A method, device, equipment and medium for predicting vertical power flow across voltage levels

Abstract

The present invention discloses a method, apparatus, device and medium for predicting vertical power flow across voltage levels, which is applied to the field of power forecasting, including fusing the operating state matrix and the network connection matrix of the target transmission and distribution network to obtain a vertical power flow prediction network; obtaining a power monitoring sample data set of the target transmission and distribution network, using the power monitoring sample data set to perform power flow prediction training on the vertical power flow prediction network, and in the actual vertical power flow prediction process, inputting the real-time power monitoring data of the target transmission and distribution network into the vertical power flow prediction network to achieve vertical power flow prediction. The present invention can comprehensively capture the complex interactive relationship between different voltage levels in the target transmission and distribution network, improve the accuracy and reliability of vertical power flow prediction, reduce the power grid operation risk caused by inaccurate vertical power flow estimation, and provide protection for the safe and stable operation of the power grid.

Description

Technical Field

[0001] The present invention relates to the technical field of power prediction, and in particular to a method, device, equipment and medium for predicting vertical power flow across voltage levels. Background Art

[0002] With the large-scale integration of new energy into the power grid, the operating mode of the power system has undergone a significant transformation. The traditional one-way power flow mode has evolved into a two-way interaction between power generation and power consumption.

[0003] This change means that the flow of electricity in the power grid is no longer limited to a single direction. In particular, vertical power flow has appeared at transformers between different voltage levels, that is, the phenomenon of power interactive feedback between different voltage levels. This complex interaction phenomenon increases the instability of the power grid. Existing vertical power flow predictions mainly rely on precise measurements and assumptions of power grid line parameters, and often model individual transformers, ignoring the complex relationship between transformers, resulting in insufficient accuracy and reliability of vertical power flow estimation.

[0004] It can be seen that how to reasonably predict the vertical power flow between different voltage levels has become a technical problem that needs to be solved urgently by those skilled in the art. Summary of the Invention

[0005] The present invention provides a method, device, equipment and medium for predicting vertical power flow across voltage levels, so as to predict vertical power flow across voltage levels.

[0006] In order to solve the above technical problems, an embodiment of the present invention provides a method for predicting vertical power flow across voltage levels, including:

[0007] An operation status matrix of the target transmission and distribution network is constructed based on the acquired historical operation status data of the target transmission and distribution network.

[0008] A network connection matrix of the target transmission and distribution network is constructed according to the acquired real-time status data of each transformer in the target transmission and distribution network.

[0009] The operation state matrix and the network connection matrix are fused through multimodal information fusion to obtain a vertical power flow prediction network.

[0010] Obtain a power monitoring sample data set of the target transmission and distribution network, and use the power monitoring sample data set to perform power flow prediction training on the vertical power flow prediction network. The power flow prediction training is configured to update the network weights of the vertical power flow prediction network according to the power monitoring sample data set until the prediction deviation value of the vertical power flow prediction network is lower than a preset deviation threshold.

[0011] In the actual vertical power flow prediction process, the real-time power monitoring data of the target transmission and distribution network is input into the vertical power flow prediction network, and the vertical power flow prediction is realized according to the output result of the vertical power flow prediction network.

[0012] Furthermore, the network connection matrix includes a transformer connection relationship matrix and a network node attribute matrix.

[0013] The step of constructing a network connection matrix of the target transmission and distribution network based on the acquired real-time status data of each transformer in the target transmission and distribution network includes:

[0014] The transformer connection relationship matrix is ​​constructed according to the spatial state data of each transformer in the target transmission and distribution network.

[0015] A network node attribute matrix is ​​constructed according to the connection attribute data of each transformer in the target transmission and distribution network.

[0016] Furthermore, the operation state matrix and the network connection matrix are spliced ​​together by multimodal information fusion to obtain a vertical power flow prediction network, including:

[0017] The operation state matrix and the network connection matrix are normalized respectively to obtain first normalized data and second normalized data.

[0018] The first normalized data and the second normalized data are multimodally spliced ​​to obtain a comprehensive feature matrix.

[0019] A vertical tidal current prediction network based on a graph neural network is constructed according to the comprehensive feature matrix.

[0020] Furthermore, normalizing the operation state matrix and the network connection matrix respectively to obtain first normalized data and second normalized data includes:

[0021] The data in the operating state matrix is ​​normalized using a z-score normalization method to obtain first normalized data.

