Virtual power plant multi-market bidding optimization method considering energy storage flexible scheduling

Abstract

The invention relates to the field of power systems, and particularly discloses a virtual power plant multi-market bidding optimization method considering energy storage flexible scheduling, which comprises the following steps: constructing a bidding decision model of a virtual power plant based on profit foreground values of the virtual power plant in an energy market, a frequency modulation market and a carbon market and the operation cost of the virtual power plant, the operation cost of the virtual power plant comprises the cost of the installed capacity of the independent energy storage of the time-sharing scheduling external operation; based on the profit foreground values of the independent energy storage in the energy market and the frequency modulation market and the profit of the installed capacity of the independent energy storage, constructing a cost optimization model of the independent energy storage; and by taking the bidding decision model as an upper layer model and the cost optimization model as a lower layer model, constructing a double-layer bidding optimization model of the virtual power plant multi-element market considering energy storage flexible scheduling, and solving to obtain a bidding optimization strategy in an equilibrium state and a scheduling strategy of the virtual power plant in the state. The scheduling capability of the virtual power plant can be remarkably improved.

Description

Technical Field

[0001] The present invention relates to the field of power systems, and in particular, to a method for optimizing the multi-market bidding of a virtual power plant considering flexible energy storage scheduling. Background Art

[0002] The proportion of new energy has been rising rapidly, and the random volatility of its output needs to be offset by flexible resources. Using market-based means to tap flexible resources on the user side is the key to meeting the system's flexibility requirements. As an aggregation carrier of user-side resources, a virtual power plant (VPP) can effectively solve the problem that it is difficult for user-side flexible resources to participate in power market transactions independently due to their dispersion and strong randomness. However, the VPP contains uncontrollable resources such as renewable energy, and there may be a problem of insufficient flexibility regulation in a short period.

[0003] In view of this, how to improve the short-term flexibility regulation ability of the virtual power plant VPP has become an urgent problem to be solved. Summary of the Invention

[0004] Therefore, the present invention provides a method for optimizing the multi-market bidding of a virtual power plant considering flexible energy storage scheduling, aiming to solve or at least alleviate the above problems.

[0005] According to one aspect of the present invention, there is provided a method for optimizing the multi-market bidding of a virtual power plant considering flexible energy storage scheduling, wherein the virtual power plant schedules the installed capacity of an independent energy storage operating externally in periods, and the method includes: constructing a bidding decision model of the virtual power plant with the maximum total revenue within the scheduling period of the virtual power plant as the objective function based on the revenue prospect values of the virtual power plant in the energy market, frequency regulation market, and carbon market and the operating cost of the virtual power plant, wherein the operating cost of the virtual power plant includes the cost of scheduling the installed capacity of the independent energy storage at different times; constructing a cost optimization model of the independent energy storage with the maximum total revenue within the scheduling period of the independent energy storage as the objective function based on the revenue prospect values of the independent energy storage in the energy market, frequency regulation market, and the revenue from renting out its installed capacity; constructing a two-layer bidding optimization model of the multi-market of the virtual power plant considering flexible energy storage scheduling with the bidding decision model as the upper model and the cost optimization model as the lower model; solving the two-layer bidding optimization model to obtain the bidding optimization strategy in the equilibrium state and the scheduling strategy of the virtual power plant in this state, wherein the bidding optimization strategy includes the declared quantities of the virtual power plant participating in the energy market, frequency regulation market, and carbon market at each moment, and the scheduling strategy includes at least the installed capacity of the independent energy storage scheduled by the virtual power plant at each moment and the power exchanged with the independent energy storage.

[0006] Optionally, in the virtual power plant multi-market bidding optimization method considering flexible energy storage scheduling according to the present invention, determining the total revenue within the scheduling period of the virtual power plant includes: obtaining the revenue prospect values of the virtual power plant in the day-ahead energy market, real-time energy market, frequency regulation capacity market, frequency regulation mileage market, and carbon market at each moment; based on the revenue prospect values of each market of the virtual power plant at each moment, obtaining the total revenue prospect value within the scheduling period of the virtual power plant, and subtracting the total operating cost within the scheduling period of the virtual power plant from it to obtain the total revenue within the scheduling period of the virtual power plant.

[0007] Optionally, in the virtual power plant multi-market bidding optimization method considering flexible energy storage scheduling according to the present invention, determining the total revenue within the scheduling period of the independent energy storage includes: obtaining the revenue prospect values of the independent energy storage in the day-ahead energy market and the frequency regulation capacity market at each moment; based on the revenue prospect values of each market of the independent energy storage at each moment, obtaining the total revenue prospect value within the scheduling period of the independent energy storage, and adding the total revenue from renting out its installed capacity within the scheduling period of the independent energy storage to it to obtain the total revenue within the scheduling period of the independent energy storage.

[0008] Optionally, in the virtual power plant multi-market bidding optimization method considering flexible energy storage scheduling according to the present invention, the revenue prospect values of each market of the virtual power plant / independent energy storage at each moment are obtained by the following formula:

[0009]

[0010] where V j,t represents the revenue prospect value of the virtual power plant / independent energy storage in the j market at time t, p i represents the occurrence probability of scenario i, w(p i ) represents the decision weight for p i , C j,i,t represents the revenue of the virtual power plant / independent energy storage in the j market at time t under scenario i, v(C j,i,t ) represents the decision maker's perception of C j,i,t , and N represents the total number of scenarios.

[0011] Optionally, in the virtual power plant multi-market bidding optimization method considering flexible energy storage scheduling according to the present invention, the objective function of the bidding decision model includes:

[0012]

[0013] C out,t = C GT,t + C DAbuy,t + C EV,t + C DR,t + C SES,t

[0014] Among them, f represents the total revenue within the dispatching period of the virtual power plant, C in,PT represents the total revenue prospect value within the dispatching period of the virtual power plant, C out,t represents the operating cost of the virtual power plant at time t, V DA,t 、V RT,t 、V fc,t 、V fp,t 、 respectively represent the revenue prospect values of the day-ahead energy market, real-time energy market, frequency regulation capacity market, frequency regulation mileage market and carbon market of the virtual power plant at time t, C GT,t represents the gas purchase cost of the gas turbine unit in the virtual power plant at time t, C DAbuy,t represents the power purchase cost of the day-ahead energy market of the virtual power plant at time t, C DR,t represents the demand response compensation cost of the virtual power plant at time t, C SES,t represents the cost of the installed capacity of the dispatching independent energy storage of the virtual power plant at time t, C EV,t represents the operating cost of the electric vehicle in the virtual power plant at time t, and T represents the total number of time instants.

[0015] Optionally, in the multi-market bidding optimization method of the virtual power plant considering flexible dispatching of energy storage according to the present invention, the cost of the installed capacity of the dispatching independent energy storage of the virtual power plant at each time instant includes:

[0016] C SES,t =m SES,t Q vpp0,t

[0017] Among them, C SES,t represents the cost of the installed capacity of the dispatching independent energy storage of the virtual power plant at time t, m SES,t represents the price per unit installed capacity of the dispatching independent energy storage of the virtual power plant at time t, Q vpp0,t represents the installed capacity of the dispatching independent energy storage of the virtual power plant at time t.

[0018] Optionally, in the multi-market bidding optimization method of the virtual power plant considering flexible dispatching of energy storage according to the present invention, the bidding decision-making model includes power balance constraint, upper and lower limits constraint of the output of the gas turbine unit and ramp rate upper and lower limits constraint, off-grid power constraint of the electric vehicle and charge-discharge power constraint, and demand response constraint.

[0019] Optionally, in the virtual power plant multi-market bidding optimization method considering flexible energy storage scheduling according to the present invention, the cost optimization model includes the capacity constraint of the independent energy storage, the exchange power constraint between the independent energy storage and the virtual power plant, the exchange power constraint between the independent energy storage and the power grid, and the cooperative operation constraint between the independent energy storage and the virtual power plant. Among them, the cooperative operation constraint between the independent energy storage and the virtual power plant includes: the revenue obtained from renting the installed capacity of the independent energy storage to the virtual power plant is not less than the revenue obtained by the same part of the installed capacity participating in the energy market and the frequency regulation market during the same period.

[0020] According to another aspect of the present invention, there is provided a computing device, including: at least one processor; and a memory storing program instructions, wherein the program instructions are configured to be executed by at least one processor, and the program instructions include instructions for executing the virtual power plant multi-market bidding optimization method considering flexible energy storage scheduling according to the present invention.

[0021] According to another aspect of the present invention, there is provided a readable storage medium storing program instructions, which when read and executed by a computing device, cause the computing device to execute the virtual power plant multi-market bidding optimization method considering flexible energy storage scheduling according to the present invention.

