Dynamic aggregation and packaging method of micro-regulation resource power plant based on high-dimensional model expression

Abstract

The present invention relates to the technical field of power systems, and specifically to a method for dynamically aggregating and encapsulating fine-tuned resources in a power plant based on high-dimensional model expression. By aggregating the feasible domains of different power generation factors, a feasible domain aggregation model of the power generation factors is obtained; distributed load resources are aggregated into virtual machine groups, and the aggregation and scheduling of power plant dynamic aggregation and encapsulation scheduling objects are dynamically optimized based on power generation characteristics and power generation economy, thereby realizing dynamic construction and aggregation of power plant dynamic aggregation and encapsulation scheduling objects; a mapping relationship between physical parameters and aggregation model parameters is established through high-dimensional model expression technology, thereby realizing real-time online identification of external characteristic parameters of aggregation units based on high-dimensional model expression technology. The present invention can dynamically aggregate and optimize resources in real time while ensuring computational efficiency and accuracy, ensuring unified regulation and optimized utilization of flexible resources, and being able to respond quickly to different scheduling requirements.

Description

Technical Field

[0001] The present invention relates to the technical field of power systems, and in particular to a method for dynamically aggregating and encapsulating micro-regulation resources in power plants based on high-dimensional model expression. Background Art

[0002] With the rapid development of renewable energy, resources such as wind and solar power have become increasingly important components of the power grid, exposing it to increasing volatility and uncertainty. To address these challenges, the "power plant" model of fine-tuning resources has emerged as a new power dispatch model. The dynamic aggregation technology of flexible resource power plants effectively improves the grid's regulation capabilities and stability. The core technology of this "power plant" approach, which dynamically aggregates and encapsulates dispatch objects, lies in the efficient and precise dynamic aggregation of distributed energy and flexible resources to meet the grid's dispatch requirements and maximize its economic benefits.

[0003] Existing aggregation methods have achieved certain results to a certain extent. However, with the continuous increase in the number of distributed energy resources, existing methods still have limitations in high-dimensional data processing, computational efficiency and accuracy. They are difficult to cope with complex and diverse resource demands and still face some challenges in practical applications. First, the existing technology is still insufficient in processing large-scale, high-dimensional data. Especially when faced with dynamic and complex distributed resources, the computational efficiency and accuracy of existing models cannot fully meet the scheduling needs of the power grid. Secondly, the current dynamic construction technology of power plant dynamic aggregation and encapsulation scheduling objects has not yet fully realized the online dynamic identification of resources, resulting in limited scheduling accuracy and flexibility after resource aggregation. Especially when faced with power dispatch of different types and demands, existing methods are unable to accurately and efficiently identify and match the external characteristic parameters of power plant dynamic aggregation and encapsulation scheduling objects in real time, affecting the scheduling response capability of power plant dynamic aggregation and encapsulation scheduling objects.

[0004] Therefore, a new technical method is urgently needed that can aggregate and optimize resources in real time and dynamically while ensuring computing efficiency and accuracy, especially in high-dimensional and large-scale data processing, to ensure that the power plant-based dynamic aggregation and encapsulation scheduling objects can respond quickly to different scheduling requirements. Summary of the Invention

[0005] To solve the above technical problems, the present invention aims to provide a method for dynamically aggregating and encapsulating micro-adjusted resources in power plants based on high-dimensional model expression. The technical solutions adopted are as follows:

[0006] In a first aspect, the present invention provides a method for dynamically aggregating and encapsulating micro-adjustment resources in a power plant based on high-dimensional model expression, comprising the following steps:

[0007] Aggregate the feasible domains of different power generation factors to obtain a feasible domain aggregation model of power generation factors;

[0008] Aggregate distributed load resources into virtual machine groups, dynamically optimize the aggregation and scheduling of power plant dynamic aggregation and encapsulation scheduling objects based on power generation characteristics and power generation economics, and realize the dynamic construction and aggregation of power plant dynamic aggregation and encapsulation scheduling objects;

[0009] Match the power generation operation mode of the power plant dynamic aggregation package scheduling object with the equipment operation mode corresponding to each response characteristic curve in the building potential characteristic library to achieve online mode matching of the external characteristic parameters of the aggregation unit;

[0010] The mapping relationship between physical parameters and aggregation model parameters is established through high-dimensional model expression technology, and real-time online identification of the external characteristic parameters of the aggregation unit based on high-dimensional model expression technology is realized.

[0011] In combination with the first aspect above, in some possible implementations, the feasible domains of different power generation factors are aggregated to obtain a feasible domain aggregation model of the power generation factors, including:

[0012] The Zeno polyhedron is used to approximate the feasible domain of power generation factors, and the feasible domains of different power generation factors are aggregated through the Minkowski sum operation of the Zeno polyhedron, so as to obtain the feasible domain aggregation model of power generation factors.

[0013] In combination with the first aspect above, in some possible implementations, a Zeno polyhedron is used to approximate the feasible region of the power generation factor, and the feasible regions of different power generation factors are aggregated through the Minkowski sum operation of the Zeno polyhedron, thereby obtaining a feasible region aggregation model of the power generation factor, including:

[0014] Use convex polyhedron to represent the feasible region of a single power generation factor:

[0015] (1)

[0016] Where, A convex polyhedron representing a single power generation factor; The power set representing the power generation factor; The dimension representing the power generation factor; is the constraint matrix; represents the constraint space; Indicates the number of constraints; is the constraint vector;

[0017] The aggregation of the feasible regions of multiple power generation factors is expressed as the Minkowski sum of all power generation factors. The feasible region after aggregation is The formula is as follows:

[0018] (2)

[0019] Where, represents the feasible domain of the aggregate; represents the feasible region of the jth power generation factor; Indicates the total number of power generation factors;

[0020] Use Zeno polyhedron to approximate the feasible region of each power generation factor. The feasible region of a single power generation factor Zeno polyhedron The approximate formula is as follows:

[0021] (3)

[0022] Where, The power vector representing the power generation factor; is the center of the Zeno polyhedron; represents the dimension of the generator matrix G; is the generator matrix; Represents the generator coefficient Dimensions; is the coefficient of the generator; is the maximum scaling factor of the generator;

[0023] Determine the optimization objective and constraints of the Zeno polyhedron approximation problem, solve the optimization objective based on the constraints, determine the feasible domain of the Zeno polyhedron approximation that meets feasibility, and merge the Zeno polyhedron approximation feasible domain by Minkowski sum operation to obtain the Zeno polyhedron representation of the aggregate, and determine the Zeno polyhedron representation of the aggregate as the feasible domain aggregation model of the power generation factor.

[0024] In conjunction with the first aspect above, in some possible implementations, the optimization objective and constraints of the Zeno polyhedron approximation problem are determined, and the corresponding calculation formula is:

[0025] (4)

[0026] st (5)

[0027] Where, Represents a Zeno polyhedron and the convex polyhedron of the feasible region In the The diameter ratio in each direction, , and Zeno polyhedrons and convex polyhedrons In the diameter in each direction; Indicates the total number of directions.

