Data-driven mass resource regulation constraint dimension reduction method and device

Abstract

The present disclosure provides a data-driven mass resource regulation constraint dimension reduction method and device, and relates to the technical field of virtual power plant regulation. The method comprises: constructing a high-dimensional regulation constraint model based on the operating parameters of distributed resources; embedding the high-dimensional regulation constraint model into an upper-layer regulation optimization model, obtaining a regulation simulation data set based on the historical operating data of the distributed resources participating in the upper-layer regulation; constructing a low-dimensional regulation constraint model according to the regulation simulation data set; embedding the low-dimensional regulation constraint model into the upper-layer regulation optimization model, and performing regulation constraint. Thus, the low-dimensional regulation constraint model can be used to approximate the original high-dimensional constraint of the mass distributed resources, greatly reducing the calculation burden and solving time of the mass resource collaborative regulation, and the data-driven technical route is adopted, which has better universality and adaptability than the traditional constraint dimension reduction method, and is helpful to better develop the flexibility of mass resources.

Description

Technical Field

[0001] This disclosure relates to the field of virtual power plant control technology, and in particular to a data-driven method and apparatus for reducing the constraints of massive resource control. Background Technology

[0002] Globally, power systems are gradually reducing their reliance on fossil fuels and significantly increasing installed capacity of wind and solar power. Against this backdrop, the supply-demand balance of power systems faces unprecedented challenges because the output of wind and solar power is affected by variable weather conditions, unlike traditional thermal power which can be flexibly adjusted. To address these challenges, a cost-effective approach is to leverage the flexibility of massive demand-side resources. However, since the capacity of individual resources is typically small, aggregation through virtual power plants is often necessary to meet market entry barriers. The upper-level grid directly optimizes control based on the feasible domain and cost characteristics of the virtual power plants, achieving deep interaction. To reduce the computational complexity of upper-level control optimization, virtual power plants need to submit their overall control constraints to the grid operator, rather than directly submitting the original control constraints of all resources; this process is the dimensionality reduction of massive resource control constraints.

[0003] In related technologies, the dimensionality reduction of resource regulation constraints essentially involves obtaining the overall regulation constraints (i.e., aggregation or dimensionality-reduced regulation constraints) from the original regulation constraints of massive resources. This process is not a simple algebraic summation of individual model parameters, but rather the summation of multiple sets, i.e., finding the Minkowski sum, for which there is currently no effective universal solution method. Therefore, common methods use internal or external approximations to obtain computationally feasible results. For example, researchers at Harvard University proposed using ellipses to approximate the regulation constraints of massive resources to accelerate subsequent aggregation. However, most methods for dimensionality reduction of regulation constraints are based on analytical methods, which have limitations in terms of universality and adaptability. Summary of the Invention

[0004] To overcome the problems existing in related technologies, this disclosure provides a data-driven method and apparatus for regulating and reducing the dimensionality of massive resources.

[0005] According to a first aspect of the present disclosure, a data-driven method for reducing the dimensionality of constraints for the regulation of massive resources is provided, comprising: constructing a high-dimensional regulation constraint model based on the operating parameters of distributed resources; embedding the high-dimensional regulation constraint model into an upper-level regulation optimization model, and obtaining a regulation simulation dataset based on historical operating data of distributed resources participating in upper-level regulation; constructing a low-dimensional regulation constraint model based on the regulation simulation dataset; and embedding the low-dimensional regulation constraint model into the upper-level regulation optimization model to perform regulation constraints.

[0006] In some embodiments of this disclosure, a high-dimensional control constraint model is embedded into an upper-level control optimization model, and a control simulation dataset is obtained based on historical operating data of distributed resources participating in upper-level control. This includes: generating parameters for multiple typical control scenarios based on historical operating data of distributed resources participating in upper-level control; embedding the high-dimensional control constraint model into the upper-level control optimization model; generating optimal control results under different typical control scenarios based on the parameters of multiple typical control scenarios; and obtaining a control simulation dataset.

[0007] In some embodiments of this disclosure, a low-dimensional control constraint model is constructed based on a control simulation dataset, including: using the optimal control result as the running result of the low-dimensional control constraint model; and fitting the parameters of the low-dimensional control constraint model to determine the low-dimensional control constraint model based on the prediction results obtained by inputting parameters of multiple typical control scenarios into the low-dimensional control constraint model and the optimal control result as the running result.

[0008] In some embodiments of this disclosure, the high-dimensional regulation constraint model includes the original regulatory feasible region P of the distributed resource; wherein, Here, P consists of a set of linear constraints about (p, r), and the variable with the subscript [T] is a T-dimensional vector, p [T] =(p1,p2,...,p T ), r [T] =(r1,r2,...,r T ), p T For the distributed resource's basic energy output in hour T, r T For the frequency regulation capability of distributed resources in hour T, y [I][T] For the I-th distributed resource, h(·) includes the power and energy constraints of the distributed resource, θ I Let I and T be the parameters of the i-th distributed resource, where I and T are both positive integers; The high-dimensional regulation constraint model is embedded in the upper-level regulation optimization model, and the regulation optimization is expressed as:

[0009]

[0010] st(p [T] ,r [T] )∈P;

[0011] in, The energy price for distributed resources, For the equivalent frequency regulation capacity price of distributed resources, s perf For performance scores, For frequency modulation capacity price, Price based on FM mileage. Let be the expected frequency modulation mileage at time t. The cost function for distributed resources;

[0012] For a given different energy frequency modulation price By calling a commercial solver to solve the above regulation optimization, the optimal regulation result (p) is obtained. [T] ,r [T] ), thus obtaining the regulation simulation dataset. Where k is the total number of indices.

[0013] In some embodiments of this disclosure, the low-dimensional control constraint model includes upper and lower limits of energy storage charging and discharging power, upper and lower limits of energy storage energy, energy storage energy conversion constraints, and initial energy, expressed as follows:

[0014]

[0015]

[0016] Where, p t,i For the output at time interval t, r t,i e is the frequency modulation capacity for time interval t. t,i Let be the energy at the end of t. The underlined parameter indicates the lower limit of the corresponding variable, and the overlined parameter indicates the upper limit. The superscript 'dis' indicates discharging into the grid, and the superscript 'ch' indicates charging from the grid. 0,i The initial energy is given by η, the charge / discharge efficiency is given by μ, and λ is the Lagrange multiplier for the corresponding constraint, respectively.

