Missing Load Recovery Method Based on Similar Block Chimerism and First-Order Polynomial Interpolation

Abstract

The present invention discloses a method for restoring missing load based on similar block chimerism and first-order polynomial interpolation. First, the load data of power users is collected and a third-order load tensor is constructed. According to the position of the missing load, a matching similar block offset is constructed, and the matching similar block offset is specifically the offset between the position of the missing load and the position of the matching similar block. The matching similar block is selected by the spreading rule and irregular search optimization, and the sum of the squared errors between the matching similar block and the data block at the missing load position is used as the selection criterion, and the optimal similar block is used to generate preliminary restored data. First-order polynomial interpolation is performed on the load data at the missing position of the unfolded tensor to construct auxiliary restored data. The preliminary restored data and the auxiliary restored data are combined to obtain the final result of restoring the missing load. The method proposed by the present invention helps to fully explore the characteristics of the load data itself, and has certain practical significance for realizing the restoration of randomly missing user load data.

Description

Technical Field

[0001] The present invention relates to the field of power big data mining, and particularly to a method for restoring missing load based on similar block chimerism and first-order polynomial interpolation. Background Art

[0002] The continuous construction of smart grids and the increase in the number of power consumers have led to the integration of a large number of intelligent measurement and real-time monitoring devices into the power network, generating huge amounts of data. These huge datasets generated by smart grids are widely used in aspects such as power system planning, power equipment status assessment, anomaly detection, fault diagnosis, and load forecasting. Accurate measurement data is the foundation for information mining and application. However, missing data is a common problem in power load data. In the distribution network and the power consumption side, due to complex distribution lines and a large user base, various factors such as equipment failures, transmission interference, and external environments may interfere during the collection, transmission, storage, and processing of user load data. These factors may cause load data to be lost or damaged, seriously impairing the integrity of load data. This will seriously affect the accuracy and effectiveness of the analysis of user load data, making a series of subsequent operations based on the analysis results unable to be accurately carried out, such as load forecasting, state estimation, operation planning, etc., thus seriously affecting the stable operation of the power system. Therefore, according to the characteristics of power user loads and the characteristic relationships of measurement data, effectively restoring missing load data is crucial for ensuring the safe and stable operation of the power system. Summary of the Invention

[0003] The purpose of the present invention is to provide a method for restoring missing load based on similar block chimerism and first-order polynomial interpolation in view of the deficiencies of the prior art. This method utilizes the local similarity and continuity existing in user load data, chimerizes the load at the missing position by matching similar blocks, and comprehensively combines the results of first-order polynomial interpolation, enabling comprehensive and efficient restoration and reconstruction of the missing load data.

[0004] The technical solution adopted by the present invention is as follows:

[0005] A method for restoring missing load based on similar block chimerism and first-order polynomial interpolation includes the following steps:

[0006] Step 1: Collect all user load data and construct a third-order load tensor based on the sampling moment, sampling date, and user. Take the sampling moment as the first mode, and expand the third-order load tensor according to the first mode to obtain an expanded tensor, highlighting the change trend and characteristics of the load in the time distribution.

[0007] Step 2: Based on the missing load position index, construct the matching similar block offset, where the matching similar block offset is specifically the offset between the missing load position and the matching similar block position; optimize the selection of the matching similar block through multiple iterations based on the spreading rule and irregular search, and use the sum of squared errors between the matching similar block and the data block at the missing load position as the selection criterion for the matching similar block; take the finally selected matching similar block as the optimal similar block, and use the optimal similar block as the preliminary recovery data for the missing load;

[0008] Step 3: Construct auxiliary recovery data for the missing position load data of the unfolded tensor based on first-order polynomial interpolation, and fuse the preliminary recovery data and the auxiliary recovery data to obtain the final missing load recovery results for all users.

[0009] In the above technical solution, further, the unfolding of the third-order load tensor according to the first mode to obtain the unfolded tensor is specifically as follows:

[0010]

[0011] In the formula, is the load matrix of user sample m, a represents the total number of sampling times, b represents the number of sampling days, and c represents the number of user samples; represents the unfolded tensor obtained by unfolding the third-order load tensor according to the sampling time. Among them, the load matrix can highlight the change trends and characteristics of the time information constructed by the data according to the sampling time in these dimensions, and retain the interaction information and correlation effects with the other two modes, which are reflected in the column distribution of the load matrix.