[0022] In summary, the present invention provides a virtual power plant multi-market bidding optimization method considering flexible energy storage scheduling. The virtual power plant flexibly schedules the installed capacity of the external independent energy storage in a time-sharing manner during its own operation and scheduling process. The independent energy storage flexibly adjusts the price of scheduling the energy storage in each period. Based on this cooperation mechanism, the present invention can achieve efficient scheduling and optimized bidding in multiple markets, significantly improve the regulation ability of the virtual power plant, and at the same time improve the market revenue of the virtual power plant and the independent energy storage. In addition, the present invention constructs a virtual power plant multi-market bidding optimization model considering flexible energy storage scheduling based on the revenue prospect value, which can make the bidding optimization strategy and scheduling strategy more accurate. BRIEF DESCRIPTION OF THE DRAWINGS

[0023] To achieve the above and related objectives, certain illustrative aspects are described herein in connection with the following description and the accompanying drawings, which indicate various ways in which the principles disclosed herein can be practiced, and all aspects and their equivalent aspects are intended to fall within the scope of the claimed subject matter. The above and other objectives, features, and advantages of the present disclosure will become more apparent by reading the following detailed description in conjunction with the accompanying drawings. Throughout the present disclosure, the same reference numerals generally refer to the same components or elements.

[0024] Figure 1 The block diagram of the computing device 100 according to an embodiment of the present invention is shown;

[0025] Figure 2Shows a flowchart of the virtual power plant multi-market bidding optimization method 200 considering flexible energy storage scheduling according to an embodiment of the present invention;

[0026] Figure 3 Shows a schematic diagram of the framework structure of virtual power plant bidding optimization and energy storage leasing pricing decision based on master-slave game according to an embodiment of the present invention;

[0027] Figure 4 Shows a schematic diagram of the wind power, photovoltaic and load prediction results according to an embodiment of the present invention;

[0028] Figure 5 Shows a schematic diagram of the predicted results of each market price according to an embodiment of the present invention;

[0029] Figure 6 Shows a schematic diagram of the scenarios generated by the Latin hypercube sampling method according to an embodiment of the present invention;

[0030] Figure 7 Shows a schematic diagram of the virtual power plant's day-ahead scheduling plan according to an embodiment of the present invention;

[0031] Figure 8 Shows a schematic diagram of the capacity scheduling plan of the independent energy storage according to an embodiment of the present invention;

[0032] Figure 9 Shows a schematic diagram of the dynamic leasing price of the energy storage according to an embodiment of the present invention;

[0033] Figure 10 Shows a schematic diagram of the intraday stage scheduling results according to an embodiment of the present invention. Detailed implementation

[0034] The exemplary embodiments of the present disclosure will be described in more detail below with reference to the accompanying drawings. Although the exemplary embodiments of the present disclosure are shown in the drawings, it should be understood that the present disclosure can be implemented in various forms and should not be limited by the embodiments set forth herein. On the contrary, these embodiments are provided so that the present disclosure can be more thoroughly understood and the scope of the present disclosure can be fully conveyed to those skilled in the art.

[0035] In view of the problem of insufficient short-term regulation ability existing in the current VPP, it can be improved by leasing external energy storage with the characteristics of rapid response and flexible call capacity. Further, in order to better improve the short-term regulation ability of the VPP, a flexible external energy storage leasing strategy can be adopted. However, although the VPP introducing flexible leasing of energy storage can improve the short-term regulation ability, its operation in the market environment will face challenges such as coordination of internal and external flexibility resources and coordination of multi-market joint bidding. Based on this, the present invention provides a multi-market bidding optimization method for a virtual power plant considering flexible scheduling (leasing) of energy storage, so as to give full play to the supporting role of flexible leasing / scheduling of energy storage for the VPP and maximize the market value of VPP cooperation.

[0036] The multi-market bidding optimization method for a virtual power plant considering flexible scheduling of energy storage of the present invention can be executed in a computing device. Figure 1 A block diagram of the physical components (i.e., hardware) of a computing device 100 is shown. In a basic configuration, the computing device 100 includes at least one processing unit 102 and a system memory 104. According to one aspect, depending on the configuration and type of the computing device, the processing unit 102 can be implemented as a processor. The system memory 104 includes, but is not limited to, volatile storage (e.g., random access memory), non-volatile storage (e.g., read-only memory), flash memory, or any combination of such memories. According to one aspect, an operating system 105 and program modules 106 are included in the system memory 104, and a bidding optimization module 120 is included in the program modules 106. The bidding optimization module 120 is configured to execute the multi-market bidding optimization method 200 for a virtual power plant considering flexible scheduling of energy storage of the present invention.

[0037] According to one aspect, the operating system 105 is suitable for controlling the operation of the computing device 100, for example. In addition, the examples are practiced in conjunction with a graphics library, other operating systems, or any other application programs, and are not limited to any specific application or system. In Figure 1 the basic configuration is shown by those components within the dashed line 108. According to one aspect, the computing device 100 has additional features or functions. For example, according to one aspect, the computing device 100 includes additional data storage devices (removable and / or non-removable), such as magnetic disks, optical disks, or magnetic tapes. Such additional storage is Figure 1 shown by removable storage 109 and non-removable storage 110.

[0038] As stated above, according to one aspect, program modules are stored in the system memory 104. According to one aspect, the program modules may include one or more applications, and the present invention does not limit the type of applications. For example, the applications may include: email and contact applications, word processing applications, spreadsheet applications, database applications, slide show applications, painting or computer-aided applications, web browser applications, etc.

[0039] According to one aspect, the examples may be practiced in a circuit including discrete electronic components, a package or integrated electronic chip containing logic gates, a circuit utilizing a microprocessor, or on a single chip containing electronic components or a microprocessor. For example, the examples may be practiced via a system on a chip (SOC) in which each or many of the components shown in Figure 1 are integrated on a single integrated circuit. According to one aspect, such an SOC device may include one or more processing units, graphics units, communication units, system virtualization units, and various application functions, all of which are integrated (or "burned") onto a chip substrate as a single integrated circuit. When operating via an SOC, the functions described herein may be operated via dedicated logic integrated with other components of the computing device 100 on a single integrated circuit (chip). Embodiments of the present invention may also be practiced using other technologies capable of performing logical operations (such as AND, OR, and NOT), including but not limited to mechanical, optical, fluidic, and quantum technologies. Additionally, embodiments of the present invention may be practiced within a general-purpose computer or in any other circuit or system.

[0040] According to one aspect, the computing device 100 may also have one or more input devices 112, such as a keyboard, mouse, pen, voice input device, touch input device, etc. An output device 114, such as a display, speaker, printer, etc., may also be included. The foregoing devices are examples and other devices may also be used. The computing device 100 may include one or more communication connections 116 that allow communication with other computing devices 118. Examples of suitable communication connections 116 include but are not limited to: RF transmitter, receiver, and / or transceiver circuits; universal serial bus (USB), parallel, and / or serial ports.

[0041] As used herein, the term computer-readable medium includes computer storage media. Computer storage media can include volatile and non-volatile, removable and non-removable media implemented in any method or technology for storing information (e.g., computer-readable instructions, data structures, or program modules). System memory 104, removable storage 109, and non-removable storage 110 are all examples of computer storage media (i.e., memory storage). Computer storage media can include random access memory (RAM), read-only memory (ROM), electrically erasable read-only memory (EEPROM), flash memory or other memory technologies, CD-ROM, digital versatile disk (DVD) or other optical storage, magnetic cassettes, magnetic tape, magnetic disk storage or other magnetic storage devices, or any other article of manufacture that can be used to store information and that can be accessed by computer device 100. According to one aspect, any such computer storage media can be part of computing device 100. Computer storage media does not include carrier waves or other propagated data signals.

[0042] According to one aspect, a communication medium is implemented with computer-readable instructions, data structures, program modules, or other data in a modulated data signal (e.g., a carrier wave or other transmission mechanism) and includes any information delivery medium. According to one aspect, the term "modulated data signal" describes a signal having one or more sets of characteristics or a signal that has been altered in such a manner as to encode information in the signal. By way of example, and not limitation, communication media includes wired media such as a wired network or direct-wired connection, and wireless media such as acoustic, radio frequency (RF), infrared, and other wireless media.

[0043] Figure 2 FIG. 200 is a flowchart of a method for optimizing multi-market bidding of a virtual power plant considering flexible scheduling of energy storage according to an embodiment of the present invention. The method 200 is adapted to be executed in a computing device (e.g., Figure 1 the computing device 100 shown).

[0044] Here, a virtual power plant will be described first. According to an embodiment of the present invention, a virtual power plant includes distributed clean energy (wind power, photovoltaic), gas turbines, electric vehicles, controllable loads, and distributed energy storage. The virtual power plant participates in the energy market, the frequency regulation market, and the carbon market.

[0045] Among them, in view of the problem that the virtual power plant has insufficient short-term regulation ability during the independent operation process, resulting in an increase in electricity costs, an independent energy storage system that operates independently is established outside the virtual power plant in this embodiment (i.e., there is an independent energy storage system that operates independently outside the virtual power plant), and a dynamic interaction mechanism between the virtual power plant and the independent energy storage is designed.

[0046] The virtual power plant can schedule the installed capacity of independently operated external energy storage in different time periods. The independent energy storage can independently participate in the energy market and the frequency regulation market, or rent its capacity to the virtual power plant. Specifically, the virtual power plant can flexibly lease the installed capacity from the independent energy storage system in different time periods throughout the day, obtain the right to use this part of the capacity, and perform charge and discharge operations on it to achieve flexible utilization of electric energy. While renting out its own installed capacity, the independent energy storage operator can also use its remaining installed capacity to participate in the energy and frequency regulation markets to obtain benefits.

