Power distribution network dispatching method and device, electronic equipment and storage medium

Abstract

The present disclosure provides a power distribution network scheduling method and device, electronic equipment and storage medium, belonging to the technical field of power processing, the method comprising: obtaining a plurality of user power datasets, processing each power consumption curve in the power dataset using tensor decomposition to obtain a plurality of power curve feature vectors; taking the power curve feature vectors of known type users as clustering centers, clustering the plurality of power curve feature vectors to obtain a plurality of clustering results; based on the plurality of clustering results, determining the user type information of each user in the plurality of users, the user type information at least including: office worker, single user, elderly user, industrial user, and commercial user. The technical solution provided by the present disclosure can greatly reduce the workload and improve the efficiency of identifying user types by using known type users to cluster unknown type users and determine the type information of unknown type users.

Description

Technical Field

[0001] This disclosure belongs to the field of power processing technology, and in particular relates to a distribution network dispatching method, device, electronic equipment and storage medium. Background Technology

[0002] In the field of power systems, with the rapid development of smart grids, it is crucial to detect users' electricity consumption types and dispatch power accordingly.

[0003] In related technologies, non-intrusive load monitoring is mainly used to identify electricity consumption data, thereby identifying the energy consumption of different appliances or users, and determining the type of electricity consumption data to achieve power dispatch. However, this traditional identification method relies on a large amount of tag data, i.e., known types of appliances or users, which has high labeling costs and cannot identify objects with small changes in power data, resulting in low identification accuracy and low identification efficiency, which in turn leads to inaccurate dispatch of the power distribution network. Summary of the Invention

[0004] This disclosure provides a solution to address the problems of low identification accuracy and low identification efficiency in related technologies, which in turn lead to inaccurate dispatching of power distribution networks.

[0005] In a first aspect, this disclosure provides a distribution network dispatching method, the method comprising:

[0006] Obtain power data sets of multiple users within a target area. The power data sets include time-aligned power consumption curves of known and unknown types of users within a preset time period.

[0007] Tensor decomposition is used to process the power consumption curves in the power dataset to obtain multiple power curve feature vectors.

[0008] Using the power curve feature vectors of known user types as cluster centers, the multiple power curve feature vectors are clustered to obtain multiple clustering results;

[0009] Based on the multiple clustering results, the user type information of each user among the multiple users is determined. The user type information includes at least: working professionals, single users, elderly users, industrial users, and business users.

[0010] Based on the user type information of each of the multiple users, the scheduling plan information of the target area is determined.

[0011] In some embodiments, tensor decomposition is used to process the power consumption curves in the power dataset to obtain multiple power curve feature vectors, including:

[0012] The power curves in the power dataset are normalized, and multiple three-dimensional tensors corresponding to the processed power curves are constructed.

[0013] Tensor decomposition is used to process the multiple three-dimensional tensors to obtain the multiple power curve feature vectors.

[0014] In some embodiments, the tensor decomposition includes CP decomposition and Tucker decomposition;

[0015] The multiple three-dimensional tensors are processed using tensor decomposition to obtain the multiple power curve feature vectors, including:

[0016] For any three-dimensional tensor among the plurality of three-dimensional tensors, the global feature vector of the three-dimensional tensor is extracted by CP decomposition, and the local feature vector of the three-dimensional tensor is extracted by Tucker decomposition.

[0017] The local feature vector and the global feature vector are fused to obtain the power curve feature vector, and then the multiple power curve feature vectors are obtained.

[0018] In some embodiments, the power curve feature vectors of known types of users are used as cluster centers to cluster the multiple power curve feature vectors, resulting in multiple clustering results, including:

[0019] For a first feature vector corresponding to any known type of user among the multiple power curve feature vectors, calculate multiple Euclidean distances between the first feature vector and the multiple power curve feature vectors corresponding to multiple unknown types of users;

[0020] Users whose Euclidean distance is less than a first threshold among the multiple Euclidean distances are assigned to the cluster to which the user corresponding to the first feature vector belongs, thus obtaining a clustering result, and then obtaining multiple clustering results.

[0021] In some embodiments, calculating multiple Euclidean distances between the first feature vector and multiple power curve feature vectors corresponding to the multiple unknown types of users includes:

[0022] Based on the Euclidean distance formula, the first feature vector, and the multiple power curve feature vectors corresponding to the multiple unknown types of users, multiple Euclidean distances are obtained between the first feature vector and the multiple power curve feature vectors corresponding to the multiple unknown types of users.

