Indirect Control Method and Apparatus for Adjustable Resource Entities Based on Peak-Shaving Market

Abstract

This invention discloses an indirect control method and apparatus for adjustable resource entities based on a peak-shaving market. The method includes: establishing an upper-level model with the goal of maximizing self-profit and based on the energy storage resource operating costs and peak-shaving market revenue of load aggregators; establishing a lower-level model based on the operating costs and power parameters of the adjustable resource entities; establishing a master-slave game model based on the upper and lower-level models; updating the electricity price population and strategy population using a heuristic algorithm, and solving the master-slave game model to obtain the optimal electricity consumption strategy; and adjusting the purchase and sale of electricity by the adjustable resource entities according to the optimal electricity consumption strategy. By employing this invention, the two-layer master-slave game model accurately reflects the operational profit demands of the upper and lower-level entities, while using a heuristic algorithm to solve the game model to prevent the two-sided model from transforming into a single-layer model, thus avoiding the aggregator acquiring all user information.

Description

Technical Field

[0001] This invention relates to the field of ancillary services in the electricity market, and more particularly to a method and apparatus for indirect control of adjustable resource entities based on the peak-shaving market. Background Technology

[0002] Numerous studies have explored how aggregators regulate lower-level resource entities. These can be broadly categorized by regulation method into direct control (centralized regulation) and indirect control (negotiated regulation). Regarding direct control, some studies have constructed electricity sales company dispatch models, directly incorporating distributed generating units and flexible loads into the company's optimized dispatch. Some existing technologies suggest that users, after signing contracts, completely relinquish control of their load to aggregators for full regulation. Clearly, direct control requires comprehensive control over the equipment information of terminal resource entities, issuing direct control signals to specific controllable devices. This approach demands complete control of lower-level user information from the aggregator, potentially compromising user privacy. Furthermore, it can lead to excessive computational burden on the aggregator when the number of lower-level entities is large. Meanwhile, game theory is widely applied to the indirect control of lower-level users by aggregators. Some studies utilize incentive compensation and electricity prices as intermediate signals to establish game models between aggregators and electric vehicle users; others construct two-layer game models for multi-regional integrated energy systems based on the control optimization characteristics of upper-level agents and the energy consumption characteristics of lower-level users. This indirect control, based on user interests, uses intermediate signals to incentivize end-users to adjust their internal power consumption plans, reasonably balancing the independent property rights of lower-level users with the interests of aggregators, making it a more practical control method. However, current solutions for game-theoretic indirect control primarily address how to transform the two-layer problem of aggregators and lower-level resource entities into a single-layer problem. This essentially ignores the micro-interaction process between the upper and lower-level entities, remaining a centralized solution that is not applicable in actual operational interactions. Summary of the Invention

[0003] This invention provides an indirect control method and apparatus for adjustable resource entities based on a peak-shaving market, in order to solve the technical problems of existing technologies that, when regulating resource entities, use centralized solutions, which neglect the operational interests of aggregators and resource entities and fail to effectively protect user privacy.

[0004] To address the aforementioned technical problems, embodiments of the present invention provide an indirect control method for adjustable resource entities based on a peak-shaving market, comprising:

[0005] With the goal of maximizing its own profits, an upper-level model is established based on the energy storage resource operating costs of the load aggregator and the peak-shaving market revenue of the load aggregator; a lower-level model is established based on the operating costs of the adjustable resource entity and the corresponding power parameters of the adjustable resource entity.

[0006] Based on the upper-level model and the lower-level model, a master-slave game model is established.

[0007] Based on the heuristic algorithm, the electricity price population of the load aggregator and the strategy population of the adjustable resource subject are updated, and the master-slave game model is solved to obtain the optimal electricity consumption strategy.

[0008] Based on the optimal power consumption strategy, adjust the purchase and sale of electricity by the adjustable resource entity.

[0009] This invention establishes an upper-level model and a lower-level model based on the respective interests of load aggregators and adjustable resource entities. The master-slave game model established based on the upper-level and lower-level models takes into account the economic benefits of aggregators and the electricity costs of resource entities. While accurately reflecting the operational interests of the upper and lower levels, it achieves indirect control over adjustable resource entities and protects user privacy through the interactive process of the game between the upper and lower levels.

[0010] Furthermore, the process of updating the electricity price population of the load aggregator and the strategy population of the adjustable resource entity based on a heuristic algorithm, and solving the master-slave game model to obtain the optimal electricity consumption strategy, specifically involves:

[0011] Initialize the grid purchase and sale price, peak-shaving parameters, initial optimal fitness, and number of iterations, and randomly generate the initial price population of the load aggregator;

[0012] Based on the power grid purchase and sale price, the peak-shaving parameters, the upper-level model, and the lower-level model, the electricity price population of the load aggregator and the strategy population of the adjustable resource subject are solved and updated until the number of iterations reaches a preset value, or when the fitness meets a preset condition, the optimal electricity consumption strategy is obtained from the updated electricity price population and the updated strategy population.

[0013] This invention is based on a heuristic algorithm. When solving the master-slave game model, it sets an iteration termination condition and continuously updates the population to ensure the two-layer structure of the master-slave game model, thereby achieving indirect control over the adjustable resource subject and protecting user privacy.