[0022] The data in the network connection matrix is ​​normalized using a maximum-minimum normalization method to obtain second normalized data.

[0023] Furthermore, the first normalized data and the second normalized data are multimodally spliced ​​to obtain a comprehensive feature matrix, including:

[0024] The data at the same node in the first normalized data and the second normalized data are spliced ​​column by column to obtain a feature vector of the corresponding node in the comprehensive feature matrix.

[0025] Furthermore, the power monitoring sample data set includes a historical power operation data subset and a historical vertical power flow data subset.

[0026] The obtaining of a power monitoring sample data set of the target transmission and distribution network includes:

[0027] Collect historical power operation data of the target transmission and distribution network under different operating conditions.

[0028] The historical power operation data is paired with the corresponding vertical power flow true value to form a power monitoring sample data set.

[0029] Furthermore, the updating of the network weight of the vertical power flow prediction network according to the power monitoring sample data set until the prediction deviation value of the vertical power flow prediction network is lower than a preset deviation threshold includes:

[0030] The data in the historical power operation data subset are input into the vertical power flow prediction network, and the output of the vertical power flow prediction network is verified according to the corresponding data in the historical vertical power flow data subset.

[0031] The network weights of the vertical power flow prediction network are continuously updated according to the verification results until the prediction deviation value of the output result of the vertical power flow prediction network reaches below a preset deviation threshold.

[0032] Another embodiment of the present invention provides a device for predicting vertical power flow across voltage levels, including:

[0033] The operation matrix construction module is used to construct the operation status matrix of the target transmission and distribution network based on the acquired historical operation status data of the target transmission and distribution network.

[0034] The connection matrix construction module is used to construct a network connection matrix of the target transmission and distribution network based on the acquired real-time status data of each transformer in the target transmission and distribution network.

[0035] The feature fusion module is used to fuse the operation state matrix and the network connection matrix through multimodal information fusion to obtain a vertical power flow prediction network.

[0036] A network training module is used to obtain a power monitoring sample data set of the target transmission and distribution network, and use the power monitoring sample data set to perform power flow prediction training on the vertical power flow prediction network. The power flow prediction training is configured to update the network weights of the vertical power flow prediction network according to the power monitoring sample data set until the prediction deviation value of the vertical power flow prediction network is lower than a preset deviation threshold.

[0037] The power flow prediction module is used to input the real-time power monitoring data of the target transmission and distribution network into the vertical power flow prediction network during the actual vertical power flow prediction process, and realize the vertical power flow prediction according to the output result of the vertical power flow prediction network.

[0038] Another embodiment of the present invention provides a computer device, including a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor, wherein the processor implements the above-mentioned method for predicting vertical power flow across voltage levels when executing the computer program.

[0039] Yet another embodiment of the present invention provides a computer-readable storage medium storing a computer program, wherein when the device where the computer-readable storage medium is located executes the computer program, the above-mentioned method for predicting vertical power flow across voltage levels is implemented.

[0040] Compared with the prior art, the embodiments of the present invention have the following advantages:

[0041] By constructing an operating status matrix and a network connection matrix, and using multimodal information fusion technology to generate a vertical power flow prediction network, it is possible to comprehensively and accurately capture the complex interactive relationships between different voltage levels in the target transmission and distribution network, improve the accuracy and reliability of vertical power flow prediction, reduce the power grid operation risks caused by inaccurate vertical power flow estimation, and provide guarantees for the safe and stable operation of the power grid. BRIEF DESCRIPTION OF THE DRAWINGS

[0042] Figure 1 is a flowchart of the steps of a method for predicting vertical power flow across voltage levels in one embodiment of the present invention;

[0043] Figure 2 is a structural block diagram of a device for predicting vertical power flow across voltage levels in one embodiment of the present invention;

[0044] Figure 3 A structural diagram of a computer device provided in an embodiment of the present invention;

[0045] Reference numerals:

[0046] 21. Operation matrix construction module; 22. Connection matrix construction module; 23. Feature fusion module; 24. Network training module; 25. Current flow prediction module. DETAILED DESCRIPTION

[0047] The following will be combined with the accompanying drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the embodiments described are only part of the embodiments of the present invention, not all of the embodiments. The purpose of providing these embodiments is to make the disclosure of the present invention more thorough and comprehensive. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts shall fall within the scope of protection of the present invention.

