A method and apparatus for controlling power flexibility resources

Abstract

The application provides a power flexibility resource control method and device, and belongs to the technical field of power utilization. After power flexibility resource data is acquired, the power flexibility resource data is preprocessed to obtain preprocessed power flexibility resource data. The preprocessed power flexibility resource data is input into a prediction function model, so that the prediction function model outputs optimal control output parameters of the power flexibility resource data. Through the function prediction model, flexible control and allocation of the power flexibility resource are realized, the controllability of the power flexibility resource is improved, and power loss and power generation cost can be effectively reduced.

Description

Technical Field

[0001] This invention belongs to the field of power application technology, and in particular relates to a method and device for controlling power flexibility resources. Background Technology

[0002] In traditional power systems, power sources are mainly thermal power units and hydropower units. However, with the continuous development of large-scale wind and solar renewable energy power generation and distributed power sources, as well as the increase in the proportion of electricity consumption in the tertiary industry and by residents, the proportion of power sources with higher uncertainty, namely power flexibility resources, in the power structure has increased significantly.

[0003] Existing technologies have strong load tracking and regulation capabilities for power sources in traditional power systems, but their controllability over power flexibility resources is very low.

[0004] Therefore, in order to cope with the increasing uncertainty, improving the controllability of power flexibility resources has become a technical problem that urgently needs to be solved by those skilled in the art. Summary of the Invention

[0005] In view of the above problems, this application proposes a method and apparatus for controlling power flexibility resources. To improve the controllability of power flexibility resources, the specific solution is as follows:

[0006] A method for controlling power flexibility resources, comprising:

[0007] Obtain power flexibility resource data;

[0008] The power flexibility resource data is preprocessed to obtain preprocessed power flexibility resource data;

[0009] The preprocessed power flexibility resource data is input into the prediction function model so that the prediction function model outputs the optimal control output parameters of the power flexibility resource data.

[0010] Optionally, the training process of the prediction function model includes:

[0011] Determine the dataset of power flexibility resources for training;

[0012] The parameter set values ​​are obtained by training using the aforementioned training power flexibility resource dataset.

[0013] The values ​​of the parameter set are optimized to obtain the prediction function model.

[0014] Optionally, after optimizing the values ​​of the parameter set to obtain the prediction function model, the method further includes:

[0015] Determine the set of power flexibility resource data for testing;

[0016] The prediction function model is evaluated using the test power flexibility resource dataset to obtain the evaluation results.

[0017] Optionally, determining the training power flexibility resource dataset includes:

[0018] Identify the relevant factors affecting the flexibility of power resources for training, including power supply side, power infrastructure, power demand side, and power storage;

[0019] The relevant factors affecting the power flexibility resources for training are quantified to determine the data set of power flexibility resources for training, x0, x1, x2...x n x0 = 1, and is defined as a column vector.

[0020] Optionally, the step of training using the training power flexibility resource dataset to obtain parameter set values ​​includes:

[0021] Determine the parameter set θ0, θ1, θ2...θ corresponding to the training power flexibility resource data set. n ;

[0022] Based on the parameter set θ0, θ1, θ2...θ n Define a data set of power flexibility resources for training, x0, x1, x2...x n The prediction function f θ (x)=θ T ·x=θ0+θ1x1+θ2x2+…θnx n ;

[0023] The prediction function is trained using the training power flexibility resource dataset to obtain parameter set values.

[0024] Optionally, training the prediction function using the training power flexibility resource data set to obtain parameter set values ​​includes:

[0025] Based on the prediction function f θ (x)=θ T ·x=θ0+θ1x1+θ2x2+…θ n x n Determine the objective function Where v = f θ (x), where y are the corresponding optimal control output parameters;

[0026] By reducing the objective function The value determines the value of the parameter group.

[0027] A control device for power flexibility resources, comprising:

[0028] The acquisition unit is used to acquire power flexibility resource data;

[0029] The processing unit is used to preprocess the power flexibility resource data to obtain preprocessed power flexibility resource data.

[0030] The input unit is used to input the preprocessed power flexibility resource data into the prediction function model so that the prediction function model outputs the optimal control output parameters of the power flexibility resource data.