[0023] The Euclidean distance formula is as follows: The Let be the eigenvalue of the power curve feature vector corresponding to the k-th user among the multiple known user types, in the i-th dimension. Let the eigenvalue of the power curve feature vector corresponding to the h-th user among the multiple unknown types of users be the eigenvalue of the i-th dimension. For feature dimensions, the The Euclidean distance is the power curve feature vector corresponding to the kth user among the plurality of known types of users and the power curve feature vector corresponding to the hth user among the plurality of unknown types of users.

[0024] In some embodiments, the power curve feature vectors of known types of users are used as cluster centers to cluster the multiple power curve feature vectors, resulting in multiple clustering results. The method further includes:

[0025] Using the power curve feature vectors of known user types as cluster centers, the multiple power curve feature vectors are clustered to obtain multiple initial clustering results, and the number of iterations is set to 1.

[0026] Based on the multiple initial clustering results, update the cluster centers in each clustering result, re-cluster, and increment the iteration count by 1;

[0027] Determine whether the number of iterations is greater than the preset number of iterations. If yes, obtain the multiple clustering results. If no, continue to update the cluster centers in each clustering result and re-cluster.

[0028] Secondly, this disclosure provides a power distribution network dispatching device, the device comprising:

[0029] The acquisition unit is used to acquire the power data set of multiple users in the target area. The power data set includes the power consumption curves of known type users and unknown type users within a preset time period, which are time-aligned.

[0030] The decomposition unit is used to process the power consumption curves in the power dataset using tensor decomposition to obtain multiple power curve feature vectors.

[0031] The clustering unit is used to cluster the multiple power curve feature vectors with the power curve feature vectors of known types of users as the cluster centers, and obtain multiple clustering results;

[0032] The determining unit is used to determine the user type information of each user among the multiple users based on the multiple clustering results. The user type information includes at least: working users, single users, elderly users, industrial users, and business users.

[0033] The determining unit is further configured to determine the scheduling plan information of the target area based on the user type information of each of the plurality of users.

[0034] Thirdly, this disclosure provides an electronic device, including:

[0035] Processor; and

[0036] Memory for storing the executable instructions of the processor;

[0037] The processor is configured to execute the first aspect or any method in a possible implementation of the first aspect by executing the executable instructions.

[0038] Fourthly, embodiments of this disclosure provide a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements any of the methods in the first aspect or any possible implementations of the first aspect.

[0039] Fifthly, embodiments of this disclosure provide a computer program product including computer instructions that, when executed by a processor, implement any method in the first aspect or any possible implementation of the first aspect.

[0040] The technical solution provided in this disclosure involves acquiring a power data set of multiple users within a target area. The power data set includes time-aligned power consumption curves of known and unknown user types within a preset time period. Tensor decomposition is used to process each power consumption curve in the power data set, resulting in multiple power curve feature vectors. Using the power curve feature vectors of known user types as cluster centers, the multiple power curve feature vectors are clustered to obtain multiple clustering results. Based on these multiple clustering results, user type information for each user among the multiple users is determined. The user type information includes at least: working professionals, single users, elderly users, industrial users, and commercial users. Based on the user type information of each user among the multiple users, scheduling plan information for the target area is determined. The technical solutions provided in the embodiments of this disclosure determine the user type of unknown users by using a small number of known user types and clustering methods. This not only improves the efficiency of user type identification but also greatly improves the accuracy of identification. Furthermore, tensor decomposition is used to process the data before clustering, which can greatly reduce the computational complexity. Based on the user type information of each user among multiple users, the scheduling plan information of the target area is determined, which can better schedule the distribution network and ensure the accuracy of scheduling. Attached Figure Description

[0041] To more clearly illustrate the technical solutions in the embodiments or related technologies of this disclosure, the accompanying drawings used in the description of the embodiments or related technologies will be briefly introduced below. Obviously, the accompanying drawings described below are some embodiments of this disclosure. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort. In the drawings:

[0042] Figure 1 A schematic flowchart illustrating a power distribution network dispatching method according to an embodiment of this disclosure;

[0043] Figure 2 This is a schematic diagram of the structure of a power distribution network dispatching device according to an embodiment of the present disclosure;

[0044] Figure 3 This is a schematic diagram of the structure of an electronic device provided in one embodiment of the present disclosure. Detailed Implementation

[0045] Embodiments of this disclosure are described in detail below, with examples of embodiments illustrated in the accompanying drawings. The embodiments described below with reference to the accompanying drawings are exemplary and intended to explain this disclosure, and should not be construed as limiting this disclosure.

