A Distributed Decision Optimization Method for Virtual Power Plants Based on McCormick Envelope

Abstract

A distributed decision optimization method for virtual power plants based on McCormick envelopes is proposed. This method includes: constructing a two-layer optimization framework comprising a power user decision optimization model and a virtual power plant operator optimization model; in the power user decision optimization model, defining power user satisfaction and optimizing the interaction power volume between power users and the virtual power plant operator based on the internal electricity price published by the virtual power plant operator; in the virtual power plant operator optimization model, performing day-ahead optimization decisions based on power users' bidding strategies; employing McCormick envelopes to relax the non-convex parts, transforming the two-layer optimization problem into a mixed-integer linear programming problem; and using an improved alternating direction multiplier method to perform distributed solution of the mixed-integer linear programming problem, iteratively solving until a preset optimal solution gap threshold is reached. This method effectively supports the optimization of internal resource scheduling within virtual power plants, achieving optimal interaction between virtual power plants and internal users, and enhancing the regulation flexibility of the power system.

Description

Technical Field

[0001] This invention belongs to the field of power system operation, specifically relating to a distributed decision optimization method for virtual power plants based on McCormick envelope. Background Technology

[0002] On the distribution side, with the increasing availability of flexible resources such as various adjustable loads, photovoltaic power generation, and energy storage systems, virtual power plants play a crucial role in integrating these resources. This not only enhances the system's regulation capabilities but also helps alleviate power supply shortages. This resource integration provides a rich operational foundation for virtual power plants and is of great significance for improving system efficiency.

[0003] How to support virtual power plants in optimizing internal resource scheduling, enhancing the regulation flexibility of the power system, maximizing the utilization of large-scale flexible resources, promoting the construction and development of virtual power plants, and enhancing the resilience and reliability of new power systems are challenges faced by existing technologies.

[0004] It should be noted that the information disclosed in the background section above is only for understanding the background of this application, and therefore may include information that does not constitute prior art known to those skilled in the art. Summary of the Invention

[0005] The main objective of this invention is to overcome the shortcomings of the aforementioned background technology and provide a distributed decision optimization method for virtual power plants based on McCormick envelope.

[0006] To achieve the above objectives, the present invention adopts the following technical solution:

[0007] A distributed decision optimization method for virtual power plants based on McCormick envelopes, the method comprising:

[0008] A two-layer optimization framework is constructed, including an optimization model for power user decision-making and an optimization model for virtual power plant operators;

[0009] In the power user decision optimization model, power user satisfaction is defined, and the interactive electricity volume between power users and virtual power plant operators is optimized based on the interactive electricity price published by the virtual power plant operator; in the virtual power plant operator optimization model, day-ahead optimization decisions are made based on the power user's bidding strategy.

[0010] We use McCormick's envelope theorem to relax the non-convex part, transforming the bi-level optimization problem into a mixed-integer linear programming problem for processing.

[0011] The mixed-integer linear programming problem is solved in a distributed manner using an improved alternating direction multiplier method, and the solution is iterated until a preset optimal solution gap threshold is reached.

[0012] In some embodiments, the improved alternating direction multiplier method includes: during the initial iteration, running a sub-optimization problem for the virtual power plant operator; updating the virtual power plant's electricity purchase and sale prices and incorporating the sub-optimization problem for the electricity user; updating the multipliers and penalty parameters; and iteratively solving until a preset optimal solution gap threshold is met.

[0013] The present invention has the following beneficial effects:

[0014] This invention proposes a distributed decision-making optimization method for virtual power plants based on McCormick envelopes. It characterizes the interaction between virtual power plant operators and resources by constructing a two-layer optimization framework involving both the virtual power plant and its internal power users. A distributed algorithm is also proposed to solve this two-layer optimization model. This invention can guide virtual power plant operators and related management platforms in optimizing the scheduling of various internal distributed resources. It can provide a reference for power companies, virtual power plant agents, and third-party load aggregators to optimize resource scheduling.