[0031] Optionally, the training process of the prediction function model includes:

[0032] The first determining subunit is used to determine the set of power flexibility resource data for training;

[0033] The training subunit is used to train using the training power flexibility resource dataset to obtain the values ​​of the parameter set.

[0034] An optimization subunit is used to optimize the values ​​of the parameter set to obtain a prediction function model.

[0035] Optionally, the device further includes:

[0036] The second determining subunit is used to determine the set of power flexibility resource data for testing;

[0037] The evaluation subunit is used to evaluate the prediction function model using the test power flexibility resource data set and obtain the evaluation result.

[0038] Optionally, the second determining subunit includes:

[0039] The third determining subunit is used to determine the relevant factors affecting the power flexibility resources for training. The relevant factors affecting the power flexibility resources for training include the power supply side, power infrastructure, power demand side, and power storage.

[0040] The quantization subunit is used to quantify the relevant factors affecting the power flexibility resources for training, and to determine the training power flexibility resource data set x0, x1, x2...x n x0 = 1, and is defined as a column vector.

[0041] Based on the above technical solution, the power flexibility resource control method and device provided by the present invention, after acquiring power flexibility resource data, preprocesses the power flexibility resource data to obtain preprocessed power flexibility resource data, and inputs the preprocessed power flexibility resource data into a prediction function model so that the prediction function model outputs the optimal control output parameters of the power flexibility resource data. Through the function prediction model, flexible control and allocation of power flexibility resources are realized, improving the controllability of power flexibility resources and effectively reducing power loss and power generation costs. Attached Figure Description

[0042] Various other advantages and benefits will become apparent to those skilled in the art upon reading the following detailed description of preferred embodiments. The accompanying drawings are for illustrative purposes only and are not intended to limit the scope of this application. Furthermore, the same reference numerals denote the same parts throughout the drawings. In the drawings:

[0043] Figure 1 A schematic diagram of the structure of a control system for power flexibility resources provided in an embodiment of the present invention;

[0044] Figure 2 A flowchart illustrating a method for controlling power flexibility resources provided in an embodiment of the present invention;

[0045] Figure 3 This is a flowchart illustrating a method for training a prediction function model disclosed in this invention.

[0046] Figure 4 This is a flowchart illustrating another method for training a prediction function model disclosed in this invention.

[0047] Figure 5 This is a flowchart illustrating a method for determining a training power flexibility resource dataset disclosed in this invention.

[0048] Figure 6 This is a flowchart illustrating a method for obtaining parameter set values ​​by training using a training power flexibility resource dataset, as disclosed in this invention.

[0049] Figure 7 This is a flowchart illustrating a method disclosed in this invention for training a prediction function using a training set of power flexibility resource data to obtain parameter set values.

[0050] Figure 8 This is a schematic diagram of the structure of a power flexibility resource control device disclosed in an embodiment of this application. Detailed Implementation

[0051] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0052] The terms "first," "second," etc., used in the specification, claims, and accompanying drawings of this invention are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such terms are interchangeable where appropriate; this is merely a way of distinguishing objects with the same attributes in the description of embodiments of the invention. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion, so that a process, method, system, product, or apparatus that comprises a series of units is not necessarily limited to those units, but may include other units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0053] To improve the controllability of power flexibility resources, this invention provides a method for controlling power flexibility resources. The method for controlling power flexibility resources provided by this invention will be further described in detail below with reference to the accompanying drawings and specific embodiments.

[0054] It should be noted that the control method for power flexibility resources provided by the present invention can be executed by the control system of power flexibility resources. For ease of understanding, the control system of power flexibility resources will be introduced first in this application.

[0055] Please see the appendix Figure 1 , Figure 1 This is a schematic diagram of a control system for power flexibility resources provided in an embodiment of the present invention. The control system for power flexibility resources includes a data acquisition module, a data preprocessing module connected to the data acquisition module, a swarm intelligence collaborative machine learning module connected to the data preprocessing module, and a model result output module connected to the swarm intelligence collaborative machine learning module.