[0028] In conjunction with the first aspect above, in some possible implementations, distributed load resources are aggregated into virtual machine groups, and the aggregation and scheduling of power plant-based dynamic aggregation and encapsulation scheduling objects are dynamically optimized based on power generation characteristics and power generation economics, thereby achieving dynamic construction and aggregation of power plant-based dynamic aggregation and encapsulation scheduling objects, including:

[0029] 1) Aggregate distributed load resources into virtual machine groups. The virtual machine group model is shown in the following formula:

[0030] (6)

[0031] Where, A model representing a group of virtual machines; and Respectively represent the upper and lower limits of the output of the virtual machine group; and They represent the maximum ramp-up and ramp-down rates of the virtual machine group respectively; and They represent the minimum running time and minimum downtime of the virtual machine group respectively; represents the cost function of the virtual machine group;

[0032] 2) Collect real-time regulation data of each central air-conditioning load in the power plant dynamic aggregation and packaging scheduling object, including regulation potential and manageable time;

[0033] 3) Preprocess the collected load data, remove noise and perform normalization to ensure data consistency and comparability;

[0034] 4) The regulation potential of each central air conditioner It is the minimum value of the adjustment amount in each control cycle within the specified control time. The calculation can be expressed as:

[0035] (7)

[0036] Where, Indicates the maximum adjustment amount of the central air conditioner; is the regulated load; is the total number of control cycles;

[0037] 5) Calculate the regulation potential of each central air conditioner, obtain its regulation characteristic curve, and fit the relationship curve between the adjustable potential and control time of a single air conditioner. The formula is as follows:

[0038] (8)

[0039] Where, Indicates control time; represents the fitting function;

[0040] 6) Compare the adjustment characteristic curves of different central air conditioners to determine their consistency and consistency The calculation formula is:

[0041] (9)

[0042] Where, and Respectively represent Central air conditioner and Central air conditioner in adjustment cycle regulatory potential when Indicates the total number of adjustment cycles;

[0043] 7) Use the agglomerative hierarchical clustering method to aggregate central air conditioners. The goal is to form a power plant-like dynamic aggregation package scheduling object that meets the scheduling requirements. The optimization objective function is shown in the following formula:

[0044] (10)

[0045] Where, It represents the number of all possible combinations of any two numbers among M numbers; Represents the number of central air conditioners in the group after aggregation; the constraint conditions are as follows:

[0046] (11)

[0047] Where, represents the load adjustment of the i-th central air conditioner; Indicates the total regulation amount of the power plant dynamic aggregation and encapsulation scheduling objects;

[0048] 8) Calculate the regulation price of each central air conditioner, which is adjusted according to the regulation potential, rated power and comfort level of the central air conditioner. The formula is as follows:

[0049] (12)

[0050] Where, Indicates the Central air conditioning adjustment quotation; Indicates the The quotation decision coefficient of central air conditioner; Indicates the Rated power of central air conditioner; Indicates the The comfort of central air conditioning;

[0051] 9) Minimize the total cost of the power plant dynamic aggregation and packaging scheduling objects. Define the total cost function of the power plant dynamic aggregation and packaging scheduling objects. The formula is as follows:

[0052] (13)

[0053] Where, represents the total cost function; Indicates control duration; Indicates the The regulation amount of the central air conditioner; N represents the total number of central air conditioners;

[0054] 10) The dispatch of the power plant dynamic aggregation and encapsulation dispatch objects must meet the total regulation requirements of the power grid dispatch department. The constraint formula is as follows:

[0055] (14)

[0056] Where, Indicates the total adjustment amount of the scheduling plan; Indicates the The load regulation potential of a central air conditioner.

[0057] In conjunction with the first aspect above, in some possible implementations, the power generation operation mode of the power plant dynamic aggregation package scheduling object is matched with the equipment operation mode corresponding to each response characteristic curve in the building potential characteristic library to achieve online mode matching of the external characteristic parameters of the aggregation unit, including:

[0058] 1) Using the maximum input power of intermittent energy sources such as wind and solar power as input, the external characteristic parameters of the computer group and the virtual power plant as a whole are obtained;

[0059] 2) Establish a normalized information model for obtaining online parameters of power plant-based dynamic aggregation and encapsulation scheduling objects. The formula is as follows:

[0060] (15)

[0061] Where, Indicates the The maximum output of each unit represents the output uncertainty of the distributed generator set; Indicates the time when the control signal is released; Indicates the Units in The winning bid amount for the time period; Indicates the Units in The amount of various types of spare winning bids during the time period; Respectively represent Dynamic ramp-up and ramp-down rate of each unit (MW / s);

[0062] 3) Model the output sequence of the power plant dynamic aggregation and packaging scheduling objects and internal power generation resources. The formula is as follows:

[0063] (16)

[0064] Where, Indicates that the power plant dynamic aggregation package scheduling object is Each moment in the period Response sequence to system output regulation signals; Indicates that the power plant dynamic aggregation package scheduling object is The basic output of the time period is also the bidding strategy of the virtual power plant in the h period; Indicates that the power plant dynamic aggregation package scheduling object is Each moment in the period output sequence;

[0065] 4) Model the maximum output of the power plant dynamic aggregation and packaging scheduling object. The formula is as follows:

[0066] (17)

[0067] Where, Indicates that the power plant dynamic aggregation package scheduling object is Each moment in the period The maximum output sequence of Indicates the first Intermittent energy units in Each moment in the time period The maximum output sequence depends on the current maximum input and control interval of the unit The product of the climbing / descending rate; Indicates the first Maximum output power of conventional units;

[0068] 5) Perform real-time standby modeling of power plant dynamic aggregation and encapsulation scheduling objects. The formula is as follows:

[0069] (18)

[0070] (19)

[0071] Where, Indicates remaining spare; Indicates that each intermittent energy unit is Each moment in the period The load shedding ratio depends on the standby control strategy of each unit and the load shedding ratio constraint; Indicates the first Conventional units in Always have a spare Indicates that the virtual power plant is in the period At each moment The reserve power;

[0072] 6) In the performance parameters of the power plant dynamic aggregation package scheduling object, the subsequent climbing / declining rate calculation formula is as follows:

[0073] (20)

[0074] Where, Indicates that under different scheduling or control instructions, The climbing / descending rate during the time period; in the subsequent climbing / descending rate calculation, Dynamically aggregate and encapsulate the output data of dispatch objects for actual power plants;

[0075] 7) In the performance parameters of the power plant dynamic aggregation package scheduling object, the calculation formula of the pre-calculation / downgrade rate is consistent with the post-calculation expression. From the above power plant dynamic aggregation and packaging scheduling object output sequence calculation model, we can get: It is an arbitrary parameter;

[0076] 8) The frequency regulation performance score of the power plant dynamic aggregation and packaging scheduling object in each time period is calculated as follows:

[0077] (twenty one)

[0078] (twenty two)

[0079] (twenty three)

[0080] (twenty four)

[0081] (25)

[0082] Where, represents the frequency modulation performance score, which is the weighted average of the three scores; DS represents the response delay score; CS represents the correlation score; PS represents the accuracy score; A, B, and C represent the weights, all of which are 1 / 3; Indicates the response delay at which CS and DS achieve their maximum values; Indicates the predetermined adjustment power; Indicates the actual regulated power; represents the correlation function; Indicates the average value of the scheduled regulated power per hour.