[0047] Based on this, the following interaction modes can be adopted between the virtual power plant and the independent energy storage: (1) During its own operation and scheduling process, the virtual power plant flexibly leases (schedules) part of the installed capacity from the external energy storage at different times to optimize its own scheduling, and the size of the leased capacity in each time period is directly determined by the virtual power plant and fed back to the independent energy storage; (2) By passively accepting the size of the capacity leased by the virtual power plant, the energy storage adjusts its output in the energy and frequency regulation markets in each time period according to its remaining installed capacity, and with the goal of maximizing its overall benefits, flexibly adjusts the lease price in each time period and feeds it back to the virtual power plant.

[0048] In summary, the specific interaction principle between the virtual power plant and the independent energy storage in this embodiment is: The virtual power plant leases the installed capacity of the independent energy storage and has the right to flexibly use it within the performance range of the energy storage device, including occupying its charge and discharge output capacity; The electric energy within the leased capacity of the virtual power plant belongs to the virtual power plant, and the virtual power plant must ensure that this part of the electric energy does not exceed the leased capacity in this time period; The energy storage operator can independently participate in the energy and frequency regulation markets on the premise of ensuring the use of the leased capacity by the virtual power plant.

[0049] After understanding the dynamic interaction mechanism between the virtual power plant and the independently operated external energy storage, it can be seen that both the virtual power plant and the independent energy storage operator pursue the maximization of their own benefits, so there is a certain game relationship between the two.

[0050] Among them, in this game process, there are two decision-making subjects at the same time. Both have the ability to make independent decisions, and the decision results of each will affect the final benefits of each other. Specifically, the virtual power plant and the independent energy storage both pursue the maximization of benefits. In the process of pursuing the maximization of benefits, the virtual power plant needs to reduce the cost of the independent energy storage, but for the independent energy storage, this is to reduce its rental income.

[0051] Therefore, the dynamic fixed-capacity pricing result of this embodiment is generated along with the decisions of both sides of the game. Specifically, during the game process, the virtual power plant determines the size of the capacity leased from the independent energy storage, and the energy storage power station determines the dynamic rental price per unit of capacity. Among them, the decision result of the virtual power plant directly affects the rental income of the independent energy storage and is in a dominant position. Therefore, it can act as the game leader in the master-slave game process and transmit the size of the leased capacity to the lower layer; the income of the independent energy storage is affected by the decision result of the upper layer and can act as the game follower, transmitting the rental price to the upper layer; this game process is continuously iterated until the optimal solution of the master-slave game is finally generated.

[0052] Based on the above analysis of the game behavior of the two, according to some embodiments of the present invention, a master-slave game model between the virtual power plant and the independent energy storage can be constructed: the upper layer is the virtual power plant multi-market bidding optimization model, and the virtual power plant operator formulates market bidding strategies including energy storage leasing strategies according to multi-market prices, energy storage leasing prices, internal energy supply and demand conditions, etc.; the lower layer is the independent energy storage cost optimization model, which optimizes the bidding for the energy storage capacity leased by the virtual power plant to participate in market transactions independently, and formulates the energy storage leasing price based on the potential market income; the virtual power plant obtains the optimal bidding strategy according to the energy storage leasing price fed back by the independent energy storage, and repeats the above game process until equilibrium is reached. As Figure 3 , which shows the framework structure diagram of the virtual power plant bidding optimization and energy storage leasing pricing decision based on the master-slave game according to an embodiment of the present invention.

[0053] Next, the virtual power plant multi-market bidding optimization method considering flexible energy storage scheduling of the present invention will be specifically described. As Figure 2 shown, the virtual power plant multi-market bidding optimization method 200 considering flexible energy storage scheduling of the present invention starts at 210.

[0054] In 210, based on the revenue prospect values of the virtual power plant in the energy market, frequency regulation market, and carbon market, as well as the operating cost of the virtual power plant, a bidding decision model of the virtual power plant is constructed with the maximum total revenue within the virtual power plant scheduling period as the objective function. Among them, the operating cost of the virtual power plant includes the cost of the installed capacity of the time-sharing scheduled independent energy storage.

[0055] Specifically, in some embodiments, the total revenue within the virtual power plant scheduling period can be determined in the following manner, and then a bidding decision model of the virtual power plant is constructed with the maximum total revenue within the virtual power plant scheduling period as the objective function.

[0056] Step 1: Obtain the revenue prospect values of the virtual power plant in the day-ahead energy market, real-time energy market, frequency regulation capacity market, frequency regulation mileage market, and carbon market at each moment. Here, it should be noted that considering a large number of uncertain factors in the system, in this embodiment, to quantitatively evaluate its impact on market revenue, based on the prospect theory, the subjective revenue of the virtual power plant operator is characterized to determine the total revenue within the dispatching cycle of the virtual power plant using the revenue prospect values of each market.

[0057] According to an embodiment of the present invention, the revenue prospect values of each market of the virtual power plant at each moment can be obtained by the following formula.

[0058]

[0059] In the formula, V j,t represents the revenue prospect value of the j-th market of the virtual power plant at time t, p i represents the occurrence probability of scenario i, w(p i ) represents the decision weight for p i , C j,i,t represents the revenue of the j-th market of the virtual power plant at time t under scenario i, v(C j,i,t ) represents the decision maker's perception of C j,i,t , and N represents the total number of scenarios.

[0060] Next, the acquisition methods of the value function v(C j,i,t ) and the decision weight w(p i ) will be specifically described.

[0061] 1) Regarding the value function v(C j,i,t ), in some embodiments, it can be obtained by the following formula.

[0062]

[0063] In the formula, r is the reference point, which is set as the bidding revenue of each market of the virtual power plant under the expected value of uncertain factors in the embodiment, and is a predicted value. Further, it is the value calculated based on the initial set of predicted values; α and β are the coefficients for perceiving gains and losses, and usually the value range is [0, 1]; θ is the loss aversion coefficient, and usually the value is greater than 1.

[0064] Among them, j represents the market type, specifically including the day-ahead energy market, real-time energy market, frequency regulation capacity market, frequency regulation mileage market, and carbon market. Next, the acquisition methods of the revenues of each market of the virtual power plant at each moment under each scenario will be described.

[0065] The energy market revenue consists of the day-ahead market revenue and the real-time market revenue. After scheduling in the day-ahead stage, the remaining electricity is declared to participate in the day-ahead market, and the electricity with deviations in the real-time stage participates in the real-time market. Among them, the day-ahead energy market revenue and the real-time energy market revenue are as follows.

[0066] The day-ahead energy market revenue of the virtual power plant at each moment in each scenario (i.e., when j is the day-ahead energy market) is obtained through the following formula.

[0067] C DA,i,t =P DA,i,t m DA,i,t (3)

[0068] In the formula, C DA,i,t represents the day-ahead energy market revenue of the virtual power plant at time t in scenario i, P DA,t represents the day-ahead energy market electricity sales volume of the virtual power plant at time t in scenario i, and m DA,t represents the day-ahead energy market electricity sales price of the virtual power plant at time t in scenario i.

[0069] The real-time energy market revenue of the virtual power plant at each moment in each scenario (i.e., when j is the real-time energy market) is obtained through the following formula.

[0070] C RT,i,t =P RT,i,t m RT,i,t (4)

[0071] In the formula, C RT,i,t represents the real-time energy market revenue of the virtual power plant at time t in scenario i, P RT,i,t represents the real-time energy market electricity sales volume of the virtual power plant at time t in scenario i, and m RT,i,t represents the real-time energy market electricity sales price at time t in scenario i.

[0072] The frequency regulation market revenue consists of frequency regulation capacity compensation, frequency regulation mileage compensation, and assessment penalties. After the virtual power plant meets its own load demand and energy market transactions, there will be a certain surplus in the system's own output capacity, and this surplus can be used to participate in the frequency regulation market for capacity bidding. After the real-time settlement is completed, the settlement compensation and assessment penalties will be carried out again according to the actual frequency regulation mileage. Among them, the frequency regulation capacity market revenue and the frequency regulation mileage market revenue are as follows.

[0073] The frequency regulation capacity market revenue of the virtual power plant at each moment in each scenario (i.e., when j is the frequency regulation capacity market) is obtained through the following formula.

[0074] C fc,i,t =P RE,i,t K t m fc,i,t (5)

[0075] Where, C fc,i,t represents the frequency regulation capacity compensation (i.e., the revenue from the frequency regulation capacity market) of the virtual power plant at time t in scenario i, P RE,i,t represents the declared frequency regulation capacity of the virtual power plant at time t in the day-ahead stage in scenario i, K t represents the frequency regulation performance parameter at time t, m fc,i,t represents the frequency regulation capacity compensation price at time t in scenario i.

[0076] The revenue of the virtual power plant from the frequency regulation mileage market at each moment in each scenario (i.e., when j is the frequency regulation mileage market) is obtained through the following formula.

[0077] C fp,i,t = λ fp,i,t P RE,i,t K t m fp,i,t + T R,i,t (6)

[0078] Where, C fp,i,t represents the frequency regulation mileage compensation of the virtual power plant at time t in scenario i, λ fp,i,t represents the actual frequency regulation mileage call coefficient at time t in scenario i, and its distribution is random, m fp,i,t represents the frequency regulation mileage compensation price at time t in scenario i, T R,i,t represents the assessment penalty at time t in scenario i.