[0015] The distributed decision optimization method for virtual power plants in this invention constructs a distributed optimization problem based on a two-layer model of virtual power plants and internal power users, thereby realizing the optimal interaction between virtual power plants and internal users. This invention can efficiently guide the optimal aggregation of internal resources in virtual power plants under various distributed resource conditions.

[0016] Other beneficial effects of the embodiments of the present invention will be further described below. Detailed Implementation

[0017] The embodiments of the present invention will be described in detail below. It should be emphasized that the following description is merely exemplary and not intended to limit the scope and application of the present invention.

[0018] This invention provides a distributed decision optimization method for virtual power plants based on McCormick envelopes, the method comprising:

[0019] A two-layer optimization framework is constructed, including an optimization model for power user decision-making and an optimization model for virtual power plant operators;

[0020] In the power user decision optimization model, power user satisfaction is defined, and the interactive electricity volume between power users and virtual power plant operators is optimized based on the interactive electricity price published by the virtual power plant operator; in the virtual power plant operator optimization model, day-ahead optimization decisions are made based on the power user's bidding strategy.

[0021] McCormick's envelope theorem is used to relax the non-convex part, transforming the bi-level optimization problem into a mixed-integer linear programming problem;

[0022] The mixed-integer linear programming problem is solved in a distributed manner using an improved alternating direction multiplier method, and the solution is iterated until a preset optimal solution gap threshold is reached.

[0023] The method of this invention can guide virtual power plant operators and related management platforms to optimize the scheduling of various distributed resources within the plant, and can provide a reference for power companies, virtual power plant agents and third-party load aggregators to optimize resource scheduling.

[0024] The following describes specific embodiments of the present invention.

[0025] The distributed decision optimization method for virtual power plants based on McCormick envelopes in this invention constructs a distributed optimization problem based on a two-layer model of the virtual power plant and internal power users, thereby achieving optimal interaction between the virtual power plant and internal users. This distributed optimization problem is a typical two-layer Stakolberg game problem.

[0026] (1) Power User Decision Optimization Model

[0027] First, an internal decision-making optimization model is constructed for electricity users who can own or contract with multiple flexible resources, such as thermal power units, renewable energy, energy storage, and adjustable loads. The virtual power plant operator (VPA) then establishes an internal market where electricity users receive buy and sell prices from the VPA, allowing the optimization model to determine the amount of electricity to buy or sell to the VPA and submit its bids.

[0028] First, define the satisfaction level of electricity users when consuming electricity:

[0029]

[0030]

[0031] In the formula, μ t υ is the preference parameter, and υ is the user characteristic coefficient. This is the user's load power. d i,t These represent the upper and lower limits of the load power, respectively.

[0032] Electricity users will submit a declaration to the virtual power plant operator, which includes: Here, S represents the price response coefficient of electricity user i at time t, and B represents the upper limit of the interaction between the electricity user and the virtual power plant. S indicates selling electricity, while B indicates buying electricity. From this reported information, the real-time interaction power between the electricity user and the virtual power plant can be obtained.

[0033]

[0034]

[0035] In the formula, This is the price at which the virtual power plant purchases its power in the internal market. This is the upper limit of the output of thermal power units contracted by electricity users. This is the upper limit of the output of new energy units. This is the upper limit of the discharge power of the energy storage system. It is the lower limit of the charging power of the energy storage system, η i It refers to charge / discharge efficiency.

[0036] The application strategy is derived from the following optimization model:

[0037]

[0038] The relevant constraints are:

[0039]

[0040]

[0041]

[0042]

[0043]

[0044]

[0045]

[0046]

[0047]

[0048]

[0049] Equation (5) demonstrates the benefits that electricity users gain by participating in the virtual power plant internal market. This represents the amount of electricity sold directly by users to the power company. For the corresponding interactive electricity price, the secondary and primary costs of thermal power units are respectively determined by... and express, The active power of the thermal power unit. Formula (6) describes the amount of electricity delivered by the electricity user to the virtual power plant operator at a specific time t. This total amount includes the output power from thermal power units and renewable energy sources. The charging and discharging power of the energy storage system Load power And the amount sold directly to the power grid company Composition. The output power of thermal power units and renewable energy units needs to meet their upper and lower limits. g i,t , RES i,t Equation (10-11) describes the climbing ability of thermal power units, U i and D i These are the upward ramp limit and the downward ramp limit, respectively. Equation (12-15) focuses on the operational constraints of the energy storage system. This refers to the real-time power of the energy storage system, and it needs to meet upper and lower limits. M is a large positive number, z i,t It is a zero-one auxiliary variable.

