Non-intrusive resident load identification method based on MTF-Vit algorithm and related equipment
By using the MTF-Vit algorithm in residential load identification, the load data is dimensioned into RGB images, and the image model training is used by Vision Transformer, the problem of feature loss in the existing technology is solved, and the accuracy of load identification is significantly improved.
Patent Information
- Application Number
- CN202411650419.1
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-11-19
- Publication Date
- 2025-03-11
AI Technical Summary
The existing non-invasive load identification methods have feature loss in the data pre-operation and conditional convolution feature extraction process, which affects the accuracy of load identification.
Using the method based on MTF-Vit algorithm, load data is used to generate RGB images through the Markov conversion field dimensionality upgrade, and image model training is used for Vision Transformer network structure to achieve load identification.
By retaining all the features of the original data and classifying the image model, the accuracy of residents' load identification is improved, reaching 94.7%.
Smart Images
Figure CN119600345A_ABST
Abstract
Description
Technical Field
[0001] The present application relates to the technical field of resident load identification, and in particular to a non-intrusive resident load identification method and related equipment based on the MTF-Vit algorithm. Background Art
[0002] The existing non-intrusive load identification method using conditional convolution feature extraction model adopts feature extraction method of data pre-operation and conditional convolution, which will cause certain feature loss of the original data, making some original data features unable to be better integrated into the subsequently generated model, thereby affecting the accuracy of the overall model load identification.
[0003] Therefore, there is an urgent need for a new resident load identification method to improve the accuracy of the model in identifying resident loads. Summary of the invention
[0004] The embodiments of the present application provide a non-intrusive residential load identification method and related equipment based on the MTF-Vit algorithm to solve the technical problems existing in the prior art.
[0005] Other features and advantages of the present application will become apparent from the following detailed description, or may be learned in part by the practice of the present application.
[0006] According to a first aspect of an embodiment of the present application, a non-intrusive residential load identification method based on an MTF-Vit algorithm is provided, characterized in that it includes:
[0007] The load data is subjected to a dimension-upgrading operation through a Markov transformation field to generate an RGB image;
[0008] Generate a data set for subsequent image model classification based on the RGB image;
[0009] Establish the Vision Transformer network structure;
[0010] Performing model training based on the Vision Transformer network structure and the data set to obtain an image model;
[0011] Load identification is performed based on the image model.
[0012] In some embodiments of the present application, based on the above scheme, the step of performing a dimension-upgrading operation on the load data through a Markov transformation field to generate an RGB image includes:
[0013] Obtain load data of different types of electrical appliances and classify the load data according to the types of electrical appliances;
[0014] Constructing a Markov transfer matrix based on the classified load data;
[0015] The Markov transfer matrix is expanded into a Markov transfer field to generate different types of RGB images.
[0016] In some embodiments of the present application, based on the aforementioned solution, generating a data set for subsequent image model classification based on the RGB image includes:
[0017] According to different types of electrical appliances, create folders named according to categories;
[0018] The generated RGB images of different types are placed in folders of corresponding types as data sets for subsequent image model classification.
[0019] In some embodiments of the present application, based on the above-mentioned solution, the establishment of the Vision Transformer network structure includes:
[0020] First, construct the Embedding layer, which is used to decompose the image into multiple small images;
[0021] Then construct the Encoder layer, which is used to perform convolution stacking processing on small blocks of images to complete the modeling of the image;
[0022] Finally, an MLP Head layer is constructed, and the MLP Head layer is used to classify the image so that the image can be classified after passing through the model.
[0023] In some embodiments of the present application, based on the aforementioned solution, the model training based on the Vision Transformer network structure and the data set includes:
[0024] The data set is input into the Vision Transformer network structure for iterative training, and the number of samples, number of iterations and learning rate used in each iteration are set.
[0025] In some embodiments of the present application, based on the above solution, the load identification based on the image model includes:
[0026] Obtain load data in real time;
[0027] The load data acquired in real time is input into the image model for recognition.
[0028] According to a second aspect of an embodiment of the present application, a non-intrusive residential load identification device based on an MTF-Vit algorithm is provided, comprising:
[0029] The first generating unit is used to perform a dimension-upgrading operation on the load data through a Markov transformation field to generate an RGB image;
[0030] A second generating unit, used to generate a data set for subsequent image model classification based on the RGB image;
[0031] Establish a unit for establishing the Vision Transformer network structure;
[0032] A training unit, used for performing model training based on the Vision Transformer network structure and the data set to obtain an image model;
[0033] An identification unit is used to perform load identification based on the image model.