[0012] Further, the matching similar block offset in Step 2 is specifically as follows:

[0013]

[0014] In the formula, represents the matching similar block offset at the similar block (i, j), including two directions of x and y; dx and dy are respectively random variables independently sampled from the uniform distribution of [-p max , p max , and p max represents the maximum allowable offset.

[0015] Further, in Step 2, the optimization of the selection of the matching similar block through multiple iterations based on the spreading rule and irregular search, and using the sum of squared errors between the matching similar block and the data block at the missing load position as the selection criterion for the matching similar block is specifically as follows:

[0016]

[0017]

[0018]

[0019]

[0020] In the formula, err cur 、err left and err up They are the similarity errors of the current similar block, the similar block on its left, and the similar block above it based on the sum of squared errors; and are the missing load location data block and the current similarity block respectively. and They are Offset in x and y directions; and They are The offset in the x and y directions, and They are The offset in the x and y directions, and are the offsets of the similar block at (i, j) and its left similar block and the similar block above it. The above process borrows the preferred matching similar block offsets from adjacent similar blocks according to the spreading rule. I (i, j) represents the matching similar block offset corresponding to the minimum sum of square errors between the similar block at (i, j) and the missing load position data block, which is the transfer optimization result of the current matching relationship.

[0021] In order to avoid local optimum, I The matching block position corresponding to (i, j) is taken as the center, and an irregular search is performed in the radius area with exponential decay until the search radius becomes a similar block:

[0022]

[0023] Where r represents the maximum search radius; β represents the specified attenuation coefficient in [0,1]; τ represents a random number that obeys a two-dimensional uniform distribution; k represents the number of irregular searches; O II (i, j) represents the offset of the matching similar block at (i, j) after k irregular searches;

[0024] If the error decreases, use O II (i,j) replaces O I (i,j), otherwise keep O I (i, j), continue iterative search until the search radius becomes a similar block;

[0025] Take the matching similar block corresponding to the offset of the matching similar block with the minimum sum of squared errors during the entire iterative search process as the optimal similar block; and take the optimal similar block as the preliminary recovery data of the missing load; the calculation formula for the preliminary recovery data of the missing load is:

[0026]

[0027] In the formula, represents the preliminary recovery data of the missing load at (i, j); is the offset of the matching similar block with the minimum sum of squared errors during the entire iterative search process, and are respectively the offsets in the x and y directions.

[0028] Furthermore, the construction of the auxiliary recovery data for the missing position load data of the unfolded tensor based on the first-order polynomial interpolation in step 3 is specifically as follows:

[0029]

[0030] In the formula, w is the row index of the load missing position, ; v is the column index of the load missing position, ; p and q represent the row indices of the non-missing positions of the load, satisfying p < w < q; represents the auxiliary recovery data of the load data at the missing position (w, v) of the unfolded tensor;

[0031] Fuse the preliminary recovery data and the auxiliary recovery data to obtain the final missing load recovery results for all users, specifically as follows:

[0032]

[0033] In the formula, represents the final missing load recovery result, and are respectively the preliminary recovery data and the auxiliary recovery data of all users.

[0034] The present invention also provides a device for recovering the missing load data of power users based on the similar block chimerism technology and the first-order polynomial interpolation, which is used to implement the above method. The device includes:

[0035] An unfolded tensor acquisition module, which is used to collect the load data of all users and construct a three-order load tensor based on the sampling time, sampling date, and users, and expand the three-order load tensor according to the first mode with the sampling time as the first mode to obtain an unfolded tensor;

[0036] The preliminary recovery data acquisition module for missing load is used to construct a matching similar block offset based on the missing load position index, where the matching similar block offset is specifically the offset between the missing load position and the matching similar block position; through multiple iterations based on the spreading rule and irregular search to optimize the selection of the matching similar block, and using the sum of squared errors between the matching similar block and the data block at the missing load position as the selection criterion for the matching similar block; taking the finally selected matching similar block as the optimal similar block, and using the optimal similar block as the preliminary recovery data for the missing load.

[0037] The missing load recovery module is used to construct auxiliary recovery data for the missing position load data of the unfolded tensor based on first-order polynomial interpolation, and fuse the preliminary recovery data and the auxiliary recovery data to obtain the final missing load recovery results for all users.

[0038] The present invention also provides an electronic device, including: one or more processors; a memory for storing one or more programs; when the one or more programs are executed by the one or more processors, enabling the one or more processors to implement the missing load recovery method based on similar block chimerism and first-order polynomial interpolation.