[0050] (2) Virtual Power Plant Operator Optimization Model

[0051] Based on the bidding strategies proposed by electricity users, virtual power plant operators make day-ahead optimization decisions. Their primary objective is to reduce the operating costs of virtual power plants (VPPs) and ensure compliance with penalties incurred due to failure to meet bid volumes caused by real-time power imbalances. When constructing the optimization model for a virtual power plant, conditional value of risk is specifically considered to better manage and mitigate potential risks.

[0052] minimize:C VPP =(1-γ)(C b -R S -R E )+γV r #(16)

[0053] in

[0054]

[0055]

[0056]

[0057]

[0058] The corresponding constraints are

[0059]

[0060]

[0061]

[0062]

[0063]

[0064]

[0065]

[0066]

[0067]

[0068] Equation (16) describes the minimum operating cost C of the virtual power plant. VPP γ represents the risk avoidance strategy adopted by the virtual power plant, s is each generated scenario, Ns is the total number of scenarios, and w s It is the probability of the scenario, V r It is conditional value at risk, C b and R s These refer to the cost of purchasing electricity and the revenue from selling electricity, as shown in (17-19). and The total volume of transactions is denoted as ζ, where ζ is the value at risk, α is the confidence coefficient, and δ is the total volume of transactions. (20) The method for calculating conditional value at risk is defined, where ζ is the value at risk, α is the confidence coefficient, and δ is the total volume of transactions. x As an auxiliary variable. (21) describes the total amount of virtual power plant bids in the market. (22) and (23) involve price restrictions in the internal market, which require the price of electricity in the market to be within the range of electricity prices. Compared with retail market prices Between. Equation (24) shows the amount of the penalty for breach of contract. The relationship between virtual power plants and value at risk. (25-26) represents the total electricity purchased and sold by virtual power plants to electricity users, and (27) depicts the penalties for default by virtual power plants. This represents the default price coefficient. (28-29) indicates that the actual electricity volume received by the virtual power plant from users is uncertain. and Total electricity volume for each uncertain scenario and This represents the probability distribution for the corresponding uncertainty scenario.

[0069] Since the upper-level optimization problem yields the electricity price traded between the virtual power plant and internal electricity users, while the lower-level problem yields the electricity traded between internal users and the virtual power plant, and these two are product of each other, they cannot be solved directly. Therefore, this invention uses McCormick's envelope theorem to relax this non-convex part:

[0070]

[0071] In the formula, and These are the auxiliary variables for buying and selling at time t. Auxiliary constraints are introduced based on the original upper and lower bound constraints of each of these variables:

[0072]

[0073]

[0074]

[0075]

[0076]

[0077]

[0078]

[0079]

[0080] Therefore, this bi-level non-convex optimization problem is transformed into a mixed-integer linear programming problem for processing.

[0081] (3) Distributed solution algorithm

[0082] Considering the specific characteristics of the problem, an improved alternating direction multiplier method is proposed. The unified augmented form of this bilevel optimization problem is as follows:

[0083]

[0084] in, Let λ be the initial value for electricity users to buy and sell electricity to virtual power plant operators, and let λ and ρ be the multiplier and penalty parameter corresponding to this constraint, respectively.