[0017] Based on the low-dimensional regulation constraint model, the constraints provided by distributed resource i for frequency regulation include non-negative capacity constraints, power capacity constraints, and sustaining time requirements, expressed as:

[0018]

[0019] Where, Δt req The time required to maintain maximum frequency modulation output for distributed resources;

[0020] The sum of bids for distributed resources is represented as:

[0021] The optimal control result in the control simulation dataset D is used as the optimal solution of the low-dimensional control constraint model. The parameters of the low-dimensional control constraint model are fitted using the control simulation dataset D, and the optimization objective is to minimize the distance to the optimal control result.

[0022]

[0023] In this context, the superscript 'a' represents the parameters of the low-dimensional control constraint model.

[0024] In some embodiments of this disclosure, the above method further includes: taking the optimal control result in the control simulation dataset D as the optimal solution of the low-dimensional control constraint model, solving it using an iterative algorithm, randomly introducing only one typical control scenario control simulation data each time, designing a learning rate to update parameters, until the parameters of the low-dimensional control constraint model converge, and obtaining the low-dimensional control constraint model.

[0025] By embedding the low-dimensional control constraint model into the upper-level control optimization model for control constraints, the scale of resource control constraints after dimensionality reduction is much smaller than the original scale, i.e., I a ≤I; The feasible region of the control constraint is:

[0026]

[0027] According to a second aspect of the present disclosure, a data-driven massive resource regulation constraint dimensionality reduction device is provided, comprising: a first model determination unit, configured to construct a high-dimensional regulation constraint model based on the operating parameters of distributed resources; a dataset acquisition unit, configured to embed the high-dimensional regulation constraint model into an upper-level regulation optimization model and acquire a regulation simulation dataset based on historical operating data of distributed resources participating in upper-level regulation; a second model determination unit, configured to construct a low-dimensional regulation constraint model based on the regulation simulation dataset; and a processing unit, configured to embed the low-dimensional regulation constraint model into the upper-level regulation optimization model for regulation constraint.

[0028] In some embodiments of this disclosure, the dataset acquisition unit is specifically used to: generate parameters for multiple typical control scenarios based on historical operating data of distributed resources participating in upper-level control; embed a high-dimensional control constraint model into the upper-level control optimization model; generate optimal control results under different typical control scenarios based on the parameters of multiple typical control scenarios; and obtain a control simulation dataset.

[0029] In some embodiments of this disclosure, the second model determining unit is specifically used to: use the optimal control result as the running result of the low-dimensional control constraint model; and based on the prediction results obtained by inputting parameters of multiple typical control scenarios into the low-dimensional control constraint model and the optimal control result as the running result, fit the parameters of the low-dimensional control constraint model to determine the low-dimensional control constraint model.

[0030] In some embodiments of this disclosure, the high-dimensional regulation constraint model includes the original regulatory feasible region P of the distributed resource; wherein,

[0031] Here, P consists of a set of linear constraints about (p, r), and the variable with the subscript [T] is a T-dimensional vector, p [T] =(p1,p2,...,p T ), r [T] =(r1,r2,...,r T), p T For the distributed resource's basic energy output in hour T, r T For the frequency regulation capability of distributed resources in hour T, y [I][T] For the I-th distributed resource, h(·) includes the power and energy constraints of the distributed resource, θ I Let T be the parameter of the i-th distributed resource, where I and T are both positive integers;

[0032] Among them, the high-dimensional regulation constraint model is embedded in the upper-level regulation optimization model, and the regulation optimization is expressed as:

[0033]

[0034] st(p [T] ,r [T] )∈P;

[0035] in, The energy price for distributed resources, For the equivalent frequency regulation capacity price of distributed resources, s perf For performance scores, For frequency modulation capacity price, Price based on FM mileage. Let be the expected frequency modulation mileage at time t. The cost function for distributed resources;

[0036] For a given different energy frequency modulation price By calling a commercial solver to solve the above regulation optimization, the optimal regulation result (p) is obtained. [T] ,r [T] ), thus obtaining the regulation simulation dataset. Where k is the total number of indices.

[0037] In some embodiments of this disclosure, the low-dimensional control constraint model includes upper and lower limits of energy storage charging and discharging power, upper and lower limits of energy storage energy, energy storage energy conversion constraints, and initial energy, expressed as follows:

[0038]

[0039] Where, p t,i For the output at time interval t, r t,i e is the frequency modulation capacity for time interval t. t,i Let be the energy at the end of t. The underlined parameter indicates the lower limit of the corresponding variable, and the overlined parameter indicates the upper limit. The superscript 'dis' indicates discharging into the grid, and the superscript 'ch' indicates charging from the grid. 0,i The initial energy is given by η, the charge / discharge efficiency is given by μ, and λ is the Lagrange multiplier for the corresponding constraint, respectively.

[0040] Based on the low-dimensional regulation constraint model, the constraints provided by distributed resource i for frequency regulation include non-negative capacity constraints, power capacity constraints, and sustaining time requirements, expressed as:

[0041]

[0042] Where, Δt req The time required to maintain maximum frequency modulation output for distributed resources;

[0043] The sum of bids for distributed resources is represented as:

[0044] The optimal control result in the control simulation dataset D is used as the optimal solution of the low-dimensional control constraint model. The parameters of the low-dimensional control constraint model are fitted using the control simulation dataset D, and the optimization objective is to minimize the distance to the optimal control result.

[0045]

[0046] In this context, the superscript 'a' represents the parameters of the low-dimensional control constraint model.

[0047] In some embodiments of this disclosure, the optimal control result in the control simulation dataset D is used as the optimal solution of the low-dimensional control constraint model. An iterative algorithm is used to solve the model. Each time, only one typical control scenario control simulation data is randomly introduced. The learning rate is designed to update the parameters until the parameters of the low-dimensional control constraint model converge, thus obtaining the low-dimensional control constraint model.

[0048] By embedding the low-dimensional control constraint model into the upper-level control optimization model for control constraints, the scale of resource control constraints after dimensionality reduction is much smaller than the original scale, i.e., I a ≤I; The feasible region of the control constraint is:

[0049]

[0050] According to a third aspect of the present disclosure, an electronic device includes: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor, when executing the computer program, implements the method as described in any one of the first aspects.