[0034] According to a second aspect of an embodiment of the present application, a computer-readable storage medium is provided, wherein the storage medium stores computer instructions, and when the computer instructions are executed on a computer, the computer executes the method according to the first aspect.
[0035] According to a fourth aspect of the embodiments of the present application, there is provided an electronic device, including a memory and a processor;
[0036] The memory is used to store computer instructions;
[0037] The processor is used to call the computer instructions stored in the memory so that the electronic device executes the method as described in the first aspect.
[0038] The technical solution of this application uses the Markov transformation field to upgrade the data dimension, retain all data features and convert them into two-dimensional RGB images. Then, the image classification algorithm of Vision Transformer is used to classify the converted two-dimensional images in the iteratively trained Vision Transformer model to achieve the purpose of load identification, so that all features of the original data are involved in the establishment of the model, thereby improving the accuracy of the model in identifying resident loads.
[0039] It should be understood that the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the present application. BRIEF DESCRIPTION OF THE DRAWINGS
[0040] The drawings herein are incorporated into the specification and constitute a part of the specification, showing embodiments consistent with the present application, and together with the specification, are used to explain the principles of the present application. Obviously, the drawings described below are only some embodiments of the present application, and for ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work. In the drawings:
[0041] Figure 1 A flow chart of a non-intrusive residential load identification method based on the MTF-Vit algorithm according to an embodiment of the present application is shown;
[0042] Figure 2 A schematic diagram of a picture after the Markov transformation field is upgraded in dimension according to an embodiment of the present application is shown;
[0043] Figure 3 A block diagram of a non-intrusive residential load identification device based on the MTF-Vit algorithm according to an embodiment of the present application is shown;
[0044] Figure 4 A block diagram of an electronic device according to an embodiment of the present application is shown;
[0045] Figure 5 A schematic diagram of the structure of a computer system suitable for implementing an electronic device of an embodiment of the present application is shown. DETAILED DESCRIPTION
[0046] Example embodiments will now be described more fully with reference to the accompanying drawings. However, example embodiments can be implemented in a variety of forms and should not be construed as limited to the examples set forth herein; rather, these embodiments are provided so that this application will be more comprehensive and complete and fully convey the concept of the example embodiments to those skilled in the art.
[0047] In addition, described feature, structure or characteristic can be combined in one or more embodiments in any suitable manner. In the following description, many specific details are provided to provide a full understanding of the embodiments of the present application. However, those skilled in the art will appreciate that the technical scheme of the present application can be put into practice without one or more of the specific details, or other methods, components, devices, steps, etc. can be adopted. In other cases, known methods, devices, realizations or operations are not shown or described in detail to avoid blurring the various aspects of the application.
[0048] The block diagrams shown in the accompanying drawings are merely functional entities and do not necessarily correspond to physically independent entities. That is, these functional entities may be implemented in software form, or in one or more hardware modules or integrated circuits, or in different networks and / or processor devices and / or microcontroller devices.
[0049] The flowcharts shown in the accompanying drawings are only exemplary and do not necessarily include all the contents and operations / steps, nor must they be executed in the order described. For example, some operations / steps can be decomposed, and some operations / steps can be combined or partially combined, so the actual execution order may change according to actual conditions.
[0050] It should be noted that the “plurality” mentioned in this article refers to two or more.
[0051] In order to make the purpose, technical solution and advantages of the present invention clearer, the technical solution in the embodiment of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiment of the present invention. Obviously, the described embodiment is only a part of the embodiment of the present invention, not all the embodiments. Based on the embodiment of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.
[0052] Some embodiments of the present application will be described in detail below in conjunction with the accompanying drawings. In the absence of conflict, the following embodiments and features in the embodiments may be combined with each other.
[0053] See also Figure 1 , shows a flow chart of a non-intrusive residential load identification method based on the MTF-Vit algorithm according to an embodiment of the present application.
[0054] like Figure 1 As shown, a non-intrusive residential load identification method based on the MTF-Vit algorithm is presented, which specifically includes steps S100 to 500.
[0055] refer to Figure 1 , step S100, performing a dimension-upgrading operation on the load data through a Markov transformation field to generate an RGB image.
[0056] It can be understood that the dimensionality increase operation can convert the load data from one dimension to two dimensions, solving the problem of feature loss and deformation in the early data processing. The generated RGB image contains the features of all source data and participates in the subsequent modeling. It has obvious advantages over traditional data preprocessing and feature extraction algorithms.
