecasting);

[0193] According to the preset dynamic priority rules, calculate the priority weights of each optimization objective function , priority weight The calculation formula is:

[0194] ,

[0195] in, For the moment The power load demand, Indicates time The state of energy storage, Indicates time The supply and demand balance is poor, Indicates time of renewable energy output; represents a weighted linear model or a nonlinear model; Represents different optimization objectives, and its values ​​are cost, balance, efficiency, and consumption;

[0196] Based on multiple optimization objective functions and the priority weights of each optimization objective function , generate the overall objective function, the expression of the overall objective function is:

[0197] ,

[0198] Solving the overall objective function and generating a prediction scheduling strategy;

[0199] When the difference between the real-time power load demand and the output power of renewable energy and the predicted power load demand and the output power of renewable energy is less than a preset error range, the prediction scheduling strategy is executed;

[0200] When the difference between the real-time power load demand and the output power of renewable energy and the predicted power load demand and the output power of renewable energy is greater than or equal to a preset error range, a real-time scheduling strategy is generated according to the real-time power load demand and the output power of renewable energy;

[0201] If the predictive scheduling strategy is executed, during the scheduling process, the output power of renewable energy, grid load and energy storage status are obtained in real time, and the predictive scheduling strategy is dynamically adjusted based on the resource conflict decoupling mechanism and collaborative optimization strategy, so that the energy storage can respond to changes in the grid load and complete the coordinated control of energy storage and renewable energy.

[0202] In the specific implementation process, the pre-built prediction model is obtained by the following method:

[0203] Step s1: Data collection and preprocessing, collect historical power load demand data, renewable energy output power data, meteorological data (such as temperature, humidity, sunshine intensity, etc.) and other related influencing factors. Preprocess these data, including data cleaning, normalization and feature selection, to ensure the quality and availability of input data. Specifically include:

[0204] First, historical power load demand data, renewable energy output power, and meteorological data, such as temperature, humidity, and sunshine intensity, are collected to form a data set;

[0205] Then, the dataset is cleaned (removing outliers and filling missing values) and normalized, and feature selection is performed to ensure the quality and consistency of the data.

[0206] Step s2: Build a prediction model and design the corresponding network structure, including input layer, hidden layer and output layer, to ensure that the model can effectively learn the temporal characteristics and nonlinear relationships of the data.

[0207] During the specific implementation process, the GRU model can be selected to predict future electricity load demand and the output power of renewable energy.

[0208] Step s3: Model training and validation

[0209] The prediction model is trained using historical data, and the generalization ability and prediction accuracy of the prediction model are ensured through cross-validation and hyperparameter tuning.

[0210] During the model training process, a loss function (such as mean square error) is used to evaluate the performance of the model, and the prediction effect of the model is verified through the test set to ensure its effectiveness in practical applications. In the specific application process, the mean square error (MSE) can be used as the loss function of the prediction model to improve the accuracy of the prediction.

[0211] Example 2

[0212] In an embodiment of the present invention, a computer-readable storage medium is provided, on which a computer program is stored. When the program is executed by a processor, the multi-dimensional balanced scheduling shared energy storage optimization method described in any one of Embodiment 1 is implemented.

[0213] Example 3

[0214] A computer program product is provided in an embodiment of the present invention, including a computer program / instruction, which, when executed by a processor, implements the multi-dimensional balanced scheduling shared energy storage optimization method described in any one of Embodiment 1.

[0215] Those skilled in the art will appreciate that the embodiments of the present application may be provided as methods, systems, or computer program products. Therefore, the present application may adopt the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Moreover, the present application may adopt the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program codes.

[0216] The present application is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0217] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0218] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process in the computer or other programmable device. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0219] The embodiments of the present invention are described above in conjunction with the accompanying drawings, but the present invention is not limited to the above-mentioned specific implementation methods. The above-mentioned specific implementation methods are merely illustrative and not restrictive. Under the enlightenment of the present invention, ordinary technicians in this field can also make many forms without departing from the scope of protection of the purpose of the present invention and the claims, which all fall within the protection of the present invention.

