Data processing method and device related to electrical device

Abstract

The embodiment of the invention provides a data processing method and device related to an electrical device, and relates to the technical field of electronic data processing, and the method comprises the steps: obtaining to-be-processed data, the to-be-processed data comprises a plurality of pieces of electrical data, and the electrical data is used for representing a certain physical property of the electrical device; on the basis of a preset sorting method, the multiple pieces of electrical data are sorted again, sorted data to be processed are obtained, and the preset sorting method is used for sorting adjacent electrical data neighbors; based on a preset compression method, performing compression processing on the sorted to-be-processed data to obtain compressed to-be-processed data; according to the compressed data to be processed, a processing result is obtained, and the processing result is used for representing the target physical property of the electrical device. According to the method, the memory consumption and the data calculation amount of the computing equipment can be reduced on the premise of ensuring the accuracy of the data processing result.

Description

Technical Field

[0001] This application relates to the technical field of electronic data processing, and in particular, to a data processing method and device related to electrical devices. Background Art

[0002] Electronic simulation is an engineering method that uses mathematical models and computer technology to simulate the actual behavior of electronic circuits or systems, aiming to verify circuit functions, optimize performance, and reduce the cost and time of physical experiments during the design stage. Electronic simulation technology can be applied in multiple fields such as circuit design, electromagnetic field analysis, and signal processing. By constructing a mathematical model of the circuit and using computer software (such as SPICE, Multisim, Proteus, etc.), the dynamic behavior of the circuit under different conditions can be simulated, and parameters such as voltage, current, and power can be analyzed without actually building a physical circuit. Among them, simulation data processing is the core link in the simulation process, involving multiple aspects such as data acquisition, cleaning, analysis, visualization, and management. That is to say, a large number of data processing tasks will be generated in the application of electronic simulation technology, such as the acquisition, cleaning, storage, analysis, and visualization of simulation data, and it may also involve the simulation data of different simulation software and the comparison of data processing processes.

[0003] Currently, the amount of simulation data generated during electronic simulation is large, and the amount of calculation increases geometrically. Correspondingly, the memory consumption of the computing device is also very large. In related technologies, the iterative method is usually used to reduce the amount of simulation data and then perform data processing. However, this method is likely to greatly reduce the accuracy of the processing result, and when the simulation data is poor, the iterative method is difficult to converge, and it is even impossible to guarantee the reliability of the finally obtained processing result.

[0004] Therefore, there is an urgent need for a data processing method and device related to electrical devices that can reduce the memory consumption of the computing device and the amount of data calculation on the premise of ensuring the accuracy of the data processing result. Summary of the Invention

[0005] The embodiments of this application provide a data processing method and device related to electrical devices, which can reduce the memory consumption of the computing device and the amount of data calculation on the premise of ensuring the accuracy of the data processing result.

[0006] In a first aspect, the embodiments of this application provide a data processing method related to electrical devices, and the method includes: Obtain data to be processed, where the data to be processed contains multiple electrical data, and the electrical data is used to characterize a certain physical property of an electrical device; Based on a preset sorting method, re-sort the multiple electrical data to obtain the sorted data to be processed, and the preset sorting algorithm is used to sort the adjacent electrical data in the vicinity; Based on a preset compression method, compress the sorted data to be processed to obtain the compressed data to be processed; According to the compressed data to be processed, obtain a processing result, where the processing result is used to characterize the target physical property of the electrical device. In a second aspect, an embodiment of the present application provides a data processing device related to an electrical device, including: An acquisition unit, configured to acquire data to be processed, where the data to be processed includes a plurality of electrical data, and the electrical data is used to characterize a certain physical property of the electrical device; A sorting unit, configured to re-sort the plurality of electrical data based on a preset sorting method to obtain the sorted data to be processed, where the preset sorting algorithm is used to sort adjacent electrical data in the vicinity; A first processing unit, configured to compress the sorted data to be processed based on a preset compression method to obtain the compressed data to be processed; A second processing unit, configured to obtain a processing result according to the compressed data to be processed, where the processing result is used to characterize the target physical property of the electrical device.