[0057] In some feasible embodiments, based on the above scheme, the step of performing a dimension-upgrading operation on the load data through a Markov transformation field to generate an RGB image includes:
[0058] Obtain load data of different types of electrical appliances and classify the load data according to the types of electrical appliances;
[0059] Constructing a Markov transfer matrix based on the classified load data;
[0060] The Markov transfer matrix is expanded into a Markov transfer field to generate different types of RGB images.
[0061] It should be noted that, in the process of classifying load data, folders corresponding to the types of electrical appliances can be generated, and the corresponding types of load data can be input into the corresponding folders for storage, thereby achieving classification and facilitating subsequent use.
[0062] Continue to refer Figure 1 , step S200, generating a data set for subsequent image model classification based on the RGB image.
[0063] In some feasible embodiments, based on the above scheme, generating a data set for subsequent image model classification based on the RGB image includes:
[0064] According to different types of electrical appliances, create folders named according to categories;
[0065] The generated RGB images of different types are placed in folders of corresponding types as data sets for subsequent image model classification.
[0066] Continue to refer Figure 1 , step S300, establish the Vision Transformer network structure.
[0067] It should be noted that the Vision Transformer network structure in this application consists of an Embedding layer, an Encoder layer, and an MLP Head layer.
[0068] In some feasible embodiments, based on the above solution, the establishment of the Vision Transformer network structure includes:
[0069] First, construct the Embedding layer, which is used to decompose the image into multiple small images;
[0070] Then construct the Encoder layer, which is used to perform convolution stacking processing on small blocks of images to complete the modeling of the image;
[0071] Finally, an MLP Head layer is constructed, and the MLP Head layer is used to classify the image so that the image can be classified after passing through the model.
[0072] It should be noted that the Embedding layer refers to the embedding layer, the Encoder layer refers to the encoder layer, and the MLP Head layer refers to the classification layer.
[0073] It should be noted that, in this embodiment, the Vision Transformer network structure is adopted, and the encoder and decoder are applied to the field of image classification. Compared with traditional ResNet, CNN, etc., the image features can be better extracted to generate the model, and the accuracy of classification can be improved, thereby achieving the improvement of the accuracy of non-invasive resident load identification.
[0074] Continue to refer Figure 1 , step S400, performing model training based on the Vision Transformer network structure and the data set to obtain an image model.
[0075] In some feasible embodiments, based on the above-mentioned scheme, the data set is input into the VisionTransformer network structure for iterative training, and the number of samples, number of iterations and learning rate used in each iteration are set.
[0076] Continue to refer Figure 1 , step S500, performing load identification based on the image model.
[0077] In some feasible embodiments, based on the above solution, the load identification based on the image model includes:
[0078] Obtain load data in real time;
[0079] The load data acquired in real time is input into the image model for recognition.
[0080] For example, a specific example of this method is provided below, and the steps are as follows:
[0081] Step 1: Classify the different types of load data obtained, and put the same type of load data into five folders named in Chinese: air conditioner, electric kettle, electric water heater, induction cooker, and hair dryer, so as to facilitate the retrieval of data and generate images during subsequent dimensionality upgrade.
[0082] Step 2: Use MTF (Markov Transformation Field) to perform two-dimensional conversion on the residential load data, that is, convert the one-dimensional waveform data into a two-dimensional RGB image with all the data features, thereby solving the problem of feature loss caused by the traditional residential load waveform decomposition algorithm. The image encoding principle takes the time series X = (xt, t = 1, 2, ..., T) as an example, and the specific steps are as follows:
[0083] Step 2.1, divide the time series X(t) into Q bins, the bins are marked as 1, 2, ..., Q, and the amount of data in each bin is the same;
[0084] Step 2.2, change each data in the time series to the serial number of its corresponding quantile bin;
[0085] Step 2.3, construct the transfer matrix W, ω ij It represents the frequency of transfer from bin i to bin j:
[0086]
[0087] Step 2.4, construct the Markov transition field M:
[0088] The Markov transformation field image of the 9 feature point data generated above is as follows Figure 2 shown.
[0089] Step 3: Build the Vision Transformer network structure. The specific steps are as follows:
[0090] Step 3.1. Plan the network structure of Vision Transformer. Vision Transformer consists of three parts: Embedding layer, Encoder layer, and MLP Head layer (classification layer).