[0220] The above shows and describes the basic principles and main features of the present invention and the advantages of the present invention. It should be understood by those skilled in the art that the present invention is not limited to the above embodiments. The above embodiments and descriptions are only for explaining the principles of the present invention. Without departing from the spirit and scope of the present invention, the present invention may have various changes and improvements, which fall within the scope of the present invention to be protected. The scope of protection of the present invention is defined by the attached claims and their equivalents.

Claims

1. A multi-dimensional balanced scheduling shared energy storage optimization method, characterized in that: include: Generate multiple optimization objective functions according to the preset multi-dimensional optimization requirements; Generate real-time scheduling strategies based on multiple optimization objective functions, combined with dynamic priority rules and multi-objective optimization algorithms; During the dispatching process, the output power of renewable energy, grid load demand and energy storage status are obtained in real time, and the real-time dispatching strategy is dynamically adjusted based on the resource conflict decoupling mechanism and collaborative optimization strategy, so that the energy storage can respond to changes in grid load; Predict future electricity load demand and renewable energy output based on pre-built forecast models; According to the preset dynamic priority rules, the priority weight w of each optimization objective function is calculated i (t), priority weight w i The calculation formula of (t) is: w i (t)=f i '(P load (t),S state (t),P balance (t),P renew (t)), Among them, P load (t) is the power load demand at time t, S state (t) is the state of energy storage at time t, P balance (t) is the balance difference between supply and demand at time t, P renew (t) is the output power of renewable energy at time t; f i '() is a weighted linear model or a nonlinear model; i is a different optimization objective, and its value can be cost, balance, efficiency, or consumption; Based on multiple optimization objective functions and the priority weight w of each optimization objective function i (t), the overall objective function is generated as: Minimize:F=w cost (t)·f cost (t)+w balance (t)·f balance (t)+w efficiency (t)·f efficiency (t)+w consumption (t)·f consumption (t), Among them, f cost (t) is the objective function of the cost model, f balance (t) is the objective function of the power supply and demand balance model; f efficiency (t) is the objective function of the energy utilization efficiency model; f consumption (t) is the objective function of the local consumption model; Solve the overall objective function and generate a predictive scheduling strategy; When the difference between the real-time power load demand and the output power of renewable energy and the predicted power load demand and the output power of renewable energy is less than the preset error range, the prediction scheduling strategy is executed; When the difference between the real-time power load demand and the output power of renewable energy and the predicted power load demand and the output power of renewable energy is greater than or equal to a preset error range, a real-time scheduling strategy is generated and executed based on the real-time power load demand and the output power of renewable energy.

2. A multi-dimensional balanced scheduling shared energy storage optimization method according to claim 1, characterized in that: The multi-dimensional optimization requirements include: cost dimension optimization requirements, power supply and demand balance optimization requirements, energy utilization efficiency optimization requirements and local consumption optimization requirements; the multiple optimization objective functions include: cost model objective function, power supply and demand balance model objective function, energy utilization efficiency model objective function and local consumption model objective function; The objective function of the cost model is f cost (t), f cost The mathematical expression of (t) is: f cost (t)=argmin(C total (t)); in: C total (t)=C charge (t)+C discharge (t)+C maintenance (t); In the formula, C total (t) is the total cost at time t, C charge (t) is the charging cost at time t, C discharge (t) is the discharge cost at time t, C maintenance (t) is the equipment maintenance cost at time t; p charge (t) and p discharge (t) are the charging price and discharging price at time t, respectively; E charge (t) and E discharge (t) are the charging energy and discharging energy at time t, respectively; k m is the maintenance cost coefficient per unit energy; The objective function of the power supply and demand balance model is f balance (t), f balance The mathematical expression of (t) is: f balance (t)=argmin(P supply (t)-P charge (t)-P demand (t)+P discharge (t)); Where P demand (t) represents the power load demand at time t, P supply (t) represents the total output power of renewable energy and other power generation resources at time t; P charge (t) and P discharge (t) are the charging power and discharging power of the energy storage system at time t; when the power supply is insufficient, the discharge power P of the energy storage is increased discharge (t); On the contrary, when the power supply is in excess, the charging power P of the energy storage is increased. charge (t); The objective function of the energy utilization efficiency model is f efficiency (t), f efficiency The mathematical expression of (t) is: f efficiency (t)=argmax(η total (t)); in: Among them, η total (t) is the total energy utilization efficiency at time t, η charge and η discharge are the charging efficiency factor and the discharging efficiency factor, respectively, 0<η charge <1,0<η discharge <1; E charge (t) and E discharge (t) are the charging energy and discharging energy at time t, respectively, and T is the time period; The objective function of the local consumption model is f consumption (t), f consumption The mathematical expression of (t) is: f consumption (t)=argmin(L trans (t)); in: P trans (t)=ΔP(t)·(1-η trans ); ΔP(t)=P renew (t)-P local (t); Among them, L trans (t) is the transmission loss at time t, d h is the distance of the reverse path h, k h is the loss coefficient of the reverse path h, P trans (t) is the reverse power at time t, η trans is the transmission efficiency factor, P local (t) is the local power load demand at time t, P renew (t) is the output power of renewable energy at time t, and ΔP(t) is the excess power at time t.

