A power transmission and transformation equipment digital twin data storage method and system

Abstract

The application discloses a power transmission and transformation equipment digital twin data storage method and system, comprising: dividing multi-modal state monitoring data into non-overlapping sampling blocks; generating basic recovery results and recovery residuals of the divided sampling blocks through a variable splitting recovery network; and determining the importance of each layer sample base by using a greedy algorithm to realize resampling and recovery of the multi-modal state monitoring data under different sampling rates. The application combines optimization theory and a convolutional neural network by means of a variable splitting technology, and a trained model can quickly solve different multi-modal data loss problems, effectively improving the accuracy and stability of power transmission and transformation equipment digital twin data; important sample bases are selected by using a greedy algorithm, the storage space of power transmission and transformation equipment multi-modal state monitoring data is effectively reduced, the effectiveness and economy of the digital twin model in a long time dimension are enhanced, and more accurate data support is provided for equipment monitoring.

Description

Technical Field

[0001] This invention relates to the field of digital twin technology, and in particular to a method and system for storing digital twin data of power transmission and transformation equipment. Background Technology

[0002] Digital twin technology for power transmission and transformation equipment has become an effective basis for equipment condition monitoring and operation and maintenance decisions. Digital twin technology utilizes real-time acquired visual images, IMU data, laser point clouds, and other multimodal data to construct virtual models of the equipment, enabling real-time monitoring and predictive maintenance of equipment status, providing strong support for more accurate operation and maintenance decisions. However, a major problem faced by existing digital twin systems in practical engineering applications is data storage. Compared to one-dimensional signals, the multimodal data acquired by digital twin technology uses two-dimensional images and three-dimensional point clouds, requiring significantly more storage space. Generally, the period for long-term equipment condition monitoring data is often several years or even decades, which is quite challenging for ordinary computer storage. Furthermore, because visual images are highly dependent on indoor or outdoor lighting conditions, data loss, such as noise and blurring, is inevitable. Noise is caused by environmental factors such as heavy rain and snowfall, as well as system factors such as camera or radar damage. Overexposure due to low shutter speed or insufficient light can cause blurring. Therefore, when further estimating equipment condition-sensitive features, the relevant solutions may be ill-conditioned and sensitive to errors caused by data loss.

[0003] Compressive sensing (CS) is a breakthrough in signal processing, capable of recovering signals at less than half the sampling rate, primarily based on the Nyquist sampling theorem. The basic idea of ​​CS is to first represent the signal in a sparse transform domain, and then recover the signal through an optimization process. CS can resample and recover one-dimensional state detection signals with considerable accuracy. Furthermore, the feasibility and efficiency of using CS to recover lost state monitoring signal data are relatively guaranteed. CS methods are divided into traditional methods and deep learning-based methods. Traditional CS methods recover complete 2D images from resampled 2D images by solving a sparse regularization optimization problem, which can be time-consuming for complex priors.

[0004] Therefore, this patent proposes a deep learning-based compressed sensing method, which aims to resample and recover multimodal state monitoring data by using compressed sensing technology. Summary of the Invention

[0005] In view of the aforementioned existing problems, the present invention is proposed.

[0006] Therefore, this invention provides a digital twin data storage method and system for power transmission and transformation equipment, which aims to solve the problems of model merging training calculation with different sampling rates and data loss of different types and levels by using compressed sensing technology to resample and recover multimodal state monitoring data.

[0007] To solve the above-mentioned technical problems, the present invention provides the following technical solution:

[0008] In a first aspect, the present invention provides a method for storing digital twin data of power transmission and transformation equipment, comprising: dividing multimodal state monitoring data into non-overlapping sampling blocks; generating basic recovery results and recovery residuals from the divided sampling blocks through a variable splitting recovery network; and using a greedy algorithm to determine the importance of each layer of sample bases in order to achieve resampling and recovery of the multimodal state monitoring data at different sampling rates.

[0009] As a preferred embodiment of the digital twin data storage method for power transmission and transformation equipment described in this invention, the step of dividing the multimodal state monitoring data into non-overlapping sampling blocks includes:

[0010] In variable-splitting scalable convolutional neural networks, a single-block sampling convolutional network is used to divide multimodal state monitoring data into blocks of size B. 2 ×l non-overlapping sampling blocks, where l represents the number of image channels and B represents the length of the vector;

[0011] The process of rewriting each sample block is embedded into a deep learning-based compressed sensing method by using convolutional layers with specific filter sizes and strides.