[0091] Step 3.2, build the Embedding layer. For the standard Transformer module, the required input is a vector sequence, that is, a two-dimensional matrix. The vector sequence of this patent has a length of 768 for each vector. For image data, its data format is [H, W, C], which is a three-dimensional matrix. Therefore, it is necessary to transform the data through an Embedding layer first. First, divide an image into a bunch of Patches of a given size. Divide the input image (224x224) into Patches of size 16x16. After division, (224 / 16)^2=196 Patches will be obtained. Then, each Patch is mapped to a one-dimensional vector through linear mapping. The shape of each Patche data is [16,16,3]. A vector of length 768 is obtained through mapping.
[0092] Step 3.3, build the Encoder layer. The Encoder layer is actually repeated stacking Encoder Block L times, which mainly consists of the following parts: Layer Norm, Multi-Head Attention, Dropout, MLP Block.
[0093] Step 3.4, build the MLP Head layer (classification layer). The output shape and input shape of the above Transformer Encoder remain unchanged. Taking ViT-B / 16 as an example, the input is [197,768] and the output is still [197,768]. Here we only need classification information, so we just need to extract the corresponding result generated by [class] token, that is, extract [1,768] corresponding to [class] token from [197,768]. Then we use MLP Head to get the final classification result.
[0094] Step 4: Input the image into the Vision Transformer network for iterative training. The specific steps are as follows:
[0095] Step 4.1. Configure the platform required for network operation. The platform used in this example is python3.7, pycharm2018, tensorflow2.0 and other operating environments.
[0096] Step 4.2: Set the program running parameters, batch_size to 8, number of iterations to 500, and learning rate to 0.001.
[0097] Step 4.3. Set the initial training model to ViT-B_16.h5. After starting training, save the final model into a folder.
[0098] Step 5: Use the final model to perform load identification.
[0099] The key point of the technical solution of this application is to identify the residential load through the MTF-Vit algorithm. First, the Markov transformation field is used to perform dimensionality increase operation to realize image encoding. This dimensionality increase algorithm is an important part of realizing load identification. It is also a more innovative and advantageous point compared with traditional data preprocessing and feature extraction algorithms. It solves the problem of feature loss and deformation in the early data processing; then the Vision Transformer is used to classify the image. The MTF-Vit algorithm not only completely retains all the features of the original data for modeling, but also uses the Vision Transformer to accurately classify the image. Compared with the traditional algorithm, the new method takes a different approach and does not model the one-dimensional original data. Instead, it identifies the load by classifying the image model after dimensionality increase. It not only retains all the original data features, but also improves the accuracy of load identification.
[0100] It should be noted that, in this application, MTF refers to Markov Transformation Field, and Vit refers to Vision Transformer.
[0101] In summary, the method proposed in this application has an accuracy of 94.7% in identifying 200 loads in the test set, which is a significant improvement over the 82% of traditional load identification algorithms DARTS, 87.6% of P-(DTW), and 91.4% of CNN, laying the foundation for the subsequent dimensionality-enhancing classification method.
[0102] The following describes an embodiment of the device of the present application, which can be used to implement a non-intrusive residential load identification method based on the MTF-Vit algorithm in the above embodiment of the present application. For details not disclosed in the embodiment of the device of the present application, please refer to the embodiment of the above method of the present application.
[0103] Reference Figure 3 As shown, according to an embodiment of the present application, a non-intrusive residential load identification device 300 based on the MTF-Vit algorithm includes:
[0104] The first generating unit 301 is used to perform a dimension-upgrading operation on the load data through a Markov transformation field to generate an RGB image;
[0105] A second generating unit 302, configured to generate a data set for subsequent image model classification based on the RGB image;
[0106] An establishing unit 303 is used to establish a Vision Transformer network structure;
[0107] A training unit 304 is used to perform model training based on the Vision Transformer network structure and the data set to obtain an image model;
[0108] The identification unit 305 is used to perform load identification based on the image model.
[0109] like Figure 4 As shown, an embodiment of the present application also provides an electronic device 400, including a memory 410, a processor 420, and a computer program 411 stored in the memory 410 and executable on the processor. When the processor 420 executes the computer program 411, the steps of the above-mentioned non-intrusive residential load identification method based on the MTF-Vit algorithm are implemented.
[0110] Since the electronic device introduced in this embodiment is a device used to implement a non-invasive residential load identification device based on the MTF-Vit algorithm in the embodiment of the present application, based on the method introduced in the embodiment of the present application, the technical personnel in this field can understand the specific implementation of the electronic device of this embodiment and its various variations. Therefore, how the electronic device implements the method in the embodiment of the present application is not introduced in detail here. As long as the equipment used by the technical personnel in this field to implement the method in the embodiment of the present application is within the scope of protection of this application.