[0048] In the description of this application, the terms "first," "second," "third," etc. are used for descriptive purposes only and should not be understood to indicate or imply relative importance or implicitly specify the number of the technical features indicated. Therefore, a feature specified as "first," "second," "third," etc. may explicitly or implicitly include one or more of the features. In the description of this application, unless otherwise specified, "plurality" means two or more.

[0049] In the description of this application, it should be noted that, unless otherwise expressly specified and limited, the terms "installed", "connected" and "connected" should be understood in a broad sense. For example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be a direct connection, or an indirect connection through an intermediate medium, or it can be a communication between the two components. The terms "vertical", "horizontal", "left", "right", "up", "down" and similar expressions used herein are for illustrative purposes only, and do not indicate or imply that the device or component referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore cannot be understood as a limitation on the present invention. The term "and / or" used herein includes any and all combinations of one or more related listed items. For those of ordinary skill in the art, the specific meanings of the above terms in this application can be understood according to specific circumstances.

[0050] In the description of this application, it should be noted that, unless otherwise defined, all technical and scientific terms used in this application have the same meanings as those commonly understood by those skilled in the art. The terms used in the specification of the present invention are only for the purpose of describing specific embodiments and are not intended to limit the present invention. For those of ordinary skill in the art, the specific meanings of the above terms in this application can be understood by those skilled in the art in specific circumstances.

[0051] With the large-scale access of new energy sources (such as wind power and solar energy), the operating mode of traditional power grids has changed significantly. The one-way power flow mode of traditional power grids has gradually been replaced by a two-way interactive mode.

[0052] Renewable energy generation is intermittent, volatile, and uncertain, and its output power is significantly affected by factors such as weather and season. For example, solar power generation significantly decreases at night and on rainy days, while wind power generation is affected by wind speed and direction.

[0053] This uncertainty complicates the interaction of power between different voltage levels, particularly at transformers between different voltage levels, where a phenomenon known as "vertical power flow" occurs. Vertical power flow refers to the feedback loop between different voltage levels. This interaction of power data increases grid instability, potentially leading to voltage fluctuations, power imbalances, and other issues, compromising the safe operation of the grid.

[0054] An embodiment of the present invention provides a method for predicting vertical power flow across voltage levels, which estimates vertical power flow at different voltage levels between transformers. For details, see Figure 1 , Figure 1 The flowchart of the method for predicting vertical power flow across voltage levels in one embodiment of the present invention includes:

[0055] S11. Constructing an operation status matrix of the target transmission and distribution network based on the acquired historical operation status data of the target transmission and distribution network.

[0056] First, we need to construct an operating status matrix for the target transmission and distribution network. This matrix is ​​based on the historical operating status data of the target transmission and distribution network. This historical operating status data includes various operating parameters and status information of the target transmission and distribution network at different times and under different operating conditions. For example, key indicators such as voltage, current, power, and frequency of the target transmission and distribution network, as well as detailed information such as equipment operating status and fault records, are included. By collecting, organizing, and analyzing this historical data, we can extract key features that reflect the grid's operating status.

[0057] These features are organized and represented in the form of a matrix to form an operating status matrix.

[0058] S12. Constructing a network connection matrix of the target transmission and distribution network based on the acquired real-time status data of each transformer in the target transmission and distribution network.

[0059] In the target transmission and distribution network, power transmission lines of different voltage levels are often connected through transformers to increase or decrease the voltage to meet the power transmission and distribution needs in different scenarios. Therefore, cross-voltage level power flow prediction requires the acquisition and processing of transformer information.

[0060] Specifically, it is necessary to construct a network connection matrix of transformers in the target distribution network through the spatial position, connection relationship and attribute information of each transformer in the entire target transmission and distribution network. The network connection matrix is ​​an important data structure that describes the connection relationship and attributes between each transformer in the target transmission and distribution network.

[0061] The network connection matrix constructed in this embodiment includes a transformer connection relationship matrix and a network node attribute matrix.

[0062] The network connection matrix of the target transmission and distribution network is constructed based on the real-time status data of each transformer in the target transmission and distribution network. The specific process is as follows:

[0063] A transformer connection relationship matrix is ​​constructed based on the spatial status data of each transformer in the target transmission and distribution network. It is mainly used to describe the physical connection relationship between transformers, including which transformers have connection relationships, the specific connection method, the location of each transformer in the target transmission and distribution network, etc. The transformer connection relationship matrix can clearly represent the topological structure of the power grid and provide basic connection information for subsequent power flow prediction.