[0014] Further, the step of solving and updating the electricity price population of the load aggregator and the strategy population of the adjustable resource entity based on the grid purchase and sale price, the peak-shaving parameters, the upper-level model, and the lower-level model, until the number of iterations reaches a preset value, or when the fitness meets a preset condition, and then obtaining the optimal electricity consumption strategy from the updated electricity price population and the updated strategy population, specifically involves:

[0015] S1. Solve the lower-level model based on the first electricity price population to obtain the strategy population of the adjustable resource subject; wherein, the first electricity price population includes: the updated electricity price population in the previous iteration or the initial electricity price population;

[0016] S2. Solve the upper-level model based on the strategy population to obtain the second fitness;

[0017] S3. Update the first electricity price population according to the heuristic algorithm to obtain the updated electricity price population;

[0018] S4. Update the iteration count and determine whether the iteration count has reached a preset value; calculate the fitness function difference based on the first fitness and the second fitness, and determine whether the fitness function difference is less than the convergence threshold; wherein, the first fitness includes: the second fitness in the previous iteration or the initial optimal fitness; when the iteration count reaches the preset value and the fitness function difference is less than or equal to the convergence threshold, execute S5; when the iteration count does not reach the preset value, or the fitness function difference is greater than the convergence threshold, return to S1;

[0019] S5. Based on the updated electricity price population, solve the lower-level model to obtain the updated strategy population, and based on the updated electricity price population and the updated strategy population, solve the master-slave game model to obtain the optimal electricity consumption strategy.

[0020] This invention continuously iterates the electricity price population of load aggregators and the strategy population of adjustable resource entities, ensuring that the master-slave game model maintains a two-layer structure during the solution process. This achieves indirect control over adjustable resource entities, avoids load aggregators obtaining all user information, protects user privacy, and improves computational efficiency.

[0021] Further, in step S2, based on the strategy population, the upper-level model is solved to obtain the second fitness, specifically as follows:

[0022] Using the grid purchase and sale price of electricity as the parameter of the upper-level model, and based on the strategy population and the peak-shaving parameter, the upper-level model is solved to obtain the first revenue corresponding to the strategy population;

[0023] Calculate the negative of the first gain, and use the negative of the first gain as the second fitness of the population corresponding to the heuristic algorithm.

[0024] Furthermore, the lower-level model is:

[0025] The operating costs of the adjustable resource entity include: the operating costs of distributed generating units, the user satisfaction costs corresponding to transferable loads, the user satisfaction costs corresponding to interruptible loads, and energy storage costs; the power parameters corresponding to the adjustable resource entity include: the power sold by the adjustable resource entity to the upper-level aggregate, the power purchased by the adjustable resource entity from the upper-level aggregate, the power generation of distributed generating units, the output of new energy resources, the adjustment amount of transferable loads, the adjustment amount of interruptible loads, the charging amount of energy storage, and the discharging amount of energy storage.

[0026]

[0027] in, The operating cost for the i-th adjustable resource entity participating in the load aggregator, where i represents the i-th adjustable resource entity and t is the time period. For the operating cost of distributed units, The cost of user satisfaction corresponding to transferable load. The cost of user satisfaction corresponding to interruptible load. For energy storage costs, The electricity price that load aggregators purchase from resource providers. The electricity price sold by load aggregators to resource entities. The amount of electricity sold by the adjustable resource entity to the upper-level aggregate. This refers to the amount of electricity that adjustable resource entities purchase from upper-level aggregates. For the decision vector of the adjustable resource subject. For the power generation of distributed generating units, Contribute to new energy resources For transferable load adjustment, This is the amount of interruptible load adjustment. For energy storage charging capacity, For energy storage discharge capacity, For electricity demand.

[0028] Furthermore, the operating cost model of the distributed unit is as follows:

[0029]

[0030] Among them, a B,i,G b B,i,G and c B,i,G These are the operating cost coefficients for distributed units. P B,i,G Lower limit constraint for the output power of distributed generating units. Limit the upper limit of output power for distributed generator units.

[0031] Furthermore, the model for the energy storage cost is as follows:

[0032]

[0033] Where, β B,i,ESS η is the energy storage degradation cost coefficient. B,i,ESSc For energy storage charging efficiency, η B,i,ESSd For energy storage and discharge efficiency, P B,i,ESSc To impose a lower limit constraint on the charging power of energy storage. Upper limit constraint on energy storage charging power, P B,i,ESSd This serves as a lower limit constraint on the energy storage discharge power. The upper limit constraint is imposed on the energy storage discharge power. In the state of energy storage charge, This is a lower limit constraint for the state of charge of energy storage. This is an upper limit constraint on the state of charge of energy storage.

[0034] Furthermore, the model for the contribution of new energy resources is as follows:

[0035]

[0036] in, To make the greatest contribution to new energy resources.

[0037] This invention constructs a lower-level model that includes the operating costs, decision vectors, and energy conservation of various energy sources by using the operating costs and energy parameters of the adjustable resource entity, thus fully reflecting the operating costs of the adjustable resource entity.

[0038] Furthermore, the upper-level model is:

[0039] The operating costs of energy storage resources for load aggregators include: the operating costs of the energy storage resources owned by the load aggregator; the peak-shaving market revenue of the load aggregator includes: the revenue from the purchase and sale of electricity by the load aggregator to various resource entities, the revenue from the purchase and sale of electricity by the load aggregator to the power grid, and the revenue from the load aggregator's participation in the peak-shaving market.

[0040]

[0041] in, This represents the revenue of a load aggregator with its own energy storage resources during time period t, where t represents the time period and i represents the i-th adjustable resource entity. The operating cost of the energy storage resources owned by the load aggregator. The revenue that load aggregators generate from the purchase and sale of electricity by various resource entities. The revenue that load aggregators receive from purchasing and selling electricity to the grid. For load aggregators to participate in the peak shaving market, The price is for peak shaving compensation. Let be the decision vector of the load aggregator. The energy storage charging capacity owned by the load aggregator. The energy storage discharge capacity owned by the load aggregator. The electricity price that load aggregators purchase from resource providers. The electricity price sold by load aggregators to resource entities. The electricity sold to the grid by load aggregators. The electricity purchased by load aggregators from the power grid. The electricity purchased by load aggregators from adjustable resource entities. Electricity sold by load aggregators to entities with adjustable resources. The amount of electricity sold by the adjustable resource entity to the upper-level aggregate. N represents the amount of electricity that adjustable resource entities purchase from the upper-level aggregate. B This represents the total number of entities with adjustable resources.