[0051] According to a fourth aspect of the present disclosure, a computer-readable storage medium is provided having a computer program stored thereon, which, when executed by a processor, implements the method as described in any one of the first aspects.

[0052] According to a fifth aspect of the present disclosure, a computer program product is provided, including a computer program that, when executed by a processor, implements the method as described in any one of the first aspects.

[0053] The technical solutions provided by the embodiments of this disclosure can include the following beneficial effects: constructing a high-dimensional control constraint model based on the operating parameters of distributed resources; embedding the high-dimensional control constraint model into an upper-level control optimization model; obtaining a control simulation dataset based on historical operating data of distributed resources participating in upper-level control; constructing a low-dimensional control constraint model based on the control simulation dataset; and embedding the low-dimensional control constraint model into the upper-level control optimization model for control constraints. Therefore, the high-dimensional constraints of the original massive distributed resources can be approximated using a low-dimensional control constraint model, greatly reducing the computational burden and solution time of massive resource collaborative control. Furthermore, the data-driven technical approach has better versatility and adaptability than traditional constraint dimensionality reduction methods, contributing to better development of the flexibility of massive resources.

[0054] It should be understood that the above general description and the following detailed description are exemplary and explanatory only, and are not intended to limit this disclosure. Attached Figure Description

[0055] The accompanying drawings, which are incorporated in and form part of this specification, illustrate embodiments consistent with the invention and, together with the description, serve to explain the principles of the invention.

[0056] Figure 1 This is a flowchart illustrating a data-driven method for regulating and reducing constraints on massive resources, according to an exemplary embodiment.

[0057] Figure 2 This is a flowchart illustrating another data-driven method for regulating and reducing constraints on massive resources, according to an exemplary embodiment.

[0058] Figure 3 This is a structural diagram illustrating a data-driven massive resource regulation and constraint dimensionality reduction device according to an exemplary embodiment;

[0059] Figure 4 This is a structural diagram of an apparatus for performing a data-driven, massive resource regulation and constraint dimensionality reduction method, according to an exemplary embodiment. Detailed Implementation

[0060] Exemplary embodiments will now be described in detail, examples of which are illustrated in the accompanying drawings. When the following description relates to the drawings, unless otherwise indicated, the same numerals in different drawings denote the same or similar elements. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with the present invention. Rather, they are merely examples of apparatuses and methods consistent with some aspects of the invention as detailed in the appended claims.

[0061] The terminology used in this disclosure is for the purpose of describing particular embodiments only and is not intended to be limiting of the present disclosure. The singular forms “a” and “the” as used in this disclosure and the appended claims are also intended to include the plural forms, unless the context clearly indicates otherwise.

[0062] It should be understood that although the terms first, second, third, etc., may be used to describe various information in embodiments of this disclosure, such information should not be limited to these terms. These terms are only used to distinguish information of the same type from one another. For example, first information may also be referred to as second information without departing from the scope of embodiments of this disclosure, and similarly, second information may also be referred to as first information. Depending on the context, the words “if” and “suppose” as used herein may be interpreted as “when”, “when”, or “in response to a determination”.

[0063] Furthermore, various forms of processes shown in the embodiments of this disclosure can be used to reorder, add, or delete steps. For example, the steps described in this application can be executed in parallel, sequentially, or in different orders, as long as the desired result of the technical solution disclosed in this disclosure can be achieved, and no limitation is imposed herein.

[0064] In related technologies, most methods for dimensionality reduction with controllability are based on analytical methods, which have some limitations in terms of universality and adaptability.

[0065] In some embodiments, regulation constraint aggregation methods typically rely on certain assumptions about the operation of various distributed resources. For example, when approximating regulation constraints for heterogeneous resources, researchers at Tsinghua University introduced the assumption that the regulation constraints of one resource are independent of other resources. However, when virtual power plants aggregate different types of resources participating in the energy and regulation service market, the energy-regulation constraints of these resources may be coupled, thus most regulation constraint aggregation methods cannot be directly applied. Secondly, typical aggregation regulation constraint dimensionality reduction methods are related to the form of each regulation constraint. For example, researchers at Harvard University used box constraints to approximate the original regulation constraints, but these constraint forms may not be consistent with the constraint forms specified by the upper-level power grid. Therefore, the adaptability of analytical aggregation regulation constraint dimensionality reduction methods is insufficient.

[0066] Some studies have attempted to use data-driven methods to reduce the dimensionality of regulatory constraints. Researchers at Virginia Tech used the original resource model to label randomly generated operational data as feasible or infeasible, and then used this data to train a convex quadratic classifier to approximate the aggregation of regulatory constraints. However, considering the temporal coupling nature of resource operations, the required amount of training data increases exponentially with the time period, and the trained classifier needs to be converted into the required bidding model form, thus introducing additional errors. Therefore, existing data-driven methods are far from practical.

[0067] In summary, data-driven methods are needed to address the universality and adaptability issues of analytical control constraint aggregation methods, while also requiring the selection of appropriate approximation models and training methods for implementation. This disclosure proposes a data-driven inverse optimization-based method for reducing the dimensionality of massive resource energy-frequency modulation aggregation control constraints. It uses the original massive resource operation model to generate optimal scheduling results under different scenarios, treats these scheduling results as the optimal solution of the dimensionality-reduced low-dimensional control constraint model, and then uses them as input to the inverse optimization problem to fit the parameters of the approximate low-dimensional control constraint model.

[0068] 1) Distributed resource regulation and constraint modeling technology: The output of distributed photovoltaic and wind power is limited by the maximum available power generation. Due to thermal inertia, the temperature of temperature-controlled loads can be analogized to the charging state in energy storage, thus establishing a battery model for temperature-controlled loads. The modeling of industrial production processes is based on a linearized state-task network model, including power constraints, material buffer constraints, initial values, and production targets for the industrial production process.

[0069] 2) Constrained Dimensionality Reduction Techniques: Constrained dimensionality reduction techniques are methods to reduce computational complexity when processing high-dimensional data. Traditional dimensionality reduction techniques such as principal component analysis and linear discriminant analysis mainly focus on the global structure of the data, but may ignore the local characteristics of the data. Recent research, such as local linear embedding and isometric mapping, better captures the inherent geometric structure of the data by preserving its local structure.