[0046] The terms "first" and "second," etc., used in the specification, claims, and drawings of this disclosure are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that embodiments of the present disclosure described herein can be implemented, for example, in orders other than those illustrated or described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover a non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0047] The power distribution network dispatching method provided in this disclosure can run on terminal devices or servers. The terminal device can be a local terminal device, including wearable devices such as VR (Virtual Reality), AR (Augmented Reality), and MR (Mixed Reality). The server can be a standalone physical server, a server cluster or distributed system composed of multiple physical servers, or a cloud server providing basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communication, middleware services, domain name services, security services, CDN, and big data and artificial intelligence platforms.

[0048] In the field of power systems, with the rapid development of smart grids, it is crucial to detect users' electricity consumption types and dispatch power accordingly.

[0049] In related technologies, non-intrusive load monitoring is mainly used to identify electricity consumption data, thereby identifying the energy consumption of different appliances or users, and determining the type of electricity consumption data to achieve power dispatch. However, this traditional identification method relies on a large amount of tag data, i.e., known types of appliances or users, which has high labeling costs and cannot identify objects with small changes in power data, resulting in low identification accuracy and low identification efficiency, which in turn leads to inaccurate dispatch of the power distribution network.

[0050] Figure 1 This is a flowchart illustrating a distribution network dispatching method provided in an exemplary embodiment of the present disclosure. The method includes at least the following steps S101-S105:

[0051] S101, Obtain the power data set of multiple users within the target area.

[0052] In some embodiments, the target area refers to an area where electricity is consumed. For example, a residential community, a building, or a commercial area. The specific settings can be determined based on actual circumstances.

[0053] In some embodiments, the power dataset includes time-aligned power consumption curves for known and unknown types of users within a preset time period.

[0054] In some embodiments, the user's power consumption curve may refer to the power consumption curve of the user's household or region within a preset time period. Specifically, the preset time period may be one day.

[0055] For example, taking a preset time period of one day and users within a residential community as an example, the system can obtain the power consumption curve of all users in the entire community on the same day. The power consumption data is collected at the same time points throughout the day, with intervals of up to fifteen minutes. Specific settings can be configured according to actual needs.

[0056] Furthermore, taking a preset time period of one day and users from different communities as an example, we obtain the power consumption curves of different communities (different communities are different users) on the same day.

[0057] In some embodiments, the user type information corresponding to the type includes at least: working professionals, single users, elderly users, industrial users, and business users.

[0058] In this embodiment, the single user can be a studio apartment, the elderly user can be a residential community with a large elderly population, and the commercial user can be a commercial area. Specific settings can be configured according to actual circumstances.

[0059] Furthermore, "known user types" refers to the situation where, during actual monitoring, the types of a subset of users can be determined through manual monitoring. For example, in a residential community, the user type information for all users in a single building can be determined first.

[0060] S102, using tensor decomposition, the power consumption curves in the power dataset are processed to obtain multiple power curve feature vectors.

[0061] In some embodiments, tensor decomposition is used to process the power consumption curves in the power dataset to obtain multiple power curve feature vectors, including steps S11-S12:

[0062] S11 normalizes each power curve in the power dataset and constructs multiple three-dimensional tensors corresponding to each power curve after processing.

[0063] S12 uses tensor decomposition to process multiple three-dimensional tensors to obtain multiple power curve feature vectors.

[0064] In this implementation, tensor decomposition is used to process multiple three-dimensional tensors to obtain multiple power curve feature vectors, including steps S121-S122:

[0065] S121: For any three-dimensional tensor among multiple three-dimensional tensors, extract the global feature vector of the three-dimensional tensor using CP decomposition, and extract the local feature vector of the three-dimensional tensor using Tucker decomposition.

[0066] In some embodiments, tensor decomposition is a lightweight structure that includes CP decomposition and Tucker decomposition.