[0042] This invention establishes an upper-level model by considering the energy storage resource operating costs and peak-shaving market revenue of load aggregators, with the goal of maximizing their own revenue, so that the upper-level model can fully reflect the economic benefit needs of load aggregators.

[0043] On the other hand, embodiments of the present invention also provide an indirect control device for adjustable resource entities based on a peak-shaving market, comprising: a first model building module, a second model building module, a model solving module, and a resource regulation module;

[0044] The first model building module is used to build an upper-level model with the goal of maximizing its own revenue, based on the energy storage resource operating cost of the load aggregator and the peak-shaving market revenue of the load aggregator; and to build a lower-level model based on the operating cost of the adjustable resource entity and the power parameters corresponding to the adjustable resource entity.

[0045] The second model building module is used to build a master-slave game model based on the upper-level model and the lower-level model;

[0046] The model solving module is used to update the electricity price population of the load aggregator and the strategy population of the adjustable resource subject according to the heuristic algorithm, and solve the master-slave game model to obtain the optimal electricity consumption strategy.

[0047] The resource regulation module is used to adjust the purchase and sale of electricity by the adjustable resource entity according to the optimal electricity consumption strategy.

[0048] This invention establishes an upper-level model and a lower-level model based on the respective interests of load aggregators and adjustable resource entities. The master-slave game model established based on the upper-level and lower-level models takes into account the economic benefits of aggregators and the electricity costs of resource entities. While accurately reflecting the operational interests of the upper and lower levels, it achieves indirect control over adjustable resource entities and protects user privacy through the interactive process of the game between the upper and lower levels. Attached Figure Description

[0049] Figure 1 This is a flowchart illustrating an embodiment of the indirect control method for adjustable resource entities based on a peak-shaving market provided by the present invention.

[0050] Figure 2 This is a flowchart illustrating another embodiment of the indirect control method for adjustable resource entities based on peak-shaving markets provided by the present invention.

[0051] Figure 3 This is a schematic diagram of an embodiment of the indirect control device for adjustable resource entities based on a peak-shaving market provided by the present invention. Detailed Implementation

[0052] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0053] With the high penetration of new energy sources and the popularization and development of distributed generation and energy storage technologies, supporting projects such as rooftop photovoltaics are constantly being promoted. The volume of distributed resources on the user side is continuously increasing, and the total load is also constantly rising. Regional electricity consumption peak-valley differences are large, and peak load problems are prominent. Therefore, in recent years, how to integrate distributed resources to effectively participate in grid peak shaving and dispatch control has received widespread attention from scholars. Local governments have also established corresponding peak shaving markets and introduced relevant policies to support this, such as the "Real-time Detailed Rules for Adjustable Load Grid Connection Operation Management and Ancillary Service Management in Southern Region." Among these, load aggregators and virtual power plants are typical third-party operating entities. As agents of distributed resource entities, they can effectively coordinate the contradictions between the upper-level grid and various lower-level resource entities, fully stimulating the value and benefits that distributed resources bring to the grid and protecting the overall rights and interests of these entities. Therefore, in the future, with the increasing number of distributed resource entities, different forms of aggregators serve as a third-party medium for achieving positive interaction between the grid and distributed resource entities.

[0054] Based on the existing research "Coordinated Optimization of Intelligent Building and Community Integrated Energy Systems Based on Chance-Constrained Programming," this paper proposes a master-slave game-based coordinated optimization scheduling strategy for intelligent building and community integrated energy systems, considering the uncertainties and differentiated interests of multiple stakeholders in the coordinated optimization process. Taking into account the differentiated interests of the community integrated energy system and the intelligent building, a master-slave game method is used to model their coordinated optimization. Subsequently, to account for the impact of the uncertainty of wind power and photovoltaic output on the coordinated optimization scheduling of the community integrated energy system and the intelligent building, a method of transforming a two-level problem into a single-level problem is further adopted to solve the master-slave game optimization model of the community integrated energy system and the intelligent building in a concentrated manner.

[0055] Based on existing research, "An Energy Sharing Method for Smart Building Clusters Considering Time-of-Use Pricing Differences and Master-Slave Game Theory," this paper constructs an energy trading framework centered on a smart building cluster operator (SBCO) equipped with an energy storage system, and proposes an energy sharing method considering time-of-use pricing differences and a master-slave game theory approach. First, considering the time-of-use pricing differences among building users with different load types, and that real-time demand response in smart buildings promotes energy sharing within the cluster, a day-ahead energy storage scheduling model for the SBCO is established. Second, taking into account the different profit-seeking characteristics of the SBCO and building users, as well as users' information privacy and requirements for electricity comfort, a real-time demand response model based on a master-slave game theory is proposed to achieve a balance between the interests of both parties and joint optimization. Finally, an enumeration method is used for solution, and practical examples demonstrate that the proposed method can effectively improve the economic benefits of the SBCO and smart buildings, and has advantages in promoting the local consumption of distributed photovoltaic power and optimizing the net load characteristics of smart building clusters.

[0056] However, the existing technologies described above transform the two-layer problem of the aggregate and the lower-level resource entities into a single-layer problem for solution. This essentially ignores the micro-interaction process between the upper and lower-level entities, and it remains a centralized problem that is not applicable in actual operational interactions. Furthermore, the convergence and efficiency of using enumeration methods to solve master-slave game theory cannot be guaranteed. To address these issues, this invention proposes an indirect control method and device for adjustable resource entities based on a peak-shaving market. Based on price signals and electricity purchase and sale information, multiple interactions between the upper-level aggregate and each lower-level resource entity achieve indirect control of each resource. This effectively protects user privacy, motivates distributed devices to participate in demand response, and ensures the effectiveness of control over each resource entity under indirect control.