[0079] The carbon market revenue consists of the carbon quota (CEA) market revenue and the voluntary emission reduction (CCER) market revenue. Under the constraint of the carbon emission right system, each entity needs to emit CO2 within its respective emission quota, and excessive emissions will be punished. Some entities cannot fully use the initial quota during the production process, so they can sell this part of the quota in the market; at the same time, due to the limited free carbon quotas initially issued, some entities can conduct transactions in the CEA and CCER markets to purchase and use them to offset emissions after fully using them. Among them, the CCER market price is generally lower than that of CEA. To ensure the coordinated development of the two markets in each region, the CCER consumption ratio is restricted, generally not exceeding 5% to 10% of the total consumption. Therefore, the carbon market revenue is as follows.

[0080] When the emissions are less than the initial carbon quota, the carbon market revenue of the virtual power plant at each moment in each scenario is obtained through the following formula.

[0081]

[0082] Where, represents the carbon market revenue of the virtual power plant at time t in scenario i, m CEA represents the unit price of carbon quota, n 0,i,t 、n GT,i,t 、nGrid,i,t respectively represent the initial carbon quota, carbon emissions of gas turbines, and indirect carbon emissions from power grid power purchases of the virtual power plant at time t in scenario i.

[0083] When the emissions are greater than the initial carbon quota, the carbon market revenue of the virtual power plant at each moment in each scenario is obtained through the following formula.

[0084]

[0085] In the formula, φ CCER represents the maximum absorption ratio of CCER stipulated locally, and m CCER represents the unit price of CCER.

[0086] Here, it should be noted that in this embodiment, considering the time division of carbon emission costs, the initial carbon quota is also allocated by time period. However, in practice, the allocation time of the initial carbon quota is based on the carbon emission compliance cycle, generally one to two years for one cycle. Therefore, the allocation here is the self - plan of the virtual power plant.

[0087] 2) Regarding the decision - making weight w(p i ), in some embodiments, it can be obtained through the following formula.

[0088]

[0089] In the formula, γ is the decision - making weight parameter. When C j,i,t - r ≥ 0, γ represents the risk - obtaining attitude coefficient; when C j,i,t - r < 0, γ represents the risk - loss attitude coefficient.

[0090] According to an embodiment of the present invention, the occurrence probability p i of each scenario can be obtained by using the Latin hypercube sampling method. Specifically, based on the probability distribution functions of each uncertainty factor, the Latin hypercube sampling method is used for scenario generation and scenario reduction to obtain multiple scenarios composed of each uncertainty factor and the probabilities of each scenario. In some embodiments, it can be set to obtain ten scenarios with the largest probabilities composed of each uncertainty factor and the probabilities of each scenario. In this case, the value of the total number of scenarios N is 10. Of course, this is only an example, and the present invention is not limited thereto. In addition, using the Latin hypercube sampling method for scenario generation is a relatively mature technical means at present, so it will not be elaborated here.

[0091] Among them, the uncertainty factors include: the output of wind power and photovoltaic power at each moment; the frequency modulation call mileage at each moment; the power purchase price and power sale price in the day - ahead energy market and real - time energy market at each moment; the declared price of the frequency modulation capacity at each moment; the load demand at each moment. Regarding each uncertainty factor, in some embodiments, different probability density functions can be used to characterize it, as follows.

[0092] The wind speed at time t follows a Weibull distribution:

[0093]

[0094] where v WT,t represents the predicted wind speed value at time t, and k and λ are the shape and scale parameters of the Weibull distribution.

[0095] The light intensity at time t follows a Beta distribution:

[0096]

[0097] where δ PV,t represents the predicted light intensity value at time t, Γ(z) is the Γ function, and α′ and β′ are parameters.

[0098] The actual frequency regulation call mileage and market price of the unit frequency regulation capacity at time t both follow a lognormal distribution:

[0099]

[0100] where μ and σ are parameters, x represents the random variable value, that is, the predicted values of the frequency regulation call mileage and market price, specifically including the frequency regulation call mileage at each moment, the declared price of the frequency regulation capacity at each moment, and the power purchase price and power sale price at each moment in the day-ahead energy market and real-time energy market.

[0101] The load demand at time t follows a standard normal distribution:

[0102]

[0103] where x ′ represents the predicted load value at time t, μ is the overall expectation, and σ is the overall standard deviation.

[0104] Step 2: Based on the revenue prospect values of each market of the virtual power plant at each moment, obtain the total revenue prospect value within the dispatching period of the virtual power plant, and subtract it from the total operating cost within the dispatching period of the virtual power plant to obtain the total revenue within the dispatching period of the virtual power plant.

[0105] Among them, the total revenue prospect value within the dispatching period of the virtual power plant can be obtained by the following formula.

[0106]

[0107] where C in,PT represents the total revenue prospect value within the dispatching period of the virtual power plant, V DA,t represents the revenue prospect value of the day-ahead energy market of the virtual power plant at time t, V RT,tRepresents the revenue prospect value of the real-time energy market of the virtual power plant at time t, V fc,t Represents the revenue prospect value of the frequency regulation capacity market of the virtual power plant at time t, V fp,t Represents the revenue prospect value of the frequency regulation mileage market of the virtual power plant at time t Represents the revenue prospect value of the carbon market of the virtual power plant at time t. T represents the total number of time periods. In some embodiments, its value can be set to 24, that is, the scheduling period is one day

[0108] After obtaining the total revenue prospect value within the scheduling period of the virtual power plant, according to an embodiment of the present invention, the total revenue within the scheduling period of the virtual power plant can be obtained by the following formula

[0109]

[0110] In the formula, f represents the total revenue within the scheduling period of the virtual power plant, C out,t Represents the operating cost of the virtual power plant at time t, and T represents the total number of time periods

[0111] Among them, the operating cost of each time period of the virtual power plant can be obtained by the following formula

[0112] C out,t =C GT,t +C DAbuy,t +C EV,t +C DR,t +C SES,t (16)

[0113] In the formula, C GT,t Represents the gas purchase cost of the gas turbine unit in the virtual power plant at time t, C DAbuy,t Represents the electricity purchase cost in the day-ahead energy market of the virtual power plant at time t, C DR,t Represents the demand response compensation cost of the virtual power plant at time t, C SES,t Represents the cost of the installed capacity of the independent energy storage dispatched by the virtual power plant at time t, C EV,t Represents the operating cost of the electric vehicle in the virtual power plant at time t

[0114] Next, the acquisition methods of the gas purchase cost of the gas turbine unit, the electricity purchase cost in the day-ahead energy market, the demand response compensation cost, the cost of the installed capacity of the independent energy storage dispatched, and the operating cost of the electric vehicle at each time period will be described separately

[0115] Gas purchase cost of the gas turbine unit at each time period

[0116] C GT,t =m gas v gas,t (17)

[0117] In the formula, m gasDenote the natural gas purchase price, v gas,t Denote the natural gas consumption of the gas turbine at time t.

[0118] The day-ahead energy market power purchase cost at each time:

[0119] C DAbuy,t = m DAbuy,t P DAbuy,t (18)

[0120] In the formula, m DAbuy,t Denote the day-ahead power purchase price at time t, P DAbuy,t Denote the day-ahead power purchase power at time t.

[0121] The operation cost of electric vehicles at each time:

[0122]

[0123] In the formula, m c Denote the charging unit price, n denote the total number of electric vehicles, p c,y,t Denote the charging power of the y-th electric vehicle at time t.

[0124] The installed capacity cost of the dispatching independent energy storage at each time:

[0125] C SES,t = m SES,t Q vpp0,t (20)

[0126] In the formula, C SES,t Denote the cost of the installed capacity of the virtual power plant dispatching (leasing) independent energy storage at time t, which is equivalent to the income of the installed capacity of the independent energy storage leased at time t, m SES,t Denote the price of the installed capacity per unit of the virtual power plant dispatching independent energy storage at time t, Q vpp0,t Denote the installed capacity of the virtual power plant dispatching independent energy storage at time t.

[0127] The demand response compensation cost at each time: Demand response can be divided into load interruption and load control. Among them, a certain compensation is required for interrupted load, so this cost consists of the interrupted load compensation cost, which is specifically as follows.

[0128] C DR,t = P cut,t m cut (21)

[0129] In the formula, C DR,t Denote the interrupted load compensation cost at time t, that is, the demand response compensation cost at time t, P cut,t Denote the interrupted load volume at time t, m cut Denote the interrupted load compensation price.

[0130] Step 3: Taking the maximum total revenue within the dispatching period of the virtual power plant as the objective function, a bidding decision-making model of the virtual power plant is constructed. Specifically, in some embodiments, the objective function of the bidding decision-making model of the virtual power plant (for the convenience of description, it can be called the first objective function) can be expressed as the following formula.