[0067] The Tucker decomposition involves modal decomposition (splitting along the channel and spatial dimensions), followed by low-rank convolution (DwConv+BN) and core tensor learning, compressing redundant parameters using the factor matrix of the Tucker decomposition. The decomposed sub-tensors are then used for feature recombination through alternating least squares (ALS) optimization. Finally, the channel dimension is restored by tensor concatenation (Concat), and channel shuffle is introduced to enhance feature interaction. The final output is a local feature vector.

[0068] The CP decomposition is processed through two paths: Path 1: CBR → Low-rank DwConv → BN → Core Tensor Projection (learning the rank-1 component of CP decomposition); Path 2: DwConv → BN → Residual Tensor Compensation (supplementing high-frequency detail features). The final output is the global feature vector.

[0069] S122, the local feature vector and the global feature vector are fused to obtain the power curve feature vector, and then multiple power curve feature vectors are obtained.

[0070] S103, using the power curve feature vectors of known types of users as cluster centers, cluster the multiple power curve feature vectors to obtain multiple clustering results.

[0071] In some implementations, each clustering result contains multiple power curve feature vectors, and the user type corresponding to each power curve feature vector is the same.

[0072] In some embodiments, using the power curve feature vectors of known types of users as cluster centers, multiple power curve feature vectors are clustered to obtain multiple clustering results, including steps S21-S22:

[0073] S21, for the first feature vector corresponding to any known type of user among the multiple power curve feature vectors, calculate multiple Euclidean distances between the first feature vector and the multiple power curve feature vectors corresponding to multiple unknown types of users.

[0074] S22, assign users whose Euclidean distance is less than the first threshold among multiple Euclidean distances to the cluster to which the user corresponding to the first feature vector belongs, to obtain a clustering result, and then obtain multiple clustering results.

[0075] In some implementations, the calculation of multiple Euclidean distances between the first feature vector and multiple power curve feature vectors corresponding to multiple unknown types of users includes: obtaining multiple Euclidean distances between the first feature vector and multiple power curve feature vectors corresponding to multiple unknown types of users based on the Euclidean distance formula, the first feature vector, and multiple power curve feature vectors corresponding to multiple unknown types of users.

[0076] The Euclidean distance formula is as follows:

[0077] Let i be the eigenvalue of the power curve eigenvector corresponding to the k-th user among multiple known user types. Let be the eigenvalue of the i-th dimension of the power curve feature vector corresponding to the h-th user among multiple unknown types of users. For feature dimension, Let Euclidean distance be the power curve feature vector corresponding to the k-th user among multiple known user types and the power curve feature vector corresponding to the h-th user among multiple unknown user types.

[0078] In some implementations, the power curve feature vectors of known user types are used as cluster centers to cluster multiple power curve feature vectors, resulting in multiple clustering results. The method also includes steps S31-S33:

[0079] S31, using the power curve feature vector of known type users as the cluster center, cluster multiple power curve feature vectors to obtain multiple initial clustering results, and set the number of iterations to 1.

[0080] S32, based on multiple initial clustering results, update the cluster centers in each clustering result, re-cluster, and increment the iteration count by 1.

[0081] S33, determine whether the number of iterations is greater than the preset number of iterations. If yes, obtain multiple clustering results. If no, continue to update the cluster centers in each clustering result and re-cluster.

[0082] In some embodiments, during the process of updating the cluster centers in each clustering result in step S32 above, the new cluster centers refer to the mean of the eigenvectors of all power curves. The calculation process of the mean is not described here.

[0083] S104, Based on the multiple clustering results, determine the user type information of each user among the multiple users.

[0084] In some embodiments, to better understand the user type identification process of this scheme, taking a community with 100 households (10 households of known type and 90 households of unknown type) and their respective power consumption curves as an example, the identification process is explained in detail:

[0085] First, obtain the power consumption curves of 100 users on the same day. Perform tensor decomposition on the 100 power consumption curves to obtain 100 power curve feature vectors. Then, using the power curve feature vectors of 10 users belonging to the known type as cluster centers, cluster all power curve feature vectors to obtain ten clustering results, thereby determining the type of all users.

[0086] In some embodiments, if a power curve feature vector does not belong to any cluster result, it means that the type of user corresponding to that power curve feature vector is uncertain and needs to be determined manually. The specific process can be set according to the actual situation.