[0057] Example 1

[0058] Please refer to Figure 1This is a flowchart illustrating an embodiment of the indirect control method for adjustable resource entities based on a peak-shaving market provided by the present invention, mainly including steps 101-104:

[0059] Step 101: With the goal of maximizing its own revenue, and based on the energy storage resource operating costs of the load aggregator and the peak-shaving market revenue of the load aggregator, establish an upper-level model; based on the operating costs of the adjustable resource entity and the corresponding power parameters of the adjustable resource entity, establish a lower-level model.

[0060] In this embodiment, the adjustable resource entity includes one or more combinations of distributed generating units, electrochemical energy storage, new energy resources, interruptible loads, or transferable loads; in addition, the lower-level model may include a distributed generating unit model, an electrochemical energy storage model, a new energy resource model, an interruptible load model, and a transferable load model.

[0061] In this embodiment, distributed resource entities participating in load aggregators need to sign contracts, the load aggregators publish electricity prices, and each resource entity can only purchase and sell electricity to the aggregator.

[0062] Step 102: Based on the upper-level model and the lower-level model, establish a master-slave game model.

[0063] In this embodiment, the master-slave game model is as follows:

[0064]

[0065] In this model, G is the master-slave game theory partner, BA is the load aggregator, and B is the adjustable resource entity. Let H be the decision vector for the load aggregator, and H be the total number of scheduling periods. For load aggregators with their own energy storage resources, the revenue during time period t is... Let be the operating cost of the i-th adjustable resource entity participating in the load aggregator.

[0066] In this embodiment, the load aggregator first determines the baseline load of the load aggregator based on historical data. The upper-level power grid then issues its own electricity price information and the planned peak-shaving power of load aggregators. After obtaining grid information, the load aggregator determines the initial operating strategy and the purchase and sale prices of electricity for lower-level resource entities, and then transmits the electricity price information and purchase price. The information is then distributed to lower-level adjustable resource entities; these entities, acting as followers, adjust their internal power generation and consumption strategies in real time based on the information distributed by the load aggregator, and also transfer the electricity purchased and sold to the load aggregator. The information is then reported. The load aggregator updates its own strategy based on the information uploaded by the adjustable resource entities, and this process is repeated until a balance is achieved.

[0067] Step 103: Based on the heuristic algorithm, update the electricity price population of the load aggregator and the strategy population of the adjustable resource subject, and solve the master-slave game model to obtain the optimal electricity consumption strategy.

[0068] In this embodiment, in the solution mechanism of the master-slave game model: for the load aggregator, its goal is to maximize its own revenue. Solving for the optimal energy storage charge and discharge rate Electricity purchased and sold from the power grid and electricity prices For entities with adjustable resources, the goal is to minimize operating costs. Determine the most reasonable electricity consumption strategy When the aggregate can accurately know the optimal response of the resource subject, the master-slave game can be transformed into a single-layer problem for solution, just as many studies have used optimal conditions to solve game equilibrium. However, this obviously requires the load aggregator to have complete knowledge of all the internal operational information of the lower-level resource subjects, which is unreasonable; on the other hand, as analyzed above, the existence and uniqueness of the equilibrium solution of the game model under this patent are difficult to prove, and the single-layer problem transformed by the optimal conditions may not be solvable. In contrast, in actual operation, the subjects may not be concerned with whether there is equilibrium, but rather with whether it is available and whether various subjects can obtain benefits.

[0069] Therefore, this application solves the master-slave game model based on a heuristic algorithm. Specifically, according to the heuristic algorithm, the electricity price population of the load aggregator and the strategy population of the adjustable resource entity are updated, and the master-slave game model is solved to obtain the optimal electricity consumption strategy.

[0070] Initialize the grid purchase and sale price, peak-shaving parameters, initial optimal fitness, and number of iterations, and randomly generate the initial price population of the load aggregator;

[0071] Based on the power grid purchase and sale price, the peak-shaving parameters, the upper-level model, and the lower-level model, the electricity price population of the load aggregator and the strategy population of the adjustable resource subject are solved and updated until the number of iterations reaches a preset value, or when the fitness meets a preset condition, the optimal electricity consumption strategy is obtained from the updated electricity price population and the updated strategy population.

[0072] In this embodiment, the power grid purchase and sale price includes: the electricity price sold by the load aggregator to the resource entity. Electricity purchase price from resource entities by load aggregators The peak-shaving parameters include: peak-shaving compensation price, baseline load, and planned peak-shaving volume. Additionally, initializeable parameters include the total number of adjustable resource entities; the initial price cluster can consist of m aggregate purchase and sale prices.

[0073] This invention is based on a heuristic algorithm. When solving the master-slave game model, it sets an iteration termination condition and continuously updates the population to ensure the two-layer structure of the master-slave game model, thereby achieving indirect control over the adjustable resource subject and protecting user privacy.

[0074] Please refer to Figure 2 The following is a flowchart illustrating another embodiment of the indirect control method for adjustable resource entities based on a peak-shaving market provided by the present invention, mainly including steps S1-S5, as follows:

[0075] In this embodiment, the step of solving and updating the electricity price population of the load aggregator and the strategy population of the adjustable resource subject based on the power grid purchase and sale price, the peak shaving parameters, the upper-level model and the lower-level model, until the number of iterations reaches a preset value, or when the fitness meets a preset condition, and then obtaining the optimal electricity consumption strategy from the updated electricity price population and the updated strategy population, specifically includes steps S1 to S5.