[0039] The present invention also provides a computer-readable storage medium, on which computer instructions are stored, characterized in that the computer instructions are used to cause a computer to execute the missing load recovery method based on similar block chimerism and first-order polynomial interpolation.

[0040] The beneficial effects of the present invention are:

[0041] Aiming at the missing completion and recovery of user load data with missing values, the present invention proposes a missing load recovery method based on similar block chimerism and first-order polynomial interpolation. The present invention optimizes the selection through the spreading rule and irregular search, uses the sum of squared errors as the optimization selection criterion, and generates preliminary recovery data with the average value of the optimal similar block. Perform first-order polynomial interpolation on the load data at the missing position of the unfolded tensor to construct auxiliary recovery data, and combine the preliminary recovery data and the auxiliary recovery data with equal weights to output the final missing load recovery results. Compared with relying solely on the first-order polynomial interpolation algorithm, this method introduces the similar block chimerism technology, which can introduce more real texture details while maintaining the structural coherence of the load matrix; at the same time, the similar block chimerism can provide better edge-preserving characteristics to make up for the blurring effect generated by the first-order polynomial interpolation at the edge. This method can fully consider the internal relationships of user load data, such as local correlation, continuity and other characteristics, and has certain practical significance for more comprehensively and effectively completing the missing values of load data. Description of the Drawings

[0042] Figure 1This is a schematic diagram of the overall step - by - step process of the missing load recovery method based on similar block chimerism and first - order polynomial interpolation of the present invention.

[0043] Figure 2 This is a comparison of the load curve results of the method of the present invention and other methods for completing load data with different missing rates. Detailed implementation manners

[0044] In order to make the objectives, technical solutions and advantages of the present invention clearer and more understandable, the present invention will be described in depth and in detail below in conjunction with the accompanying drawings and implementation cases. It should be understood that the specific implementation cases described herein are only used to explain the present invention and are not used to limit the invention.

[0045] The present invention proposes a missing load recovery method based on similar block chimerism and first - order polynomial interpolation. As Figure 1 shown, according to a specific embodiment of the present invention, the implementation process of this method includes the following steps:

[0046] Step 1: Collect all user load data and construct a third - order load tensor based on sampling time, sampling date and users. Take the sampling time as the first mode, and expand the third - order load tensor according to the first mode to obtain an expanded tensor, highlighting the change trend and characteristics of the load in the time distribution. The specific implementation method of this step is as follows:

[0047]

[0048] In the formula, is the load matrix of user sample m, a represents the total number of sampling times, b represents the number of sampling days, and c represents the number of user samples; represents the expanded tensor obtained by expanding the third - order load tensor according to the sampling time.

[0049] Step 2: Based on the missing load position index, construct a matching similar block offset. The matching similar block offset is specifically the offset between the missing load position and the matching similar block position; through multiple iterations based on the spreading rule and irregular search to optimize the selection of the matching similar block, and use the sum of squared errors between the matching similar block and the data block at the missing load position as the selection criterion for the matching similar block; take the finally selected matching similar block as the optimal similar block, and use the optimal similar block as the preliminary recovery data of the missing load. Among them, the data block at the missing load position can specifically be composed of the missing load position and its 5 surrounding data (up, down, left, and right). Then the finally selected optimal similar block will include 5 data. At this time, the preliminary recovery data of the missing load can be obtained by calculating the average value of the optimal similar block (optimal similar block / 5). The specific implementation method of this step is as follows:

[0050]

[0051]

[0052]

[0053]

[0054]

[0055] wherein, represents the matching similar block offset at the similar block (i,j), including the x and y directions; dx and dy are respectively random variables sampled independently from the uniform distribution in [-p max , p max , and p max represents the maximum allowable offset; err cur , err left and err up are respectively the similarity errors based on the sum of squared errors of the current similar block, its left adjacent similar block, and its upper adjacent similar block; and are respectively the missing load position data block and the current similar block, and are respectively the offsets in the x and y directions; and are respectively the offsets in the x and y directions, and are respectively the offsets in the x and y directions, and are respectively the offsets between the similar block at (i,j) and its left adjacent similar block and its upper adjacent similar block; O I (i,j) represents the matching similar block offset corresponding to the minimum sum of squared errors between the similar block at (i,j) and the missing load position data block.