[0083] In conjunction with the first aspect above, in some possible implementations, a mapping relationship between physical parameters and aggregation model parameters is established using high-dimensional model expression technology to achieve real-time online identification of external characteristic parameters of aggregation units based on high-dimensional model expression technology, including:

[0084] 1) The influence of uncertainty in the physical parameters of air conditioning load on the aggregation model parameters is analyzed through high-dimensional model expression. The calculation formula is as follows:

[0085] (26)

[0086] Where, represents a constant term; represents the first-order component; represents the second-order component; represents the order of the component; the system input is , the output is , and so on;

[0087] 2) Generate Group input samples, the input includes multiple physical parameters of air conditioning load;

[0088] 3) Each set of input samples is simulated to obtain the aggregate model parameters, thereby obtaining output samples, as shown in the following formula:

[0089] (27)

[0090] Where, Indicates the Group input samples The corresponding aggregation model output; Indicates the Group input samples; Indicates the The value of the output quantity;

[0091] 4) Standardize each input quantity using the following formula:

[0092] (28)

[0093] Where, and Represent the actual value and standard value of the input variable respectively; and Respectively represent the minimum and maximum values of the input;

[0094] 5) Use the input and output values of N0 to establish a high-dimensional model expression model of the air conditioning group and calculate the correlation coefficient , And the global sensitivity of the physical parameters of the air conditioning load is as follows:

[0095] (29)

[0096] (30)

[0097] Where, 、 、 、 、 and All of them represent the physical parameters of the air conditioning load. By analogy, we can obtain higher-order coefficients of the first-order component function of the input quantity and the second-order component function of the input quantity; represents the first-order component function;

[0098] 6) Calculate the global sensitivity of the independent and coupled inputs;

[0099] 7) Establish a high-dimensional model expression model database for air conditioning aggregation model parameters at different time scales.

[0100] In combination with the first aspect above, in some possible implementations, the Quasi Monte Calculator method is used to generate The input samples of the group air-conditioning system can be expressed as:

[0101] (31)

[0102] Where, 、 、 、 、 、 and Both represent the physical parameters of air conditioning load.

[0103] In a second aspect, the present invention further provides a system for dynamically aggregating and packaging micro-regulated resources for power plants based on high-dimensional model expression, comprising a memory and a processor. The memory is configured to store executable program code, and the processor is configured to retrieve and execute the executable program code from the memory, causing the device to execute the method of the first aspect or any possible implementation of the first aspect.

[0104] In a third aspect, the present invention further provides a computer program product, which includes: computer program code, which, when running on a computer, enables the computer to execute the method in the above-mentioned first aspect or any possible implementation of the first aspect.

[0105] In a fourth aspect, the present invention further provides a computer-readable storage medium, which stores a computer program code. When the computer program code runs on a computer, the computer executes the method in the above-mentioned first aspect or any possible implementation of the first aspect.

[0106] The present invention has the following beneficial effects: the present invention obtains a feasible domain aggregation model of power generation factors by aggregating the feasible domains of different power generation factors; aggregates distributed load resources into virtual machine groups, dynamically optimizes the aggregation and scheduling of power plant dynamic aggregation and encapsulation scheduling objects based on power generation characteristics and power generation economy, and realizes the dynamic construction and aggregation of power plant dynamic aggregation and encapsulation scheduling objects; establishes a mapping relationship between physical parameters and aggregation model parameters through high-dimensional model expression technology, and realizes real-time online identification of external characteristic parameters of aggregation units based on high-dimensional model expression technology. The present invention can aggregate and optimize resources in real time and dynamically while ensuring computing efficiency and accuracy, especially in high-dimensional and large-scale data processing, ensure unified regulation and optimized utilization of flexible resources, and respond quickly to different scheduling requirements. BRIEF DESCRIPTION OF THE DRAWINGS

[0107] In order to more clearly illustrate the technical solutions and advantages of the embodiments of the present invention or the prior art, the following briefly introduces the drawings required for use in the embodiments or the prior art descriptions. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0108] Figure 1 This is a flowchart of the steps of a method for dynamically aggregating and packaging fine-tuning resources into power plants based on high-dimensional model expression according to an embodiment of the present invention;

[0109] Figure 2 A flowchart for dynamically constructing a power plant-based dynamic aggregation and encapsulation scheduling object according to an embodiment of the present invention;

[0110] Figure 3 A schematic diagram of a high-dimensional model expression modeling process according to an embodiment of the present invention;

[0111] Figure 4 This is a flow chart of a real-time identification framework of air conditioning load aggregation model parameters according to an embodiment of the present invention;

[0112] Figure 5A comparison chart of the power pulse capacity obtained by independent simulation and high-dimensional model expression according to an embodiment of the present invention;

[0113] Figure 6 Schematic diagram of the relative error comparison of parameter identification results based on high-dimensional model expression according to an embodiment of the present invention. DETAILED DESCRIPTION

[0114] In order to clearly illustrate the technical features of this solution, the present invention is described in detail below through specific implementation methods in conjunction with the accompanying drawings.

[0115] Embodiments of the present invention will be described in more detail below with reference to the accompanying drawings. Although certain embodiments of the present invention are shown in the accompanying drawings, it should be understood that the present invention can be implemented in various forms and should not be construed as limited to the embodiments described herein. Rather, these embodiments are provided to provide a more thorough and complete understanding of the present invention. It should be understood that the drawings and embodiments of the present invention are for illustrative purposes only and are not intended to limit the scope of protection of the present invention.

[0116] It should be understood that the various steps described in the method embodiments of the present invention may be performed in different orders and / or in parallel. In addition, the method embodiments may include additional steps and / or omit the steps shown. The scope of the present invention is not limited in this respect.

[0117] As used herein, the term "including" and its variations are open-ended, i.e., "including but not limited to." The term "based on" means "based, at least in part, on." The term "one embodiment" means "at least one embodiment," the term "another embodiment" means "at least one additional embodiment," and the term "some embodiments" means "at least some embodiments." Other terms are defined in the following description.

[0118] It should be noted that the concepts of "first" and "second" mentioned in the present invention are only used to distinguish different devices, modules or units, and are not used to limit the order or interdependence of the functions performed by these devices, modules or units.