[0056] In this application, a unified data storage module can be added between the data preprocessing module and the swarm intelligence collaborative machine learning module to uniformly store the data directly generated from the collection and the preprocessed receipts, and to support unified access by the swarm intelligence collaborative machine learning module.

[0057] It's important to note that swarm intelligence refers to the intelligent behavior exhibited by individuals within a group, through interaction, that far exceeds their individual capabilities. Ants foraging for food, bee colonies building nests, schools of fish avoiding predators, manufacturing, market economies, and even human civilization are all typical examples of swarm intelligence in nature and human society. By leveraging swarm intelligence technology, performance issues in cloud environments, such as high latency, network instability, and low bandwidth on the user end, can be addressed, providing end users with fast, stable, and scalable enhanced cloud services.

[0058] Based on the flow monitoring system provided in the above embodiments, the control method for power flexibility resources provided in the embodiments of the present invention is described as follows:

[0059] Please see the appendix Figure 2 , Figure 2 This is a flowchart illustrating a control method for power flexibility resources provided in an embodiment of the present invention. The method is applied to a control system for power flexibility resources and may include the following steps:

[0060] Step S101: Obtain power flexibility resource data.

[0061] In this step, the power flexibility resource data can be acquired by the data acquisition module.

[0062] In this application, power flexibility resource data can be obtained from the power resource terminal.

[0063] It should be noted that power flexibility resource data can be sensor data.

[0064] Step S102: Preprocess the power flexibility resource data to obtain preprocessed power flexibility resource data.

[0065] In this step, the data preprocessing module can preprocess the power flexibility resource data to obtain preprocessed power flexibility resource data.

[0066] In this application, missing values ​​are filled in for power flexibility resource data with missing values, and power flexibility resource data that is not standardized is standardized.

[0067] Step S103: Input the preprocessed power flexibility resource data into the prediction function model so that the prediction function model outputs the optimal control output parameters of the power flexibility resource data.

[0068] In this step, the swarm intelligence collaborative machine learning module establishes a prediction function, which is responsible for supervising the training of the prediction function model by inputting the preprocessed power flexibility resource data into the prediction function model. After completion, the processed power flexibility resource data is passed to the model result output module so that the prediction function model outputs the optimal control output parameters of the power flexibility resource data through the model result output module.

[0069] In summary, the power flexibility resource control method provided by this invention, after acquiring power flexibility resource data, preprocesses the data to obtain preprocessed power flexibility resource data, and inputs the preprocessed power flexibility resource data into a prediction function model so that the prediction function model outputs the optimal control output parameters of the power flexibility resource data. Through the function prediction model, flexible control and allocation of power flexibility resources are realized, improving the controllability of power flexibility resources and effectively reducing power loss and power generation costs.

[0070] Based on the embodiments disclosed in the present invention above, in another embodiment of the present invention, the specific implementation of the training process of the prediction function model is described in detail.

[0071] As one possible implementation method, please refer to the appendix. Figure 3 This is a flowchart illustrating a method for training a prediction function model disclosed in this invention. The method may include the following steps:

[0072] Step S201: Determine the set of power flexibility resource data for training.

[0073] In this step, the set of power flexibility resource data for training can be determined by the swarm intelligence collaborative machine learning module.

[0074] In this application, the K-fold method is used to divide the original power flexibility resource dataset into K equal parts, with the first part serving as the test set and the rest as the training set.

[0075] Step S202: Use the training power flexibility resource dataset to train and obtain the values ​​of the parameter set.

[0076] In this step, the swarm intelligence collaborative machine learning module can be trained using a training power flexibility resource dataset to obtain the values ​​of the parameter set.

[0077] In this application, the values ​​of unknown parameter sets are determined through a large amount of known training data. This will be explained in detail through the following embodiments and will not be elaborated here.

[0078] Step S203: Optimize the values ​​of the parameter set to obtain the prediction function model.

[0079] In this step, the values ​​of the parameter set can be optimized by the swarm intelligence collaborative machine learning module to obtain the prediction function model.

[0080] In this application, the values ​​of the parameter set can be optimized by verifying the data using a set of power flexibility resource data.