[0012] As a preferred embodiment of the digital twin data storage method for power transmission and transformation equipment described in this invention, the variable splitting and recovery network includes:

[0013] From the damaged observation data of the multimodal state monitoring data Recover clean image blocks I R (i,t), different types of data loss problems can be generated by specifying different corruption matrices;

[0014] When using optimization theory, the objective function for a clean image patch is:

[0015]

[0016] in, λ is the fidelity term, r(·) is the regularization term, and λ is a parameter used to control the relative weights between the fidelity term and the regularization term.

[0017] In deep learning-based methods, the relevant objective function is:

[0018]

[0019] Where Lo is the loss function and Θ is the prior parameter.

[0020] As a preferred embodiment of the digital twin data storage method for power transmission and transformation equipment described in this invention, the variable splitting and recovery network further includes:

[0021] To decouple the regularization and fidelity terms, and to adaptively address different types and levels of data loss through integration with HQS, an auxiliary variable w is introduced into HQS. The objective function for the clean image patch can be reformulated as:

[0022]

[0023] HQS attempts to minimize the following loss function:

[0024]

[0025] Where μ is a penalty parameter used to control The relative weights;

[0026] The loss function is solved using the following iterative formula:

[0027]

[0028] The regularization term and the fidelity term are decoupled into two separate subproblems. The fidelity term is a quadratic regularized least squares problem with the following solution:

[0029]

[0030] Where E is the identity matrix;

[0031] The regularization term can be restated as follows:

[0032]

[0033] As a preferred embodiment of the digital twin data storage method for power transmission and transformation equipment described in this invention, the architecture of the variable splitting and recovery network includes seven operations. The first operation consists of a dilated convolutional layer and a subsequent modified linear unit activation layer. The second to sixth operations each include a dilated convolutional layer, a batch normalization layer, and a ReLU activation layer. The last operation only includes a dilated convolutional layer.

[0034] As a preferred embodiment of the digital twin data storage method for power transmission and transformation equipment described in this invention, the step of using a greedy algorithm to determine the importance of each layer of sample bases includes:

[0035] Given a sampling rate S;

[0036] When the sampling rate S is less than the sampling rate S of the base layer or the h-th reinforcement layer h However, it is greater than the sampling rate S of the base layer or the (h-1)th reinforcement layer. h-1 It is necessary to identify and select the most important sample base.

[0037] As a preferred embodiment of the digital twin data storage method for power transmission and transformation equipment described in this invention, it is assumed that the h-th enhancement layer uses a sample basis set G, denoted as G = {1, 2, ..., Num}. G When certain sample bases and their connections to the recovery network are removed, the remaining samples should generate the most accurate recovery results possible, i.e., solve the following optimization problem:

[0038]

[0039] Among them, V j For a validation set, and I (h-1) These are the sample and recovery result of the (h-1)th enhancement layer, respectively. and These are the sample and recovery result of the h-th enhancement layer after removing samples from the subset Q of set G, respectively;

[0040] Greedy algorithms are used to determine and select a set of important sample bases so that the solution to the optimization problem can provide the highest average peak signal-to-noise ratio.

[0041] In a second aspect, the present invention provides a digital twin data storage system for power transmission and transformation equipment, comprising:

[0042] The sampling block partitioning module is used to divide the multimodal state monitoring data into non-overlapping sampling blocks;

[0043] The data recovery module is used to generate basic recovery results and recovery residuals from the divided sampling blocks through a variable splitting recovery network;

[0044] The resampling and recovery module is used to determine the importance of each layer of sample bases using a greedy algorithm, so as to achieve resampling and recovery of the multimodal state monitoring data at different sampling rates.

[0045] Thirdly, the present invention provides an electronic device, comprising:

[0046] Memory and processor;

[0047] The memory is used to store computer-executable instructions, and the processor is used to execute the computer-executable instructions. When the computer-executable instructions are executed by the processor, the steps of the method for storing digital twin data of power transmission and transformation equipment are implemented.

[0048] Fourthly, the present invention provides a computer-readable storage medium storing computer-executable instructions, which, when executed by a processor, implement the steps of the digital twin data storage method for power transmission and transformation equipment.