[0070] 3) Virtual Power Plant Control and Optimization Technology: Control and optimization technology mainly focuses on how to effectively dispatch distributed energy resources to meet the needs of the power grid. Traditional optimization methods, such as mixed-integer linear programming and mixed-integer nonlinear programming, may encounter computational inefficiencies when dealing with large-scale problems. With the development of artificial intelligence and machine learning technologies, some new optimization methods, such as reinforcement learning and deep learning, have begun to be applied to virtual power plant control and optimization.

[0071] The purpose of this disclosure is to address the problem of dimensionality reduction of high-dimensional control constraints for massive distributed resources when virtual power plants need to submit their overall control constraints to higher-level control agencies. The original high-dimensional control constraints of massive distributed resources are embedded into the upper-level control optimization model problem to generate optimal control results under different control scenarios, forming a control simulation dataset. The optimal control results in the control simulation dataset are considered as the running results based on the dimensionality-reduced approximate low-dimensional control constraint model, thereby fitting the parameters of the approximate low-dimensional control constraint model. The low-dimensional control constraint model fitted based on the simulation data is then embedded into the upper-level control optimization model to achieve dimensionality reduction of control constraints. This approach can approximate the original high-dimensional constraints of massive distributed resources using a low-dimensional control constraint model, significantly reducing the computational burden and solution time for the coordinated control of massive distributed resources. Furthermore, the data-driven approach offers better versatility and adaptability than traditional constraint dimensionality reduction methods, contributing to better development of the flexibility of massive resources.

[0072] Figure 1 This is a flowchart illustrating a data-driven, massive resource regulation and constraint dimensionality reduction method according to an exemplary embodiment, such as... Figure 1 As shown, it should be noted that the data-driven massive resource regulation and constraint dimensionality reduction method of this disclosure embodiment is applied in a data-driven massive resource regulation and constraint dimensionality reduction device. For example... Figure 1 As shown, the method may include the following steps:

[0073] S1: Construct a high-dimensional control constraint model based on the operating parameters of distributed resources.

[0074] In this embodiment of the disclosure, the original operating parameters of each distributed resource are obtained, including the upper limit of the output of distributed generation resources, the power and energy capacity of energy storage resources, the adjustment range of adjustable loads, and energy demand.

[0075] Among them, based on the operating parameters of each distributed resource, a high-dimensional regulation and constraint model (or refined regulation and constraint model) of massive distributed resources is established, including power constraints, energy constraints, energy demand constraints, and capacity constraints involved in regulation.

[0076] S2: Embed the high-dimensional regulation constraint model into the upper-level regulation optimization model, and obtain the regulation simulation dataset based on the historical operation data of distributed resources participating in the upper-level regulation.

[0077] In some embodiments, S2: embedding the high-dimensional regulation constraint model into the upper-level regulation optimization model, and obtaining the regulation simulation dataset based on the historical operation data of distributed resources participating in the upper-level regulation, including: generating multiple typical regulation scenario parameters based on the historical operation data of distributed resources participating in the upper-level regulation; embedding the high-dimensional regulation constraint model into the upper-level regulation optimization model, generating the optimal regulation results under different typical regulation scenarios according to the multiple typical regulation scenario parameters, and obtaining the regulation simulation dataset.

[0078] In this embodiment of the disclosure, based on the historical operation data of massive distributed resources participating in upper-level regulation, parameters for multiple typical regulation scenarios are generated; the original high-dimensional regulation constraints of massive resources are embedded into the upper-level regulation optimization model problem to generate the optimal regulation results under different typical regulation scenarios, including the optimal energy consumption arrangement and frequency regulation capacity, thus forming a regulation simulation dataset.

[0079] Based on historical operational data involving massive resources in upper-level regulation, parameters for multiple typical regulation scenarios are generated. Theoretically, to obtain the scheduling results after submitting the original feasible domain, it is necessary to model all units and loads within the scope of the power grid operator. However, in reality, the capacity of resource clusters is very small, so they can be modeled as price takers. Therefore, under the assumption of perfect competition, the scheduling scenario can be represented by an energy frequency regulation price scenario.

[0080] In this embodiment, the original high-dimensional control constraints of massive resources are embedded into the upper-level control optimization model problem to generate the optimal control results under typical control scenarios, including the optimal energy consumption arrangement and frequency regulation capacity, thus forming a control simulation dataset.

[0081] In form, the original feasible domain for the regulation of massive resources, i.e. the decision space where the regulation variables are constrained, includes the original feasible domain P for the regulation of distributed resources in some embodiments of this disclosure.

[0082] in,

[0083] Here, P consists of a set of linear constraints about (p, r), and the variable with the subscript [T] is a T-dimensional vector, p [T] =(p1,p2,...,p T ), r [T] =(r1,r2,...,r T ), p T For the distributed resource's basic energy output in hour T, r T For the frequency regulation capability of distributed resources in hour T, y [I][T] For the I-th distributed resource, h(·) includes the power and energy constraints of the distributed resource, θ ILet T be the parameter of the i-th distributed resource, where I and T are both positive integers.

[0084] Among them, the high-dimensional regulation constraint model is embedded in the upper-level regulation optimization model, and the regulation optimization is expressed as:

[0085]

[0086] st(p [T] ,r [T] )∈P;

[0087] in, The energy price for distributed resources, For the equivalent frequency regulation capacity price of distributed resources, s perf For performance scores, For frequency modulation capacity price, Price based on FM mileage. Let be the expected frequency modulation mileage at time t. The cost function for distributed resources;

[0088] For a given different energy frequency modulation price By calling a commercial solver to solve the above regulation optimization, the optimal regulation result (p) is obtained. [T] ,r [T] ), thus obtaining the regulation simulation dataset. Where k is the total number of indices.

[0089] S3: Construct a low-dimensional control constraint model based on the control simulation dataset.

[0090] In some embodiments, S3: Constructing a low-dimensional control constraint model based on the control simulation dataset includes: using the optimal control result as the running result of the low-dimensional control constraint model; fitting the parameters of the low-dimensional control constraint model based on the prediction results obtained by inputting parameters of multiple typical control scenarios into the low-dimensional control constraint model and the optimal control result as the running result, and determining the low-dimensional control constraint model.