[0119] Although operations or steps are described in a particular order in the drawings in the embodiments of the present invention, this should not be understood as requiring that these operations or steps be performed in the particular order shown or in a serial order, or that all of the operations or steps shown be performed to obtain a desired result. In the embodiments of the present invention, these operations or steps may be performed serially; they may also be performed in parallel; or a portion of these operations or steps may be performed.

[0120] At the same time, it is understood that the data involved in the technical solutions of the present invention (including but not limited to the data itself, the acquisition or use of the data) must comply with the requirements of relevant laws, regulations and relevant provisions. Unless otherwise defined, all technical and scientific terms used in this invention have the same meanings as those commonly understood by those skilled in the art to which this invention belongs, and all parameters or indicators in the formulas involved in this invention are normalized values to eliminate dimension effects.

[0121] In order to overcome the shortcomings of the existing technology, an embodiment of the present invention provides a method for dynamic aggregation and encapsulation of fine-tuning resources in power plants based on high-dimensional model expression, aiming to solve the problems of complex resource characteristics, large differences in principles, and difficult external characteristics. It is a new development in flexibility resource management and provides basic support for the unified regulation and optimal utilization of flexibility resources.

[0122] The following will introduce in detail a method for dynamic aggregation and packaging of fine-tuning resources in a power plant based on high-dimensional model expression provided by an embodiment of the present invention in conjunction with the accompanying drawings.

[0123] Figure 1 The basic flow chart of a method for dynamically aggregating and packaging micro-adjusted resources into power plants based on high-dimensional model expression provided by an embodiment of the present invention is shown as follows: Figure 1 As shown, the method specifically includes the following steps:

[0124] Step S100: using Zeno polyhedron to approximate the feasible domain of power generation factors, and performing Minkowski sum operation of Zeno polyhedron to aggregate the feasible domain of power generation factors.

[0125] Specifically, based on the Minkowski sum combined power generation factor constraint space, the Zeno polyhedron is used to approximate the power generation factor feasible region, and aggregation is performed through the Minkowski sum operation of the Zeno polyhedron, thereby realizing the standardized power generation factor aggregation. The specific implementation steps include:

[0126] 1) The power feasible region of a single power generation factor can be represented by a convex polyhedron The specific formula is as follows:

[0127] (1)

[0128] Where, A convex polyhedron representing a single power generation factor; The power set representing the power generation factor; The dimension representing the power generation factor; is the constraint matrix; represents the constraint space; Indicates the number of constraints; is the constraint vector;

[0129] 2) The aggregation of the feasible regions of multiple power generation factors can be expressed as the Minkowski sum of these power generation factors. The feasible region after aggregation is The formula is as follows:

[0130] (2)

[0131] Where, represents the power feasible region of the aggregate; represents the power feasible region of the jth power generation factor; Indicates the total number of power generation factors.

[0132] 3) Use Zeno polyhedron to approximate the feasible region of each power generation factor. The feasible region of a single power generation factor Zeno polyhedron The approximate formula is as follows:

[0133] (3)

[0134] Where, The power vector representing the power generation factor; is the center of the Zeno polyhedron; represents the dimension of the generator matrix G; is the generator matrix; Represents the generator coefficient Dimensions; is the coefficient of the generator; is the maximum scaling factor of the generator;

[0135] 4) Design an optimization problem, the optimization goal is to maximize the number of Zeno polyhedra and original convex polyhedra The similarity between them is subject to the constraints of the Zeno polyhedron Must be located on a convex polyhedron Internally, the formula is as follows:

[0136] (4)

[0137] The similarity is defined as the diameter ratio of two feasible regions in a certain direction:

[0138] (5)

[0139] Where, and Zeno polyhedrons and convex polyhedrons In the Directions The diameter on.

[0140] 5) Define the optimization objective formula for the Zeno polyhedron approximation problem as shown below:

[0141] (6)

[0142] Where, Indicates the total number of directions.

[0143] The constraint formula is as follows:

[0144] (7)

[0145] The approximate feasible region of the Zeno polyhedron that satisfies feasibility is obtained.

[0146] 6) Combine the feasible regions of multiple Zeno polyhedra through the Minkowski sum operation to obtain the Zeno polyhedron representation of the aggregate, as shown in the following formula:

[0147] (8)

[0148] (9)

[0149] (10)

[0150] in, represents the power generation factor cluster; Represents a unique operation, used to remove duplicate generators; Indicates the Generator matrices of Zeno polyhedra; represents the center of the merged Zeno polyhedron; Indicates the the center of a Zeno polyhedron;

[0151] Step S200: Aggregate distributed load resources into virtual machine groups, dynamically optimize the aggregation and scheduling of power plant dynamic aggregation and encapsulation scheduling objects based on power generation characteristics and power generation economy, and realize dynamic construction and aggregation of power plant dynamic aggregation and encapsulation scheduling objects.

[0152] Specifically, Figure 2 The following figure shows a flow chart for the dynamic construction of a power plant-based dynamic aggregation and encapsulation scheduling object. This is achieved by aggregating distributed load resources into virtual machine groups, defining performance parameters, and dynamically optimizing the aggregation and scheduling of power plant-based dynamic aggregation and encapsulation scheduling objects based on power generation characteristics and economic efficiency. The specific implementation steps include:

[0153] 1) Aggregate distributed load resources into virtual machine groups. The virtual machine group model is shown in the following formula:

[0154] (11)

[0155] Where, A model representing a group of virtual machines; and Respectively represent the upper and lower limits of the output of the virtual machine group; and They represent the maximum ramp-up and ramp-down rates of the virtual machine group respectively; and They represent the minimum running time and minimum downtime of the virtual machine group respectively; Represents the cost function for a group of virtual machines.

[0156] 2) Collect real-time regulation data of each central air-conditioning load in the power plant dynamic aggregation and packaging scheduling object, including regulation potential and controllable time.

[0157] 3) Preprocess the collected load data, remove noise and perform normalization to ensure data consistency and comparability.

[0158] 4) The regulation potential of each central air conditioner It is the minimum value of the adjustment amount in each control cycle within the specified control time. The calculation can be expressed as:

[0159] (12)

[0160] Where, Indicates the maximum adjustment amount of the central air conditioner; is the regulated load; is the total number of control cycles.

[0161] 5) Calculate the regulation potential of each central air conditioner, obtain its regulation characteristic curve, and fit the relationship curve between the adjustable potential and control time of a single air conditioner. The formula is as follows:

[0162] (13)

[0163] Where, Indicates control time; represents the fitting function.

[0164] 6) Compare the adjustment characteristic curves of different central air conditioners to determine their consistency and consistency The calculation formula is:

[0165] (14)

[0166] Where, and Respectively represent Central air conditioner and Central air conditioner in adjustment cycle regulatory potential when Indicates the total number of adjustment cycles.