[0049] Compared with existing technologies, the beneficial effects of this invention are as follows: This invention provides a digital twin data storage method and system for power transmission and transformation equipment. By combining optimization theory with convolutional neural networks using variable splitting technology, the trained model can quickly solve different multimodal data loss problems, effectively improving the accuracy and stability of digital twin data for power transmission and transformation equipment. By selecting important sample bases through a greedy algorithm, a model can resample and recover multimodal state monitoring data at different sampling rates, effectively reducing the storage space of multimodal state monitoring data for power transmission and transformation equipment, enhancing the effectiveness and economy of the digital twin model over long periods, and providing more accurate data support for equipment monitoring. Attached Figure Description

[0050] To more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings used in the following description of the embodiments will be briefly introduced. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0051] Figure 1 This is a schematic diagram of the overall process logic of the digital twin data storage method for power transmission and transformation equipment according to an embodiment of the present invention;

[0052] Figure 2 This is a diagram of the variable splitting scalable convolutional neural network structure of the digital twin data storage method for power transmission and transformation equipment according to an embodiment of the present invention;

[0053] Figure 3 This is a network architecture diagram of the variable splitting and recovery method for digital twin data storage of power transmission and transformation equipment according to an embodiment of the present invention. Detailed Implementation

[0054] To make the above-mentioned objects, features, and advantages of the present invention more apparent and understandable, specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of the present invention, and not all of them. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the protection scope of the present invention.

[0055] Example 1

[0056] Reference Figures 1-3 As one embodiment of the present invention, a method for storing digital twin data of power transmission and transformation equipment is provided, such as... Figure 1 The specific steps shown are as follows:

[0057] S100: Divide the multimodal state monitoring data into non-overlapping sampling blocks;

[0058] S200: The divided sampling blocks are used to generate basic recovery results and recovery residuals through a variable splitting recovery network;

[0059] S300: A greedy algorithm is used to determine the importance of each layer of sample bases in order to achieve resampling and recovery of multimodal state monitoring data at different sampling rates.

[0060] It should be noted that this invention provides a method and system for storing digital twin data of power transmission and transformation equipment. By combining optimization theory with convolutional neural networks using variable splitting technology, the trained model can quickly solve different multimodal data loss problems, effectively improving the accuracy and stability of digital twin data of power transmission and transformation equipment. By selecting important sample bases through a greedy algorithm, a model can resample and recover multimodal state monitoring data at different sampling rates, effectively reducing the storage space of multimodal state monitoring data of power transmission and transformation equipment, enhancing the effectiveness and economy of the digital twin model in the long-term dimension, and providing more accurate data support for equipment monitoring.

[0061] Before introducing the specific implementation plan, the compressed sensing method will be introduced as follows:

[0062] According to the Nyquist sampling theorem, to recover the complete signal, the sampling rate of the sampled signal should be greater than half, and the sampled signal should contain at least half of the information of the complete signal. This is to avoid mode aliasing. Therefore, the CS method was proposed, which has the maximum sampling rate while still accurately recovering the complete signal. The mathematical theory is introduced as follows:

[0063] Assuming the length of the complete signal x is N, and the length of the sampled signal y is M, and M << N, the relationship between signals x and y is:

[0064] y = φx

[0065] Where Φ is a sampling matrix of size M×N, and S=M / N is the sampling rate;

[0066] The CS method assumes that x can be represented by a basis Ψ of size N×N, i.e. if If x has only K non-zero coefficients, and K << N, then x is K-sparse in the basis Ψ. If the matrix Φ satisfies the restricted isometric property, then it can be obtained from a matrix of length Ψ. To recover the K-sparse signal x from the sampled signal y, that is:

[0067]

[0068] When p is 0, it is used to characterize the sparsity of the signal; however, if it is equal to 0, solving the optimization problem becomes complex and a nondeterministic polynomial problem. Therefore, it is usually transformed into a simpler convex problem by setting p≤1, and if x is sufficiently sparse, accurate recovery may be achieved even when S is less than 0.5.