[0064] A network node attribute matrix is ​​constructed based on the connection attribute data of each transformer in the target transmission and distribution network. The network node attribute matrix further represents the attribute information of each node (i.e., transformer) in the target transmission and distribution network, such as the voltage level, power capacity, and load of each transformer. It reflects the differences in the function, status, or characteristics of each transformer, thereby more accurately predicting vertical power flow.

[0065] S13. The operation state matrix and the network connection matrix are fused through multimodal information fusion to obtain a vertical power flow prediction network.

[0066] The network connection matrix is ​​used as the framework matrix of the connection relationship and attributes of each transformer in the target transmission and distribution network, and the operation status matrix is ​​used as the power data matrix of the transmission data operation in the power lines in the target transmission and distribution network. It is necessary to fuse the operation status matrix and the network connection matrix to obtain the comprehensive feature matrix of the target transmission and distribution network.

[0067] The operating status matrix contains dynamic operating data for power lines in the target transmission and distribution network, including real-time operating parameters such as voltage, current, power, and frequency, as well as equipment operating status and fault records. The network connection matrix, on the other hand, contains the connection relationships and attribute information for each transformer in the target transmission and distribution network, such as the physical connections between transformers, voltage levels, power capacity, and load conditions. The two matrices contain different data types and both contain data from multiple modalities, necessitating a fusion between them.

[0068] Specifically, the z-score normalization method is used to normalize the data in the operating status matrix. By converting the data into a standard normal distribution (mean is 0, standard deviation is 1), the dimensional differences between different features and the skewness of the data distribution are eliminated, and the first normalized data are obtained, making the data comparable.

[0069] The data in the network connection matrix is ​​normalized using the maximum and minimum normalization method. Maximum and minimum normalization scales the data to the interval [0, 1], retaining the original distribution of the data and obtaining the second normalized data.

[0070] Features of different modes (such as voltage, current, power, etc.) may have different dimensions and orders of magnitude. Directly using this data will cause certain features to dominate the subsequently constructed model, affecting the training effect of the model. Through normalization, this difference can be eliminated so that all features can be compared and processed on the same scale.

[0071] For the same node, the normalized operation status matrix data and the normalized network connection matrix data are spliced ​​column by column, and the feature information from different sources is integrated together to form the feature vector of the corresponding node in the comprehensive feature matrix.

[0072] Graph neural networks can effectively process graph structured data and capture complex relationships between nodes. They are more suitable for scenarios with clear topological structures such as transformer topology. Therefore, this embodiment constructs a vertical power flow prediction network based on graph neural networks according to the comprehensive feature matrix, thereby predicting the relationship between power operation data and vertical power flow.

[0073] S14. Obtain a power monitoring sample data set of the target transmission and distribution network, and use the power monitoring sample data set to perform power flow prediction training on the vertical power flow prediction network. The power flow prediction training is configured to update the network weights of the vertical power flow prediction network according to the power monitoring sample data set until the prediction deviation value of the vertical power flow prediction network is lower than a preset deviation threshold.

[0074] By collecting a large amount of historical power operation data and the corresponding true vertical power flow values, the model can learn the grid's operating modes and vertical power flow patterns under different operating conditions, thereby improving prediction accuracy. Therefore, this embodiment collects subsets of historical power operation data for the target transmission and distribution network under different operating conditions. Each set of historical power operation data includes real-time operating parameters such as voltage, current, power, and frequency, as well as equipment operating status and fault records. This data reflects the dynamic operation of the grid at different times and under different operating conditions.

[0075] At the same time, it is necessary to obtain the true value of the vertical flow corresponding to each set of historical power operation data to obtain a subset of historical vertical flow data.

[0076] Among them, the historical power operation data subset corresponds to the data in the historical vertical flow data subset one by one, and each set of historical power operation data has the corresponding true value of historical vertical flow data.

[0077] The data in the historical power operation data subset are input into the vertical power flow prediction network to train the vertical power flow prediction network. The vertical power flow prediction network is trained with these input data, learns the relationship between power operation data and vertical power flow, and adjusts its network weights during the training process.

