A method for power distribution data compression based on singular value decomposition
The distribution data is compressed and reconstructed through the singular value decomposition method, which solves the problem of large amount of distribution data transmission, realizes efficient data transmission and flexible processing of information loss, and meets the requirements of different applications.
Patent Information
- Application Number
- CN202210972390.3
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2022-08-15
- Publication Date
- 2024-11-08
- Estimated Expiration
- 2042-08-15
AI Technical Summary
The prior art is difficult to achieve high compression ratio in the transmission of power distribution data, while ensuring that information loss does not affect the requirements of target applications, and the existing communication network has limited data transmission volume.
The distribution data matrix is compressed by the singular value decomposition method, and the amount of data transmitted is reduced by retaining the singular value with a large value, and reconstructing it in the background to calculate the average absolute value error and the average percentage error to evaluate information loss. The number of singular values needs to be adjusted to meet the balance of compression ratio and information loss.
It realizes the transmission of a larger amount of power distribution data in the existing communication infrastructure, flexibly handles the relationship between compression ratio and information loss, meets the information loss requirements of different applications, and improves data transmission efficiency.
Smart Images

Figure CN115914407B_ABST
Abstract
Description
Technical Field
[0001] The invention belongs to the technical field of data compression, and relates to a method for compressing power distribution data, in particular to a method for compressing power distribution data based on singular value decomposition. Background Art
[0002] The development of metering system infrastructure and the continuous deployment of smart meters have resulted in a large amount of power distribution and consumption data being generated every day. This data can help power companies perform subsequent advanced applications such as topology identification, power theft analysis, and coordinated scheduling. However, this goal is constrained by the limitations of communication networks. Many different data compression techniques have been proposed by some scholars to reduce the amount of data transmission. In smart grid computing and control, some data compression methods are used for power quality analysis, aiming to compress the electrical signal waveform related to the transient response after a given system disturbance. M. Ringwelski, C. Renner, A. Reinhardt, A. Weigel, and V. Turau, "The Hitchhiker's guide to choosing the compression algorithm for your smart meter data," in Proc. IEEE Int. Energy Conf. Exhibit., Florence, Italy, 2012, pp. 935–940 proposed a lossless compression algorithm for smart meters and compared it with existing algorithms. In terms of compression ratio, the best performance is achieved by the Lempel–Ziv–Markov chain Huffman coding algorithm, resulting in average compression ratios ranging from 4:1 to 20:1. Compression methods tailored to the requirements of load curve data transmission in smart metering are shown in A. Unterweger and D. Engel, “Resumable load data compression in smart grids,” IEEE Trans. Smart Grid, vol. 6, no. 2, pp. 919–929, Mar. 2015. S. Brindha and D. Sundararajan, “Power quality monitoring and compression using the discrete wavelet transform,” in Proc. Int. Conf. Adv. Comput. Commun. Syst., Coimbatore, India, 2013, pp. 1–6 use biorthogonal filters to compress power system waveforms, achieving compression ratios up to 8:1.J. Ning, J. Wang, W. Gao, and C. Liu, “A wavelet-based data compression technique for smart grid,” IEEE Trans. Smart Grid, vol. 2, no. 1, pp. 212–218, Mar. 2011, used multi-scale analysis based on wavelet transform to perform compression of interference signals in smart grid environments, achieving a compression ratio of 5.4:1.
[0003] Most of the works found in the literature show relatively good data compression results, however, considering the amount of data available for transmission and storage in the power distribution system, a higher compression ratio may be required. In addition, it is also necessary to consider whether the information loss after data compression is suitable for the requirements of the target application and to handle the relationship between the compression ratio and the reconstruction information loss more flexibly.
[0004] After searching, no patent documents of the prior art that are identical or similar to the present invention were found. Summary of the invention
[0005] The purpose of the present invention is to solve the deficiencies of the prior art and to provide a method for compressing power distribution data based on singular value decomposition, which can solve the problem of excessive transmission volume of power distribution data.