[0069] Resampling and reconstruction of multimodal data signals differs from that of one-dimensional signals, which are based on block-by-block processing. Therefore, in acquired multimodal data, the frame I(t) at time t to be resampled and reconstructed should be divided into non-overlapping blocks of size B×B. The i-th block at time t is denoted by I(I,t), where i = 1,...,n. B n B This refers to the number of blocks. In the calculation, I(I,t) is compressed into a block of length B. 2 The vector, and I R (i,t) is a vector M that has been compressed into a vector of length M. B The sampling blocks. Therefore, the sampling matrix is ​​a diagonal matrix:

[0070]

[0071] Where, Φ B A block I(I,t) of size M B ×B 2 The sampling matrix has a correlation sampling rate of S = M B / B 2 Therefore, the relationship between signals x and y can be rewritten in a block-by-block manner:

[0072] I R (i,t)=Φ B I(i,t)

[0073] Therefore, the relevant optimization problem for resampling and recovering multimodal information is:

[0074]

[0075] There are two approaches to solving the aforementioned optimization problem: traditional mathematical methods and deep learning-based methods. Traditional mathematical methods recover multimodal information by solving a sparse regularization optimization problem. Although traditional methods can achieve resampling and recovery at different sampling rates, their iterative nature leads to high computational burden. In contrast, deep learning methods based on convolutional neural networks (CNNs) can achieve acceptable accuracy with lower computational burden compared to traditional methods.

[0076] In this embodiment, addressing the problem of merging training computations and handling different types and levels of data loss in deep learning-based CS methods for models with different sampling rates, a deep learning CS method using a variable-splitting scalable convolutional neural network is proposed. This method leverages the advantages of optimization theory and integrates it as a modular component with deep learning-based CS methods. The variable-splitting scalable convolutional neural network is designed as a single-block sampling convolutional network plus a hierarchical recovery network. By using the single-block sampling convolutional network, multimodal state monitoring data is divided into non-overlapping sampling blocks. The hierarchical recovery network consists of a base layer (BL) and multiple enhancement layers (ELs). Figure 2 As shown, the sampling blocks generated by the block sampling convolutional network are treated as multi-band (MB) feature maps and divided into multiple groups marked with different colors. The base layer (BL) uses one set of sampling blocks to generate the basic reconstruction result. Each enhancement layer (EL) uses one set of sampling blocks to generate the reconstruction residual and references the results of the lower layers to improve the quality of the basic reconstruction. The variable splitting reconstruction network is specifically designed to address different types and degrees of data loss. All three reconstruction networks are supervised by the mean squared error (MSE) loss function. Finally, a greedy algorithm is used to determine the importance of the sample bases at each layer, and a model is used to select important sample bases for resampling and reconstruction at different sampling rates.

[0077] In this embodiment of the application, the step S100 of dividing the multimodal state monitoring data into non-overlapping sampling blocks includes:

[0078] In variable-splitting scalable convolutional neural networks, a single-block sampling convolutional network is used to divide multimodal state monitoring data into blocks of size B. 2 ×l non-overlapping sampling blocks, where l represents the number of image channels and is set to 1, and B represents the length of the vector;

[0079] By embedding the process of rewriting each sampling block into a deep learning-based compressed sensing method using convolutional layers with specific filter sizes and strides, the formula is expressed as:

[0080] I R (i,t)=F s *I(i,t)

[0081] Among them, F s It is M B A filter, which will have a size of B 2 Each block I(i,t) of length M is resampled. B Block I R (i,t), this block sampling convolutional network is represented as Conv(B) 2 ,l,M BThe convolutional layer is denoted as Conv(z,in,out) in the following description, where z is the spatial size of the filter, in is the number of input channels, and out is the number of output channels. To achieve non-overlapping resampling at a fixed sampling rate, the stride of this convolutional layer is B×B. The convolutional layer allows for optimization of the sampling convolutional network and subsequent recovery network to learn the sampling process.

[0082] It should be noted that step S100 above ensures that the data within each sampling block has a high degree of correlation and consistency, thereby reducing redundancy and interference between data and improving the efficiency and accuracy of data processing. Organized data segmentation not only facilitates parallel processing and accelerates computation but also provides structured input for subsequent steps such as the generation of variable decomposition and recovery networks, enabling the system to more accurately capture the inherent connections and change patterns between different modalities. Furthermore, non-overlapping sampling blocks help maintain the time-series characteristics of the data, ensuring accurate state monitoring and fault detection at different sampling rates and time scales, thus enhancing the stability and reliability of the entire system.

[0083] In this embodiment of the application, the variable splitting recovery network in step S200 above includes:

[0084] Damaged observation data from multimodal state monitoring data Recover clean image blocks I R (i,t):

[0085]

[0086] Here, H is a corruption matrix, and v is additive Gaussian noise with a standard deviation of σ; different types of data loss problems are generated by specifying different corruption matrices;

[0087] When using optimization theory, the objective function for a clean image patch is:

[0088]

[0089] in, Let r(·) be the fidelity term, r(·) be the regularization term, and λ be a parameter used to control the relative weights between the fidelity and regularization terms. The fidelity term ensures that the solution conforms to the associated damage condition, while the regularization term restricts the characteristics required for recovery. When applying optimization theory, the solution to the objective function of a clean image patch is usually based on time-consuming iterative estimation.