[0007] Optionally, the preset sorting method is KD-Tree. Based on the preset sorting method, the sorting unit is specifically configured to Perform index processing on the plurality of electrical data, and construct a KD-Tree according to the coordinate data of each of the plurality of electrical data; Based on the KD-Tree, re-sort the plurality of electrical data to obtain the sorted data to be processed.

[0008] Optionally, the plurality of electrical data in the sorted data to be processed is an electrical data matrix, and the preset compression method is a low-rank matrix method; the first processing unit is specifically configured to Adopt a low-rank matrix method to perform hierarchical decomposition processing on the electrical data matrix to obtain a plurality of sub-matrices; For each sub-matrix, perform compression processing using a corresponding adaptive compression threshold to obtain a corresponding compressed sub-matrix until the compressed sub-matrices corresponding to each of the plurality of sub-matrices form the compressed data to be processed.

[0009] Optionally, the first processing unit is specifically configured to If the sub-matrix has a symmetric feature, perform compression processing on the sub-matrix with the symmetric feature using a corresponding adaptive compression threshold to obtain a compressed sub-matrix of the sub-matrix with the symmetric feature; If the sub-matrix has an asymmetric feature, based on the adaptive cross approximation method, a corresponding adaptive compression threshold is used to compress the sub-matrix with the asymmetric feature, and a compressed sub-matrix of the sub-matrix without the symmetric feature is obtained.

[0010] Optionally, the first processing unit is specifically configured to, if the electrical data matrix is a symmetric matrix, based on the low-rank matrix method, use the Cholesky decomposition method to hierarchically decompose the electrical data matrix to obtain the multiple sub-matrices; If the electrical data matrix is an asymmetric matrix, based on the low-rank matrix method, use the LU decomposition method to hierarchically decompose the electrical data matrix to obtain the multiple sub-matrices.

[0011] Optionally, the second processing unit is further configured to use a preset iterative algorithm to correct the processing result to obtain a corrected processing result.

[0012] In a third aspect, an embodiment of the present application provides a computer-readable storage medium. The computer-readable storage medium stores a program. When the program runs on a computer, the computer is caused to execute the data processing method in any possible design of the first aspect.

[0013] In a fourth aspect, an embodiment of the present application provides a computer device, including: a memory and a processor; the memory is used to store a computer program; the processor is used to call the computer program stored in the memory, so that the computer device executes the data processing method in any possible design of the first aspect.

[0014] In a fifth aspect, an embodiment of the present application provides a computer program product. When the computer program product runs on a computer, the computer is caused to execute the data processing method in any possible design of the first aspect.

[0015] Advantages of the present application: In the data processing method related to electrical devices provided by the embodiments of the present application, by reordering a plurality of electrical data, the electrical data with adjacent positions are sorted in the nearest neighbor manner to obtain the to-be-processed data after sorting, and then the to-be-processed data after sorting is compressed to obtain the to-be-processed data after compression. In this way, the main features of the to-be-processed data after compression will not be lost during the compression process. Correspondingly, the accuracy of the processing result obtained based on the to-be-processed data after compression is ensured, and the data volume can also be reduced due to compression, thereby accelerating the data processing speed and reducing the memory consumption.

[0016] These implementation manners or other implementation manners of the present application will be more concise and easy to understand in the following description of the embodiments. Description of the Drawings

[0017] To more clearly illustrate the technical solutions in the embodiments of the present application, the following briefly introduces the accompanying drawings required for the description of the embodiments. Obviously, the accompanying drawings in the following description are only some embodiments of the present application. For those of ordinary skill in the art, without creative efforts, other accompanying drawings can be obtained based on these drawings.

[0018] Figure 1 A schematic flowchart of a data processing method related to an electrical device provided by an embodiment of the present application; Figure 2 A schematic flowchart of a data processing method related to an electrical device provided by an embodiment of the present application; Figure 3 A schematic diagram of a data processing device related to an electrical device provided by an embodiment of the present application. Detailed implementation manners

[0019] In order to make the objectives, technical solutions, and advantages of the present application clearer, the following will further describe the present application in detail with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative efforts belong to the scope of protection of the present application.