[0131]

[0132] So far, the objective function of the bidding decision-making model has been obtained. Further, in some embodiments, the bidding decision-making model of the virtual power plant also includes constraint conditions corresponding to the first objective function, which are called the first constraint conditions for the convenience of description. Among them, the first constraint conditions include power balance constraint, upper and lower limits of the output of gas turbines, upper and lower limits of ramping, off-grid power constraint of electric vehicles, charging and discharging power constraint, and demand response constraint. Next, each constraint will be described.

[0133] Power balance constraint: For the virtual power plant, the total output of equipment at time t should be equal to the total load, which can be specifically expressed as the following formula.

[0134] P PV,t +P WT,t +P GT,t +P EV,t +P SES,t +P DAbuy,t +P RTbuy,t =P load1,t +P DA,t +P RT,t +λ fp,t P RE,t (23)

[0135] In the formula, P PV,t 、P WT,t 、P GT,t respectively represent the output of photovoltaic, wind power, and gas turbines at time t, P EV,t 、P SES,t respectively represent the exchange power of electric vehicles and leased energy storage at time t, P DAbuy,t 、P RTbuy,t respectively represent the power purchase from the power grid at time t for the day-ahead and real-time, P load1,t represents the total load after demand response at time t, P DA,t 、P RT,t respectively represent the power selling of the day-ahead energy market and real-time energy market at time t, P RE,t represents the declared frequency regulation capacity at time t in the day-ahead stage, and λ fp,t represents the actual frequency regulation mileage call coefficient at time t.

[0136] Output upper and lower limit constraints of gas turbines: Constraints can be imposed through the upper and lower limits of natural gas consumption, which can be specifically expressed by the following formula.

[0137] v gas,min ≤v gas,t ≤v gas,max (24)

[0138] In the formula, v gas,min and v gas,max respectively represent the minimum and maximum natural gas consumption of the gas turbine, and v gas,t represents the natural gas consumption of the gas turbine at time t.

[0139] Ramp-up and ramp-down upper and lower limit constraints of gas turbines: Constraints can be imposed through the change in natural gas consumption, which can be specifically expressed by the following formula.

[0140] v gas,cdmax ≤v gas,t -v gas,t-1 ≤v gas,cumax (25)

[0141] In the formula, v gas,cdmax and v gas,cumax respectively represent the maximum change in natural gas consumption of the gas turbine in the cooling and ramp-up states, and v gas,t-1 represents the natural gas consumption of the gas turbine at time (t - 1).

[0142] Off-grid power constraints of electric vehicles: At the moment when an electric vehicle goes off-grid, the battery power of the electric vehicle should not be lower than a fixed threshold to ensure its use after going off-grid, which can be specifically expressed by the following formula.

[0143]

[0144] In the formula, SOC out is the minimum SOC (State Of Charge) of the off-grid battery of the electric vehicle specified by this system, is the battery power of the y-th electric vehicle at the off-grid time t, and δ EVmax is the battery capacity of the electric vehicle.

[0145] Charging and discharging power constraints of electric vehicles: It can be specifically expressed by the following formula.

[0146] p dmax ≤p c,y,t ≤p cmax (27)

[0147] In the formula, p cmax and p dmax respectively represent the maximum charging power and discharging power of a single electric vehicle, and p c,y,tRepresents the charging power of the y-th electric vehicle at time t.

[0148] Demand response constraint: The demand response of the virtual power plant needs to satisfy that the total change in controlled load is 0, and at the same time, the interrupted load cannot exceed a certain proportion, which can be specifically expressed by the following formula.

[0149]

[0150] |P con,t |≤θ con P DR,t (29)

[0151] P cut,t ≤θ cut P DR,t (30)

[0152] P DR,t ≤θ DR P load0,t (31)

[0153] θ con +θ cut ≤1 (32)

[0154] In the formula, P con,t represents the controlled load at time t, θ con represents the maximum proportion of the controlled load in the total demand response, P DR,t represents the absolute value of the total amount of load response at time t, θ cut represents the maximum proportion of the interrupted load in the total demand response, θ DR represents the maximum proportion of the demand response load in the original load, P load0,t represents the total load before demand response at time t.

[0155] So far, the construction of the bidding decision-making model of the virtual power plant has been completed. Next, the construction of the cost optimization model of the independent energy storage will be described as follows.

[0156] In 220, based on the revenue prospect values of the independent energy storage in the energy market and the frequency regulation market, as well as the revenue from renting out its installed capacity, a cost optimization model of the independent energy storage is constructed with the maximum total revenue within the independent energy storage scheduling period as the objective function.

[0157] Specifically, according to an embodiment of the present invention, the total revenue within the independent energy storage scheduling period can be determined in the following manner, and then a cost optimization model of the independent energy storage is constructed with the maximum total revenue within the independent energy storage scheduling period as the objective function.

[0158] Step 1: Obtain the revenue prospect values of the independent energy storage in the day-ahead energy market and the frequency regulation capacity market at each moment. Specifically, the above formula (1) can be used to obtain them.

[0159] To distinguish from the revenue prospect values of the day-ahead energy market and the frequency regulation capacity market of the virtual power plant at each moment, in this embodiment, U is adopted DA,t to represent the revenue prospect value of the day-ahead energy market of the independent energy storage at each moment, and U is adopted fc,t to represent the revenue prospect value of the frequency regulation capacity market of the independent energy storage at each moment. Based on this, the above formula (1) can be transformed into the following formula.

[0160]

[0161] In the formula, U j,t represents the revenue prospect value of the j market of the independent energy storage at time t, p i represents the occurrence probability of scenario i, w(p i ) represents the decision weight for p i , r j,i,t represents the revenue of the j market of the independent energy storage at time t under scenario i, v(r j,i,t ) represents the decision maker's perception of r j,i,t , and N represents the total number of scenarios.

[0162] For the independent energy storage, j only takes the day-ahead energy market and the frequency regulation capacity market. Among them, the revenue r of the day-ahead energy market of the independent energy storage at each moment under each scenario DA,i,t (that is, when j is the day-ahead energy market) is: the product of the exchange power between the independent energy storage and the day-ahead energy market at time t under scenario i and the trading price (purchase price or selling price) of the day-ahead energy market at time t; the revenue r of the frequency regulation capacity market of the independent energy storage at each moment under each scenario fc,i,t (that is, when j is the frequency regulation capacity market) is: the product of the electricity quantity of the independent energy storage participating in the frequency regulation capacity market at time t under scenario i and the compensation price of the frequency regulation capacity market at time t. Regarding the decision weight and the value function, they will not be elaborated here, and specific details can be referred to the previous description.

[0163] Step 2: Based on the revenue prospect values of each market of the independent energy storage at each moment, obtain the total revenue prospect value within the scheduling period of the independent energy storage, and add it to the total revenue from renting out its installed capacity within the scheduling period of the independent energy storage to obtain the total revenue within the scheduling period of the independent energy storage. Specifically, the total revenue within the scheduling period of the independent energy storage can be obtained through the following formula.

[0164]

[0165] In the formula, g represents the total revenue within the scheduling period of the independent energy storage, r DA,PT , r fc,PT respectively represent the revenue prospect values of the day-ahead energy market and the frequency regulation capacity market within the scheduling period of the independent energy storage, C SES,tIt represents the revenue of the installed capacity of the independent energy storage leased to the virtual power plant at time t.

[0166] Step 3: With the goal of maximizing the total revenue within the independent energy storage dispatching period, construct a cost optimization model for the independent energy storage. Specifically, in some embodiments, the objective function of the cost optimization model of the independent energy storage (for the sake of easy description, it can be called the second objective function) can be expressed as the following formula.

[0167]

[0168] Thus, the objective function of the cost optimization model is obtained. Further, in some embodiments, the cost optimization model of the independent energy storage also includes constraint conditions corresponding to the second objective function, which are called the second constraint conditions for the sake of easy description. Among them, the second constraint conditions include the capacity constraint of the independent energy storage, the exchange power constraint between the independent energy storage and the virtual power plant, the exchange power constraint between the independent energy storage and the power grid, the cooperative operation constraint between the independent energy storage and the virtual power plant, the installed capacity constraint of the independent energy storage leased to the virtual power plant, and the installed capacity constraint of the independent energy storage participating in the market. Next, each constraint condition will be described.

[0169] The capacity constraint of the independent energy storage can be specifically expressed as the following formula.

[0170] 0≤Q vpp0,t +Q bid0,t ≤Q max (38)

[0171] In the formula, Q vpp0,t represents the installed capacity of the virtual power plant leasing the independent energy storage at time t, Q bid0,t represents the installed capacity of the independent energy storage participating in the market at time t, and Q max represents the maximum available capacity of the independent energy storage.

[0172] The exchange power constraint between the independent energy storage and the virtual power plant and the exchange power constraint between the independent energy storage and the power grid can be specifically expressed as the following formula.

[0173]

[0174] In the formula, respectively represent the maximum charging power and the maximum discharging power of the independent energy storage at time t, and P vpp,t represents the exchange power between the independent energy storage and the virtual power plant at time t, and P bid,t represents the exchange power between the independent energy storage and the power grid at time t.