[0087] In some embodiments, to improve the accuracy of clustering results, the clustering results can be verified multiple times. The specific steps are as follows: obtain the power consumption curves of known type users and unknown type users within another preset time period, calculate multiple new clustering results according to the steps of the above embodiment, and then compare the multiple clustering results in step S103 above. If the similarity is greater than the qualified threshold, the multiple clustering results are determined to be qualified.

[0088] Specifically, continuing with the example of a community with 100 households (10 households of known type and 90 households of unknown type), the electricity consumption curves of each household are as follows:

[0089] Obtain the power consumption curves of 100 users on the first and second days, and then obtain the clustering results for the first and second days according to the classification process described above.

[0090] To increase the accuracy of clustering results, the number of preset time periods can be increased. For example, the first day, the second day, the third day, etc. The specific number can be set according to the actual situation.

[0091] S105, Based on the user type information of each of the multiple users, determine the scheduling plan information of the target area.

[0092] In some embodiments, based on the obtained scheduling plan information, the power distribution within the target area can be adjusted, thereby making the power consumption within the target area more reasonable.

[0093] The technical solution provided in this disclosure involves acquiring a power data set of multiple users within a target area. The power data set includes time-aligned power consumption curves of known and unknown user types within a preset time period. Tensor decomposition is used to process each power consumption curve in the power data set, resulting in multiple power curve feature vectors. Using the power curve feature vectors of known user types as cluster centers, the multiple power curve feature vectors are clustered to obtain multiple clustering results. Based on these multiple clustering results, user type information for each user among the multiple users is determined. The user type information includes at least: working professionals, single users, elderly users, industrial users, and commercial users. Based on the user type information of each user among the multiple users, scheduling plan information for the target area is determined. The technical solutions provided in the embodiments of this disclosure determine the user type of unknown users by using a small number of known user types and clustering methods. This not only improves the efficiency of user type identification but also greatly improves the accuracy of identification. Furthermore, tensor decomposition is used to process the data before clustering, which can greatly reduce the computational complexity. Based on the user type information of each user among multiple users, the scheduling plan information of the target area is determined, which can better schedule the distribution network and ensure the accuracy of scheduling.

[0094] Figure 2 A schematic diagram of the structure of a power distribution network dispatching device provided as an exemplary embodiment of this disclosure;

[0095] The device includes: an acquisition unit 201, a decomposition unit 202, a clustering unit 203, and a determination unit 204;

[0096] The acquisition unit 201 is used to acquire the power data set of multiple users in the target area. The power data set includes the power consumption curves of known type users and unknown type users within a preset time period, which are time-aligned.

[0097] Decomposition unit 202 is used to process the power consumption curves in the power dataset using tensor decomposition to obtain multiple power curve feature vectors.

[0098] Clustering unit 203 is used to cluster multiple power curve feature vectors with the power curve feature vectors of known types of users as cluster centers to obtain multiple clustering results;

[0099] The determining unit 204 is used to determine the user type information of each user among multiple users based on multiple clustering results. The user type information includes at least: working users, single users, elderly users, industrial users, and business users.

[0100] The determining unit 204 is further configured to determine the scheduling plan information of the target area based on the user type information of each user among the plurality of users.

[0101] In some embodiments, the apparatus is used to process the power consumption curves in the power dataset using tensor decomposition to obtain multiple power curve feature vectors. Specifically, the apparatus is used to:

[0102] Normalize the power curves in the power dataset and construct multiple three-dimensional tensors corresponding to the processed power curves.

[0103] Tensor decomposition is used to process multiple three-dimensional tensors to obtain multiple power curve feature vectors.

[0104] In some embodiments, tensor decomposition includes CP decomposition and Tucker decomposition;

[0105] The device is used to process multiple three-dimensional tensors using tensor decomposition to obtain multiple power curve feature vectors. Specifically, the device is used for:

[0106] For any three-dimensional tensor among multiple three-dimensional tensors, the global feature vector of the three-dimensional tensor is extracted by CP decomposition, and the local feature vector of the three-dimensional tensor is extracted by Tucker decomposition.

[0107] By fusing local and global feature vectors, a power curve feature vector is obtained, which in turn yields multiple power curve feature vectors.

[0108] In some embodiments, the apparatus is used to cluster multiple power curve feature vectors using the power curve feature vectors of known types of users as cluster centers, to obtain multiple clustering results. Specifically, the apparatus is used to:

[0109] For a first feature vector corresponding to any known type of user among multiple power curve feature vectors, calculate multiple Euclidean distances between the first feature vector and multiple power curve feature vectors corresponding to multiple unknown types of users.