3. A multi-dimensional balanced scheduling shared energy storage optimization method according to claim 2, characterized in that: The method for generating the real-time scheduling strategy includes: According to the preset dynamic priority rules, the priority weight w of each optimization objective function is calculated i (t), priority weight w i The calculation formula of (t) is: w i (t)=f i '(P load (t),S state (t),P balance (t),P renew (t)), Among them, P load (t) is the power load demand at time t, S state (t) represents the state of energy storage at time t, P balance (t) represents the balance difference between supply and demand at time t, P renew (t) represents the output power of renewable energy at time t; f i '() represents a weighted linear model or a nonlinear model; the value of i is cost, balance, efficiency, consumption; Based on multiple optimization objective functions and the priority weight w of each optimization objective function i (t), generate the overall objective function, the expression of the overall objective function is: Minimize:F=w cost (t)·f cost (t)+w balance (t)·f balance (t)+w efficiency (t)·f efficiency (t)+w consumption (t)·f consumption (t), Solve the overall objective function and generate a real-time scheduling strategy.

4. A multi-dimensional balanced scheduling shared energy storage optimization method according to claim 3, characterized in that: The method for adjusting the real-time scheduling strategy includes: Generate agents corresponding to each optimization objective function, and set corresponding constraints for each agent, and the objective function of each agent is the corresponding optimization objective function; Based on the objective function of each agent, a global objective function is generated, and the expression of the global objective function is: Minimize:F=w cost (t)·f cost (t)+w balance (t)·f balance (t)+w efficiency (t)·f efficiency (t)+w consumption (t)·f consumption (t); Based on the resource conflict decoupling mechanism and the collaborative optimization strategy, the global objective function is solved to obtain a new scheduling strategy to adjust the real-time scheduling strategy; The resource conflict decoupling mechanism refers to introducing a reconciliation function to adjust the priority weight of the objective function of each agent if a conflict is identified between the agents; The collaborative optimization strategy refers to the interactive decision-making of each agent based on game theory and collaborative evolution algorithm to achieve collaborative optimization in the form of Nash equilibrium.