[0111] During the specific implementation process, when the computer program 411 is executed by a processor, any implementation method in the embodiments corresponding to the first aspect can be implemented.
[0112] Figure 5 A schematic diagram of the structure of a computer system suitable for implementing an electronic device of an embodiment of the present application is shown.
[0113] It should be noted that Figure 5 The computer system 500 of the electronic device shown is only an example and should not bring any limitation to the functions and scope of use of the embodiments of the present application.
[0114] like Figure 5 As shown, the computer system 500 includes a central processing unit (CPU) 501, which can perform various appropriate actions and processes according to the program stored in the read-only memory (ROM) 502 or the program loaded from the storage part 508 to the random access memory (RAM) 503, such as executing the method described in the above embodiment. In the RAM 503, various programs and data required for system operation are also stored. The CPU 501, the ROM 502 and the RAM 503 are connected to each other through the bus 504. The input / output (I / O) interface 505 is also connected to the bus 504.
[0115] The following components are connected to the I / O interface 505: an input section 506 including a keyboard, a mouse, etc.; an output section 507 including a cathode ray tube (CRT), a liquid crystal display (LCD), etc., and a speaker; a storage section 508 including a hard disk, etc.; and a communication section 509 including a network interface card such as a LAN (Local Area Network) card, a modem, etc. The communication section 509 performs communication processing via a network such as the Internet. A drive 510 is also connected to the I / O interface 505 as needed. A removable medium 511, such as a magnetic disk, an optical disk, a magneto-optical disk, a semiconductor memory, etc., is installed on the drive 510 as needed so that a computer program read therefrom is installed into the storage section 508 as needed.
[0116] In particular, according to an embodiment of the present application, the process described above with reference to the flowchart can be implemented as a computer software program. For example, an embodiment of the present application includes a computer program product, which includes a computer program carried on a computer-readable medium, and the computer program includes a program code for executing the method shown in the flowchart. In such an embodiment, the computer program can be downloaded and installed from a network through a communication section 509, and / or installed from a removable medium 511. When the computer program is executed by a central processing unit (CPU) 501, various functions defined in the system of the present application are executed.
[0117] It should be noted that the computer-readable medium shown in the embodiment of the present application may be a computer-readable signal medium or a computer-readable storage medium or any combination of the above two. The computer-readable storage medium may be, for example, - but not limited to - an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device or device, or any combination of the above. More specific examples of computer-readable storage media may include, but are not limited to: an electrical connection with one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM), a flash memory, an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above. In the present application, a computer-readable storage medium may be any tangible medium containing or storing a program, which may be used by an instruction execution system, device or device or used in combination with it. In the present application, a computer-readable signal medium may include a data signal propagated in a baseband or as part of a carrier wave, wherein a computer-readable program code is carried. Such propagated data signals may take a variety of forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination of the above. Computer-readable signal media may also be any computer-readable medium other than computer-readable storage media, which may send, propagate, or transmit programs for use by or in conjunction with an instruction execution system, apparatus, or device. The program code contained on the computer-readable medium may be transmitted using any appropriate medium, including but not limited to: wireless, wired, etc., or any suitable combination of the above.
[0118] The flowchart and block diagram in the accompanying drawings illustrate the possible architecture, functions and operations of the system, method and computer program product according to various embodiments of the present application. Wherein, each box in the flowchart or block diagram can represent a module, a program segment, or a part of the code, and the above-mentioned module, program segment, or a part of the code contains one or more executable instructions for realizing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the box can also occur in a different order from the order marked in the accompanying drawings. For example, two boxes represented in succession can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each box in the block diagram or flowchart, and the combination of boxes in the block diagram or flowchart can be implemented with a dedicated hardware-based system that performs a specified function or operation, or can be implemented with a combination of dedicated hardware and computer instructions.
[0119] The units involved in the embodiments described in this application may be implemented by software or hardware, and the units described may also be set in a processor. The names of these units do not, in some cases, constitute limitations on the units themselves.
[0120] As another aspect, the present application further provides a computer program product or a computer program, the computer program product or the computer program including computer instructions, the computer instructions being stored in a computer-readable storage medium. A processor of a computer device reads the computer instructions from the computer-readable storage medium, and the processor executes the computer instructions, so that the computer device executes a non-intrusive residential load identification method based on the MTF-Vit algorithm described in the above embodiment.