[0078] After the data in the historical power operation data subset are input into the vertical flow prediction network, the vertical flow prediction network will calculate and output the vertical flow based on the historical power operation data, and compare the output result with the corresponding vertical flow true value in the historical vertical flow data subset to obtain the prediction deviation value of the vertical flow prediction network.

[0079] If the prediction deviation value is less than the preset deviation threshold, the training is terminated and a trained vertical current prediction network is obtained.

[0080] If the predicted deviation value is still greater than the preset deviation threshold, the vertical power flow prediction network is iteratively trained using historical power operation data until the predicted deviation value reaches below the preset deviation threshold or the number of iterations reaches the preset maximum number of training times.

[0081] S15. In the actual vertical power flow prediction process, the real-time power monitoring data of the target transmission and distribution network is input into the vertical power flow prediction network, and the vertical power flow prediction is realized according to the output result of the vertical power flow prediction network.

[0082] In the actual vertical power flow prediction process, the real-time power monitoring data of the target transmission and distribution network is input into the vertical power flow prediction network to obtain the vertical power flow prediction results obtained by real-time monitoring.

[0083] The vertical power flow prediction results can reflect the power interaction between different voltage levels in the target transmission and distribution network in real time, helping grid operation and maintenance personnel to promptly understand the operating status of the grid and quickly identify potential operating risks such as voltage fluctuations and power imbalance, so that timely measures can be taken to make adjustments.

[0084] The cross-voltage level vertical power flow prediction method of the present invention constructs an operating status matrix and a network connection matrix, and uses multimodal information fusion technology to generate a vertical power flow prediction network. It can comprehensively and accurately capture the complex interactive relationship between different voltage levels in the target transmission and distribution network, improve the accuracy and reliability of vertical power flow prediction, reduce the power grid operation risk caused by inaccurate vertical power flow estimation, and provide protection for the safe and stable operation of the power grid.

[0085] The embodiment of the present invention further provides a device for predicting vertical power flow across voltage levels, which is used to execute the above-mentioned method for predicting vertical power flow across voltage levels. Figure 2 This is a structural block diagram of a device for predicting vertical power flow across voltage levels according to an embodiment of the present invention. The device includes:

[0086] The operation matrix construction module 21 is used to construct the operation status matrix of the target transmission and distribution network according to the acquired historical operation status data of the target transmission and distribution network.

[0087] The connection matrix construction module 22 is configured to construct a network connection matrix of the target transmission and distribution network based on the acquired real-time status data of each transformer in the target transmission and distribution network.

[0088] The feature fusion module 23 is used to fuse the operation state matrix and the network connection matrix through multimodal information fusion to obtain a vertical power flow prediction network.

[0089] The network training module 24 is used to obtain the power monitoring sample data set of the target transmission and distribution network, and use the power monitoring sample data set to perform power flow prediction training on the vertical power flow prediction network. The power flow prediction training is configured to update the network weights of the vertical power flow prediction network according to the power monitoring sample data set until the prediction deviation value of the vertical power flow prediction network is lower than a preset deviation threshold.

[0090] The power flow prediction module 25 is used to input the real-time power monitoring data of the target transmission and distribution network into the vertical power flow prediction network during the actual vertical power flow prediction process, and realize the vertical power flow prediction according to the output result of the vertical power flow prediction network.

[0091] The technical features and technical effects of the device proposed in the embodiment of the present invention are the same as those of the method proposed in the embodiment of the present invention and are not described in detail here. Each module in the above-mentioned device can be implemented in whole or in part by software, hardware, or a combination thereof. Each of the above-mentioned modules can be embedded in or independent of the processor in the computer device in the form of hardware, or can be stored in the memory of the computer device in the form of software so that the processor can call and execute the operations corresponding to the above modules.

[0092] See also Figure 3, which is a structural block diagram of a computer device provided in an embodiment of the present invention. The computer device provided in an embodiment of the present invention includes a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor. When the processor executes the computer program, the steps in the embodiment of the method for predicting vertical power flow across voltage levels described above are implemented, for example Figure 1 or, when the processor executes the computer program, the functions of the modules in the above-mentioned device embodiments are realized, such as module 21 to module 25 of the device for predicting vertical power flow across voltage levels.