[0020] With the continuous expansion of application scenarios and the increasing improvement of computational complexity, the integral equation method faces double challenges of balancing computational efficiency and accuracy in engineering practice. Currently, the mainstream solution techniques can be divided into two categories: one is the direct method of matrix decomposition (such as LU decomposition, QR decomposition), and the other is the iterative method of fast algorithms (such as the fast multipole method FMM, adaptive cross approximation ACA). Among them, although the direct method can ensure high computational accuracy, due to the characteristics of the dense matrix generated by the integral equation method, its memory consumption and computational amount increase quadratically with the problem scale, severely restricting the application of this method in large-scale problems. In contrast, the iterative method significantly reduces the consumption of computational resources through fast algorithms and retains data characteristics to a certain extent; however, when the matrix condition number is poor, this method often has problems of difficult convergence, which in turn affects the reliability of the final calculation results.

[0021] In view of this, the embodiments of the present application provide a data processing method related to an electrical device, as Figure 1 shown, including: Step 101, obtain the data to be processed. The data to be processed contains multiple electrical data, and the electrical data is used to characterize a certain physical property of the electrical device; In one embodiment, the electrical device can be a resistor, a capacitor, an inductor, a sensor, etc., as well as a diode, a transistor, a field effect transistor, a microprocessor, an operational amplifier, a digital logic chip, etc., and also a connector, a socket, a printed circuit board (PCB), an antenna, a transformer, a relay, etc. There is no specific limitation on the electrical device here, and it can be set according to needs.

[0022] In one embodiment, the electrical data can be current, voltage, resistance, capacitance, inductance, conductivity, resistivity, dielectric constant, breakdown strength, frequency response, impedance, power loss, transmission characteristics, charge carrier concentration (n) and mobility (μ), and so on.

[0023] Step 102: Based on a preset sorting method, re-sort multiple electrical data to obtain the to-be-processed data after sorting. The preset sorting algorithm is used to sort adjacent electrical data in the neighborhood.

[0024] In one embodiment, the preset sorting method can adopt the K-nearest neighbor algorithm, Ball tree accelerated neighbor search, etc. There is no specific limitation on the method specifically adopted for the preset sorting method here.

[0025] Step 103: Based on a preset compression method, perform compression processing on the sorted to-be-processed data to obtain the compressed to-be-processed data.

[0026] In one embodiment, the preset compression method can adopt singular value decomposition (SVD), principal component analysis (PCA), robust PCA (RPCA), non-negative matrix factorization (NMF), etc. There is no specific limitation on the method specifically adopted for the preset compression method here.

[0027] Step 104: Obtain a processing result according to the compressed to-be-processed data. The processing result is used to characterize the target physical property of the electrical device.

[0028] In one embodiment, the processing method for the compressed to-be-processed data here can be processing such as taking the mean value or dimensionality reduction. There is no specific limitation on the data processing method here, and it can be set according to the processing method required by the specific electrical data.

[0029] In the above method, by processing in the way of re-sorting and compressing multiple electrical data, compared with the traditional direct solution of electrical data, the present application can, on the premise of ensuring the calculation accuracy, reduce the matrix filling complexity from to , and reduce the solution complexity from to (where: N is the matrix dimension, and K is the maximum rank of the low-rank approximation). Compared with the fast iterative algorithm, the present application can stably obtain a solution that meets the accuracy requirements. In addition, due to the adoption of a general processing method based on linear algebra, the present application is applicable to all matrix solution problems with non-diagonal low-rank characteristics.)

[0030] Based on the above Figure 1 method flow, an embodiment of the present application provides a data sorting method. The preset sorting method is KD-Tree. In step 102, based on the preset sorting method, multiple electrical data are re-sorted to obtain the to-be-processed data after sorting, including: Step 1021: Perform index processing on multiple electrical data, and construct a KD-Tree according to the coordinate data of each of the multiple electrical data.

[0031] Step 1022: Based on the KD-Tree, re-sort the multiple electrical data to obtain the to-be-processed data after sorting.

[0032] In one embodiment, the electrical data can be represented in a function form, and the function form can include integral equations.

[0033] In one embodiment, multiple electrical data in function form are processed in matrix form.