[0006] The present invention solves the technical problem by the following technical solutions:
[0007] A method for compressing power distribution data based on singular value decomposition, characterized in that it comprises the following steps:
[0008] Step 1: The power distribution and utilization data of m different meters that need to be transmitted through the communication network within a time period of t sampling points are used as the data to be compressed, and a matrix X of the data to be compressed is established.
[0009]
[0010] Among them, X 11 It represents the power distribution data value obtained by the first meter at the first sampling time, X t1 represents the power distribution data value obtained by the first meter at the tth sampling time, X 1m represents the power distribution data value obtained by the mth meter at the first sampling time, X tm It represents the power distribution data value obtained by the mth meter at the tth sampling time;
[0011] Step 2: compress the data matrix X to be compressed established in step 1 based on singular value decomposition, and transmit it to the background through the communication network of the power distribution system;
[0012] Step 3, recover the compressed data transmitted in step 2 in the background to obtain reconstructed data X';
[0013] Step 4: Calculate the mean absolute error MAE and the mean percentage error MPE to evaluate the information loss of the reconstructed data X' obtained in step 3 compared to the original data X in step 1.
[0014]
[0015]
[0016] Where X(i,j) and X'(i,j) are the elements at the i-th row and j-th column of X and X' respectively, t is the sampling number of a single meter, and m is the number of meters that need to transmit power distribution data through the communication network;
[0017] Step 5, check whether the information loss after data compression obtained in step 4 is suitable for the target application. If it is unacceptable, increase the number of retained singular values r and repeat steps 3 to 5 to perform data compression again.
[0018] Moreover, the method of compressing the to-be-compressed data matrix X established in step 1 based on singular value decomposition in step 2 and transmitting it to the background through the communication network of the power distribution system includes the following steps:
[0019] Step 2.1, by applying singular value decomposition, decompose the data matrix X into three matrices as follows
[0020] X=U∑V T
[0021] Among them, V T is the transposed matrix of V; U and V are orthogonal matrices, containing XX respectively T and X T The eigenvectors of X; ∑ is a diagonal matrix whose elements are the singular values, where the singular values are XX in descending order T or X T The square root of the eigenvalue of X; X T is the transposed matrix of X.
[0022] Step 2.2, only retain the first r singular values with larger values in the matrix ∑ to reduce the amount of data that needs to be transmitted;
[0023] The specific steps are as follows:
[0024] First, the number of singular values r that need to be retained is calculated from the target compression ratio (CR) according to the following formula to form a submatrix and
[0025]
[0026] The compression ratio is calculated by comparing the amount of data after compression to the amount of data before compression, indicating the degree of reduction in the amount of data before and after compression.
[0027] Submatrix∑ 11 ,U 11 ,U 21 ,V 11 and V 12 The position in the original matrix is shown as follows:
[0028]
[0029]
[0030]
[0031] Then, the submatrix ∑ 11 ,U 11 ,U 21 ,V 11 and V 12 Transfer back to the background to reconstruct X.
[0032] Furthermore, the method of recovering the compressed data transmitted in step 2 in the background to obtain the reconstructed data X' in step 3 specifically includes the following steps:
[0033] Only the first r singular values of the matrix ∑ are retained to obtain ∑', as shown in the following formula:
[0034]
[0035] And in step 2.1, replace ∑ with ∑' to restore the original matrix X. The specific method is to recover the original matrix X according to the submatrix ∑ transmitted to the background. 11 ,U 11 ,U 21 ,V 11 and V 12 , the reconstruction matrix X' is obtained by the following formula, which is used to approximate the data matrix X to be compressed.