[0167] 7) Agglomerative Hierarchical Clustering is used to cluster central air conditioners. The goal is to form a power plant-like dynamic aggregate encapsulated scheduling object that meets scheduling requirements. The optimization objective function is shown in the following formula:

[0168] (15)

[0169] Where, It represents the number of all possible combinations of any two numbers among M numbers; Represents the number of central air conditioners in the group after aggregation. The constraint conditions are as follows:

[0170] (16)

[0171] Where, represents the load adjustment of the i-th central air conditioner; Indicates the total regulation amount of the power plant dynamic aggregation and encapsulation scheduling objects.

[0172] 8) Calculate the regulation price of each central air conditioner, which is adjusted according to the regulation potential, rated power and comfort level of the central air conditioner. The formula is as follows:

[0173] (17)

[0174] Where, Indicates the Central air conditioning adjustment quotation; Indicates the The quotation decision coefficient of central air conditioner; Indicates the Rated power of central air conditioner; Indicates the The comfort of central air conditioning.

[0175] 9) Minimize the total cost of the power plant dynamic aggregation and packaging scheduling objects. Define the total cost function of the power plant dynamic aggregation and packaging scheduling objects. The formula is as follows:

[0176] (18)

[0177] Where, represents the total cost function; Indicates control duration; Indicates the The regulation capacity of a central air conditioner; N represents the total number of central air conditioners.

[0178] 10) The dispatch of the power plant dynamic aggregation and encapsulation dispatch objects must meet the total regulation requirements of the power grid dispatch department. The constraint formula is as follows:

[0179] (19)

[0180] Where, Indicates the total adjustment amount of the scheduling plan; Indicates the The load regulation potential of a central air conditioner; N represents the total number of central air conditioners.

[0181] Step S300: Match the power generation operation mode of the power plant dynamic aggregation package scheduling object with the equipment operation mode corresponding to each response characteristic curve in the building potential characteristic library to achieve online mode matching of the external characteristic parameters of the aggregation unit.

[0182] Specifically, the power generation operation mode of the power plant dynamic aggregation package scheduling object is matched with the equipment operation mode corresponding to each response characteristic curve in the building potential characteristic library to achieve online mode matching of the external characteristic parameters of the aggregation unit. The specific implementation steps include:

[0183] 1) Taking the maximum input power of intermittent energy sources such as wind and solar power as input, the external characteristic parameters of the computer group and the virtual power plant are obtained. The external characteristic parameters of the virtual power plant include: rated power, upper and lower limits of economic power, dead band of unit control error, up and down ramp rate, speed regulation droop of the governor, upper and lower limits of secondary frequency regulation (AGC), minimum time interval between unit shutdown and restart, and cost-capacity function.

[0184] 2) Establish a normalized information model for obtaining online parameters of power plant-based dynamic aggregation and encapsulation scheduling objects. The formula is as follows:

[0185] (20)

[0186] Where, Indicates the The maximum output of each unit represents the output uncertainty of the distributed generator set; Indicates the time when the control signal is released; Indicates the Units in The winning bid amount for the time period; Indicates the Units in The amount of various types of spare winning bids during the time period; Respectively represent Dynamic ramp-up and ramp-down rate of each unit (MW / s).

[0187] 3) Model the output sequence of the power plant dynamic aggregation and packaging scheduling objects and internal power generation resources. The formula is as follows:

[0188] (twenty one)

[0189] Where, Indicates that the power plant dynamic aggregation package scheduling object is Each moment in the period Response sequence to system output regulation signals; Indicates that the power plant dynamic aggregation package scheduling object is Basic output during the time period; Indicates that the power plant dynamic aggregation package scheduling object is Each moment in the period output sequence;

[0190] 4) Model the maximum output of the power plant dynamic aggregation and packaging scheduling object. The formula is as follows:

[0191] (twenty two)

[0192] Where, Indicates that the power plant dynamic aggregation package scheduling object is Each moment in the period The maximum output sequence of Indicates the first Intermittent energy units in Each moment in the time period The maximum output sequence depends on the current maximum input and control interval of the unit The product of the climbing / descending rate; Indicates the first The maximum output power of a conventional unit.

[0193] 5) Perform real-time standby modeling of power plant dynamic aggregation and encapsulation scheduling objects. The formula is as follows:

[0194] (twenty three)

[0195] (twenty four)

[0196] Where, Indicates remaining spare; Indicates that each intermittent new energy unit is Each moment in the period The load shedding ratio depends on the standby control strategy of each unit and the load shedding ratio constraint; Indicates the first Conventional units in Time slot backup; Indicates that the virtual power plant is in the period At each moment of standby power.

[0197] 6) In the performance parameters of the power plant dynamic aggregation package scheduling object, the subsequent climbing / declining rate calculation formula is as follows:

[0198] (25)

[0199] Where, Indicates that under different scheduling or control instructions, The climbing / descending rate during the time period; in the subsequent climbing / descending rate calculation, Dynamically aggregate and encapsulate the output data of scheduling objects for actual power plants.

[0200] 7) In the performance parameters of the power plant dynamic aggregation package scheduling object, the calculation formula of the pre-calculation / downgrade rate is consistent with the post-calculation expression. From the above power plant dynamic aggregation and packaging scheduling object output sequence calculation model, we can get: It is an arbitrary parameter.

[0201] 8) The frequency regulation performance score of the power plant dynamic aggregation and packaging scheduling object in each time period is calculated as follows:

[0202] (26)

[0203] (27)

[0204] (28)

[0205] (29)

[0206] (30)

[0207] Where, represents the frequency modulation performance score, which is the weighted average of the three scores; DS represents the response delay score; CS represents the correlation score; PS represents the accuracy score; A, B, and C represent the weights, all of which are 1 / 3; Indicates the response delay at which CS and DS achieve their maximum values; Indicates the predetermined adjustment power; Indicates the actual regulated power; represents the correlation function; Indicates the average value of the scheduled regulated power per hour.

[0208] Step S400: establishing a mapping relationship between physical parameters and aggregation model parameters through high-dimensional model expression technology, and realizing real-time online identification of aggregation unit external characteristic parameters based on high-dimensional model expression technology.

[0209] Specifically, Figure 3 Shows a schematic diagram of the high-dimensional model expression modeling process, Figure 4 The flowchart of the real-time identification framework of air conditioning load aggregation model parameters is shown. The mapping relationship between physical parameters and aggregation model parameters is established through high-dimensional model expression technology, and the real-time online identification of the external characteristic parameters of the aggregation unit based on high-dimensional model expression technology is realized. The specific implementation steps include:

[0210] 1) The influence of uncertainty in the physical parameters of air conditioning load on the aggregation model parameters is analyzed through high-dimensional model expression. The calculation formula is as follows:

[0211] (31)

[0212] Where, represents a constant term; represents the first-order component; represents the second-order component; represents the order of the component; the system input is , the output is , and so on.