[0091] In this embodiment of the disclosure, a low-dimensional regulation constraint model is designed, and an inverse optimization problem is constructed based on the regulation results to deduce the model parameters. As the default approximate low-dimensional regulation constraint model, a low-order virtual energy storage model can be used to approximate the original regulation constraints, including the upper and lower limits of energy storage charging and discharging power, the upper and lower limits of energy storage energy, and energy storage energy conversion constraints.

[0092] Among them, a low-dimensional regulation constraint model is designed as the default approximate regulation constraint model. A low-order virtual energy storage model can be used to approximate the original regulation constraints, including the upper and lower limits of energy storage charging and discharging power, the upper and lower limits of energy storage energy, energy storage energy conversion constraints, and initial energy.

[0093] Based on the virtual energy storage model, resource i provides frequency regulation constraints including non-negative capacity, power capacity limit, and sustainment time requirement, where Δt req This represents the time required for the resource to maintain its maximum frequency modulation output, for example, 15 minutes.

[0094] S4: Embed the low-dimensional regulation constraint model into the upper-level regulation optimization model to implement regulation constraints.

[0095] In some embodiments of this disclosure, the low-dimensional control constraint model includes upper and lower limits of energy storage charging and discharging power, upper and lower limits of energy storage energy, energy storage energy conversion constraints, and initial energy, expressed as follows:

[0096]

[0097]

[0098] Where, p t,i For the output at time interval t, r t,i e is the frequency modulation capacity for time interval t. t,i Let be the energy at the end of t. The underlined parameter indicates the lower limit of the corresponding variable, and the overlined parameter indicates the upper limit. The superscript 'dis' indicates discharging into the grid, and the superscript 'ch' indicates charging from the grid. 0,i The initial energy is given by η, the charge / discharge efficiency is given by μ, and λ is the Lagrange multiplier for the corresponding constraint, respectively.

[0099] Based on the low-dimensional regulation constraint model, the constraints provided by distributed resource i for frequency regulation include non-negative capacity constraints, power capacity constraints, and sustaining time requirements, expressed as:

[0100]

[0101] Where, Δt req The time required to maintain maximum frequency modulation output for distributed resources;

[0102] The sum of bids for distributed resources is represented as:

[0103] The optimal control result in the control simulation dataset D is used as the optimal solution of the low-dimensional control constraint model. The parameters of the low-dimensional control constraint model are fitted using the control simulation dataset D, and the optimization objective is to minimize the distance to the optimal control result.

[0104]

[0105] In this context, the superscript 'a' represents the parameters of the low-dimensional control constraint model.

[0106] This involves constructing an inverse optimization problem that uses the control results to deduce the parameters of a low-dimensional control constraint model. The approach is to treat the scheduling results generated in dataset D as the optimal solution to an optimization problem defined by an approximate model (with some noise added), and then use D to fit the problem parameters. This can be represented as a data-driven inverse optimization problem, with constraints including strong duality conditions, feasibility of the primal problem, feasibility of the dual problem, etc., and the optimization objective being to minimize the distance to the optimal control result.

[0107] The optimal control result in the control simulation dataset is regarded as the result of the operation based on the dimensionality-reduced approximate low-dimensional control constraint model (also known as the approximate control constraint model), thereby fitting the parameters of the approximate low-dimensional control constraint model; the low-dimensional control constraint model obtained by fitting the simulation data is embedded into the upper-level control optimization model to achieve dimensionality reduction of control constraints.

[0108] In this embodiment, the optimal control result in the control simulation dataset is regarded as the result of the operation based on the approximate low-dimensional control constraint model after dimensionality reduction, thereby fitting the parameters of the approximate low-dimensional control constraint model. Among them, the optimization problem is large in scale and there are nonlinear terms in the strong duality condition, which are difficult to solve. Therefore, an iterative algorithm is used to solve it. Each time, only one typical scenario of control simulation data is randomly introduced, and the learning rate is designed to update the parameters until the model parameters converge. The resulting low-dimensional control constraint model can be used to approximate the original high-dimensional control constraint model.

[0109] In some embodiments of this disclosure, the above method further includes: taking the optimal control result in the control simulation dataset D as the optimal solution of the low-dimensional control constraint model, solving it using an iterative algorithm, randomly introducing only one typical control scenario control simulation data each time, designing a learning rate to update parameters, until the parameters of the low-dimensional control constraint model converge, and obtaining the low-dimensional control constraint model.

[0110] By embedding the low-dimensional control constraint model into the upper-level control optimization model for control constraints, the scale of resource control constraints after dimensionality reduction is much smaller than the original scale, i.e., I a ≤I; The feasible region of the control constraint is:

[0111]

[0112] For example, such as Figure 2As shown in this embodiment, a data-driven method for reducing the dimensionality of massive resource regulation constraints is provided. First, the original operating parameters of each distributed resource are obtained, and a refined regulation constraint model is established (see the high-dimensional regulation constraint model in the above method). Then, regulation simulation results under different typical scenarios are generated based on the high-dimensional regulation constraints. After that, a low-dimensional regulation constraint model is designed (see the low-dimensional regulation constraint model in the above method), and an inverse optimization problem of the derived parameters is constructed. The parameters of the low-dimensional regulation constraint model are fitted based on the regulation simulation results. Finally, the obtained low-dimensional regulation constraints are embedded into the upper-level regulation optimization model.

[0113] This disclosure, based on distributed resource regulation constraint modeling and massive constraint dimensionality reduction techniques in virtual power plant control technology, establishes a data-driven method for massive resource regulation constraint dimensionality reduction. Compared to methods in related technologies, this method utilizes a data-driven approach to address the universality and adaptability issues of analytical regulation constraint aggregation methods, while also requiring the selection of appropriate approximation models and training methods for implementation. Applying this disclosure, a low-dimensional regulation constraint model can approximate the high-dimensional constraints of the original massive distributed resources, thereby significantly reducing the computational burden and solution time for optimizing and controlling massive distributed resources. Therefore, it has significant practical implications and promising application prospects.