[0081] In summary, the method for training the prediction function model provided by this invention obtains the values ​​of the parameter array through the training set, optimizes the values ​​of the parameter array, constructs the prediction function model, and thereby realizes flexible control and allocation of power flexibility resources.

[0082] Based on the embodiments disclosed in the present invention above, in another embodiment of the present invention, the specific implementation method after optimizing the values ​​of the parameter set to obtain the prediction function model is described in detail.

[0083] As one possible implementation method, please refer to the appendix. Figure 4 This is a flowchart illustrating another method for training a prediction function model disclosed in this invention. The method may include the following steps:

[0084] Step S301: Determine the set of power flexibility resource data for testing.

[0085] In this step, the set of power flexibility resource data for testing can be determined by the swarm intelligence collaborative machine learning module.

[0086] In this application, the K-fold method is used to divide the original power flexibility resource dataset into K equal parts, with the first part serving as the test set and the rest as the training set.

[0087] Step S302: Evaluate the prediction function model using the test power flexibility resource dataset and obtain the evaluation results.

[0088] In this step, the swarm intelligence collaborative machine learning module can evaluate the prediction function model using a test power flexibility resource dataset to obtain the evaluation results.

[0089] In this application, the fit of the trained model to the test data can be checked through the test data. The smaller the error, the higher the accuracy and the better the model fits.

[0090] In summary, the method for training another prediction function model provided by this invention evaluates the fitting of the trained prediction function model based on the obtained training model, thus avoiding underfitting or overfitting.

[0091] Based on the embodiments disclosed in the present invention above, in another embodiment of the present invention, the specific implementation of step S201, determining the training power flexibility resource data set, is described in detail.

[0092] As one possible implementation method, please refer to the appendix. Figure 5 This is a flowchart illustrating a method for determining a training power flexibility resource dataset disclosed in this invention. The method may include the following steps:

[0093] Step S401: Identify the relevant factors affecting the power flexibility resources for training. The relevant factors affecting the power flexibility resources for training include the power supply side, power infrastructure, power demand side, and power storage.

[0094] It should be noted that the power supply side mainly includes different power source types such as thermal power and hydropower, and further subdivides them to analyze the parameters affecting power generation output; the power infrastructure includes multiple parameters that mainly involve equipment in different functional parts of the transmission network system; the power demand side includes multiple parameters that mainly involve different management strategies for different types of users; the power storage includes multiple parameters that mainly involve battery energy storage, pumped hydro storage, flywheel energy storage, compressed air energy storage, etc., among which pumped hydro storage is currently the main flexibility resource of the power system.

[0095] Step S402: Quantify the relevant factors affecting the power flexibility resources for training, and determine the data set of power flexibility resources for training x0, x1, x2...x n x0 = 1, and is defined as a column vector.

[0096] It should be noted that setting X0 to 1 has no actual physical meaning and is the default column vector.

[0097] In summary, the present invention provides a method for determining a training power flexibility resource dataset, which accurately identifies relevant factors of power flexibility resources and finds quantitative relationships between them, which is beneficial for determining the prediction function model.

[0098] Based on the embodiments disclosed in the present invention above, in another embodiment of the present invention, the specific implementation method of step S202, which uses a training power flexibility resource data set to train and obtain parameter group values, is described in detail.

[0099] As one possible implementation method, please refer to the appendix. Figure 6 This is a flowchart illustrating a method for obtaining parameter set values ​​by training using a training power flexibility resource dataset, as disclosed in this invention. The method may include the following steps:

[0100] Step S501: Determine the parameter set θ0, θ1, θ2...θ corresponding to the training power flexibility resource data set. n .

[0101] It should be noted that the parameter set θ(θ0, θ1, θ2...θ) n () refers to the different calculation standards used for different variables.

[0102] Step S502: Based on the parameter set θ0, θ1, θ2...θ n Define a dataset of power flexibility resources for training, x0, x1, x2...x n The prediction function f θ (x)=θ T ·x=θ0+θ1x1+θ2x2+…θ n x n .