[0076] S1: Solve the lower-level model based on the first electricity price population to obtain the strategy population of the adjustable resource subject; wherein, the first electricity price population includes: the updated electricity price population in the previous iteration or the initial electricity price population.

[0077] In this embodiment, the strategy population is used to upload to the load aggregator.

[0078] S2: Solve the upper-level model based on the strategy population to obtain the second fitness.

[0079] In this embodiment, step S2, solving the upper-level model based on the strategy population to obtain the second fitness, specifically involves:

[0080] Using the grid purchase and sale price of electricity as the parameter of the upper-level model, and based on the strategy population and the peak-shaving parameter, the upper-level model is solved to obtain the first revenue corresponding to the strategy population;

[0081] Calculate the negative of the first gain, and use the negative of the first gain as the second fitness of the population corresponding to the heuristic algorithm.

[0082] In this embodiment, solving the upper-level model based on the strategy population and the peak-shaving parameters specifically involves: obtaining a first population composed of the electricity purchased and sold by several adjustable resource entities to load aggregators in the strategy population; and solving the upper-level model based on the first population and the peak-shaving parameters.

[0083] S3: Update the first electricity price population according to the heuristic algorithm to obtain the updated electricity price population.

[0084] In this embodiment, the updated electricity price population is used to distribute to the adjustable resource entity.

[0085] S4: Update the iteration count and determine whether the iteration count has reached a preset value; calculate the fitness function difference based on the first fitness and the second fitness, and determine whether the fitness function difference is less than the convergence threshold; wherein, the first fitness includes: the second fitness in the previous iteration or the initial optimal fitness; when the iteration count reaches the preset value and the fitness function difference is less than or equal to the convergence threshold, execute S5; when the iteration count does not reach the preset value, or the fitness function difference is greater than the convergence threshold, return to S1.

[0086] In this embodiment, updating the iteration count specifically involves adding a first value to the iteration count of the previous iteration; calculating the fitness function difference based on the first fitness and the second fitness specifically involves obtaining the first fitness. First fitness function Acquiring the second fitness The second fitness function Calculate the difference between the second fitness function and the first fitness function.

[0087] S5: Based on the updated electricity price population, solve the lower-level model to obtain the updated strategy population, and based on the updated electricity price population and the updated strategy population, solve the master-slave game model to obtain the optimal electricity consumption strategy.

[0088] In this embodiment, the updated electricity price population is the optimal agent electricity price of the load aggregator.

[0089] This invention continuously iterates the electricity price population of load aggregators and the strategy population of adjustable resource entities, ensuring that the master-slave game model maintains a two-layer structure during the solution process. This achieves indirect control over adjustable resource entities, avoids load aggregators obtaining all user information, protects user privacy, and improves computational efficiency.

[0090] Step 104: Adjust the purchase and sale of electricity by the adjustable resource entity according to the optimal electricity consumption strategy.

[0091] In this embodiment, the optimal electricity consumption strategy includes the optimal agency electricity price of the load aggregator, the reasonable purchase and sale volume of the load aggregator, and the reasonable purchase and sale volume of the adjustable resource entity. According to the optimal electricity consumption strategy, the actual purchase and sale volume of the adjustable resource entity can be adjusted.

[0092] In this embodiment, the lower-level model is:

[0093] The operating costs of the adjustable resource entity include: the operating costs of distributed generating units, the user satisfaction costs corresponding to transferable loads, the user satisfaction costs corresponding to interruptible loads, and energy storage costs; the power parameters corresponding to the adjustable resource entity include: the power sold by the adjustable resource entity to the upper-level aggregate, the power purchased by the adjustable resource entity from the upper-level aggregate, the power generation of distributed generating units, the output of new energy resources, the adjustment amount of transferable loads, the adjustment amount of interruptible loads, the charging amount of energy storage, and the discharging amount of energy storage.

[0094]

[0095] in, The operating cost for the i-th adjustable resource entity participating in the load aggregator, where i represents the i-th adjustable resource entity and t is the time period. For the operating cost of distributed units, The cost of user satisfaction corresponding to transferable load. The cost of user satisfaction corresponding to interruptible load. For energy storage costs, The electricity price that load aggregators purchase from resource providers. The electricity price sold by load aggregators to resource entities. The amount of electricity sold by the adjustable resource entity to the upper-level aggregate. This refers to the amount of electricity that adjustable resource entities purchase from upper-level aggregates. For the decision vector of the adjustable resource subject. For the power generation of distributed generating units, Contribute to new energy resources For transferable load adjustment, This is the amount of interruptible load adjustment. For energy storage charging capacity, For energy storage discharge capacity, For electricity demand.

[0096] In this embodiment, the operating cost model of the distributed unit is as follows:

[0097]

[0098] Among them, a B,i,G b B,i,G and c B,i,G These are the operating cost coefficients for distributed units. P B,i,G Lower limit constraint for the output power of distributed generating units. Limit the upper limit of output power for distributed generator units.

[0099] In this embodiment, the model for energy storage cost is as follows:

[0100]

[0101] Where, β B,i,ESS η is the energy storage degradation cost coefficient. B,i,ESSc For energy storage charging efficiency, η B,i,ESSd For energy storage and discharge efficiency, P B,i,ESSc To impose a lower limit constraint on the charging power of energy storage. Upper limit constraint on energy storage charging power, P B,i,ESSd This serves as a lower limit constraint on the energy storage discharge power. The upper limit constraint is imposed on the energy storage discharge power. In the state of energy storage charge, This is a lower limit constraint for the state of charge of energy storage. This is an upper limit constraint on the state of charge of energy storage.

[0102] In this embodiment, the model for the output of new energy resources is as follows:

[0103]

[0104] in, To make the greatest contribution to new energy resources.