[0093] For example, the computer program may be divided into one or more modules, which are stored in the memory and executed by the processor to implement the present invention. The one or more modules may be a series of computer program instruction segments capable of implementing specific functions, and the instruction segments are used to describe the execution process of the computer program in the computer device.

[0094] The computer device may include, but is not limited to, a processor and a memory. Those skilled in the art will appreciate that the schematic diagram is merely an example of a computer device and does not limit the computer device. The computer device may include more or fewer components than shown, or a combination of certain components, or different components. For example, the computer device may also include input and output devices, network access devices, buses, etc.

[0095] The processor may be a central processing unit (CPU), other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor may be a microprocessor or any conventional processor, etc. The processor is the control center of the computer device, connecting various parts of the entire computer device using various interfaces and lines.

[0096] The memory can be used to store the computer programs and / or modules. The processor implements the various functions of the computer device by running or executing the computer programs and / or modules stored in the memory and accessing the data stored in the memory. The memory may primarily include a program storage area and a data storage area. The program storage area may store an operating system and at least one application required for a function (such as a sound playback function or an image playback function); the data storage area may store data generated based on the use of the mobile phone (such as audio data, a phone book, etc.). Furthermore, the memory may include high-speed random access memory and non-volatile memory, such as a hard disk, internal memory, a plug-in hard disk, a smart media card (SMC), a secure digital (SD) card, a flash card, at least one disk storage device, a flash memory device, or other volatile solid-state storage device.

[0097] If the module integrated into the computer device is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the present invention can also implement all or part of the process steps in the above-mentioned method embodiments by instructing the relevant hardware through a computer program. The computer program can be stored in a computer-readable storage medium. When executed by a processor, the computer program can implement the steps of each of the above-mentioned method embodiments. The computer program includes computer program code, which can be in source code form, object code form, executable file, or some intermediate form. The computer-readable medium can include: any entity or device capable of carrying the computer program code, recording medium, USB flash drive, mobile hard drive, magnetic disk, optical disk, computer memory, read-only memory (ROM), random access memory (RAM), electrical carrier signal, telecommunication signal, and software distribution medium.

[0098] Those skilled in the art will appreciate that all or part of the processes in the above-described method embodiments can be implemented by instructing the relevant hardware through a computer program. The program can be stored in a computer-readable storage medium, and when executed, the program can include the processes in the above-described method embodiments. The storage medium can be a magnetic disk, an optical disk, a read-only memory (ROM), or a random access memory (RAM).

[0099] Accordingly, an embodiment of the present invention provides a computer-readable storage medium, wherein the computer-readable storage medium includes a stored computer program, wherein when the computer program is executed, the device where the computer-readable storage medium is located is controlled to perform the steps of the method for predicting vertical power flow across voltage levels in the above embodiment, for example Figure 1 Steps S11 to S15 described in .

[0100] In summary, compared with the prior art, the method, apparatus, computer device, and computer-readable storage medium for predicting vertical power flows across voltage levels provided by the embodiments of the present invention have the following beneficial effects:

[0101] By constructing an operating status matrix and a network connection matrix, and using multimodal information fusion technology to generate a vertical power flow prediction network, it is possible to comprehensively and accurately capture the complex interactive relationships between different voltage levels in the target transmission and distribution network, improve the accuracy and reliability of vertical power flow prediction, reduce the power grid operation risks caused by inaccurate vertical power flow estimation, and provide guarantees for the safe and stable operation of the power grid.

[0102] The above-described embodiments merely illustrate several implementations of the present invention, and while their descriptions are relatively specific and detailed, they should not be construed as limiting the scope of the present invention. It should be noted that a person skilled in the art would be able to make numerous variations and improvements without departing from the spirit of the present invention, all of which fall within the scope of protection of the present invention. Therefore, the scope of protection of the present invention shall be determined by the appended claims.