[0175] Installed capacity constraint for independent energy storage leased to a virtual power plant: During the capacity lease process, the electric energy stored in the independent energy storage system from the virtual power plant at the end of the previous period will still occupy this part of the capacity. Therefore, the leased capacity at each moment must be no less than the remaining power inside at that moment, nor less than the remaining power inside at the end of the previous moment. That is to say, the installed capacity of the independent energy storage leased to the virtual power plant at time t is greater than or equal to the electric energy stored from the virtual power plant in the independent energy storage at time t and (t - 1). Specifically, it can be expressed as the following formula.

[0176] Q vpp0,t ≥max(Q vpp,t ,Q vpp,t-1 ) (43)

[0177] In the formula, Q vpp0,t represents the installed capacity of the independent energy storage leased to the virtual power plant at time t, and Q vpp,t , Q vpp,t-1 respectively represent the electric energy stored from the virtual power plant in the independent energy storage at time t and (t - 1).

[0178] Installed capacity constraint for independent energy storage participating in the market: The capacity participating in the market at each moment must be no less than the remaining power inside at that moment and the previous moment. That is to say, the installed capacity of the independent energy storage participating in the market at time t is greater than or equal to the electric energy stored by itself in the independent energy storage at time t and (t - 1) (that is, the remaining power except for that from the virtual power plant). Specifically, it can be expressed as the following formula.

[0179] Q bid0,t ≥max(Q bid,t ,Q bid,t-1 ) (44)

[0180] In the formula, Q bid0,t represents the installed capacity of the independent energy storage participating in the market at time t, and Q bid,t , Q bid,t-1 respectively represent the electric energy stored by itself in the independent energy storage at time t and (t - 1).

[0181] Cooperation operation constraint between independent energy storage and virtual power plant: According to the opportunity cost theory, this constraint is specifically that the revenue obtained from the installed capacity of the independent energy storage leased to the virtual power plant must not be lower than the revenue obtained from participating in the day-ahead energy market and frequency regulation market during the same period for this part of the installed capacity. Otherwise, the independent energy storage will refuse to cooperate. In some embodiments, this constraint can be expressed as the following formula.

[0182]

[0183] In the formula, C SES,t represents the revenue of the installed capacity of the independent energy storage leased to the virtual power plant at time t, and C bid,tDenote the revenue of the independent energy storage's remaining capacity participating in the market at time t as C bid0,t Denote the revenue of the independent energy storage participating in the market at time t without capacity leasing.

[0184] Thus far, the construction of the bidding decision model for the virtual power plant and the cost optimization model for the independent energy storage has been completed. Next, proceed to 230 to construct a two-layer bidding optimization model for the multi-market of the virtual power plant considering flexible energy storage scheduling, with the bidding decision model of the virtual power plant as the upper-layer model and the cost optimization model of the independent energy storage as the lower-layer model. That is, the two-layer bidding optimization model is obtained by taking the bidding decision model as the upper-layer model and the cost optimization model as the lower-layer model.

[0185] Subsequently, proceed to 240 to solve the constructed two-layer bidding optimization model to obtain the bidding optimization strategy in the equilibrium state and the scheduling strategy of the virtual power plant in this state. Among them, the bidding optimization strategy includes the declared volumes of the virtual power plant participating in the energy market, frequency regulation market, and carbon market at each moment, and the scheduling strategy includes the output of the gas turbine units at each moment, the charging power and discharging power of the electric vehicles at each moment, the demand response volume at each moment, the installed capacity of the independent energy storage leased by the virtual power plant at each moment, and the power exchanged between it and the independent energy storage.

[0186] Regarding the two-layer bidding optimization model for the multi-market of the virtual power plant considering flexible energy storage scheduling, in some embodiments, it can be converted into a single-layer model for solution based on the Karush-Kuhn-Tucker (KKT) conditions. Specifically, the two-layer bidding optimization model can be transformed into a single-layer model through the following steps based on the KKT conditions.

[0187] Step 1, standardize the second constraint condition as follows.

[0188] For Equation (38), split it into two inequalities to obtain the following constraints:

[0189] Q vpp0,t +Q bid0,t -Q max ≤0 (46)

[0190] -Q vpp0,t -Q bid0,t ≤0 (47)

[0191] Similarly, Equations (39) to (41) can be split into the following:

[0192]

[0193]

[0194] For Equation (42), removing the absolute value signs from the two variables results in the following four equations:

[0195]

[0196] For Equation (43) and Equation (44), removing the maximum value signs results in the following four equations:

[0197]

[0198] where P vpp,d and P bid,d represent the exchange power between the independent energy storage and the virtual power plant at time d and the exchange power between the independent energy storage and the power grid at time d, respectively.

[0199] For Equation (45), taking out the variable Q vpp0,t gives:

[0200]

[0201] where m DAsell,p represents the predicted highest electricity selling price.

[0202] In the lower - layer model, the variables P vpp,t and Q vpp0,t are actually conduction variables determined by the upper - layer. The lower - layer model can only passively accept them. Therefore, although they still need to be considered as variables in their constraints, they can be incorporated as constants into the KKT condition solving process.

[0203] After such arrangement, the lower - layer model altogether contains three variables: P bid,t 、Q bid0,t and m SES,t . And Q bid0,t can be obtained by summing up P bid,t . Therefore, there are 13 constraint conditions in Equations (46) - (49), (52) - (57), (60) - (62), and there are no equality constraints.

[0204] Step 2: Based on the 13 constraint conditions obtained after arrangement, construct the Lagrangian function. Specifically, each constraint can be set as h f (x), where f represents the constraint number, and the corresponding Lagrange multiplier is λ f . Then the following Lagrangian function can be constructed:

[0205]

[0206] where x represents the two variables P bid,t and m SES,λ in the lower - layer model.

[0207] Step 3: Take the partial derivatives of the two variables respectively, and the following formula must be satisfied:

[0208]

[0209] Step 4: Constrain its complementary slackness, and the following formula must be satisfied:

[0210] λ f ⊥h f (x) (66)

[0211] Step 5: For the Lagrange multiplier, the non-negativity must be satisfied:

[0212] λ f ≥0 (67)

[0213] In this way, the lower-level model is transformed into the following formula through the KKT conditions:

[0214]

[0215] Step 6: Introduce the above formula (68) as a constraint condition into the upper-level optimization function to complete the transformation of the two-layer model into a single-layer optimization.

[0216] So far, the two-layer bidding optimization model of the virtual power plant's multi-market considering flexible energy storage scheduling has been transformed into a single-layer model. Next, use Matlab to solve the transformed single-layer model, and then the bidding optimization strategy in the equilibrium state and the scheduling strategy of the virtual power plant in this state can be obtained.

[0217] In addition, it should be noted here that for some non-linear terms in the model, the Big-M method can be used for linearization of the non-linear terms, which specifically includes the following two parts.

[0218] 1) Product of two linear variables

[0219] Taking the multiplication of the independent energy storage capacity Q vpp0,t leased by the virtual power plant and the independent energy storage lease price m SES,t as an example. First, perform a binary expansion on it. Among them, the variable Q vpp0,t has a value range of [0, Q max . Introduce a 0-1 variable x q,t , and expand it into the following formula.

[0220]

[0221] Multiply both sides of formula (69) by the variable m SES,t , then the multiplication result of the two can be transformed into the multiplication result of a continuous variable and a 0-1 variable, as can be seen in the following formula.

[0222]

[0223] Define the variable mx q,t = m SES,t x q,t (In this formula, mx represents a variable), and transform formula (71) into the following formula.

[0224]

[0225] Then, according to the Big-M method, express it as the following formula.

[0226] 0 ≤ m SES,t -mx q,t ≤ M(1 - x q,t ) (73)

[0227] 0 ≤ mx q,t ≤ Mx q,t (74)

[0228] In summary, the linearization of the product of two continuous variables is completed.

[0229] 2) The product of the linear variable and the Lagrange multiplier in the KKT conditions

[0230] When the Lagrange multiplier λ is multiplied by the inequality relationship, when the inequality h(x) = 0, λ can be any non-negative number; when h(x) ≠ 0, it must be that λ = 0. Then, according to the above relationship, the original equality relationship can be regarded as the product of a continuous variable and a 0-1 variable. Introduce the 0-1 variable τ, and λx = 0 can be expanded as the following formula:

[0231] 0 ≤ x ≤ τM (75)

[0232] 0 ≤ λ ≤ (1 - τ)M (76)

[0233] In summary, the linearization of the product of the linear variable and the Lagrange multiplier is completed.

[0234] The above is the virtual power plant multi-market bidding optimization method considering flexible energy storage scheduling of the present invention. Further, the present invention also gives an example. Specifically, a virtual power plant including distributed new energy power generation, three small gas turbines, more than 2,000 orderly schedulable electric vehicles, a 40 MWh distributed energy storage, and adjustable load is selected as the research object. Among them, the wind power, photovoltaic, and load prediction results are as Figure 4 shown, and the predicted results of each market price are as Figure 5 shown. The detailed technical parameters of the main equipment are shown in Table 1 below.

[0235] Table 1

[0236]

[0237] 1. Uncertain data set description

[0238] In this example, wind power, photovoltaics, load, and various market prices have a certain degree of randomness, which will have a certain impact on the prediction results. Considering this uncertainty, in this example, different probability distribution functions are fitted based on the historical data of each group of results, and the Latin hypercube sampling method is used to generate 1000 scenarios for each object. Then, the 10 scenarios with the highest probability are selected, as Figure 6 shown. Among them, the upper and lower bounds of the uncertainty set are taken as the maximum and minimum outputs of each period after reduction, and it is assumed that the load power fluctuation is 10% of the predicted value.