[0036]
[0037] The advantages and beneficial effects of the present invention are:
[0038] This power distribution data compression method based on singular value decomposition can transmit a larger amount of power distribution data in the existing communication infrastructure. Through the mechanism of weighing the degree of data compression and the degree of information loss, considering whether the information loss after data compression is suitable for the requirements of the target application, the proposed method can more flexibly handle the relationship between compression ratio and reconstruction information loss. BRIEF DESCRIPTION OF THE DRAWINGS
[0039] Figure 1 is a flow chart of the method of the present invention;
[0040] Figure 2 This is a flow chart of the specific processing process of data compression and reconstruction of the present invention. DETAILED DESCRIPTION
[0041] The present invention is further described in detail below through specific examples. The following examples are only illustrative and not restrictive, and the protection scope of the present invention cannot be limited thereto.
[0042] A data compression method for intelligent power distribution system based on singular value decomposition technology, such as Figure 1 As shown, the following steps are included:
[0043] Step 1: Take the power distribution and utilization data of m different meters that need to be transmitted through the communication network within t sampling points as the target data, and establish the data matrix X to be compressed, as shown in formula (1).
[0044]
[0045] Among them, each column of X is the measurement value obtained by a given instrument at each sampling time, X tm It represents the power distribution data value obtained by the mth meter at the tth sampling time.
[0046] Step 2: compress the data matrix established in step 1 according to a given compression ratio, and transmit it to the background through the communication network of the power distribution system.
[0047] like Figure 2 As shown, the specific method of step 2 is as follows:
[0048] Step 2.1, by applying singular value decomposition, the data matrix X is decomposed into three matrices, as shown in formula (2).
[0049] X=U∑V T (2)
[0050] Among them, V T is the transposed matrix of V; and is an orthogonal matrix, containing XX T and X T The eigenvector of X; is a diagonal matrix whose elements are singular values, and the singular values are XX in descending order. T or X T The square root of the eigenvalue of X; X T is the transposed matrix of X.
[0051] Step 2.2, only retain the first r singular values with larger values in the matrix ∑ to reduce the amount of data that needs to be transmitted. The specific steps are as follows:
[0052] First, according to formula (3), the number of singular values r that need to be retained is calculated from the target compression ratio (CR) to form a submatrix and The compression ratio is calculated by dividing the amount of data after compression by the amount of data before compression, indicating the degree of reduction in the amount of data before and after compression. 11 ,U 11 ,U 21 ,V 11 and V 12 The position of in the original matrix is shown in formula (4). Then, the submatrix ∑ 11 ,U 11 ,U 21 ,V 11 and V 12 Transfer back to the background to reconstruct X.
[0053]
[0054]
[0055]
[0056]
[0057] Step 3: Restore the compressed data transmitted in step 2 in the background to obtain reconstructed data X'.
[0058] The specific steps of step 3 are as follows:
[0059] Only the first r singular values of ∑ are retained to obtain ∑', as shown in equation (5), and ∑' is used to replace ∑ in (2) to restore the original matrix X. The specific method is as follows: according to the submatrix ∑ transmitted to the background 11 ,U 11 ,U 21 ,V 11 and V 12 , we can get the reconstructed matrix X' by calculating formula (6), and use it to approximate the matrix X.
[0060]
[0061]
[0062] Step 4: Calculate the mean absolute error MAE and the mean percentage error MPE to evaluate the information loss of the reconstructed data X' obtained in step 3 compared to the original data X in step 1. The calculation methods are as shown in equations (7) and (8) respectively.
[0063]
[0064]
[0065] Where X(i,j) and X'(i,j) are the elements at the i-th row and j-th column of X and X' respectively.
[0066] Step 5: Check whether the information loss after data compression obtained in step 4 is suitable for the target application. If it is unacceptable, increase the number of retained singular values r and repeat steps 3 to 5 to perform data compression again.
[0067] Since in many practical applications, the number of singular values in descending order decreases rapidly, only a small number of singular values need to be retained to effectively compress data. The present invention compresses power distribution data based on the singular value decomposition method, takes the number of singular values to be retained as the only parameter that needs to be set, and proposes a scheme to explore different trade-offs between data CR and information loss. It is simple and effective, and can help communication infrastructure cope with the challenges of large-scale power distribution data transmission in smart grids.