[0105] The model for the user satisfaction cost corresponding to the transferable load is as follows:

[0106]

[0107] Among them, a B,i,TL b B,i,TL and c B,i,TL These are the user satisfaction cost coefficients corresponding to transferable loads. This is a lower limit constraint on the amount of transferable load. This represents the upper limit constraint on the transferable load. H is the total control period.

[0108] In this embodiment, the model for the user satisfaction cost corresponding to the interruptible load is as follows:

[0109]

[0110] Among them, a B,i,IL b B,i,IL and c B,i,IL These are the user satisfaction cost coefficients corresponding to interruptible loads. This is an upper limit constraint on the interruptible load.

[0111] This invention constructs a lower-level model that includes the operating costs, decision vectors, and energy conservation of various energy sources by using the operating costs and energy parameters of the adjustable resource entity, thus fully reflecting the operating costs of the adjustable resource entity.

[0112] In this embodiment, the upper-layer model is:

[0113] The operating costs of energy storage resources for load aggregators include: the operating costs of the energy storage resources owned by the load aggregator; the peak-shaving market revenue of the load aggregator includes: the revenue from the purchase and sale of electricity by the load aggregator to various resource entities, the revenue from the purchase and sale of electricity by the load aggregator to the power grid, and the revenue from the load aggregator's participation in the peak-shaving market.

[0114]

[0115] in, This represents the revenue of a load aggregator with its own energy storage resources during time period t, where t represents the time period and i represents the i-th adjustable resource entity. The operating cost of the energy storage resources owned by the load aggregator. The revenue that load aggregators generate from the purchase and sale of electricity by various resource entities. The revenue that load aggregators receive from purchasing and selling electricity to the grid. For load aggregators to participate in the peak shaving market, The price is for peak shaving compensation. Let be the decision vector of the load aggregator. The energy storage charging capacity owned by the load aggregator. The energy storage discharge capacity owned by the load aggregator. The electricity price that load aggregators purchase from resource providers. The electricity price sold by load aggregators to resource entities. The electricity sold to the grid by load aggregators. The electricity purchased by load aggregators from the power grid. The electricity purchased by load aggregators from adjustable resource entities. Electricity sold by load aggregators to entities with adjustable resources. The amount of electricity sold by the adjustable resource entity to the upper-level aggregate. N represents the amount of electricity that adjustable resource entities purchase from the upper-level aggregate. B This represents the total number of entities with adjustable resources.

[0116] In this embodiment, in addition to coordinating with downstream resource entities by publishing electricity purchase and sale prices, load aggregators also possess a certain degree of independent energy storage to ensure the power supply and peak-shaving response capabilities of the aggregator. Agents need to rationally formulate electricity purchase and sale prices and charging / discharging plans based on grid price information, thereby interacting with the upper-level grid and downstream adjustable resource entities to generate revenue.

[0117] In this embodiment, the model for the operating cost of the energy storage resources owned by the load aggregator is as follows:

[0118]

[0119] Where, β ESS η is the energy storage degradation cost coefficient for load aggregators. BA,ESSc For the energy storage charging efficiency of load aggregators, η BA,ESSd The energy storage discharge efficiency of the load aggregator. P BA,ESSc The lower limit constraint on the energy storage charging power of load aggregators. The upper limit constraint on the energy storage charging power of load aggregators. P BA,ESSd The lower limit constraint of energy storage discharge power for load aggregators The upper limit constraint on the energy storage discharge power of load aggregators. The state of charge of the energy storage of the load aggregator. This is a lower bound constraint on the state of charge of energy storage for load aggregators. This is an upper limit constraint on the state of charge of the energy storage of the load aggregator.

[0120] In this embodiment, the revenue model for the load aggregator's purchase and sale of electricity to various resource entities is as follows:

[0121]

[0122] in, The baseline load size reported by load aggregators to the peak-shaving market. This represents the actual peak-shaving power output of the load aggregator. This refers to the effective peak-shaving power of the load aggregator.

[0123] In this embodiment, the model for the effective peak-shaving power is:

[0124]

[0125] in, For the planned peak-shaving electricity allocated to the peak-shaving market, when and When the ratio is greater than 1.3, the effective peak-shaving power is determined to be... When the ratio is within the range of 0.8-1.3, the effective peak-shaving power is determined as follows: When the ratio is less than 0.8, it is considered an invalid response, and the effective peak-shaving power is determined to be 0, with no subsidy provided. The model for the effective peak-shaving power can be transformed into a form containing binary variables using the Big M method.

[0126] In this embodiment, the power balance constraint of the upper-level model is: Considering all relevant constraints of the load aggregator and minimizing the load aggregator's revenue, we arrive at the economic optimization operation model for aggregators containing autonomous energy storage resources in the peak-shaving market.

[0127] This invention establishes an upper-level model by considering the energy storage resource operating costs and peak-shaving market revenue of load aggregators, with the goal of maximizing their own revenue, so that the upper-level model can fully reflect the economic benefit needs of load aggregators.

[0128] Please refer to Figure 3 This is a schematic diagram of an embodiment of the indirect control device for adjustable resource entities based on the peak-shaving market provided by the present invention, which mainly includes a first model establishment module 301, a second model establishment module 302, a model solving module 303, and a resource regulation module 304.

[0129] In this embodiment, the first model building module 301 is used to build an upper-level model with the goal of maximizing its own revenue, based on the energy storage resource operating cost of the load aggregator and the peak-shaving market revenue of the load aggregator; and to build a lower-level model based on the operating cost of the adjustable resource entity and the power parameters corresponding to the adjustable resource entity.

[0130] The second model building module 302 is used to build a master-slave game model based on the upper-level model and the lower-level model.