5. A multi-dimensional balanced scheduling shared energy storage optimization method according to claim 4, characterized in that: The constraints corresponding to the settings for each proxy include: The agent corresponding to the objective function of the cost model is defined as the cost agent. The constraints of the cost agent include: Charging constraints: If C charge (t)<C discharge (t),P charge (t)>0,P discharge (t)=0; Discharge constraints: If C charge (t)>C discharge (t),P charge (t)=0,P discharge (t)>0; Charging power constraints: 0≤P charge (t)≤P charge,max ; Discharge power constraint: 0≤P discharge (t)≤P discharge,max ; In the formula, C charge (t) is the charging electricity price at time t; C discharge (t) is the discharge electricity price at time t; P charge (t) is the charging power at time t; P discharge (t) is the discharge power at time t; P charge,max is the maximum charging power; P discharge,max is the maximum discharge power; The agent corresponding to the objective function of the energy utilization efficiency model is defined as the efficiency agent. The constraints of the efficiency agent include: min(∑(Loss charge (t)+Loss discharge (t))); in, Loss charge (t)=P charge (t)-η charge (t)·P charge (t); Loss discharge (t)=P discharge (t)-η discharge (t)·P discharge (t); In the formula, Loss charge (t) and Loss discharge (t) are the power loss during charging and the power loss during discharging at time t respectively; The agent corresponding to the objective function of the power supply and demand balance model is defined as the supply and demand balance agent, and the constraints of the supply and demand balance agent include: P charge (t)-P discharge (t)=D(t)-P renewable (t)-P other (t); Where D(t) is the power load demand at time t; P renewable (t) is the output power of renewable energy at time t; P other (t) is the output power of other power generation resources at time t; The agent corresponding to the objective function of the local consumption model is defined as the local consumption agent, and the constraints of the local consumption agent include: ∣P renewable (t)-P demand (t)∣≤∈∈ Among them, ∈ is the fluctuation threshold.

6. A multi-dimensional balanced scheduling shared energy storage optimization method according to claim 4, characterized in that: The collaborative optimization strategy includes the following steps: Step (1): In the initial stage of the game, each agent selects a preliminary charging and discharging strategy according to its own optimization goal, and calculates the corresponding performance index according to its objective function and constraints; Step (2): Each agent sends its own charging and discharging strategy and corresponding performance indicators to other agents; Step (3): Each agent generates an optimal charging and discharging strategy based on the charging and discharging strategies and corresponding performance indicators sent by other agents, according to the game theory algorithm, and its own objective function and constraints; (4) Repeat steps (2) and (3) until the charging and discharging strategies of each agent are stable and coordinated optimization among the agents is achieved.

7. A multi-dimensional balanced scheduling shared energy storage optimization method according to claim 4, characterized in that: The conflicts between agents are ij (t) is calculated using the following formula: Where Cov represents the objective function f of the i-th agent at time t i (t) and the objective function f of the jth agent j The covariance of (t), the larger the covariance, the smaller the conflict; represents the objective function f of the i-th agent at time t i The standard deviation of represents the objective function f of the jth agent at time t j The standard deviation of Use the conflict to calculate the priority weight w for each agent i (t), the formula for priority weight is: The expression of the harmonic function is: Where H(t) is the harmonic function at time t, which is the sum of the ratios of the priority weights of all agents and the objective function; f i (t) is the objective function of the ith agent, and n represents the total number of agents; Each agent uses the following formula to calculate the optimal charging and discharging strategy: In the formula, represents the optimal charging and discharging strategy of the ith agent; P i represents the current charging and discharging strategy of the i-th agent; P j ,... represents the charging and discharging strategies of other agents except the i-th agent. The i-th agent generates its own optimal charging and discharging strategy given the strategies of other agents.

8. A multi-dimensional balanced scheduling shared energy storage optimization method according to claim 1, characterized in that: The multi-dimensional balanced scheduling shared energy storage optimization method also includes: If the prediction and scheduling strategy is executed, the output power of renewable energy, grid load demand and energy storage status are obtained in real time, and the prediction and scheduling strategy is dynamically adjusted based on the resource conflict decoupling mechanism and collaborative optimization strategy, so that the energy storage can respond to changes in grid load and complete the coordinated control of energy storage and renewable energy.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the program is executed by a processor, the multi-dimensional balanced scheduling shared energy storage optimization method described in any one of claims 1 to 8 is implemented.

10. A computer program product comprising a computer program / instructions, characterized in that When the computer program / instruction is executed by a processor, the multi-dimensional balanced scheduling shared energy storage optimization method described in any one of claims 1 to 8 is implemented.

Images