[0103] It should be noted that the prediction function f θ (x)=θ T ·x=θ0+θ1x1+θ2x2+…θ n x n It is a linear equation with multiple variables in x, and its value represents the optimal control output parameter. This parameter is affected by all the explanatory variables in x.

[0104] Step S503: Train the prediction function using the training power flexibility resource dataset to obtain parameter set values.

[0105] In this application, the prediction function is trained using a large amount of known training data to determine the values ​​of unknown parameter sets. This will be explained in detail in the following embodiments and will not be elaborated here.

[0106] Based on the embodiments disclosed in the present invention above, in another embodiment of the present invention, the specific implementation method of step S503, which uses a training power flexibility resource data set to train the prediction function and obtain parameter set values, is described in detail.

[0107] As one possible implementation method, please refer to the appendix. Figure 7 This is a flowchart illustrating a method disclosed in this invention for training a prediction function using a training set of power flexibility resource data to obtain parameter set values. The method may include the following steps:

[0108] Step S601: Based on the prediction function f θ (x)=θ T ·x=θ0+θ1x1+θ2x2+…θ n x n Determine the objective function Where v = f θ (x), where y is the corresponding optimal control output parameter.

[0109] Step S602: By reducing the objective function The value determines the value of the parameter group.

[0110] In this application, based on the steepest descent method or stochastic gradient descent method, the value of the objective function E(θ) is continuously reduced to decrease f. θ (x) and the true optimal control output parameter y are used to determine the parameters θ0, θ1, θ2...θ n The value of .

[0111] Since there are multiple variables, partial derivative processing of the objective function is required to obtain the final θ. j These are the optimal control output parameters.

[0112]

[0113]

[0114]

[0115]

[0116]

[0117] The methods described in the above-disclosed embodiments of the present invention are detailed. The methods of the present invention can be implemented by various forms of devices. Therefore, the present invention also discloses a control device for power flexibility resources. Specific embodiments are given below for detailed description.

[0118] Please see the appendix Figure 8 , Figure 8 This is a schematic diagram of the structure of a control device for power flexibility resources disclosed in an embodiment of this application. The control system for power flexibility resources includes a data acquisition module, a data preprocessing module connected to the data acquisition module, a swarm intelligence collaborative machine learning module connected to the data preprocessing module, and a model result output module connected to the swarm intelligence collaborative machine learning module. The device may include:

[0119] Acquisition unit 11 is used to acquire power flexibility resource data;

[0120] Processing unit 12 is used to preprocess the power flexibility resource data to obtain preprocessed power flexibility resource data;

[0121] Input unit 13 is used to input the preprocessed power flexibility resource data into the prediction function model so that the prediction function model outputs the optimal control output parameters of the power flexibility resource data.

[0122] As one possible implementation, the training process of the prediction function model includes:

[0123] The first determining subunit is used to determine the set of power flexibility resource data for training;

[0124] The training subunit is used to train using the training power flexibility resource dataset to obtain the values ​​of the parameter set.

[0125] An optimization subunit is used to optimize the values ​​of the parameter set to obtain a prediction function model.

[0126] As one possible implementation, the device further includes:

[0127] The second determining subunit is used to determine the set of power flexibility resource data for testing;

[0128] The evaluation subunit is used to evaluate the prediction function model using the test power flexibility resource data set and obtain the evaluation result.

[0129] As one possible implementation, the second determining subunit includes:

[0130] The third determining subunit is used to determine the relevant factors affecting the power flexibility resources for training. The relevant factors affecting the power flexibility resources for training include the power supply side, power infrastructure, power demand side, and power storage.

[0131] The quantization subunit is used to quantify the relevant factors affecting the power flexibility resources for training, and to determine the training power flexibility resource data set x0, x1, x2...x n x0 = 1, and is defined as a column vector.

[0132] The various embodiments in this specification are described in a progressive manner, with each embodiment focusing on its differences from other embodiments. Similar or identical parts between embodiments can be referred to interchangeably. For the apparatus disclosed in the embodiments, since they correspond to the methods disclosed in the embodiments, the description is relatively simple; relevant parts can be referred to the method section.