[0239] 2. Result Analysis

[0240] This example compares the conventional bidding results under the deterministic scenario and the bidding results under the risk aversion strategy based on prospect theory when considering the uncertainty scenario.

[0241] 1) Conventional Strategy

[0242] The day-ahead scheduling plan of the virtual power plant is as Figure 7 shown. It is calculated that the virtual power plant obtains a total revenue of 139,000 yuan in the day-ahead stage, including 169,000 yuan in the energy market revenue (including the power purchase cost and the power sales revenue), 46,000 yuan in the frequency regulation market revenue, and 2,000 yuan in the carbon market revenue; the operating cost (including the gas cost, the maintenance cost of each device, etc.) is 31,000 yuan, the demand response cost is 10,000 yuan, and the energy storage lease cost is 37,000 yuan.

[0243] From Figure 7It can be seen that from 0 to 5, the photovoltaic power does not output, and the wind power output level cannot meet the load demand. However, at this time, the electricity price is relatively low. The virtual power plant adopts the strategy of increasing power purchase from the grid to meet the load demand, flexibly invokes distributed energy storage to further reduce the power purchase cost, and at the same time invokes load demand response to shift part of the expected load peak to this period; from 6 to 7, the day-ahead electricity price reaches the lowest level. The virtual power plant starts to lease energy storage and adopts a large-scale power purchase strategy at this time, charging the leased energy storage and distributed energy storage for subsequent invocation and sale; from 8 to 11, the photovoltaic power output gradually increases, and the demand response moves the load peak to this period, and the new energy output basically meets the load demand; from 12 to 15, the day-ahead electricity price reaches the highest level. The virtual power plant adopts a large-scale power sale strategy to sell the electric energy in the leased energy storage and distributed energy storage, and the gas turbine starts to output power to obtain the maximum profit. At the same time, due to the increase in electricity consumption cost, the demand response has shifted the load peak to the previous period, and the load after this period is significantly lower than that before the response; after 16:00, the photovoltaic power output gradually decreases to 0, but the wind power, distributed energy storage and gas turbine can meet the load after demand response. However, at night, the wind power output level decreases, but the day-ahead power purchase price also drops to a relatively low level, so a large-scale power purchase strategy is adopted. Generally speaking, the virtual power plant preferentially absorbs clean energy, and the overall absorption rate reaches 100%; the demand response effect is significant, shifting the load peak from 20:00 with a relatively high electricity price and relatively insufficient new energy output to 11:00 with a relatively low electricity price and a large amount of new energy output; the energy storage lease plays a good regulatory role, storing electric energy during low-price periods and invoking or selling it during high-price periods, greatly improving the flexibility and revenue level of the virtual power plant.

[0244] The capacity scheduling plan of the independent energy storage is as Figure 8 shown. It can be seen from this figure that the virtual power plant leases the most capacity around 11:00 - 13:00. This is because the electricity price level is relatively low before this period, and the electric energy can be gradually stored in the external energy storage. The capacity of this part remains unchanged from 11:00 to 13:00 when the electricity price level is relatively high. After 14:00, the electricity price level reaches the highest. Part of the leased energy storage releases the electric energy during this period to meet the load demand and sell electricity to the grid. Correspondingly, the leased capacity also gradually decreases as the SOC decreases. By 16:00, the discharge speed slows down. Since the electricity price is at a relatively high level thereafter, no more power is purchased to charge the external energy storage, and the remaining electricity is consumed in turn when the electricity price returns to the highest level from 20:00 to 21:00. No more capacity is leased thereafter.

[0245] For independent energy storage, the electricity price is relatively low from 06:00 to 07:00 and at 16:00. The energy storage purchases electricity for charging with the unsold capacity and uses it to participate in the energy market. At 08:00 - 09:00 and 20:00, when the electricity price level rises relatively, this part of the electricity is sold. Under the result of flexible leasing of energy storage, the revenue generated from selling at the peak price is lower than the revenue from leasing the capacity to the virtual power plant. Therefore, electricity is sold in advance to obtain partial revenue. Judging from the results, flexible leasing of energy storage improves the revenue levels of the virtual power plant and energy storage, enhances the flexibility of the virtual power plant, and at the same time improves the utilization rate of energy storage.

[0246] The dynamic leasing price of energy storage is as Figure 9 shown. It can be seen from this figure that for the first time, before the electricity price peak appears, the price change level shows an opposite trend to the electricity price change. This is because the capacity leased out at a lower electricity price has a relatively high opportunity cost for energy storage. This part of the capacity could have been used to purchase electricity at a low price and sold at a higher electricity price level. When the electricity price is high, the profit margin from buying and selling electricity is not large, and accordingly, the leasing price decreases. When the electricity purchase price is higher than the target electricity selling price, it is not profitable to participate in the power market during this period, so the leasing price drops to the lowest, only needing to pay the operation and maintenance costs.

[0247] 2) Risk aversion strategy

[0248] During the intraday stage, there are differences between the wind power and photovoltaic power outputs, load, energy, frequency regulation market price, and frequency regulation mileage coefficient and the day-ahead forecast values. Latin hypercube sampling is used to generate ten scenarios with the highest probabilities. On this basis, a bidding strategy is generated with the maximum prospect value as the goal based on the prospect theory, and the final cost is calculated based on the scenario with the highest occurrence probability as the actual situation. At the same time, the rationality of the selected risk preference degree in this example is calculated under two scenarios of risk-loving preference and risk-averse preference.

[0249] The dispatching results during the intraday stage are as Figure 10 shown, where the fluctuations generated during the intraday stage are compensated by energy storage, electric vehicles, gas turbines, and the real-time energy market. The day-ahead dispatching results and intraday dispatching results under each risk preference are shown in Table 2 below.

[0250] Table 2

[0251]

[0252] When the loss aversion coefficient θ of the decision maker is 1.5, the virtual power plant prefers risks and is insufficiently sensitive to losses. In the day-ahead stage, the power purchase and adjustable output levels will be slightly increased, but this part is not enough to cope with the actual fluctuations generated in the intraday stage. When the loss aversion coefficient θ of the decision maker reaches 4.5, the virtual power plant extremely dislikes risks and is extremely sensitive to losses. In the day-ahead stage, the levels of power purchase, adjustable output, etc. are increased, and this part cannot be consumed when meeting the load demand of the virtual power plant. It can only be sold in the spot market, and the revenue cannot cover the power purchase and generation costs. Therefore, the final cost increases. In this study, when θ is set to 2.25, it can not only cope with the influence of uncertain factors, but also avoid over-response caused by being too sensitive to losses, thus controlling the cost at the lowest level and greatly improving the economic operation of the virtual power plant.

[0253] 3. Comparative analysis

[0254] To verify the dynamic capacity determination and pricing strategy of the energy storage power station, this example evaluated five different situations as shown in Table 3 below: Case1 represents the pricing strategy of this example, Case2 only considers dynamic pricing, that is, the single-day rental capacity remains unchanged; Case3 only considers dynamic capacity determination, that is, the rental price remains unchanged; Case4 is the long-term agreement rental strategy, that is, the long-term rental capacity and price both remain unchanged; Case5 does not consider the capacity rental of the energy storage power station. The bidding results under different scenarios are shown in Table 4 below. Among them, the rental capacity of Case2 and the rental price of Case3 are generated by the game, and the rental price of Case4 is formulated according to the current long-term rental market price.

[0255] Table 3

[0256]

[0257] Table 4

[0258]

[0259] By comparing the total costs of the virtual power plant in each scenario, it shows that the dynamic capacity determination and pricing strategy proposed in the present invention reduces the operating cost of the virtual power plant to the greatest extent.

[0260] By comparing Case1 with Case2 and Case3, it can be seen that when one of the factors of rental capacity and rental price is fixed, the cost of the virtual power plant will increase. First, under the fixed capacity, after discharging during the peak period, the SOC of the energy storage remains at a low level. In fact, most of the capacity is idle, but the virtual power plant still needs to pay the rental cost for this part of the capacity. And under the fixed price, the new energy output level is relatively high and the electricity price is relatively low in some periods, but the rental price is relatively high, and the cost of the virtual power plant to store the excess electricity increases, resulting in a decrease in the rental willingness of the virtual power plant in this period. Instead, part of the cost is shifted to the power grid for power purchase, causing an increase in the cost of the virtual power plant.

[0261] Comparing Case 1 and Case 4, since both the capacity and price do not change with time, the overall cost increases to a large extent.

[0262] In Case 5, since energy storage leasing is not carried out, the electricity price during the valley period cannot be transferred to the peak period, resulting in a significant increase in the electricity purchase cost.

[0263] In Case 2, since the leased capacity remains unchanged, if leased at the maximum capacity, there will be a large amount of idle capacity during the idle period, resulting in some capacity waste; while when reducing the leased capacity, it will lead to insufficient transfer of the electricity during the valley period, and the role of energy storage is limited. Generally speaking, fixed-capacity leasing lacks flexibility, resulting in increased costs.