[0121] As another aspect, the present application also provides a computer-readable medium, which may be included in the electronic device described in the above embodiment; or may exist independently without being assembled into the electronic device. The above computer-readable medium carries one or more programs, and when the above one or more programs are executed by an electronic device, the electronic device implements a non-intrusive residential load identification method based on the MTF-Vit algorithm described in the above embodiment.
[0122] It should be noted that, although several modules or units of the equipment for action execution are mentioned in the above detailed description, this division is not mandatory. In fact, according to the embodiments of the present application, the features and functions of two or more modules or units described above can be embodied in one module or unit. On the contrary, the features and functions of one module or unit described above can be further divided into being embodied by multiple modules or units.
[0123] Through the description of the above implementation methods, it is easy for those skilled in the art to understand that the example implementation methods described here can be implemented by software or by combining software with necessary hardware. Therefore, the technical solution according to the implementation methods of the present application can be embodied in the form of a software product, which can be stored in a non-volatile storage medium (which can be a CD-ROM, a USB flash drive, a mobile hard disk, etc.) or on a network, and includes several instructions to enable a computing device (which can be a personal computer, a server, a touch terminal, or a network device, etc.) to execute the method according to the implementation methods of the present application.
[0124] Those skilled in the art will readily appreciate other embodiments of the present application after considering the specification and practicing the embodiments disclosed herein. The present application is intended to cover any variations, uses or adaptations of the present application, which follow the general principles of the present application and include common knowledge or customary techniques in the art that are not disclosed in the present application. It should be understood that the present application is not limited to the precise structures described above and shown in the accompanying drawings, and various modifications and changes can be made without departing from the scope thereof. The scope of the present application is limited only by the appended claims.
Claims
1. A non-intrusive residential load identification method based on MTF-Vit algorithm, characterized in that: include: The load data is subjected to a dimension-upgrading operation through a Markov transformation field to generate an RGB image; Generate a data set for subsequent image model classification based on the RGB image; Establish the Vision Transformer network structure; Performing model training based on the Vision Transformer network structure and the data set to obtain an image model; Load identification is performed based on the image model.
2. The method according to claim 1, characterized in that The step of performing a dimension-up operation on the load data through a Markov transformation field to generate an RGB image includes: Obtain load data of different types of electrical appliances and classify the load data according to the types of electrical appliances; Constructing a Markov transfer matrix based on the classified load data; The Markov transfer matrix is expanded into a Markov transfer field to generate different types of RGB images.
3. The method according to claim 2, characterized in that The step of generating a data set for subsequent image model classification based on the RGB image comprises: According to different types of electrical appliances, create folders named according to categories; The generated RGB images of different types are placed in folders of corresponding types as data sets for subsequent image model classification.
4. The method according to claim 1, characterized in that: The Vision Transformer network structure is established, including: First, construct the Embedding layer, which is used to decompose the image into multiple small images; Then construct the Encoder layer, which is used to perform convolution stacking processing on small blocks of images to complete the modeling of the image; Finally, an MLP Head layer is constructed, and the MLP Head layer is used to classify the image so that the image can be classified after passing through the model.
5. The method according to claim 1, characterized in that The model training based on the Vision Transformer network structure and the data set includes: The data set is input into the Vision Transformer network structure for iterative training, and the number of samples, number of iterations and learning rate used in each iteration are set.
6. The method according to claim 1, characterized in that The load identification based on the image model includes: Obtain load data in real time; The load data acquired in real time is input into the image model for recognition.
7. A non-intrusive residential load identification device based on MTF-Vit algorithm, characterized in that: include: The first generating unit is used to perform a dimension-upgrading operation on the load data through a Markov transformation field to generate an RGB image; A second generating unit, used to generate a data set for subsequent image model classification based on the RGB image; Establish a unit for establishing the Vision Transformer network structure; A training unit, used for performing model training based on the Vision Transformer network structure and the data set to obtain an image model; An identification unit is used to perform load identification based on the image model.
8. A computer-readable storage medium, characterized in that: The storage medium stores computer instructions, and when the computer instructions are executed on a computer, the computer is enabled to execute the method according to any one of claims 1 to 6.
9. An electronic device, characterized in that: including memory and processor; The memory is used to store computer instructions; The processor is used to call the computer instructions stored in the memory so that the electronic device executes the method as described in any one of claims 1-6.