[0090] In deep learning-based methods, the relevant objective function is:

[0091]

[0092] Where Lo is the loss function and Θ is the prior parameter;

[0093] The aforementioned objective function can be restated as follows:

[0094]

[0095] The solution to the above formula is achieved by optimizing the loss function, using multimodal data containing both corrupted and clean data for training; however, the optimization method has complex priors, making it difficult to train with CNNs. By employing variable splitting techniques, such as Alternating Multiple-Pin Orientation Method (ADMM) and Hierarchical Semi-Quadratic Splitting (HQS), the regularization and fidelity terms in the optimization method can be decoupled, enabling the integration of optimization theory with CNNs.

[0096] To decouple the regularization and fidelity terms, and to adaptively address different types and levels of data loss through integration with HQS, an auxiliary variable w is introduced into HQS. The objective function for a clean image patch can be reformulated as:

[0097]

[0098] HQS attempts to minimize the following loss function:

[0099]

[0100] Where μ is a penalty parameter used to control The relative weights;

[0101] The loss function is solved using the following iterative scheme:

[0102]

[0103] Where k is the iteration index;

[0104] The regularization term and the fidelity term are decoupled into two separate subproblems. The fidelity term is a quadratic regularized least squares problem with the following solution:

[0105]

[0106] Where E is the identity matrix;

[0107] The regularization term can be restated as:

[0108]

[0109] To enable the CNN to learn the above regularization term, it can be restated as follows:

[0110]

[0111] Therefore, HQS provides a feasible solution that integrates optimization theory with CNN, namely, integrating the learned regularization term formula model with the fidelity term formula.

[0112] Furthermore, such as Figure 3 The architecture of the variable decomposition and recovery network shown comprises seven operations. The first operation consists of a dilated convolutional layer followed by a modified linear unit activation layer. Operations 2 through 6 each contain a dilated convolutional layer, a batch normalization layer, and a ReLU activation layer. The last operation contains only a dilated convolutional layer. The dilation factors of the 3×3 dilated convolutions from the first to the last operation are set to 1, 2, 3, 4, 3, 2, and 1, respectively. The equivalent receptive fields corresponding to each operation are 3, 5, 7, 9, 7, 5, and 3, respectively, and the receptive field of the entire network is 33×33.

[0113] It should be noted that step S200 not only effectively separates the main features and subtle differences in the data, ensuring high-fidelity recovery of the original signal, but also provides valuable foundational and supplementary information for subsequent analysis. By using a variable-splitting recovery network, the system can more accurately capture complex and changing data patterns, enhancing its ability to identify potential faults or anomalies. Furthermore, the generated basic recovery results and recovery residuals provide rich material for data analysis and model training at different levels, helping to further optimize the data processing flow, improve the overall system's response speed and decision-making accuracy, and thus play a crucial role in the real-time monitoring of power systems.

[0114] In this embodiment, step S300 uses a greedy algorithm to determine the importance of each layer of sample bases, so as to achieve resampling and recovery of multimodal state monitoring data at different sampling rates, including:

[0115] Given a sampling rate S;

[0116] When the sampling rate S is less than the sampling rate S of the base layer or the h-th reinforcement layer h However, it is greater than the sampling rate S of the base layer or the (h-1)th reinforcement layer. h-1 The most important sample base needs to be identified and selected. Since the base layer (BL) can be regarded as a specific reinforcement layer (EL), both the base layer (BL) and the reinforcement layer (EL) are uniformly labeled as the reinforcement layer (EL).

[0117] Suppose that the h-th enhancement layer uses a sample basis set G, denoted as G = {1, 2, ..., Num}. G When certain sample bases and their connections to the recovery network are removed, the remaining samples should generate the most accurate recovery results possible, i.e., solve the following optimization problem:

[0118]

[0119] Among them, V j For a validation set, and I (h-1) These are the sample and recovery result of the (h-1)th enhancement layer, respectively. and These are the sample and recovery result of the h-th enhancement layer after removing samples from the subset Q of set G, respectively;

[0120] Greedy algorithms are used to determine and select a set of important sample bases so that the solution to the optimization problem can provide the highest average peak signal-to-noise ratio.