[0085] Therefore, first, when the iteration number k is 0, substitute... Sub-optimization problem of running a virtual power plant operator:

[0086]

[0087] Then, incorporate the updated virtual power plant electricity purchase and sale prices into the sub-optimization problem for electricity users:

[0088]

[0089] Further update the multipliers:

[0090]

[0091] Simultaneously updated Iterative solution until

[0092]

[0093]

[0094] ∈ represents the minimum difference between the iteration point and the optimal solution, which is set to 0.005 in the preferred embodiment of the present invention.

[0095] In summary, the method of the present invention can effectively support the optimization of internal resource scheduling in virtual power plants, realize optimal interaction between virtual power plants and internal users, enhance the regulation flexibility of the power system, maximize the utilization of large-scale flexible resources, promote the construction and development of virtual power plants, and enhance the resilience and reliability of new power systems.

[0096] This invention also provides a storage medium for storing a computer program, which, when executed, performs at least the methods described above.

[0097] This invention also provides a control device, including a processor and a storage medium for storing a computer program; wherein the processor executes the computer program by performing at least the method described above.

[0098] This invention also provides a processor that executes a computer program, at least performing the methods described above.

[0099] The storage medium can be implemented by any type of non-volatile storage device, or a combination thereof. The non-volatile memory can be read-only memory (ROM), programmable read-only memory (PROM), erasable programmable read-only memory (EPROM), electrically erasable programmable read-only memory (EEPROM), magnetic random access memory (FRAM), flash memory, magnetic surface memory, optical disc, or compact disc read-only memory (CD-ROM); the magnetic surface memory can be a disk drive or magnetic tape drive. The storage media described in the embodiments of this invention are intended to include, but are not limited to, these and any other suitable types of memory.

[0100] In the several embodiments provided by this invention, it should be understood that the disclosed systems and methods can be implemented in other ways. The device embodiments described above are merely illustrative. For example, the division of units is only a logical functional division, and in actual implementation, there may be other division methods, such as: multiple units or components can be combined, or integrated into another system, or some features can be ignored or not executed. In addition, the coupling, direct coupling, or communication connection between the various components shown or discussed can be through some interfaces, and the indirect coupling or communication connection between devices or units can be electrical, mechanical, or other forms.

[0101] The units described above as separate components may or may not be physically separate. The components shown as units may or may not be physical units, that is, they may be located in one place or distributed across multiple network units. Some or all of the units may be selected to achieve the purpose of this embodiment according to actual needs.

[0102] In addition, in the various embodiments of the present invention, each functional unit can be integrated into one processing unit, or each unit can be a separate unit, or two or more units can be integrated into one unit; the integrated unit can be implemented in hardware or in the form of hardware plus software functional units.

[0103] Those skilled in the art will understand that all or part of the steps of the above method embodiments can be implemented by hardware related to program instructions. The aforementioned program can be stored in a computer-readable storage medium. When the program is executed, it performs the steps of the above method embodiments. The aforementioned storage medium includes various media capable of storing program code, such as mobile storage devices, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0104] Alternatively, if the integrated units of this invention are implemented as software functional modules and sold or used as independent products, they can also be stored in a computer-readable storage medium. Based on this understanding, the technical solutions of the embodiments of this invention, or the parts that contribute to the prior art, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the methods described in the various embodiments of this invention. The aforementioned storage medium includes various media capable of storing program code, such as mobile storage devices, ROM, RAM, magnetic disks, or optical disks.

[0105] The methods disclosed in the several method embodiments provided by this invention can be arbitrarily combined without conflict to obtain new method embodiments.

[0106] The features disclosed in the several product embodiments provided by this invention can be arbitrarily combined without conflict to obtain new product embodiments.

[0107] The features disclosed in the several method or device embodiments provided by the present invention can be arbitrarily combined without conflict to obtain new method or device embodiments.

[0108] The above description, in conjunction with specific preferred embodiments, provides a further detailed explanation of the present invention. It should not be construed that the specific implementation of the present invention is limited to these descriptions. For those skilled in the art, various equivalent substitutions or obvious modifications can be made without departing from the concept of the present invention, and all such modifications, achieving the same performance or application, should be considered within the scope of protection of the present invention.