[0068] Those skilled in the art will appreciate that the embodiments of the present application may be provided as methods, systems, or computer program products. Therefore, the present application may adopt the form of a complete hardware embodiment, a complete software embodiment, or an embodiment in combination with software and hardware. Moreover, the present application may adopt the form of a computer program product implemented in one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) that include computer-usable program code.
[0069] The present application is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.
[0070] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.
[0071] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.
[0072] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the above embodiments, ordinary technicians in the relevant field should understand that the specific implementation methods of the present invention can still be modified or replaced by equivalents. Any modification or equivalent replacement that does not depart from the spirit and scope of the present invention should be covered within the scope of protection of the claims of the present invention.
Claims
1. A method for compressing power distribution data based on singular value decomposition, characterized in that: The following steps are involved: Step 1: The power distribution and utilization data of m different meters that need to be transmitted through the communication network within a time period of t sampling points are used as the data to be compressed, and a matrix X of the data to be compressed is established. Among them, X 11 It represents the power distribution data value obtained by the first meter at the first sampling time, X t1 represents the power distribution data value obtained by the first meter at the tth sampling time, X 1m represents the power distribution data value obtained by the mth meter at the first sampling time, X tm It represents the power distribution data value obtained by the mth meter at the tth sampling time; Step 2: compress the data matrix X to be compressed established in step 1 based on singular value decomposition, and transmit it to the background through the communication network of the power distribution system; Step 3, recover the compressed data transmitted in step 2 in the background to obtain reconstructed data X'; Step 4: Calculate the mean absolute error MAE and the mean percentage error MPE to evaluate the information loss of the reconstructed data X' obtained in step 3 compared to the original data X in step 1. Where X(i,j) and X'(i,j) are the elements at the i-th row and j-th column of X and X' respectively, t is the sampling number of a single meter, and m is the number of meters that need to transmit power distribution data through the communication network; Step 5, check whether the information loss after data compression obtained in step 4 is suitable for the target application. If it is unacceptable, increase the number of retained singular values r and repeat steps 3 to 5 to perform data compression again.
2. The method for compressing power distribution data based on singular value decomposition according to claim 1, characterized in that: The method of step 2 compressing the to-be-compressed data matrix X established in step 1 based on singular value decomposition and transmitting it to the background through the communication network of the power distribution system includes the following steps: Step 2.1, by applying singular value decomposition, decompose the data matrix X into three matrices as follows X=UΣV T Among them, V T is the transposed matrix of V; U and V are orthogonal matrices, containing XX respectively T and X T The eigenvectors of X; Σ is a diagonal matrix whose elements are the singular values, where the singular values are XX in descending order T or X T The square root of the eigenvalue of X; X T is the transposed matrix of X; Step 2.2, only retain the first r singular values with larger values in the matrix Σ to reduce the amount of data that needs to be transmitted; The specific steps are as follows: First, the number of singular values r that need to be retained is calculated from the target compression ratio (CR) according to the following formula to form a submatrix and The compression ratio is calculated by comparing the amount of data after compression to the amount of data before compression, indicating the degree of reduction in the amount of data before and after compression. Submatrix∑ 11 ,U 11 ,U 21 ,V 11 and V 12 The position in the original matrix is shown as follows: Then, the submatrix ∑ 11 ,U 11 ,U 21 ,V 11 and V 12 Transfer back to the background to reconstruct X.
3. The method for compressing power distribution data based on singular value decomposition according to claim 2, characterized in that: The method of step 3 recovering the compressed data transmitted in step 2 in the background to obtain reconstructed data X' specifically includes the following steps: Only the first r singular values of the matrix Σ are retained to obtain Σ', as shown below: And in step 2.1, replace Σ with Σ' to restore the original matrix X. The specific method is to recover the original matrix X according to the submatrix Σ transmitted to the background. 11 ,U 11 ,U 21 ,V 11 and V 12 , the reconstruction matrix X' is obtained by the following formula, which is used to approximate the data matrix X to be compressed.