[0213] 2) Generate The input samples include multiple physical parameters of air conditioning load. In specific implementation, the Quasi Monte Carlo method can be used to generate Group input samples.

[0214] 3) Each set of input samples is simulated to obtain the aggregate model parameters, thereby obtaining output samples, as shown in the following formula:

[0215] (32)

[0216] Where, Indicates the Group input samples The corresponding aggregation model output; Indicates the Group input samples; Indicates the The value of the output quantity;

[0217] 4) Standardize each input quantity using the following formula:

[0218] (33)

[0219] Where, and Represent the actual value and standard value of the input variable respectively; and Represents the minimum and maximum values of the input respectively.

[0220] 5) Use the input and output values of N0 to establish a high-dimensional model expression model of the air conditioning group and calculate the correlation coefficient , And the global sensitivity of the physical parameters of the air conditioning load is as follows:

[0221] (34)

[0222] (35)

[0223] Where, 、 、 、 、 and etc. all represent the physical parameters of the air conditioning load. By analogy, we can obtain higher-order coefficients of the first-order component function of the input quantity and the second-order component function of the input quantity; represents the first-order component function.

[0224] 6) Based on the values of higher-order coefficients α and β and the HDMR model of the conditioning group, the global sensitivity of the input variables to independent and coupled effects is calculated.

[0225] 7) Establish a high-dimensional model expression model database for air conditioning aggregation model parameters at different time scales, see Figure 4 .

[0226] 8) The above is generated using the Quasi Monte Calculator method The input samples of the group air-conditioning system can be expressed as:

[0227] (36)

[0228] Where, 、 、 、 、 、 、 、 、 、 、 and In order to verify the effectiveness of the dynamic aggregation and encapsulation method of micro-regulation resource power plant based on high-dimensional model expression, the above method is put into practical application and a modified real example is selected.

[0229] This example mainly verifies the identification results of the air conditioning load aggregation model parameters based on the high-dimensional model. The number of fixed-frequency and variable-frequency air conditioners is 10,000 respectively. Assuming that the air conditioning load parameters are normally distributed, the distribution of the physical parameters of a certain air conditioning group is shown in Table 1 below. In Table 1, R μ 、C a , P, T o ,δ,η μ 、P min 、P max , k1, l1, k2, l2 respectively represent the response characteristics of air conditioning load, heat capacity or load, power or load, set temperature, temperature variation range or control error, air conditioning efficiency, minimum power, maximum power, adjustment coefficient, nonlinear control coefficient, adjustment coefficient and control error correlation coefficient; μ represents the mean; χ represents the standard deviation; before the air conditioning load participates in demand response, the entire air conditioning group operates in a stable state, and each air conditioning load operates at the set temperature. For fixed-frequency air conditioning, the indoor temperature is [ ] are evenly distributed within.

[0230] Table 1 Parameter distribution of air conditioning load

[0231] parameter μ χ <![CDATA[R μ ]]> 2 0.2 <![CDATA[C a ]]> 2 0.2 P 5.6 0.56 <![CDATA[T set ]]> 22.5 2.25 δ 0.3 0.03 <![CDATA[η μ ]]> 2.5 0.25 <![CDATA[P min ]]> 0.5 0.05 <![CDATA[P max ]]> 5 0.5 <![CDATA[k1]]> 0.03 0.003 <![CDATA[l1]]> -0.4 0.04 <![CDATA[k2]]> 0.06 0.006 <![CDATA[l2]]> -0.3 0.03

[0232] First, we take the example of fixed-frequency air conditioning load generating minute-level power pulses to analyze the high-dimensional model expression modeling results of its power capacity. In this example, only three physical parameters of the air conditioning load have a significant impact on the capacity of the power pulse, which are , and According to the high-dimensional model expression modeling results, the capacity of the power pulse The relationship between it and the physical parameters of air conditioning load is:

[0233] (37)

[0234] Where, 、 and Respectively represent the standardized values; 、 Represent the first and second order component functions respectively.

[0235] The distribution of physical parameters of an air conditioning group is shown in Table 2. While the distribution of physical parameters remains unchanged, 100 sets of air conditioning group parameter samples are generated according to the distribution of the physical parameters of the air conditioning group. The actual capacity of 100 power pulses is obtained by independent simulation.

[0236] Table 2 Global sensitivity of physical parameters of air conditioning load

[0237] parameter #timg# #timg# #timg# Global sensitivity 0.5164 0.3902 0.0445

[0238] Figure 5 A comparison of the power pulse capacity of the air conditioning load obtained through independent simulation and calculation is shown. As can be seen, the difference between the power capacity calculated using the high-dimensional model and independent simulation is very small, with the calculated average relative error being only 4.38%. This demonstrates that the high-dimensional model can accurately establish a mapping relationship between power capacity and the physical parameters of the air conditioning load.

[0239] In addition, the multivariate linear regression method was used to establish the relationship between the power pulse capacity of the air conditioning load and its physical parameters, and the modeling results were compared with the modeling results expressed by the high-dimensional model. When the scheduling period is 20 minutes, the modeling results of the multivariate linear regression are:

[0240] (38)

[0241] The modeling results expressed by the high-dimensional model are:

[0242] (39)

[0243] Where, 、 、 Represent the functions of each order component respectively.

[0244] Verify the modeling results expressed by the high-dimensional model by Figure 6 The simulation comparison shows that the average relative errors of the two methods are 15.03% and 5.52% respectively. In terms of parameter identification of air-conditioning load aggregation model, the high-dimensional model expression has higher modeling accuracy.

[0245] In the above-mentioned dynamic aggregation and packaging method for fine-tuning resources based on high-dimensional model expression, the mathematical model and properties of Minkowski sums and Zeno polyhedra are used to express the feasible domain of power generation factors as a convex polyhedron. The constraint space of the power aggregation power of power generation factors is then expressed as the Minkowski sum of the constraint spaces of each power generation factor, and a standardized power generation factor aggregation method is proposed. Based on the constructed power plant dynamic aggregation and packaging scheduling object unit model, a dynamic construction technology for power plant dynamic aggregation and packaging scheduling objects is proposed from the perspectives of power generation characteristics and power generation economy, significantly improving aggregation efficiency and accuracy. A system of external characteristic parameter indicators for aggregation units is proposed to match the power generation operation mode of the aggregation unit with the equipment operation mode corresponding to each response characteristic curve in the resource potential characteristic library, realizing online dynamic identification of the external characteristic parameters of the aggregation unit. This method can aggregate and optimize resources in real time and dynamically while ensuring computational efficiency and accuracy. Especially in high-dimensional and large-scale data processing, it can ensure the unified regulation and optimized utilization of flexible resources and can respond quickly to different scheduling needs.

[0246] Based on the same inventive concept, an embodiment of the present invention also provides a system for dynamically aggregating and packaging fine-tuning resources for power plants based on high-dimensional model expression, the system comprising: a memory, a processor, and a computer program stored in the memory and running on the processor, wherein when the processor executes the computer program, the system can execute any one of the methods for dynamically aggregating and packaging fine-tuning resources for power plants based on high-dimensional model expression described above.