[0056] To prevent local optimality, an irregular search is performed within an exponentially decaying radius region centered at the matching similar block position corresponding to O I (i,j) until the search radius becomes one similar block:

[0057]

[0058] wherein, r represents the maximum search radius; β represents a specified decay coefficient within [0,1]; τ represents a random number obeying a two-dimensional uniform distribution; k represents the number of irregular search times; O II(i, j) represents the offset of the matching similar block corresponding to the matching similar block at (i, j) after k irregular searches;

[0059] If the error decreases, then use O II to replace O with (i, j) I (i, j), otherwise retain O I (i, j), and continue the iterative search until the search radius becomes one similar block;

[0060] Take the matching similar block corresponding to the offset of the matching similar block with the minimum sum of squared errors during the entire iterative search process as the optimal similar block; and use the optimal similar block as the preliminary recovery data for the missing load; The calculation formula for the preliminary recovery data of the missing load is:

[0061]

[0062] In the formula, represents the preliminary recovery data of the missing load at (i, j); is the offset of the matching similar block with the minimum sum of squared errors during the entire iterative search process, and are respectively the offsets in the x and y directions.

[0063] Step 3: Construct auxiliary recovery data for the missing position load data of the unfolded tensor based on first-order polynomial interpolation, and fuse the preliminary recovery data and the auxiliary recovery data to obtain the final missing load recovery results for all users. The specific implementation method of this step is as follows:

[0064] The construction of auxiliary recovery data for the missing position load data of the unfolded tensor based on first-order polynomial interpolation is specifically:

[0065]

[0066] In the formula, w is the row index of the load missing position, ; v is the column index of the load missing position, ; p and q represent the row indices of the non-missing positions of the load, satisfying p < w < q; represents the auxiliary recovery data of the load data at the missing position (w, v) of the unfolded tensor;

[0067] Fusing the preliminary recovery data and the auxiliary recovery data to obtain the final missing load recovery results for all users is specifically:

[0068]

[0069] In the formula, represents the final missing load recovery result, and are the preliminary recovery data and the auxiliary recovery data for all users respectively.

[0070] The following takes the load data of users in a certain area's power distribution station as an example for illustration:

[0071] To verify the effectiveness and superiority of the proposed missing load recovery method based on similar block chimerism and first-order polynomial interpolation, the load voltage data of users in a certain area's power distribution station from March 1st to 28th for a total of 28 days is selected as the complete original data sample. The data has a sampling interval of 15 minutes, and there are 96 sampling points per day. For the complete original load data, 5%, 10%, 15%, 20%, 25%, 30%, 35%, and 40% of the load data are randomly missing respectively to establish load data analysis samples with different missing rates.

[0072] Taking the user load data with different missing rates above as input samples, the load data completion results of the present invention are compared with those of the first-order polynomial interpolation method and the cubic spline interpolation method. Among them, the first-order polynomial interpolation connects two known points and estimates the value at any position between these two points using a linear equation; the cubic spline interpolation constructs a cubic polynomial curve between multiple known points to ensure that the curve is continuous and smooth at these points to estimate the value at any position.

[0073] Table 1 shows the root mean square error between the completion results of three different methods for load data with different missing rates and the complete load data. Table 2 shows the coefficient of determination between the completion results of three different methods for load data with different missing rates and the complete load data. Table 3 shows the mean absolute percentage error between the completion results of three different methods for load data with different missing rates and the complete load data. It can be seen that the present invention shows the best effect among the three error analysis indicators. The load data curve results before and after missing data recovery are as Figure 2 shown, where Figure 2 in (a), (b), and (c) are the comparison effects of the load recovered by the method proposed in the present invention, the first-order polynomial interpolation, and the cubic spline interpolation in the load data with the same missing degree respectively. It can be seen that the method proposed in the present invention is closer to the original load data curve in the overall trend, and the load data completion effect is better.

[0074] Table 1

[0075] Missing rate of load data First-order polynomial interpolation Cubic spline interpolation Method proposed by the present invention 5% 0.7078 1.1517 0.7015 10% 0.7223 1.3378 0.7152 15% 0.7318 4.6441 0.7241 20% 0.7233 2.4270 0.7161 25% 0.7330 1.8353 0.7257 30% 0.7698 3.2275 0.7603 35% 0.7792 3.4881 0.7690 40% 0.7800 6.6110 0.7700

[0076] Table 2

[0077] Missing rate of load data First-order polynomial interpolation Cubic spline interpolation Method proposed by the present invention 5% 0.8774 0.7264 0.8784 10% 0.8711 0.6583 0.8722 15% 0.8682 0.1408 0.8692 20% 0.8711 0.3660 0.8722 25% 0.8672 0.5019 0.8682 30% 0.8554 0.2447 0.8569 35% 0.8523 0.2143 0.8539 40% 0.8516 0.0721 0.8531