[0121] It should be noted that the steps of the greedy algorithm are shown in Table 1.

[0122] Table 1: Greedy algorithm for sample base selection.

[0123]

[0124]

[0125] It should be noted that step S300 not only ensures the efficiency and accuracy of data recovery but also optimizes resource utilization, enabling the system to flexibly respond to various sampling rate requirements while maintaining data integrity and detail. Dynamically selecting the most important sample base enhances the system's adaptability, maintaining stable performance in complex and ever-changing monitoring environments. Step S300 effectively improves the overall efficiency and reliability of multimodal condition monitoring data processing, providing strong support for real-time monitoring and fault early warning of power systems, and ensuring the system's operational stability and security.

[0126] The above is an illustrative scheme of a digital twin data storage method for power transmission and transformation equipment according to this embodiment. It should be noted that the technical solution of this digital twin data storage system for power transmission and transformation equipment belongs to the same concept as the technical solution of the aforementioned digital twin data storage method for power transmission and transformation equipment. Details not described in detail in the technical solution of the digital twin data storage system for power transmission and transformation equipment in this embodiment can be found in the description of the technical solution of the aforementioned digital twin data storage method for power transmission and transformation equipment.

[0127] This embodiment also provides a digital twin data storage system for power transmission and transformation equipment, including:

[0128] The sampling block partitioning module is used to divide the multimodal state monitoring data into non-overlapping sampling blocks;

[0129] The data recovery module is used to generate basic recovery results and recovery residuals by dividing the sample blocks into a variable split recovery network;

[0130] The resampling and recovery module is used to determine the importance of each layer of sample basis using a greedy algorithm, so as to realize the resampling and recovery of multimodal state monitoring data under different sampling rates.

[0131] The above-mentioned unit modules can be embedded in the processor of the computer device in hardware form or independent of it, or they can be stored in the memory of the computer device in software form, so that the processor can call and execute the corresponding operations of the above modules.

[0132] This embodiment also provides an electronic device, which includes a processor, a memory, a communication interface, a display screen, and an input device connected via a system bus. The processor of this computer device provides computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and internal memory. The non-volatile storage medium stores an operating system and computer programs. The internal memory provides an environment for the operation of the operating system and computer programs in the non-volatile storage medium. The communication interface of the computer device is used for wired or wireless communication with external terminals. Wireless communication can be achieved through Wi-Fi, carrier networks, NFC (Near Field Communication), or other technologies. When the computer program is executed by the processor, it implements a digital twin data storage method for power transmission and transformation equipment. The display screen of the computer device can be a liquid crystal display screen or an e-ink display screen. The input device of the computer device can be a touch layer covering the display screen, or buttons, a trackball, or a touchpad mounted on the computer device casing, or an external keyboard, touchpad, or mouse, etc.

[0133] This embodiment also provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the method proposed in the above embodiments.

[0134] The storage medium proposed in this embodiment belongs to the same inventive concept as the method proposed in the above embodiments. Technical details not described in detail in this embodiment can be found in the above embodiments, and this embodiment has the same beneficial effects as the above embodiments.

[0135] Based on the above description of the implementation methods, those skilled in the art can clearly understand that the present invention can be implemented using software and necessary general-purpose hardware, and of course, it can also be implemented using hardware, but in many cases the former is a better implementation method. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as a computer floppy disk, read-only memory (ROM), random access memory (RAM), flash memory, hard disk, or optical disk, etc., including several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute the methods of the various embodiments of the present invention.

[0136] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention, and all such modifications or substitutions should be covered within the scope of the claims of the present invention.

[0137] Those skilled in the art will understand that embodiments of this application can be provided as methods, systems, or computer program products. Therefore, this application can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this application can take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code. The solutions in the embodiments of this application can be implemented in various computer languages, such as the object-oriented programming language Java and the interpreted scripting language JavaScript.

[0138] This application is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of this application. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart... Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0139] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0140] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0141] Although preferred embodiments of this application have been described, those skilled in the art, upon learning the basic inventive concept, can make other changes and modifications to these embodiments. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments as well as all changes and modifications falling within the scope of this application.

[0142] Obviously, those skilled in the art can make various modifications and variations to this application without departing from the spirit and scope of this application. Therefore, if such modifications and variations fall within the scope of the claims of this application and their equivalents, this application also intends to include such modifications and variations.