[0247] In embodiments of the present invention, the system can be divided into functional modules based on the above-described method examples. For example, these modules can correspond to individual functional modules, or two or more functions can be integrated into a single processing module. The integrated modules can be implemented in hardware. It should be noted that the module division in this embodiment is illustrative and represents only a logical functional division. In actual implementation, other division methods may be used.

[0248] Based on the same inventive concept, an embodiment of the present invention also provides a computer program product, which includes: computer program code, which, when running on a computer, enables the computer to execute any one of the aforementioned methods for dynamic aggregation and packaging of fine-tuning resource power plants based on high-dimensional model expression.

[0249] Based on the same inventive concept, an embodiment of the present invention also provides a computer-readable storage medium, which stores computer program code. When the computer program code runs on a computer, the computer executes any one of the aforementioned methods for dynamic aggregation and packaging of fine-tuning resource power plants based on high-dimensional model expression.

[0250] It should be noted that the above-described embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit the same. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. These modifications or replacements do not deviate the essence of the corresponding technical solutions from the scope of the technical solutions of the various embodiments of the present invention, and should all be included in the protection scope of the present invention.

Claims

1. A dynamic aggregation and packaging method for fine-tuning resource power plants based on high-dimensional model expression, characterized in that: The following steps are involved: Aggregate the feasible domains of different power generation factors to obtain a feasible domain aggregation model of power generation factors; Aggregate distributed load resources into virtual machine groups, dynamically optimize the aggregation and scheduling of power plant dynamic aggregation and encapsulation scheduling objects based on power generation characteristics and power generation economics, and realize the dynamic construction and aggregation of power plant dynamic aggregation and encapsulation scheduling objects; Match the power generation operation mode of the power plant dynamic aggregation package scheduling object with the equipment operation mode corresponding to each response characteristic curve in the building potential characteristic library to achieve online mode matching of the external characteristic parameters of the aggregation unit; The mapping relationship between the physical parameters of the air conditioning load and the aggregation model parameters is established through high-dimensional model expression technology, and the real-time online identification of the external characteristic parameters of the aggregation unit based on high-dimensional model expression technology is realized; Match the power generation operation mode of the power plant dynamic aggregation package scheduling object with the equipment operation mode corresponding to each response characteristic curve in the building potential characteristic library to achieve online mode matching of the external characteristic parameters of the aggregation unit, including: 1) Using the maximum input power of intermittent energy sources such as wind and solar power as input, the external characteristic parameters of the computer group and the virtual power plant as a whole are obtained; 2) Establish a normalized information model for obtaining online parameters of power plant-based dynamic aggregation and encapsulation scheduling objects. The formula is as follows: (15) Where, Indicates the The maximum output of each unit represents the output uncertainty of the distributed generator set; Indicates the time when the control signal is released; Indicates the Units in The winning bid amount for the time period; Indicates the Units in The amount of various types of spare winning bids during the time period; Respectively represent Dynamic climbing and descending rate of each unit; 3) Model the output sequence of the power plant dynamic aggregation and packaging scheduling objects and internal power generation resources. The formula is as follows: (16) Where, Indicates that the power plant dynamic aggregation package scheduling object is Each moment in the period Response sequence to system output regulation signals; Indicates that the power plant dynamic aggregation package scheduling object is The basic output of the time period is also the bidding strategy of the virtual power plant in the h period; Indicates that the power plant dynamic aggregation package scheduling object is Each moment in the period output sequence; 4) Model the maximum output of the power plant dynamic aggregation and packaging scheduling object. The formula is as follows: (17) Where, Indicates that the power plant dynamic aggregation package scheduling object is Each moment in the period The maximum output sequence of Indicates the first Intermittent energy units in Each moment in the time period The maximum output sequence depends on the current maximum input and control interval of the unit The product of the climbing / descending rate; Indicates the first Maximum output power of conventional units; 5) Perform real-time standby modeling of power plant dynamic aggregation and encapsulation scheduling objects. The formula is as follows: (18) (19) Where, Indicates remaining spare; Indicates that each intermittent energy unit is Each moment in the period The load shedding ratio depends on the standby control strategy of each unit and the load shedding ratio constraint; Indicates the first Conventional units in Time slot backup; Indicates that the virtual power plant is in the period At each moment The reserve power; 6) In the performance parameters of the power plant dynamic aggregation package scheduling object, the subsequent climbing / declining rate calculation formula is as follows: (20) Where, Indicates that under different scheduling or control instructions, The climbing / descending rate during the time period; in the subsequent climbing / descending rate calculation, Dynamically aggregate and encapsulate the output data of dispatch objects for actual power plants; 7) In the performance parameters of the power plant dynamic aggregation package scheduling object, the calculation formula of the pre-calculation / downgrade rate is consistent with the post-calculation expression. From the above power plant dynamic aggregation and packaging scheduling object output sequence calculation model, we can get: It is an arbitrary parameter; 8) The frequency regulation performance score of the power plant dynamic aggregation and packaging scheduling object in each time period is calculated as follows: (21) (22) (23) (24) (25) Where, represents the frequency modulation performance score, which is the weighted average of the three scores; DS represents the response delay score; CS represents the correlation score; PS represents the accuracy score; A, B, and C represent the weights, all of which are 1 / 3; It represents the response delay for CS and DS to achieve the maximum value; Reg represents the predetermined adjustment power; Indicates the actual regulated power; n indicates the total number of predetermined regulated powers; represents the correlation function; Indicates the average value of the scheduled regulated power per hour.

2. The method for dynamic aggregation and packaging of fine-tuning resources into power plants based on high-dimensional model expression according to claim 1 is characterized in that: The feasible domains of different power generation factors are aggregated to obtain the feasible domain aggregation model of power generation factors, including: The Zeno polyhedron is used to approximate the feasible domain of power generation factors, and the feasible domains of different power generation factors are aggregated through the Minkowski sum operation of the Zeno polyhedron, so as to obtain the feasible domain aggregation model of power generation factors.