[0078] Table 3

[0079] Missing rate of load data <![CDATA[First-order polynomial interpolation (×10 -3 )]]> <![CDATA[Cubic Spline Interpolation (×10 -3 )]]> <![CDATA[The method proposed by the present invention (×10 -3 )]]> 5% 2.043 2.758 2.041 10% 2.080 2.860 2.072 15% 2.104 3.202 2.097 20% 2.091 3.176 2.085 25% 2.123 3.174 2.116 30% 2.185 3.582 2.176 35% 2.212 3.789 2.203 40% 2.232 4.016 2.222

[0080] An embodiment of the present invention further provides a power user missing load data recovery device based on similar block chimerism technology and first-order polynomial interpolation for implementing the above method. The device includes:

[0081] An unfolded tensor acquisition module, configured to collect all user load data and construct a third-order load tensor based on the sampling time, sampling date, and user, and use the sampling time as the first mode to unfold the third-order load tensor according to the first mode to obtain an unfolded tensor;

[0082] A preliminary recovery data acquisition module for missing loads, configured to construct a matching similar block offset based on the missing load position index. The matching similar block offset is specifically the offset between the missing load position and the matching similar block position; optimize the selection of the matching similar block through multiple iterations based on the spreading rule and irregular search, and use the sum of the squared errors between the matching similar block and the missing load position data block as the selection criterion for the matching similar block; use the finally selected matching similar block as the optimal similar block, and use the optimal similar block as the preliminary recovery data for the missing load;

[0083] A missing load recovery module, configured to construct auxiliary recovery data for the missing position load data of the unfolded tensor based on first-order polynomial interpolation, and fuse the preliminary recovery data and the auxiliary recovery data to obtain the final missing load recovery results of all users.

[0084] An embodiment of the present invention further provides an electronic device, including: one or more processors; a memory for storing one or more programs; when the one or more programs are executed by the one or more processors, the one or more processors implement the missing load recovery method based on similar block chimerism and first-order polynomial interpolation.

[0085] An embodiment of the present invention further provides a computer-readable storage medium, on which computer instructions are stored, and characterized in that the computer instructions are used to cause a computer to execute the missing load recovery method based on similar block chimerism and first-order polynomial interpolation.

[0086] The above has introduced in detail the missing load recovery method based on similar block chimerism and first-order polynomial interpolation provided by the present invention. Specific examples are used in the present invention to elaborate on the principle and implementation manner of the present invention. The description of the above embodiments is only used to help understand the method and its core idea of the present invention. It should be noted that for those of ordinary skill in the art in this technical field, without departing from the principle of the present invention, several improvements and modifications can be made to the present invention, and these improvements and modifications also fall within the protection scope of the claims of the present invention.

Claims

1. A missing load recovery method based on similar block chimerism and first-order polynomial interpolation, characterized in that Including the following steps: Step 1: Collect all user load data and construct a third-order load tensor based on the sampling moment, sampling date, and user. Take the sampling moment as the first mode, and expand the third-order load tensor according to the first mode to obtain an expanded tensor; Step 2: Based on the missing load position index, construct a matching similar block offset, where the matching similar block offset is specifically the offset between the missing load position and the matching similar block position; optimize the selection of the matching similar block through multiple iterations based on the spreading rule and irregular search, and use the sum of squared errors between the matching similar block and the data block at the missing load position as the selection criterion for the matching similar block; take the finally selected matching similar block as the optimal similar block, and use the optimal similar block as the preliminary recovery data for the missing load; Step 3: Construct auxiliary recovery data for the missing position load data of the expanded tensor based on first-order polynomial interpolation, and fuse the preliminary recovery data and the auxiliary recovery data to obtain the final missing load recovery results for all users.

2. The method for missing load recovery based on similar block chimerism and first-order polynomial interpolation according to claim 1, wherein The specific method of expanding the third-order load tensor according to the first mode in Step 1 to obtain an expanded tensor is: ； In the formula, is the load matrix of the user sample m, a represents the total number of sampling times, b represents the number of sampling days, and c represents the user sample size; represents the unfolded tensor obtained by unfolding the third-order load tensor according to the sampling time.