Claims

1. A method for storing digital twin data of power transmission and transformation equipment, characterized in that, include: The multimodal state monitoring data is divided into non-overlapping sampling blocks; The divided sampling blocks are used to generate basic recovery results and recovery residuals through a variable splitting recovery network; A greedy algorithm is used to determine the importance of each layer of sample bases in order to achieve resampling and recovery of the multimodal state monitoring data at different sampling rates; The process of dividing the multimodal state monitoring data into non-overlapping sampling blocks includes: In variable-splitting scalable convolutional neural networks, a single-block sampling convolutional network is used to divide multimodal state monitoring data into blocks of size [missing information]. The non-overlapping sampling blocks, where, l Indicates the number of image channels. B Indicates the length of the vector; The process of rewriting each sampling block is embedded into a deep learning-based compressed sensing method by using convolutional layers with specific filter sizes and strides. The variable decomposition recovery network includes: From the damaged observation data of the multimodal state monitoring data Recover clean image patches Different types of data loss problems can be generated by specifying different corruption matrices; When using optimization theory, the objective function for a clean image patch is: in, To ensure fidelity, For regularization terms, These are parameters used to control the relative weights between the fidelity term and the regularization term; In deep learning-based methods, the relevant objective function is: in, For loss function, These are prior parameters; The variable splitting recovery network also includes: To decouple the regularization and fidelity terms, and to adaptively address different types and levels of data loss through integration with HQS, an auxiliary variable w is introduced into HQS. The objective function for the clean image patch can be reformulated as: HQS attempts to minimize the following loss function: in, This is a penalty parameter used to control... The relative weights; The loss function is solved using the following iterative formula: The regularization term and the fidelity term are decoupled into two separate subproblems. The fidelity term is a quadratic regularized least squares problem with the following solution: Where E is the identity matrix; The regularization term can be restated as follows: 。 2. The method for storing digital twin data of power transmission and transformation equipment as described in claim 1, characterized in that, The architecture of the variable splitting and recovery network includes seven operations. The first operation consists of a dilated convolutional layer followed by a modified linear unit activation layer. Operations 2 through 6 each include a dilated convolutional layer, a batch normalization layer, and a ReLU activation layer. The last operation contains only a dilated convolutional layer.

3. The method for storing digital twin data of power transmission and transformation equipment as described in claim 2, characterized in that, The method of using a greedy algorithm to determine the importance of each layer of sample bases includes: Given a sampling rate S; When the sampling rate S is less than the sampling rate of the base layer or the h-th reinforcement layer However, it is greater than the sampling rate of the base layer or the (h-1)th reinforcement layer. It is necessary to identify and select the most important sample base.

4. The method for storing digital twin data of power transmission and transformation equipment as described in claim 3, characterized in that, Assume the h-th enhancement layer uses the sample basis set. G , recorded as When certain sample bases and their connections to the recovery network are removed, the remaining samples should generate accurate recovery results, i.e., the following optimization problem needs to be solved: in, For a validation set, and These are the sample and recovery result of the (h-1)th enhancement layer, respectively. and In the removal of the set respectively G subset of Q The sample after the first h Samples and recovery results of each enhancement layer; Greedy algorithms are used to determine and select a set of important sample bases so that the solution to the optimization problem can provide the highest average peak signal-to-noise ratio.

5. A digital twin data storage system for power transmission and transformation equipment, employing the digital twin data storage method for power transmission and transformation equipment as described in any one of claims 1 to 4, characterized in that, include: The sampling block partitioning module is used to divide the multimodal state monitoring data into non-overlapping sampling blocks; The data recovery module is used to generate basic recovery results and recovery residuals from the divided sampling blocks through a variable splitting recovery network; The resampling and recovery module is used to determine the importance of each layer of sample bases using a greedy algorithm, so as to achieve resampling and recovery of the multimodal state monitoring data at different sampling rates.

6. An electronic device, comprising: Memory and processor; The memory is used to store computer-executable instructions, and the processor is used to execute the computer-executable instructions. When the computer-executable instructions are executed by the processor, they implement the steps of the digital twin data storage method for power transmission and transformation equipment as described in any one of claims 1 to 4.

7. A computer-readable storage medium storing computer-executable instructions, which, when executed by a processor, implement the steps of the digital twin data storage method for power transmission and transformation equipment as described in any one of claims 1 to 4.

Images