[0131] The model solving module 303 is used to update the electricity price population of the load aggregator and the strategy population of the adjustable resource subject according to the heuristic algorithm, and solve the master-slave game model to obtain the optimal electricity consumption strategy.

[0132] The resource regulation module 304 is used to adjust the purchase and sale of electricity by the adjustable resource entity according to the optimal power consumption strategy.

[0133] This invention establishes an upper-level model and a lower-level model based on the respective interests of load aggregators and adjustable resource entities. The master-slave game model established based on the upper-level and lower-level models takes into account the economic benefits of aggregators and the electricity costs of resource entities. While accurately reflecting the operational interests of the upper and lower levels, it achieves indirect control over adjustable resource entities and protects user privacy through the interactive process of the game between the upper and lower levels.

[0134] The specific embodiments described above further illustrate the purpose, technical solution, and beneficial effects of the present invention. It should be understood that the above descriptions are merely specific embodiments of the present invention and are not intended to limit the scope of protection of the present invention. In particular, it should be noted that any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention for those skilled in the art.

Claims

1. A method for indirect control of adjustable resource entities based on peak-shaving markets, characterized in that, The application relates to a method for establishing an optimal electricity consumption strategy for a load aggregator and a controllable resource subject. The upper-layer model is established by maximizing self-benefit and according to the operation cost of the energy storage resource of the load aggregator and the peak shaving market benefit of the load aggregator. The lower-layer model is established according to the operation cost of the controllable resource subject and the corresponding power parameter of the controllable resource subject. The master-slave game model is established according to the upper-layer model and the lower-layer model. The electricity price population of the load aggregator and the strategy population of the controllable resource subject are updated according to the heuristic algorithm, and the master-slave game model is solved to obtain the optimal electricity consumption strategy. The electricity purchase and sale quantity of the controllable resource subject is adjusted according to the optimal electricity consumption strategy. The electricity price population of the load aggregator and the strategy population of the controllable resource subject are updated according to the heuristic algorithm, and the master-slave game model is solved to obtain the optimal electricity consumption strategy. The grid electricity purchase and sale price, the peak shaving parameter, the initial optimal fitness and the iteration number are initialized, and the initial electricity price population of the load aggregator is randomly generated. The electricity price population of the load aggregator and the strategy population of the controllable resource subject are solved and updated according to the grid electricity purchase and sale price, the peak shaving parameter, the upper-layer model and the lower-layer model until the iteration number reaches a preset value or the fitness meets a preset condition, and the optimal electricity consumption strategy is obtained from the updated electricity price population and the updated strategy population. The electricity price population of the load aggregator and the strategy population of the controllable resource subject are solved and updated according to the grid electricity purchase and sale price, the peak shaving parameter, the upper-layer model and the lower-layer model until the iteration number reaches a preset value or the fitness meets a preset condition, and the optimal electricity consumption strategy is obtained from the updated electricity price population and the updated strategy population. S1, the lower-layer model is solved according to the first electricity price population to obtain the strategy population of the controllable resource subject; wherein the first electricity price population comprises the updated electricity price population in the last iteration or the initial electricity price population. S2, the upper-layer model is solved according to the strategy population to obtain a second fitness. S3, the first electricity price population is updated according to the heuristic algorithm to obtain the updated electricity price population. S4, the iteration number is updated, and it is judged whether the iteration number reaches a preset value; a fitness function difference value is calculated according to a first fitness and the second fitness, and it is judged whether the fitness function difference value is less than a convergence threshold value; wherein the first fitness comprises the second fitness in the last iteration or an initial optimal fitness; when the iteration number reaches the preset value and the fitness function difference value is less than or equal to the convergence threshold value, S5 is executed; when the iteration number does not reach the preset value or the fitness function difference value is greater than the convergence threshold value, S1 is returned. S5, the lower-layer model is solved according to the updated electricity price population to obtain the updated strategy population, and the master-slave game model is solved according to the updated electricity price population and the updated strategy population to obtain the optimal electricity consumption strategy.

2. The method for indirect control of a dispatchable resource subject based on a peak- shaving market as claimed in claim 1, wherein, S2, according to the strategy population, solving the upper model, get the second fitness, specifically: The grid electricity purchase and sale price is used as the parameter of the upper model, and the first income corresponding to the strategy population is obtained by solving the upper model according to the strategy population and the peak regulation parameter; The opposite number of the first income is calculated, and the opposite number of the first income is used as the second fitness of the heuristic algorithm corresponding population.

3. The method for indirect control of a dispatchable resource subject based on a peak- shaving market according to any one of claims 1-2, characterized in that, The lower model is: Wherein, the operation cost of the adjustable resource subject includes: the operation cost of the distributed unit, the user satisfaction cost corresponding to the transferable load, the user satisfaction cost corresponding to the interruptible load and the energy storage cost; The power parameter corresponding to the adjustable resource subject includes: the adjustable resource subject to the upper layer aggregation body, the adjustable resource subject to the upper layer aggregation body, the distributed unit, the new energy resource output, the transferable load adjustment amount, the interruptible load adjustment amount, the energy storage charging amount and the energy storage discharging amount; ； wherein, is the operation cost of the ith adjustable resource subject participating in the operation under the load aggregator, i represents the ith adjustable resource subject, and t is a time period, is the operation cost of the distributed generator, is the user satisfaction cost corresponding to the transferable load, is the user satisfaction cost corresponding to the interruptible load, is the energy storage cost, is the electricity purchase price of the load aggregator to the resource subject, is the electricity sale price of the load aggregator to the resource subject, is the electricity sale amount of the adjustable resource subject to the upper layer aggregator, is the electricity purchase amount of the adjustable resource subject to the upper layer aggregator, is the decision vector of the adjustable resource subject, is the power generation amount of the distributed generator, is the power output amount of the new energy resource, is the adjustment amount of the transferable load, is the adjustment amount of the interruptible load, is the charging amount of the energy storage, is the discharging amount of the energy storage, is the electricity demand.