[0114] This embodiment of the disclosure constructs a high-dimensional control constraint model based on the operating parameters of distributed resources; embeds the high-dimensional control constraint model into an upper-level control optimization model; obtains a control simulation dataset based on historical operating data of distributed resources participating in upper-level control; constructs a low-dimensional control constraint model based on the control simulation dataset; and embeds the low-dimensional control constraint model into the upper-level control optimization model for control constraints. Therefore, the high-dimensional constraints of the original massive distributed resources can be approximated using the low-dimensional control constraint model, greatly reducing the computational burden and solution time of massive resource collaborative control. Furthermore, the data-driven approach offers better versatility and adaptability than traditional constraint dimensionality reduction methods, contributing to better development of the flexibility of massive resources.

[0115] Figure 3 This is a structural diagram illustrating a data-driven, massive resource regulation and constraint dimensionality reduction device 1 according to an exemplary embodiment. (Refer to...) Figure 3 The device 1 includes a first model determination unit 11, a dataset acquisition unit 12, a second model determination unit 13, and a processing unit 14.

[0116] The first model determination unit 11 is used to construct a high-dimensional control constraint model based on the operating parameters of distributed resources.

[0117] The dataset acquisition unit 12 is used to embed the high-dimensional regulation constraint model into the upper-level regulation optimization model and acquire the regulation simulation dataset based on the historical operation data of the distributed resources participating in the upper-level regulation.

[0118] The second model determination unit 13 is used to construct a low-dimensional control constraint model based on the control simulation dataset.

[0119] Processing unit 14 is used to embed the low-dimensional regulation constraint model into the upper-level regulation optimization model for regulation constraint.

[0120] In some embodiments, the dataset acquisition unit 12 is specifically used to: generate parameters for multiple typical control scenarios based on historical operating data of distributed resources participating in upper-level control; embed a high-dimensional control constraint model into the upper-level control optimization model; generate optimal control results under different typical control scenarios based on the parameters of multiple typical control scenarios; and obtain a control simulation dataset.

[0121] In some embodiments, the second model determining unit 13 is specifically used to: take the optimal control result as the running result of the low-dimensional control constraint model; and, based on the prediction results obtained by inputting parameters of multiple typical control scenarios into the low-dimensional control constraint model and the optimal control result as the running result, fit the parameters of the low-dimensional control constraint model to determine the low-dimensional control constraint model.

[0122] In some embodiments of this disclosure, the high-dimensional regulation constraint model includes the original regulatory feasible region P of the distributed resource; wherein,

[0123] Here, P consists of a set of linear constraints about (p, r), and the variable with the subscript [T] is a T-dimensional vector, p [T] =(p1,p2,...,p T ), r [T] =(r1,r2,...,r T ), p T For the distributed resource's basic energy output in hour T, r T For the frequency regulation capability of distributed resources in hour T, y [I][T] For the I-th distributed resource, h(·) includes the power and energy constraints of the distributed resource, θ I Let T be the parameter of the i-th distributed resource, where I and T are both positive integers;

[0124] Among them, the high-dimensional regulation constraint model is embedded in the upper-level regulation optimization model, and the regulation optimization is expressed as:

[0125]

[0126] st(p [T] ,r[T] )∈P;

[0127] in, The energy price for distributed resources, For the equivalent frequency regulation capacity price of distributed resources, s perf For performance scores, For frequency modulation capacity price, Price based on FM mileage. Let be the expected frequency modulation mileage at time t. The cost function for distributed resources;

[0128] For a given different energy frequency modulation price By calling a commercial solver to solve the above regulation optimization, the optimal regulation result (p) is obtained. [T] ,r [T] ), thus obtaining the regulation simulation dataset. Where k is the total number of indices.

[0129] In some embodiments of this disclosure, the low-dimensional control constraint model includes upper and lower limits of energy storage charging and discharging power, upper and lower limits of energy storage energy, energy storage energy conversion constraints, and initial energy, expressed as follows:

[0130]

[0131] Where, p t,i For the output at time interval t, r t,i e is the frequency modulation capacity for time interval t. t,i Let be the energy at the end of t. The underlined parameter indicates the lower limit of the corresponding variable, and the overlined parameter indicates the upper limit. The superscript 'dis' indicates discharging into the grid, and the superscript 'ch' indicates charging from the grid. 0,i The initial energy is given by η, the charge / discharge efficiency is given by μ, and λ is the Lagrange multiplier for the corresponding constraint, respectively.

[0132] Based on the low-dimensional regulation constraint model, the constraints provided by distributed resource i for frequency regulation include non-negative capacity constraints, power capacity constraints, and sustaining time requirements, expressed as:

[0133]

[0134] Where, Δt req The time required to maintain maximum frequency modulation output for distributed resources;

[0135] The sum of bids for distributed resources is represented as:

[0136] The optimal control result in the control simulation dataset D is used as the optimal solution of the low-dimensional control constraint model. The parameters of the low-dimensional control constraint model are fitted using the control simulation dataset D, and the optimization objective is to minimize the distance to the optimal control result.

[0137]

[0138] In this context, the superscript 'a' represents the parameters of the low-dimensional control constraint model.

[0139] In some embodiments of this disclosure, the optimal control result in the control simulation dataset D is used as the optimal solution of the low-dimensional control constraint model. An iterative algorithm is used to solve the model. Each time, only one typical control scenario control simulation data is randomly introduced. The learning rate is designed to update the parameters until the parameters of the low-dimensional control constraint model converge, thus obtaining the low-dimensional control constraint model.

[0140] By embedding the low-dimensional control constraint model into the upper-level control optimization model for control constraints, the scale of resource control constraints after dimensionality reduction is much smaller than the original scale, i.e., I a ≤I; The feasible region of the control constraint is:

[0141]

[0142] By implementing the embodiments of this disclosure, a data-driven massive resource regulation constraint dimensionality reduction device includes a first model determination unit for constructing a high-dimensional regulation constraint model based on the operating parameters of distributed resources; a dataset acquisition unit for embedding the high-dimensional regulation constraint model into an upper-level regulation optimization model and acquiring a regulation simulation dataset based on historical operating data of distributed resources participating in upper-level regulation; a second model determination unit for constructing a low-dimensional regulation constraint model based on the regulation simulation dataset; and a processing unit for embedding the low-dimensional regulation constraint model into the upper-level regulation optimization model to perform regulation constraints. This device can approximate the high-dimensional constraints of the original massive distributed resources using a low-dimensional regulation constraint model, significantly reducing the computational burden and solution time for massive resource collaborative regulation. Furthermore, by adopting a data-driven technical approach, it has better versatility and adaptability than traditional constraint dimensionality reduction methods, contributing to better development of the flexibility of massive resources.