[0110] Users whose Euclidean distance is less than a first threshold among multiple Euclidean distances are assigned to the cluster to which the user corresponding to the first feature vector belongs, thus obtaining a clustering result, and then multiple clustering results.

[0111] In some embodiments, the apparatus is configured to calculate multiple Euclidean distances between a first feature vector and multiple power curve feature vectors corresponding to multiple users of unknown types, specifically for:

[0112] Based on the Euclidean distance formula, the first feature vector, and the feature vectors of multiple power curves corresponding to multiple users of multiple unknown types, multiple Euclidean distances between the first feature vector and the feature vectors of multiple power curves corresponding to multiple users of multiple unknown types are obtained.

[0113] The Euclidean distance formula is as follows: , Let i be the eigenvalue of the power curve eigenvector corresponding to the k-th user among multiple known user types. Let be the eigenvalue of the i-th dimension of the power curve feature vector corresponding to the h-th user among multiple unknown types of users. For feature dimension, Let Euclidean distance be the power curve feature vector corresponding to the k-th user among multiple known user types and the power curve feature vector corresponding to the h-th user among multiple unknown user types.

[0114] In some embodiments, the apparatus is used to cluster multiple power curve feature vectors using the power curve feature vectors of known types of users as cluster centers, to obtain multiple clustering results. Specifically, the apparatus is used to:

[0115] Using the power curve feature vectors of known user types as cluster centers, multiple power curve feature vectors are clustered to obtain multiple initial clustering results, and the number of iterations is set to 1.

[0116] Based on multiple initial clustering results, update the cluster centers in each clustering result, re-cluster, and increment the iteration count by 1;

[0117] Determine if the number of iterations is greater than the preset number of iterations. If yes, obtain multiple clustering results. If no, continue to update the cluster centers in each clustering result and re-cluster.

[0118] The technical solution provided in this disclosure involves acquiring a power data set of multiple users within a target area. The power data set includes time-aligned power consumption curves of known and unknown user types within a preset time period. Tensor decomposition is used to process each power consumption curve in the power data set, resulting in multiple power curve feature vectors. Using the power curve feature vectors of known user types as cluster centers, the multiple power curve feature vectors are clustered to obtain multiple clustering results. Based on these multiple clustering results, user type information for each user among the multiple users is determined. The user type information includes at least: working professionals, single users, elderly users, industrial users, and commercial users. Based on the user type information of each user among the multiple users, scheduling plan information for the target area is determined. The technical solutions provided in the embodiments of this disclosure determine the user type of unknown users by using a small number of known user types and clustering methods. This not only improves the efficiency of user type identification but also greatly improves the accuracy of identification. Furthermore, tensor decomposition is used to process the data before clustering, which can greatly reduce the computational complexity. Based on the user type information of each user among multiple users, the scheduling plan information of the target area is determined, which can better schedule the distribution network and ensure the accuracy of scheduling.

[0119] It should be understood that the device embodiments and method embodiments can correspond to each other, and similar descriptions can be referred to the method embodiments. To avoid repetition, they will not be repeated here. Specifically, the device can execute the above method embodiments, and the foregoing and other operations and / or functions of each module in the device correspond to the corresponding processes in the various methods in the above method embodiments, which will not be repeated here for the sake of brevity.

[0120] The apparatus of this disclosure embodiment has been described above from the perspective of functional modules in conjunction with the accompanying drawings. It should be understood that this functional module can be implemented in hardware, in software instructions, or in a combination of hardware and software modules. Specifically, the steps of the method embodiments in this disclosure can be completed by integrated logic circuits in the processor's hardware and / or by software instructions. The steps of the method disclosed in this disclosure embodiment can be directly embodied as being executed by a hardware decoding processor, or by a combination of hardware and software modules in the decoding processor. Optionally, the software module can reside in a mature storage medium in the art, such as random access memory, flash memory, read-only memory, programmable read-only memory, electrically erasable programmable memory, registers, etc. This storage medium is located in memory, and the processor reads information from the memory and, in conjunction with its hardware, completes the steps in the above method embodiments.