4. The method for indirect control of a dispatchable resource subject based on a peak- shaving market as claimed in claim 3, wherein, The model of the operation cost of the distributed unit is: ； wherein, , and are the operating cost coefficients of the distributed units, respectively, is the lower limit constraint of the output power of the distributed units, is the upper limit constraint of the output power of the distributed units.

5. The method for indirect control of a dispatchable resource subject based on a peak- shaving market as claimed in claim 3, wherein, The model of the energy storage cost is: ； wherein, is a storage degradation cost coefficient, is a storage charging efficiency, is a storage discharging efficiency, is a lower bound constraint on storage charging power, is an upper bound constraint on storage charging power, is a lower bound constraint on storage discharging power, is an upper bound constraint on storage discharging power, is a storage state of charge, is a lower bound constraint on storage state of charge, is an upper bound constraint on storage state of charge.

6. The method for indirect control of a dispatchable resource subject based on a peak- shaving market as claimed in claim 3, wherein, The model of the new energy resource output is: ； wherein, is the maximum power output of the new energy resource.

7. The method for indirect control of a dispatchable resource subject based on a peak- shaving market according to any one of claims 1-2, wherein, The upper model is: Wherein, the energy storage resource operation cost of the load aggregator includes: the operation cost of the energy storage resource owned by the load aggregator; The peak regulation market income of the load aggregator includes: the income of the load aggregator to each resource subject, the income of the load aggregator to the grid, and the income of the load aggregator participating in the peak regulation market; ； wherein, is the profit of the load aggregator with autonomous energy storage resources at time period t, denotes time period, denotes the ith adjustable resource subject, is the operating cost of the energy storage resources owned by the load aggregator, is the profit of the load aggregator from buying and selling electricity to each resource subject, is the profit of the load aggregator from buying and selling electricity to the grid, is the profit of the load aggregator from participating in the peak shaving market, is the peak shaving compensation price, is the decision vector of the load aggregator, is the amount of energy storage charging owned by the load aggregator, is the amount of energy storage discharging owned by the load aggregator, is the electricity purchase price of the load aggregator to the resource subject, is the electricity sale price of the load aggregator to the resource subject, is the amount of electricity sold by the load aggregator to the grid, is the amount of electricity purchased by the load aggregator from the grid, is the amount of electricity purchased by the load aggregator from the adjustable resource subject, is the amount of electricity sold by the load aggregator to the adjustable resource subject, is the amount of electricity sold by the adjustable resource subject to the upper layer aggregator, is the amount of electricity purchased by the adjustable resource subject to the upper layer aggregator, is the total number of adjustable resource subjects.

8. An indirect control device for adjustable resource entities based on a peak-shaving market, characterized in that, It includes: The first model establishing module, the second model establishing module, the model solving module and the resource regulation module; Wherein, the first model establishing module is used to maximize the self income as the target, and according to the energy storage resource operation cost of the load aggregator and the peak regulation market income of the load aggregator, the upper model is established; And according to the operation cost of the adjustable resource subject and the power parameter corresponding to the adjustable resource subject, the lower model is established; The second model establishing module is used to establish the master-slave game model according to the upper model and the lower model; The model solving module is used to update the electricity price population of the load aggregator and the strategy population of the adjustable resource subject according to the heuristic algorithm, and solve the master-slave game model to obtain the optimal power utilization strategy; The resource regulation module is used to adjust the purchase and sale power of the adjustable resource subject according to the optimal power utilization strategy; According to the heuristic algorithm, the electricity price population of the load aggregator and the strategy population of the adjustable resource subject are updated, and the optimal power utilization strategy is obtained by solving the master-slave game model, specifically: Initialize the grid electricity purchase and sale price, the peak regulation parameter, the initial optimal fitness and the iteration number, and randomly generate the initial electricity price population of the load aggregator; According to the grid electricity purchase and sale price, the peak regulation parameter, the upper model and the lower model, the electricity price population of the load aggregator and the strategy population of the adjustable resource subject are solved and updated until the iteration number reaches the preset value, or the fitness meets the preset condition, and the optimal power utilization strategy is obtained from the updated electricity price population and the updated strategy population. The power price population of the load aggregator and the strategy population of the adjustable resource subject are solved and updated according to the grid buying and selling power price, the peak regulation parameter, the upper model and the lower model until the iteration number reaches a preset value, or the fitness meets a preset condition, and the optimal power utilization strategy is obtained from the updated power price population and the updated strategy population, specifically as follows: S1. According to the first power price population, the lower model is solved to obtain the strategy population of the adjustable resource subject; wherein the first power price population includes the updated power price population in the last iteration or the initial power price population; S2. According to the strategy population, the upper model is solved to obtain a second fitness; S3. According to a heuristic algorithm, the first power price population is updated to obtain an updated power price population; S4. The iteration number is updated, and it is judged whether the iteration number reaches a preset value; according to the first fitness and the second fitness, a fitness function difference value is calculated, and it is judged whether the fitness function difference value is less than a convergence threshold; wherein the first fitness includes the second fitness in the last iteration or an initial optimal fitness; when the iteration number reaches the preset value and the fitness function difference value is less than or equal to the convergence threshold, S5 is executed; when the iteration number does not reach the preset value or the fitness function difference value is greater than the convergence threshold, S1 is returned; S5. According to the updated power price population, the lower model is solved to obtain an updated strategy population, and according to the updated power price population and the updated strategy population, the master-slave game model is solved to obtain an optimal power utilization strategy.

Images