[0143] Regarding the apparatus in the above embodiments, the specific manner in which each module performs its operation has been described in detail in the embodiments related to the method, and will not be elaborated upon here.

[0144] Figure 4This is a block diagram illustrating an apparatus 700 for performing a data-driven, massive resource regulation constraint dimensionality reduction method according to an exemplary embodiment. For example, apparatus 700 may be an electronic device, such as a mobile phone, computer, digital broadcasting terminal, messaging device, game console, tablet device, medical device, fitness equipment, personal digital assistant, etc.

[0145] Reference Figure 4 The device 700 may include one or more of the following components: a processing component 702, a memory 704, a power component 706, a multimedia component 708, an audio component 710, an input / output (I / O) interface 712, a sensor component 714, and a communication component 716.

[0146] Processing component 702 typically controls the overall operation of device 700, such as operations associated with display, telephone calls, data communication, camera operation, and recording. Processing component 702 may include one or more processors 720 to execute instructions to complete all or part of the steps of the methods described above. Furthermore, processing component 702 may include one or more modules to facilitate interaction between processing component 702 and other components. For example, processing component 702 may include a multimedia module to facilitate interaction between multimedia component 708 and processing component 702.

[0147] Memory 704 is configured to store various types of data to support the operation of device 700. Examples of this data include instructions for any application or method operating on device 700, contact data, phonebook data, messages, pictures, videos, etc. Memory 704 can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic storage, flash memory, magnetic disk, or optical disk.

[0148] The power supply component 706 provides power to the various components of the device 700. The power supply component 706 may include a power management system, one or more power sources, and other components associated with generating, managing, and distributing power to the device 700.

[0149] Multimedia component 708 includes a screen that provides an output interface between the device 700 and the user. In some embodiments, the screen may include a liquid crystal display (LCD) and a touch panel (TP). If the screen includes a touch panel, the screen may be implemented as a touchscreen to receive input signals from the user. The touch panel includes one or more touch sensors to sense touches, swipes, and gestures on the touch panel. The touch sensors may sense not only the boundaries of the touch or swipe action but also the duration and pressure associated with the touch or swipe operation. In some embodiments, multimedia component 708 includes a front-facing camera and / or a rear-facing camera. When the device 700 is in an operating mode, such as a shooting mode or a video mode, the front-facing camera and / or the rear-facing camera may receive external multimedia data. Each front-facing camera and rear-facing camera may be a fixed optical lens system or have focal length and optical zoom capabilities.

[0150] Audio component 710 is configured to output and / or input audio signals. For example, audio component 710 includes a microphone (MIC) configured to receive external audio signals when device 700 is in an operating mode, such as call mode, recording mode, and voice recognition mode. The received audio signals may be further stored in memory 704 or transmitted via communication component 716. In some embodiments, audio component 710 also includes a speaker for outputting audio signals.

[0151] I / O interface 712 provides an interface between processing component 702 and peripheral interface modules, such as keyboards, click wheels, buttons, etc. These buttons may include, but are not limited to, home buttons, volume buttons, power buttons, and lock buttons.

[0152] Sensor assembly 714 includes one or more sensors for providing status assessments of various aspects of device 700. For example, sensor assembly 714 may detect the on / off state of device 700, the relative positioning of components such as the display and keypad of device 700, changes in the position of device 700 or a component of device 700, the presence or absence of user contact with device 700, the orientation or acceleration / deceleration of device 700, and temperature changes of device 700. Sensor assembly 714 may include a proximity sensor configured to detect the presence of nearby objects without any physical contact. Sensor assembly 714 may also include a light sensor, such as a CMOS or CCD image sensor, for use in imaging applications. In some embodiments, sensor assembly 714 may also include an accelerometer, a gyroscope, a magnetometer, a pressure sensor, or a temperature sensor.

[0153] Communication component 716 is configured to facilitate wired or wireless communication between device 700 and other devices. Device 700 can access wireless networks based on communication standards, such as WiFi, 2G, or 3G, or combinations thereof. In one exemplary embodiment, communication component 716 receives broadcast signals or broadcast-related information from an external broadcast management system via a broadcast channel. In one exemplary embodiment, communication component 716 also includes a near-field communication (NFC) module to facilitate short-range communication. For example, the NFC module may be implemented based on radio frequency identification (RFID) technology, Infrared Data Association (IrDA) technology, ultra-wideband (UWB) technology, Bluetooth (BT) technology, and other technologies.

[0154] In an exemplary embodiment, the apparatus 700 may be implemented by one or more application-specific integrated circuits (ASICs), digital signal processors (DSPs), digital signal processing devices (DSPDs), programmable logic devices (PLDs), field-programmable gate arrays (FPGAs), controllers, microcontrollers, microprocessors, or other electronic components to perform the methods described above.

[0155] In an exemplary embodiment, a non-transitory computer-readable storage medium including instructions is also provided, such as a memory 704 including instructions, which can be executed by a processor 720 of the device 700 to perform the above-described method. For example, the non-transitory computer-readable storage medium may be a ROM, random access memory (RAM), CD-ROM, magnetic tape, floppy disk, and optical data storage device, etc.

[0156] In an exemplary embodiment, a computer program product is also provided, including a computer program that implements the above-described method when executed by the processor 720 of the device 700.

[0157] Other embodiments of the invention will readily occur to those skilled in the art upon consideration of the specification and practice of the invention disclosed herein. This disclosure is intended to cover any variations, uses, or adaptations of the invention that follow the general principles of the invention and include common knowledge or customary techniques in the art not disclosed herein. The specification and examples are to be considered exemplary only, and the true scope and spirit of the invention are indicated by the following claims.

[0158] It should be understood that the present invention is not limited to the precise structure described above and shown in the accompanying drawings, and various modifications and changes can be made without departing from its scope. The scope of the invention is limited only by the appended claims.