[0121] Figure 3 This is a schematic block diagram of an electronic device provided in an embodiment of this disclosure. The electronic device may include:

[0122] The system includes a memory 301 for storing computer programs and a processor 302 for transferring program code to the processor 302. In other words, the processor 302 can retrieve and run the computer program from the memory 301 to implement the methods described in this embodiment.

[0123] For example, the processor 302 can be used to execute the above-described method embodiments according to instructions in the computer program.

[0124] In some embodiments of this disclosure, the processor 302 may include, but is not limited to:

[0125] General-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc.

[0126] In some embodiments of this disclosure, the memory 301 includes, but is not limited to:

[0127] Volatile memory and / or non-volatile memory. Non-volatile memory can be read-only memory (ROM), programmable read-only memory (PROM), erasable programmable read-only memory (EPROM), electrically erasable programmable read-only memory (EEPROM), or flash memory. Volatile memory can be random access memory (RAM), used as an external cache. By way of example, but not limitation, many forms of RAM are available, such as Static RAM (SRAM), Dynamic RAM (DRAM), Synchronous DRAM (SDRAM), Double Data Rate SDRAM (DDR SDRAM), Enhanced SDRAM (ESDRAM), Synchronous Link DRAM (SLDRAM), and Direct Rambus RAM (DR RAM).

[0128] In some embodiments of this disclosure, the computer program may be divided into one or more modules, which are stored in the memory 301 and executed by the processor 302 to perform the method provided in this disclosure. The one or more modules may be a series of computer program instruction segments capable of performing a specific function, which describe the execution process of the computer program in the electronic device.

[0129] like Figure 3 As shown, the electronic device may also include:

[0130] Transceiver 303, which can be connected to processor 302 or memory 301.

[0131] The processor 302 can control the transceiver 303 to communicate with other devices; specifically, it can send information or data to other devices or receive information or data sent by other devices. The transceiver 303 may include a transmitter and a receiver. The transceiver 303 may further include antennas, and the number of antennas may be one or more.

[0132] It should be understood that the various components in the electronic device are connected through a bus system, which includes a data bus, a power bus, a control bus, and a status signal bus.

[0133] This disclosure also provides a computer storage medium storing a computer program thereon, which, when executed by a computer, enables the computer to perform the methods of the above-described method embodiments. Alternatively, embodiments of this disclosure also provide a computer program product containing instructions that, when executed by a computer, cause the computer to perform the methods of the above-described method embodiments.

[0134] When implemented using software, it can be implemented entirely or partially as a computer program product. This computer program product includes one or more computer instructions. When these computer program instructions are loaded and executed on a computer, all or part of the flow or function according to embodiments of this disclosure is generated. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, the computer instructions can be transmitted from one website, computer, server, or data center to another via wired (e.g., coaxial cable, fiber optic, digital subscriber line (DSL)) or wireless (e.g., infrared, wireless, microwave, etc.) means. The computer-readable storage medium can be any available medium accessible to a computer or a data storage device such as a server or data center that integrates one or more available media. The available medium can be a magnetic medium (e.g., floppy disk, hard disk, magnetic tape), an optical medium (e.g., digital video disc (DVD)), or a semiconductor medium (e.g., solid-state disk (SSD)).

[0135] Those skilled in the art will recognize that the modules and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this disclosure.

[0136] In the embodiments provided in this disclosure, it should be understood that the disclosed systems, apparatuses, and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of modules is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple modules or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection between apparatuses or modules may be electrical, mechanical, or other forms.

[0137] The modules described as separate components may or may not be physically separate. The components shown as modules may or may not be physical modules; 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. For example, the functional modules in the various embodiments of this disclosure may be integrated into one processing module, or each module may exist physically separately, or two or more modules may be integrated into one module.

[0138] The above are merely specific embodiments of this disclosure, but the scope of protection of this disclosure is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this disclosure should be included within the scope of protection of this disclosure. Therefore, the scope of protection of this disclosure should be determined by the scope of the claims.