[0264] In Case 3, since the hourly leasing price remains unchanged, the role of energy storage in different periods cannot be reflected; and in order to ensure its own revenue level, the energy storage power station will maintain the price at a certain level. At this level, if the virtual power plant leases according to the original plan, the generated cost will increase, which also leads to the virtual power plant changing the plan, reducing the leased capacity in each period to a certain extent, and ultimately resulting in a decline in the dispatching ability of the virtual power plant and an increase in the total cost.

[0265] Case 4 lacks flexibility because both the capacity and price do not change with time, and the overall cost increases to a certain extent compared with Case 2 and Case 3.

[0266] In summary, the present invention considers the multi-market operation mechanism of VPP with the maximization of energy storage dispatching benefits and multiple psychological factors of decision-makers, and establishes a bidding optimization model for the multi-market of VPP considering flexible energy storage dispatching. First, considering the flexibility and benefit maximization requirements of VPP leasing external energy storage, a flexible capacity leasing interaction mechanism between VPP and energy storage operators is proposed to realize dynamic fixed-capacity pricing of energy storage leasing. Second, probability distributions are used to describe the uncertain factors in the operation of VPP, and the risk preferences of decision-makers are characterized by prospect theory. On this basis, a bidding optimization and flexible energy storage leasing pricing decision model for the multi-market of VPP based on the master-slave game is proposed to obtain the optimal bidding optimization strategy in the equilibrium state and the dispatching strategy of the virtual power plant in this state. Based on this, the present invention can significantly improve the flexibility of dispatching, while also improving market revenue and avoiding installation waste.

[0267] The various techniques described herein can be implemented in combination with hardware or software, or a combination thereof. Thus, the methods and apparatuses of the present invention, or certain aspects or portions of the methods and apparatuses of the present invention, may take the form of program code (i.e., instructions) embedded in a tangible medium, such as a removable hard disk, a USB flash drive, a floppy disk, a CD-ROM, or any other machine-readable storage medium, wherein when the program is loaded into and executed by a machine such as a computer, the machine becomes an apparatus for practicing the present invention.

[0268] In the specification provided herein, a number of specific details are set forth. However, it will be understood that embodiments of the present invention may be practiced without these specific details. In some instances, well-known methods, structures, and techniques have not been shown in detail so as not to obscure an understanding of this description.

[0269] It should be understood that, in order to streamline this disclosure and assist in understanding one or more of the various inventive aspects, in the foregoing description of exemplary embodiments of the present invention, various features of the present invention are sometimes grouped together in a single embodiment, figure, or description thereof. However, the disclosed method should not be construed as reflecting an intention that the claimed invention requires more features than are expressly recited in each claim.

[0270] Moreover, unless otherwise specified, the use of ordinal terms such as "first", "second", "third", etc. to describe ordinary objects merely indicates different instances of similar objects and is not intended to imply that the objects so described must have a given order in terms of time, space, ranking, or any other manner.

[0271] Although the present invention has been described in terms of a limited number of embodiments, those skilled in the art, having the benefit of the foregoing description, will appreciate that other embodiments can be contemplated within the scope of the invention as thus described. Further, it should be noted that the language used in this specification has been principally selected for readability and instructional purposes and not for the purpose of explaining or limiting the subject matter of the invention. Thus, many modifications and variations will be apparent to those of ordinary skill in the art without departing from the scope and spirit of the appended claims. The disclosure of the present invention is illustrative, not restrictive, and the scope of the invention is defined by the appended claims.

Claims

1. A multi-market bidding optimization method for a virtual power plant considering flexible energy storage dispatching, wherein, The virtual power plant schedules the installed capacity of the independently operated energy storage externally in time periods, and the method includes: Based on the revenue prospect values of the virtual power plant in the energy market, frequency regulation market, and carbon market, as well as the operating cost of the virtual power plant, with the maximum total revenue within the virtual power plant scheduling period as the objective function, a bidding decision model of the virtual power plant is constructed, where the operating cost of the virtual power plant includes the cost of time-sharing scheduling the installed capacity of the independent energy storage; Based on the revenue prospect values of the independent energy storage in the energy market and frequency regulation market, as well as the revenue from renting out its installed capacity, with the maximum total revenue within the independent energy storage scheduling period as the objective function, a cost optimization model of the independent energy storage is constructed; Taking the bidding decision model as the upper-layer model and the cost optimization model as the lower-layer model, a two-layer bidding optimization model for the multi-market of the virtual power plant considering flexible energy storage scheduling is constructed; Solve the two-layer bidding optimization model to obtain the bidding optimization strategy in the equilibrium state and the scheduling strategy of the virtual power plant in this state, where the bidding optimization strategy includes the declared volumes of the virtual power plant participating in the energy market, frequency regulation market, and carbon market at each moment, and the scheduling strategy at least includes the installed capacity of the independent energy storage scheduled by the virtual power plant at each moment and the power exchanged with the independent energy storage.

2. The method according to claim 1, wherein Determine the total revenue within the virtual power plant scheduling period, including: Obtain the revenue prospect values of the virtual power plant in the day-ahead energy market, real-time energy market, frequency regulation capacity market, frequency regulation mileage market, and carbon market at each moment; Based on the revenue prospect values of each market of the virtual power plant at each moment, obtain the total revenue prospect value within the virtual power plant scheduling period, and subtract the total operating cost within the virtual power plant scheduling period from it to obtain the total revenue within the virtual power plant scheduling period.

3. The method according to claim 1 or 2, wherein Determine the total revenue within the independent energy storage scheduling period, including: Obtain the revenue prospect values of the independent energy storage in the day-ahead energy market and frequency regulation capacity market at each moment; Based on the revenue prospect values of each market of the independent energy storage at each moment, obtain the total revenue prospect value within the independent energy storage scheduling period, and add it to the total revenue from renting out its installed capacity within the independent energy storage scheduling period to obtain the total revenue within the independent energy storage scheduling period.

4. The method according to any one of claims 1-3, wherein The revenue prospect values of each market of the virtual power plant / independent energy storage at each moment are obtained through the following formula: Among them, V j,t represents the revenue prospect value of the virtual power plant / independent energy storage in the j market at time t, p i represents the occurrence probability of scenario i, w(p i ) represents the decision weight for p i , C j,i,t represents the revenue of the virtual power plant / independent energy storage in the j market at time t under scenario i, v(C j,i,t ) represents the decision maker's perception of C j,i,t , and N represents the total number of scenarios.

5. The method according to any one of claims 1-4, wherein The objective function of the bidding decision model includes: C out,t = C GT,t + C DAbuy,t + C EV,t + C DR,t + C SES,t Among them, f represents the total revenue within the dispatching period of the virtual power plant, C in,PT represents the total revenue prospect value within the dispatching period of the virtual power plant, C out,t represents the operating cost of the virtual power plant at time t, V DA,t 、V RT,t 、V fc,t 、V fp,t 、 respectively represent the revenue prospect values of the day-ahead energy market, real-time energy market, frequency regulation capacity market, frequency regulation mileage market, and carbon market of the virtual power plant at time t, C GT,t represents the gas purchase cost of the gas turbine unit in the virtual power plant at time t, C DAbuy,t represents the power purchase cost of the day-ahead energy market of the virtual power plant at time t, C DR,t represents the demand response compensation cost of the virtual power plant at time t, C SES,t represents the cost of the installed capacity of the dispatching independent energy storage of the virtual power plant at time t, C EV,t represents the operating cost of the electric vehicle in the virtual power plant at time t, and T represents the total number of time periods.

6. The method according to claim 5, wherein, The cost of the virtual power plant scheduling the installed capacity of the independent energy storage at each moment includes: C SES,t = m SES,t Q vpp0,t Among them, C SES,t represents the cost of the installed capacity of the virtual power plant for scheduling independent energy storage at time t, m SES,t represents the price per unit installed capacity of the virtual power plant for scheduling independent energy storage at time t, Q vpp0,t represents the installed capacity of the virtual power plant for scheduling independent energy storage at time t.

7. The method according to any one of claims 1-6, wherein The bidding decision model includes power balance constraints, upper and lower limits constraints on the output of gas turbines and ramp-up and ramp-down limits constraints, off-grid power constraints of electric vehicles, and charge and discharge power constraints, demand response constraints.

8. The method according to any one of claims 1-7, wherein, The cost optimization model includes capacity constraints of the independent energy storage, exchange power constraints between the independent energy storage and the virtual power plant, exchange power constraints between the independent energy storage and the power grid, and cooperative operation constraints between the independent energy storage and the virtual power plant. The cooperative operation constraints between the independent energy storage and the virtual power plant include: the revenue obtained from renting the installed capacity of the independent energy storage to the virtual power plant is not less than the revenue obtained from participating in the energy market and the frequency regulation market by this part of the installed capacity during the same period.

9. A computing device, comprising: at least one processor; and a memory storing program instructions, wherein the program instructions are configured to be executed by the at least one processor, and the program instructions include instructions for executing the method according to any one of claims 1-8.

10. A readable storage medium storing program instructions, which, when read and executed by a computing device, cause the computing device to execute the method according to any one of claims 1-8.

Images