[0133] It should also be noted that the device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate, and the components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs. Furthermore, in the accompanying drawings of the device embodiments provided by this invention, the connection relationships between modules indicate that they have communication connections, which can be specifically implemented as one or more communication buses or signal lines. Those skilled in the art can understand and implement this without any creative effort.

[0134] Through the above description of the embodiments, those skilled in the art can clearly understand that the present invention can be implemented by means of software plus necessary general-purpose hardware, and of course, it can also be implemented by special hardware including application-specific integrated circuits, special CPUs, special memory, special components, etc. Generally, any function performed by a computer program can be easily implemented by corresponding hardware, and the specific hardware structure used to implement the same function can also be diverse, such as analog circuits, digital circuits, or special-purpose circuits. However, for the present invention, software program implementation is more often the preferred implementation method. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product is stored in a readable storage medium, such as a computer floppy disk, USB flash drive, portable hard disk, read-only memory (ROM), random access memory (RAM), magnetic disk, or optical disk, etc., including several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute the methods described in the various embodiments of the present invention.

[0135] In summary, the above embodiments are only used to illustrate the technical solutions of the present invention, and are not intended to limit it. Although the present invention has been described in detail with reference to the above embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the above embodiments, or equivalent substitutions can be made to some of the technical features. These modifications or substitutions do not cause the essence of the corresponding technical solutions to depart from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for controlling power flexibility resources, characterized in that, include: Obtain power flexibility resource data; The power flexibility resource data is preprocessed to obtain preprocessed power flexibility resource data; The preprocessed power flexibility resource data is input into the prediction function model so that the prediction function model outputs the optimal control output parameters of the power flexibility resource data. The training process of the prediction function model includes: Determining the training power flexibility resource dataset includes: identifying relevant factors affecting the training power flexibility resources, including power supply-side, power infrastructure, power demand-side, and power storage; quantifying the relevant factors affecting the training power flexibility resources; and determining the training power flexibility resource dataset. And defined in the form of column vectors ; Training using the training power flexibility resource dataset to obtain parameter set values ​​includes: determining the parameter set corresponding to the training power flexibility resource dataset. Based on the parameter set Define a dataset of power flexibility resources for the training. Prediction function Based on the prediction function Determine the objective function ,in The corresponding optimal control output parameters are obtained by reducing the objective function. The value determines the parameter group value; The values ​​of the parameter set are optimized to obtain the prediction function model.

2. The method according to claim 1, characterized in that, After optimizing the values ​​of the parameter set to obtain the prediction function model, the method further includes: Determine the set of power flexibility resource data for testing; The prediction function model is evaluated using the test power flexibility resource dataset to obtain the evaluation results.

3. A control device for power flexibility resources, characterized in that, include: The acquisition unit is used to acquire power flexibility resource data; The processing unit is used to preprocess the power flexibility resource data to obtain preprocessed power flexibility resource data. The input unit is used to input the preprocessed power flexibility resource data into the prediction function model so that the prediction function model outputs the optimal control output parameters of the power flexibility resource data. The training process of the prediction function model includes: The first determining subunit is used to determine the training power flexibility resource data set, including: determining the relevant factors affecting the training power flexibility resources, wherein the relevant factors affecting the training power flexibility resources include power supply side, power infrastructure, power demand side, and power storage; quantifying the relevant factors affecting the training power flexibility resources, and determining the training power flexibility resource data set. And defined in the form of column vectors ; The training subunit, used to train using the training power flexibility resource data set to obtain the values ​​of the parameter set, includes: determining the parameter set corresponding to the training power flexibility resource data set. Based on the parameter set Define a dataset of power flexibility resources for the training. Prediction function Based on the prediction function Determine the objective function ,in The corresponding optimal control output parameters are obtained by reducing the objective function. The value determines the parameter group value; An optimization subunit is used to optimize the values ​​of the parameter set to obtain a prediction function model.

4. The apparatus according to claim 3, characterized in that, The device further includes: The second determining subunit is used to determine the set of power flexibility resource data for testing; The evaluation subunit is used to evaluate the prediction function model using the test power flexibility resource data set and obtain the evaluation result.

Images