Claims

1. A distribution network dispatching method, characterized in that, The method includes: Obtain power data sets of multiple users within a target area. The power data sets include time-aligned power consumption curves of known and unknown types of users within a preset time period. Tensor decomposition is used to process the power consumption curves in the power dataset to obtain multiple power curve feature vectors. Using the power curve feature vectors of known user types as cluster centers, the multiple power curve feature vectors are clustered to obtain multiple clustering results; Based on the multiple clustering results, the user type information of each user among the multiple users is determined. The user type information includes at least: working professionals, single users, elderly users, industrial users, and business users. Based on the user type information of each user among the multiple users, the scheduling plan information of the target area is determined; using tensor decomposition, the power consumption curves in the power dataset are processed to obtain multiple power curve feature vectors, including: The power curves in the power dataset are normalized, and multiple three-dimensional tensors corresponding to the processed power curves are constructed. Tensor decomposition is used to process the multiple three-dimensional tensors to obtain the multiple power curve feature vectors; the tensor decomposition includes CP decomposition and Tucker decomposition; The multiple three-dimensional tensors are processed using tensor decomposition to obtain the multiple power curve feature vectors, including: For any three-dimensional tensor among the plurality of three-dimensional tensors, the global feature vector of the three-dimensional tensor is extracted by CP decomposition, and the local feature vector of the three-dimensional tensor is extracted by Tucker decomposition. The local feature vector and the global feature vector are fused to obtain the power curve feature vector, and then the multiple power curve feature vectors are obtained.

2. The distribution network dispatching method according to claim 1, characterized in that, Using the power curve feature vectors of known user types as cluster centers, clustering is performed on the multiple power curve feature vectors to obtain multiple clustering results, including: For a first feature vector corresponding to any known type of user among the multiple power curve feature vectors, calculate multiple Euclidean distances between the first feature vector and the multiple power curve feature vectors corresponding to multiple unknown types of users; Users whose Euclidean distance is less than a first threshold among the multiple Euclidean distances are assigned to the cluster to which the user corresponding to the first feature vector belongs, thus obtaining a clustering result, and then obtaining multiple clustering results.

3. The distribution network dispatching method according to claim 2, characterized in that, Calculating multiple Euclidean distances between the first feature vector and multiple power curve feature vectors corresponding to the multiple unknown types of users includes: Based on the Euclidean distance formula, the first feature vector, and the multiple power curve feature vectors corresponding to the multiple unknown types of users, multiple Euclidean distances are obtained between the first feature vector and the multiple power curve feature vectors corresponding to the multiple unknown types of users. The Euclidean distance formula is as follows: The Let be the eigenvalue of the power curve feature vector corresponding to the k-th user among the multiple known user types, in the i-th dimension. Let the eigenvalue of the power curve feature vector corresponding to the h-th user among the multiple unknown types of users be the eigenvalue of the i-th dimension. For feature dimensions, the The Euclidean distance is the power curve feature vector corresponding to the kth user among the plurality of known types of users and the power curve feature vector corresponding to the hth user among the plurality of unknown types of users.

4. The distribution network dispatching method according to claim 1, characterized in that, Using the power curve feature vectors of known user types as cluster centers, the method clusters the multiple power curve feature vectors to obtain multiple clustering results. The method further includes: Using the power curve feature vectors of known user types as cluster centers, the multiple power curve feature vectors are clustered to obtain multiple initial clustering results, and the number of iterations is set to 1. Based on the multiple initial clustering results, update the cluster centers in each clustering result, re-cluster, and increment the iteration count by 1; Determine whether the number of iterations is greater than the preset number of iterations. If yes, obtain the multiple clustering results. If no, continue to update the cluster centers in each clustering result and re-cluster.

5. A power distribution network dispatching device, characterized in that, The apparatus is configured to perform the method as described in any one of claims 1-4, the apparatus comprising: The acquisition unit is used to acquire the power data set of multiple users in the target area. The power data set includes the power consumption curves of known type users and unknown type users in a time sequence aligned within a preset time period. The decomposition unit is used to process the power consumption curves in the power dataset using tensor decomposition to obtain multiple power curve feature vectors. The clustering unit is used to cluster the multiple power curve feature vectors with the power curve feature vectors of known types of users as the cluster centers, and obtain multiple clustering results; The determining unit is used to determine the user type information of each user among the multiple users based on the multiple clustering results. The user type information includes at least: working users, single users, elderly users, industrial users, and business users. The determining unit is further configured to determine the scheduling plan information of the target area based on the user type information of each of the plurality of users.

6. An electronic device, characterized in that, include: processor; as well as Memory for storing the executable instructions of the processor; The processor is configured to execute the method of any one of claims 1-4 by executing the executable instructions.

7. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it implements the method described in any one of claims 1-4.

Images