Claims

1. A distributed decision optimization method for virtual power plants based on McCormick envelopes, characterized in that, The method includes: A two-layer optimization framework is constructed, including an optimization model for power user decision-making and an optimization model for virtual power plant operators; In the power user decision optimization model, power user satisfaction is defined, and the interactive electricity volume between power users and virtual power plant operators is optimized based on the interactive electricity price published by the virtual power plant operator; in the virtual power plant operator optimization model, day-ahead optimization decisions are made based on the power user's bidding strategy. The McCormick envelope theorem is used to relax the non-convex part, transforming the bi-level optimization problem into a mixed-integer linear programming problem for processing. The mixed-integer linear programming problem is solved in a distributed manner using an improved alternating direction multiplier method, and the solution is iterated until a preset optimal solution gap threshold is reached. The power user decision optimization model includes: determining the user's satisfaction with electricity consumption based on the user's preference parameters, user characteristic coefficients, and load power, wherein the load power does not exceed the set upper and lower limits of load power; determining the real-time interaction power between the power user and the virtual power plant based on the difference between the internal electricity price published by the virtual power plant operator and the grid benchmark electricity price, combined with the price response coefficient and interaction power upper limit set by the power user; the determination of the interaction power upper limit takes into account the operational constraints of distributed resources on the power user side, including the load power range, the output range of renewable energy units, the output range and ramping limit of traditional units, and the charging and discharging power limit of the energy storage system.

2. The distributed decision optimization method for virtual power plants as described in claim 1, characterized in that, The power user decision optimization model includes: Define the satisfaction level of electricity users when consuming electricity. : In the formula, It is a preference parameter. It is a user characteristic coefficient. It is the load power of the i-th user at time t. Indicates load, These are the upper and lower limits of the load power, respectively; Electricity users submit application information to the virtual power plant operator. This application information includes: , , respectively, represent the price response coefficient of electricity user i at time t, and the upper limit of the interaction between electricity user and virtual power plant. S indicates selling electricity, and if the superscript is B, it indicates buying electricity. The real-time interaction power between electricity user and virtual power plant is obtained from this declaration information: In the formula, This is the price at which the virtual power plant purchases its power in the internal market. It is the grid electricity price at time t. This is the upper limit of the output of thermal power units contracted by electricity users. This is the upper limit of the output of new energy generating units. This is the upper limit of the discharge power of the energy storage system. It is the lower limit of the charging power of the energy storage system. It refers to charge / discharge efficiency.

3. The distributed decision optimization method for virtual power plants as described in claim 2, characterized in that, The application strategy is derived from the following optimization model: maximize: , indicating variables The optimization objective is to maximize the total revenue of users. Formula (5) is used as the objective function, and formulas (6)-(15) are used as constraints. The maximum value of the objective function is found within the constraints, which is also called "optimization model solution". The relevant constraints are: Equation (5) demonstrates the benefits that electricity users gain by participating in the virtual power plant internal market. This represents the amount of electricity sold directly by users to the power company. For the corresponding interactive electricity price, the secondary and primary costs of thermal power units are respectively determined by... and express, The active power of the thermal power unit; Formula (6) describes the amount of electricity submitted by the power user to the virtual power plant operator at a specific time t, which includes the active power from the thermal power unit. Output power of renewable energy Discharge power of energy storage system Charging power of energy storage system Load power Battery level directly interacted with other users And the amount sold directly to the power grid company The output power of thermal power units and renewable energy units meets their upper and lower limits. Equations (10)-(11) describe the climbing ability of thermal power units. and These are respectively uphill and downhill restrictions. Indicates time The previous adjacent time; formulas (12)-(15) focus on the operational constraints of the energy storage system, This represents the real-time power of the energy storage system and meets the upper and lower limits. M is a large positive number. It is a zero-one auxiliary variable.