3. The method for restoring missing load based on similar block fitting and first-order polynomial interpolation according to claim 2, wherein The specific value of the matching similar block offset in Step 2 is: ； wherein, represents the matching similar block offset at the similar block (i, j), including the x and y directions; dx and dy are respectively independent random variables sampled from a uniform distribution of max , p max , and p max represents the maximum allowable offset.

4. The method for restoring missing load based on similar block chimerism and first-order polynomial interpolation according to claim 3, characterized in that, In Step 2, the method of optimizing the selection of the matching similar block through multiple iterations based on the spreading rule and irregular search, and using the sum of squared errors between the matching similar block and the data block at the missing load position as the selection criterion for the matching similar block is specifically: ； ； ； ； where err cur , err left and err up are the similarity errors based on the sum of squared errors of the current similar block, its left similar block, and its upper similar block, respectively; and are the missing load position data block and the current similar block, respectively, and are the offsets in the x and y directions, respectively; and are the offsets in the x and y directions, respectively, and are the offsets in the x and y directions, respectively, and are the offsets of the similar block at (i, j) from its left similar block and upper similar block, respectively; O I (i, j) represents the offset of the matching similar block corresponding to the minimum sum of squared errors between the similar block at (i, j) and the missing load position data block; To prevent local optimality, taking O I (i, j) as the center of the corresponding matching similar block position, perform irregular search within the exponentially decaying radius region until the search radius becomes a similar block: ； Wherein, r represents the maximum search radius; β represents a specified attenuation coefficient within [0, 1]; τ represents a random number subject to a two-dimensional uniform distribution; k represents the number of irregular searches; O II (i, j) represents the offset of the matching similar block corresponding to the matching similar block at (i, j) after k irregular searches; If the error decreases, then use O II to replace O with (i, j) I (i, j), otherwise retain O I (i, j), and continue iterative search until the search radius becomes a similar block; Take the matching similar block corresponding to the matching similar block offset with the minimum sum of squared errors in the entire iterative search process as the optimal similar block; and use the optimal similar block as the preliminary recovery data for the missing load; the calculation formula for the preliminary recovery data of the missing load is: ； Wherein, represents the preliminary recovery data of the missing load at (i,j); is the offset of the matching similar block with the minimum sum of squared errors in the entire iterative search process, and are respectively the offsets in the x and y directions.

5. The method for missing load recovery based on similar block chimerism and first-order polynomial interpolation according to claim 4, characterized in that In Step 3, the method of constructing auxiliary recovery data for the missing position load data of the expanded tensor based on first-order polynomial interpolation is specifically: ； where w is the row index of the load missing position, ; v is the column index of the load missing position, ; p and q represent the row indices of the non-missing load positions, satisfying p < w < q; represents the auxiliary recovery data of the load data at the missing position (w, v) of the unfolded tensor; The method of fusing the preliminary recovery data and the auxiliary recovery data to obtain the final missing load recovery results for all users is specifically: ； wherein, represents the final missing load restoration result, and are respectively the preliminary restoration data and the auxiliary restoration data of all users.

6. A power user missing load data recovery device based on similar block chimerism technology and first-order polynomial interpolation, characterized in that, For implementing the method according to any one of claims 1-5, the device includes: An expanded tensor acquisition module, configured to collect all user load data and construct a third-order load tensor based on the sampling moment, sampling date, and user. Take the sampling moment as the first mode, and expand the third-order load tensor according to the first mode to obtain an expanded tensor; A preliminary recovery data acquisition module for the missing load, configured to construct a matching similar block offset based on the missing load position index, where the matching similar block offset is specifically the offset between the missing load position and the matching similar block position; optimize the selection of the matching similar block through multiple iterations based on the spreading rule and irregular search, and use the sum of squared errors between the matching similar block and the data block at the missing load position as the selection criterion for the matching similar block; take the finally selected matching similar block as the optimal similar block, and use the optimal similar block as the preliminary recovery data for the missing load; A missing load recovery module, which is used to construct auxiliary recovery data for the missing position load data of the unfolded tensor based on first-order polynomial interpolation, and fuse the preliminary recovery data and the auxiliary recovery data to obtain the final missing load recovery results of all users.

7. An electronic device, characterized in that, Comprising: One or more processors; A memory for storing one or more programs; When the one or more programs are executed by the one or more processors, the one or more processors implement the method according to any one of claims 1-5.

8. A computer-readable storage medium, characterized in that, Stored thereon are computer instructions, characterized in that the computer instructions are used to cause a computer to execute the steps of the method according to any one of claims 1-5.

Images