Claims

1. A data-driven mass resource regulation constraint dimension reduction method, characterized in that, The method comprises the following steps: constructing a high-dimensional regulation constraint model based on operation parameters of distributed resources; embedding the high-dimensional regulation constraint model into an upper-layer regulation optimization model, and obtaining a regulation simulation data set based on historical operation data of the distributed resources participating in upper-layer regulation; constructing a low-dimensional regulation constraint model according to the regulation simulation data set, wherein, based on the historical operation data of the distributed resources participating in upper-layer regulation, a plurality of typical regulation scenario parameters are generated, the high-dimensional regulation constraint model is embedded into the upper-layer regulation optimization model, optimal regulation results in different typical regulation scenarios are generated according to the plurality of typical regulation scenario parameters, and the regulation simulation data set is obtained; the optimal regulation results are taken as operation results of the low-dimensional regulation constraint model; parameters of the low-dimensional regulation constraint model are fitted based on predicted results obtained by inputting the plurality of typical regulation scenario parameters into the low-dimensional regulation constraint model and the optimal regulation results as the operation results, and the low-dimensional regulation constraint model is determined; embedding the low-dimensional regulation constraint model into the upper-layer regulation optimization model to perform regulation constraint.

2. The method of claim 1, wherein, The high-dimensional regulation constraint model includes a raw regulation feasible region of distributed resources P ; wherein, ； in, P From a set of about (p, r) The linear constraints are composed of variables whose subscripts [T] are T-dimensional vectors. , , For the distributed resource number T Basic energy output per hour For the distributed resource number T Hourly frequency modulation capability, For the first I The runtime variables of a distributed resource h (·) includes the power and energy constraints of the distributed resources. For the parameters of the i-th distributed resource, I,T All are positive integers; The high-dimensional regulation constraint model is embedded into the upper-layer regulation optimization model, and regulation optimization is represented as: ； ； wherein, is the energy price for the distributed resource, is the equivalent frequency regulation capacity price for the distributed resource, is the performance score, is the frequency regulation capacity price, is the frequency regulation mileage price, is the expected frequency regulation mileage at time t, is the cost function for the distributed resource; For a given different energy frequency price (Pf) , , the optimal control result is obtained by calling a commercial solver to solve the above control optimization （ ） , and the control simulation data set is obtained , wherein, k is the total number of indexes.

3. The method of claim 2, wherein, The low-dimensional regulation constraint model comprises upper and lower limits of energy storage charging and discharging power, upper and lower limits of energy storage energy, energy storage energy conversion constraints and initial energy, and is represented as: ； ； ； ； = : t=1; wherein is the output of the time interval t , is the frequency regulation capacity of the time interval t , is the energy at the end of the time interval t , the underlined parameters represent the lower limit of the respective variable, the overlined parameters represent the upper limit of the respective variable, the superscript dis indicates discharging to the grid, the superscript ch indicates charging from the grid, is the initial energy, is the charge / discharge efficiency, and are the Lagrange multipliers of the respective constraints, respectively; Based on the low-dimensional regulation constraint model, the constraint of frequency regulation provided by the distributed resource i comprises non-negative capacity constraint, power capacity constraint and maintenance time requirement, and is represented as: ； ； ； ； ； wherein, is the time required to maintain the maximum frequency modulation output for the distributed resource; wherein the sum of the bids of the distributed resources is denoted as: ； The optimal regulation result in the regulation simulation dataset is taken as the optimal solution of the low-dimensional regulation constraint model, and the parameters of the low-dimensional regulation constraint model are fitted using the regulation simulation dataset , and the optimization target is the minimum distance of the optimal regulation result: ; wherein, the superscript a represents parameters of the low-dimensional regulation constraint model.

4. The method of claim 3, wherein, The method further comprises: The optimal regulation result in the regulation simulation data set The optimal regulation result in the regulation simulation data set The optimal regulation result in the regulation simulation data set The optimal regulation result in the regulation simulation data set The optimal regulation result in the regulation simulation data set The optimal regulation result in the regulation simulation data set The optimal regulation result in the regulation simulation data set The optimal regulation result in the regulation simulation data set The optimal regulation result in the regulation simulation data set The optimal regulation result in the regulation simulation data set The optimal regulation result in the regulation simulation data set The optimal regulation result in the regulation simulation data set < The low-dimensional regulation constraint model is embedded into the upper regulation optimization model to perform regulation constraint, and the scale of the resource regulation constraint after dimension reduction is far smaller than the original scale, that is ; and the feasible region of the regulation constraint is 。 5. A data-driven mass resource regulation constraint dimension reduction apparatus, characterized in that, The method comprises the following steps: a first model determination unit configured to construct a high-dimensional regulation constraint model based on operation parameters of distributed resources; a data set acquisition unit configured to embed the high-dimensional regulation constraint model into an upper-layer regulation optimization model, and obtain a regulation simulation data set based on historical operation data of the distributed resources participating in upper-layer regulation; a second model determination unit configured to construct a low-dimensional regulation constraint model according to the regulation simulation data set, wherein, based on the historical operation data of the distributed resources participating in upper-layer regulation, a plurality of typical regulation scenario parameters are generated, the high-dimensional regulation constraint model is embedded into the upper-layer regulation optimization model, optimal regulation results in different typical regulation scenarios are generated according to the plurality of typical regulation scenario parameters, and the regulation simulation data set is obtained; the optimal regulation results are taken as operation results of the low-dimensional regulation constraint model; parameters of the low-dimensional regulation constraint model are fitted based on predicted results obtained by inputting the plurality of typical regulation scenario parameters into the low-dimensional regulation constraint model and the optimal regulation results as the operation results, and the low-dimensional regulation constraint model is determined; a processing unit configured to embed the low-dimensional regulation constraint model into the upper-layer regulation optimization model to perform regulation constraint.

6. An electronic device, comprising: The method comprises the following steps: a memory, a processor and a computer program stored in the memory and executable on the processor, wherein the processor implements the method according to any one of claims 1 to 4 when executing the computer program.

7. A computer-readable storage medium having stored thereon a computer program, characterized in that The computer program, which when executed by a processor, implements the method of any one of claims 1 to 4.

8. A computer program product comprising a computer program, characterized in that, The computer program, which when executed by a processor, implements the method of any one of claims 1 to 4.

Images