Claims

1. A method for predicting vertical power flow across voltage levels, characterized in that: include: Constructing an operation status matrix of the target transmission and distribution network based on the acquired historical operation status data of the target transmission and distribution network; Constructing a network connection matrix of the target transmission and distribution network according to the acquired real-time status data of each transformer in the target transmission and distribution network; Specifically comprising: constructing the transformer connection relationship matrix according to the spatial state data of each transformer in the target transmission and distribution network; constructing the network node attribute matrix according to the connection attribute data of each transformer in the target transmission and distribution network; the network connection matrix includes the transformer connection relationship matrix and the network node attribute matrix; The operation state matrix and the network connection matrix are fused through multimodal information fusion to obtain a vertical power flow prediction network; specifically, the method includes: normalizing the operation state matrix and the network connection matrix respectively to obtain first normalized data and second normalized data; performing multimodal splicing on the first normalized data and the second normalized data to obtain a comprehensive feature matrix; and constructing a vertical power flow prediction network based on a graph neural network according to the comprehensive feature matrix; Obtaining a power monitoring sample data set of the target transmission and distribution network, and using the power monitoring sample data set to perform power flow prediction training on the vertical power flow prediction network, wherein the power flow prediction training is configured to update the network weights of the vertical power flow prediction network according to the power monitoring sample data set until a prediction deviation value of the vertical power flow prediction network is lower than a preset deviation threshold; In the actual vertical power flow prediction process, the real-time power monitoring data of the target transmission and distribution network is input into the vertical power flow prediction network, and the vertical power flow prediction is realized according to the output result of the vertical power flow prediction network.

2. The method for predicting vertical power flow across voltage levels according to claim 1, wherein: Normalizing the operation state matrix and the network connection matrix respectively to obtain first normalized data and second normalized data includes: Normalizing the data in the operating state matrix using a z-score normalization method to obtain first normalized data; The data in the network connection matrix is ​​normalized using a maximum-minimum normalization method to obtain second normalized data.

3. The method for predicting vertical power flow across voltage levels according to claim 1, wherein: The multimodal concatenation of the first normalized data and the second normalized data to obtain a comprehensive feature matrix includes: The data at the same node in the first normalized data and the second normalized data are spliced ​​column by column to obtain a feature vector of the corresponding node in the comprehensive feature matrix.

4. The method for predicting vertical power flow across voltage levels according to claim 1, wherein: The power monitoring sample data set includes a historical power operation data subset and a historical vertical power flow data subset; The obtaining of a power monitoring sample data set of the target transmission and distribution network includes: Collecting historical power operation data of the target transmission and distribution network under different operating conditions; The historical power operation data is paired with the corresponding vertical power flow true value to form a power monitoring sample data set.

5. The method for predicting vertical power flow across voltage levels according to claim 4, wherein: The updating of the network weight of the vertical power flow prediction network according to the power monitoring sample data set until the prediction deviation value of the vertical power flow prediction network is lower than a preset deviation threshold includes: inputting data from the historical power operation data subset into the vertical power flow prediction network, and verifying the output of the vertical power flow prediction network based on corresponding data from the historical vertical power flow data subset; The network weights of the vertical power flow prediction network are continuously updated according to the verification results until the prediction deviation value of the output result of the vertical power flow prediction network reaches below a preset deviation threshold.

6. A device for predicting vertical power flow across voltage levels, used to implement the method for predicting vertical power flow across voltage levels according to any one of claims 1 to 5, characterized in that: include: An operation matrix construction module is used to construct an operation status matrix of the target transmission and distribution network based on the acquired historical operation status data of the target transmission and distribution network; a connection matrix construction module, configured to construct a network connection matrix of the target transmission and distribution network based on the acquired real-time status data of each transformer in the target transmission and distribution network; A feature fusion module, configured to fuse the operating state matrix and the network connection matrix through multimodal information fusion to obtain a vertical power flow prediction network; a network training module, configured to obtain a power monitoring sample data set of the target transmission and distribution network, and perform power flow prediction training on the vertical power flow prediction network using the power monitoring sample data set, wherein the power flow prediction training is configured to update the network weights of the vertical power flow prediction network according to the power monitoring sample data set until a prediction deviation value of the vertical power flow prediction network is lower than a preset deviation threshold; The power flow prediction module is used to input the real-time power monitoring data of the target transmission and distribution network into the vertical power flow prediction network during the actual vertical power flow prediction process, and realize the vertical power flow prediction according to the output result of the vertical power flow prediction network.

7. A computer device, characterized in that: The method comprises a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor, wherein when the processor executes the computer program, the method for predicting vertical power flow across voltage levels according to any one of claims 1 to 5 is implemented.

8. A computer-readable storage medium, characterized in that The computer-readable storage medium stores a computer program, wherein when the device where the computer-readable storage medium is located executes the computer program, the cross-voltage level vertical power flow prediction method according to any one of claims 1 to 5 is implemented.

Images