4. The distributed decision optimization method for virtual power plants as described in claim 3, characterized in that, The virtual power plant operator optimization model is based on the following formula: in The corresponding constraints are Equation (16) describes the minimum operating cost of the virtual power plant. , The risk mitigation strategy adopted by the virtual power plant is represented by 's', where 's' is the risk mitigation strategy for each generated scenario. The total number of scenes, It's the probability of the scenario. It is conditional risk value. and These refer to the cost of purchasing electricity and the revenue from selling electricity, respectively. The revenue obtained by the virtual power plant from participating in the electricity market transaction is shown in equations (17)-(19). For electricity purchases under specific scenario x, For the electricity sales volume in a specific scenario x, T refers to the total duration of the optimization time period. It is the electricity price in the internal market; Equation (20) defines the method for calculating conditional risk value. Value at risk, It is the confidence coefficient. The disturbance parameter is used as an auxiliary variable; Equation (21) describes the total bid volume of the virtual power plant in the market. , For the total purchased electricity volume, For the total electricity sales, equations (22) and (23) involve price restrictions in the internal market, which are based on the electricity market price. Compared with retail market prices Between; Equation (24) indicates the amount of penalty for breach of contract. The relationship between and value at risk As auxiliary variables; equations (25)-(26) represent the total electricity purchased and sold by the virtual power plant to electricity users. For the i-th user, the amount of electricity sold to the internal market LM. For the j-th user, the electricity purchased from the internal market LM, For the group of electricity buyers, equation (27) depicts the penalty for default by the virtual power plant. This is the default price coefficient. Let x be the net transaction volume under a specific scenario; Equations (28)-(29) show that the actual electricity volume received by the virtual power plant from users is uncertain. For electricity sales users, and Total electricity volume for each uncertain scenario and This represents the probability distribution for the corresponding uncertainty scenario.

5. The distributed decision optimization method for virtual power plants as described in claim 4, characterized in that, For the bilayer optimization problem, McCormick's envelope theorem is used to relax the non-convex part: In the formula, and These are the auxiliary variables for buying and selling at time t, and auxiliary constraints are introduced based on the original upper and lower bound constraints of each of these variables: This represents the upper limit for the electricity sales volume of virtual power plants. The upper limit for the amount of electricity that virtual power plants can purchase. To determine the total electricity purchased by the virtual power plant, the bi-level non-convex optimization problem is transformed into a mixed-integer linear programming problem.

6. The distributed decision optimization method for virtual power plants as described in claim 5, characterized in that, The improved alternating direction multiplier method includes: during the initial iteration, running a sub-optimization problem of the virtual power plant operator; updating the virtual power plant's electricity purchase and sale prices and incorporating the sub-optimization problem of the electricity user; updating the multipliers and penalty parameters; and iteratively solving until a preset optimal solution gap threshold is met.

7. The distributed decision optimization method for virtual power plants as described in claim 6, characterized in that, The distributed solution process includes: The unified augmented form of the bilevel optimization problem is: in, To optimize the objective function, Basic income item, Estimate the amount of electricity sold by the i-th user to the internal market LM. For the estimated electricity purchase amount of the i-th user from the internal market LM, and The initial value is 0. and These are the multipliers corresponding to different constraints; and These are the penalty parameters corresponding to different constraints; First, when the initial value of the iteration number k is 0, substitute it into... and Given an initial value of 0, the sub-optimization problem of running a virtual power plant operator is as follows: The electricity purchase price in the virtual power plant's internal market during the (k+1)th iteration. With electricity sales price , ":" indicates that it is in the corresponding round. Then, incorporate the updated virtual power plant electricity purchase and sale prices into the sub-optimization problem for electricity users: Further update the multipliers: Simultaneously updated ， Iterative solution until and These are the deviations between the electricity sold and purchased by users to the virtual power plant's internal market and their respective forecasts. and They are respectively and The allowable deviation threshold.

8. The distributed decision optimization method for virtual power plants as described in claim 7, characterized in that, The minimum difference between the iteration point and the optimal solution is set to 0.

005.

9. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed by a processor, it implements the method as described in any one of claims 1-8.

10. A computer program product, characterized in that, The computer program product is executed by a processor to implement the method as described in any one of claims 1-8.

Images