3. The method for dynamic aggregation and packaging of fine-tuning resource power plants based on high-dimensional model expression according to claim 2 is characterized in that: The feasible domain of power generation factors is approximated by Zeno polyhedron, and the feasible domains of different power generation factors are aggregated through the Minkowski sum operation of Zeno polyhedron, thereby obtaining the feasible domain aggregation model of power generation factors, including: Use convex polyhedron to represent the feasible region of a single power generation factor: (1) Where, A convex polyhedron representing a single power generation factor; The power set representing the power generation factor; The dimension representing the power generation factor; is the constraint matrix; represents the constraint space; Indicates the number of constraints; is the constraint vector; The aggregation of the feasible regions of multiple power generation factors is expressed as the Minkowski sum of all power generation factors. The feasible region after aggregation is The formula is as follows: (2) Where, represents the feasible domain of the aggregate; represents the feasible region of the jth power generation factor; Indicates the total number of power generation factors; Use Zeno polyhedron to approximate the feasible region of each power generation factor. The feasible region of a single power generation factor Zeno polyhedron The approximate formula is as follows: (3) Where, The power vector representing the power generation factor; is the center of the Zeno polyhedron; represents the dimension of the generator matrix G; is the generator matrix; Represents the generator coefficient Dimensions; is the coefficient of the generator; is the maximum scaling factor of the generator; Determine the optimization objective and constraints of the Zeno polyhedron approximation problem, solve the optimization objective based on the constraints, determine the feasible domain of the Zeno polyhedron approximation that meets feasibility, and merge the Zeno polyhedron approximation feasible domain by Minkowski sum operation to obtain the Zeno polyhedron representation of the aggregate, and determine the Zeno polyhedron representation of the aggregate as the feasible domain aggregation model of the power generation factor.

4. The method for dynamic aggregation and packaging of fine-tuning resource power plants based on high-dimensional model expression according to claim 3 is characterized in that: Determine the optimization objective and constraints of the Zeno polyhedron approximation problem. The corresponding calculation formula is: (4) s.t. (5) Where, Represents a Zeno polyhedron and the convex polyhedron of the feasible region In the The diameter ratio in each direction, , and Zeno polyhedrons and convex polyhedrons In the diameter in each direction; Indicates the total number of directions.

5. The method for dynamic aggregation and packaging of fine-tuning resources into power plants based on high-dimensional model expression according to claim 1 is characterized in that: Aggregate distributed load resources into virtual machine groups, dynamically optimize the aggregation and scheduling of power plant-based dynamic aggregation and encapsulation scheduling objects based on power generation characteristics and power generation economics, and realize the dynamic construction and aggregation of power plant-based dynamic aggregation and encapsulation scheduling objects, including: 1) Aggregate distributed load resources into virtual machine groups. The virtual machine group model is shown in the following formula: (6) Where, A model representing a group of virtual machines; and Respectively represent the upper and lower limits of the output of the virtual machine group; and They represent the maximum ramp-up and ramp-down rates of the virtual machine group respectively; and They represent the minimum running time and minimum downtime of the virtual machine group respectively; represents the cost function of the virtual machine group; 2) Collect real-time regulation data of each central air-conditioning load in the power plant dynamic aggregation and packaging scheduling object, including regulation potential and manageable time; 3) Preprocess the collected load data, remove noise and perform normalization to ensure data consistency and comparability; 4) Regulation potential of each central air conditioner It is the minimum value of the adjustment amount in each control cycle within the specified control time. The calculation can be expressed as: (7) Where, Indicates the maximum adjustment amount of the central air conditioner; is the regulated load; is the total number of control cycles; 5) Calculate the regulation potential of each central air conditioner, obtain its regulation characteristic curve, and fit the relationship curve between the adjustable potential and control time of a single air conditioner. The formula is as follows: (8) Where, Indicates control time; represents the fitting function; 6) Compare the adjustment characteristic curves of different central air conditioners to determine their consistency and consistency The calculation formula is: (9) Where, and Respectively represent Central air conditioner and Central air conditioner in adjustment cycle regulatory potential when Indicates the total number of adjustment cycles; 7) Use the agglomerative hierarchical clustering method to aggregate central air conditioners. The goal is to form a power plant-like dynamic aggregation package scheduling object that meets the scheduling requirements. The optimization objective function is shown in the following formula: (10) Where, It represents the number of all possible combinations of any two numbers among M numbers; Represents the number of central air conditioners in the group after aggregation; the constraint conditions are as follows: (11) Where, represents the load adjustment of the i-th central air conditioner; Indicates the total regulation amount of the power plant dynamic aggregation and encapsulation scheduling objects; 8) Calculate the regulation price of each central air conditioner, which is adjusted according to the regulation potential, rated power and comfort level of the central air conditioner. The formula is as follows: (12) Where, Indicates the Central air conditioning adjustment quotation; Indicates the The quotation decision coefficient of central air conditioner; Indicates the Rated power of central air conditioner; Indicates the The comfort of central air conditioning; 9) Minimize the total cost of the power plant dynamic aggregation and packaging scheduling objects. Define the total cost function of the power plant dynamic aggregation and packaging scheduling objects. The formula is as follows: (13) Where, represents the total cost function; Indicates control duration; Indicates the The regulation amount of the central air conditioner; N represents the total number of central air conditioners; 10) The dispatch of the power plant dynamic aggregation and encapsulation dispatch objects must meet the total regulation requirements of the power grid dispatch department. The constraint formula is as follows: (14) Where, Indicates the total adjustment amount of the scheduling plan; Indicates the The load regulation potential of a central air conditioner.

6. The method for dynamic aggregation and packaging of fine-tuning resources into power plants based on high-dimensional model expression according to claim 1 is characterized in that: The mapping relationship between physical parameters and aggregation model parameters is established through high-dimensional model expression technology, and the real-time online identification of the external characteristic parameters of the aggregation unit based on high-dimensional model expression technology is realized, including: 1) The influence of uncertainty in the physical parameters of air conditioning load on the aggregation model parameters is analyzed through high-dimensional model expression. The calculation formula is as follows: (26) Where, represents a constant term; represents the first-order component; represents the second-order component; represents the order of the component; the system input is , the output is , and so on; 2) Generate Group input samples, the input includes multiple physical parameters of air conditioning load; 3) Each set of input samples is simulated to obtain the aggregate model parameters, thereby obtaining output samples, as shown in the following formula: (27) Where, Indicates the Group input samples The corresponding aggregation model output; Indicates the Group input samples; Indicates the The value of the output quantity; 4) Standardize each input quantity using the following formula: (28) Where, and Represent the actual value and standard value of the input variable respectively; and Respectively represent the minimum and maximum values of the input; 5) Use the input and output values of N0 to establish a high-dimensional model expression model of the air conditioning group and calculate the correlation coefficient , And the global sensitivity of the physical parameters of the air conditioning load is as follows: (29) (30) Where, 、 、 、 、 and All of them represent the physical parameters of the air conditioning load. By analogy, we can obtain higher-order coefficients of the first-order component function of the input quantity and the second-order component function of the input quantity; represents the first-order component function; 6) Calculate the global sensitivity of the independent and coupled inputs; 7) Establish a high-dimensional model expression model database for air conditioning aggregation model parameters at different time scales.

7. The method for dynamic aggregation and packaging of fine-tuning resource power plants based on high-dimensional model expression according to claim 6 is characterized in that: Generated using the Quasi Monte Calculator method The input sample of the group air-conditioning system can be expressed as: (31) Where, 、 、 、 、 、 and Both represent the physical parameters of air conditioning load.

Images