[0034] In one embodiment, based on the index re-sorting strategy of the KD-Tree, determine the discrete unit indexes involved in each integral equation in the electrical data matrix, construct a KD-Tree according to the geometric space coordinates of the electrical data matrix, and re-sort the row and column indexes of the electrical data matrix accordingly to improve the numerical calculation efficiency of the electrical data matrix structure.

[0035] Based on the above Figure 1 method flow, an embodiment of the present application provides a data compression method. The multiple electrical data in the to-be-processed data after sorting are an electrical data matrix, and the preset compression method is a low-rank matrix method; in step 103, based on the preset compression method, perform compression processing on the to-be-processed data after sorting to obtain the to-be-processed data after compression, including: Step 1031: Use the low-rank matrix method to perform hierarchical decomposition processing on the electrical data matrix to obtain multiple sub-matrices.

[0036] In one embodiment, based on the matrix block division strategy of the integral equation type, that is, according to the integral equation type corresponding to the electrical property data, the matrix block division strategy can be obtained, and then the electrical property data matrix is block-divided according to the type of the integral equation: if the electrical property data matrix is composed of two or more integral equations, it is divided into several sub-blocks. Among them, the diagonal blocks are square matrices (corresponding to a single type of integral equation), and the non-diagonal blocks can be non-square matrices (reflecting the coupling relationship between different integral equations).

[0037] Step 1032: For each sub-matrix, perform compression processing using the corresponding adaptive compression threshold to obtain the corresponding compressed sub-matrix until the compressed data to be processed formed by the compressed sub-matrices corresponding to each of the multiple sub-matrices is obtained.

[0038] In one embodiment, based on the hierarchical construction method of the hierarchical off-diagonal low-rank matrix (HODLR), multiple sub-matrices after the decomposition processing of the electrical property data matrix are obtained, and then based on the matrix structure after block division, the HODLR matrix is constructed using a hierarchical strategy. During the construction process, an adaptive compression threshold is set for different levels. When the matrix block at a certain level has a symmetric property, only the upper triangular part of it and its sub-level matrices is constructed.

[0039] Based on the above data compression method flow, an embodiment of the present application provides a data compression method. In step 1032, for each sub-matrix, compression processing is performed using the corresponding adaptive compression threshold to obtain the corresponding compressed sub-matrix, including: If the sub-matrix has a symmetric feature, use the corresponding adaptive compression threshold to perform compression processing on the sub-matrix with the symmetric property to obtain the compressed sub-matrix of the sub-matrix with the symmetric property.

[0040] If the sub-matrix has an asymmetric feature, based on the adaptive cross approximation method, use the corresponding adaptive compression threshold to perform compression processing on the sub-matrix with the asymmetric feature to obtain the compressed sub-matrix of the sub-matrix without the symmetric property.

[0041] In one embodiment, based on the compression strategy of the adaptive cross approximation (ACA), for the non-diagonal matrix block - sub-matrix, low-rank compression is performed using the ACA with the adaptive compression threshold.

[0042] Based on the above data compression method flow, an embodiment of the present application provides a data compression method. In step 1031, using the low-rank matrix method, perform hierarchical decomposition processing on the electrical property data matrix to obtain multiple sub-matrices, including: If the electrical property data matrix is a symmetric matrix, based on the low-rank matrix method, use the Cholesky decomposition method to perform hierarchical decomposition processing on the electrical property data matrix to obtain multiple sub-matrices; If the electrical property data matrix is an asymmetric matrix, based on the low-rank matrix method, the LU decomposition method is used to hierarchically decompose the electrical property data matrix to obtain multiple sub-matrices.

[0043] Based on the above Figure 1 method flow, an embodiment of the present application provides a data correction method. After obtaining the processing result according to the compressed data to be processed in step 104, it further includes: Using a preset iterative algorithm to correct the processing result to obtain a corrected processing result.

[0044] In one embodiment, as Figure 2 shown, it is a data processing method flow related to an electrical device provided by an embodiment of the present application, including: (a) Start processing to obtain the data to be processed, and the data to be processed can be an electrical property data matrix.

[0045] (b) Each electrical property in the electrical property data matrix is characterized by an integral equation type, and the electrical property-integral equation indexes are re-sorted.

[0046] (c) Decompose the sorted data to be processed - the sorted electrical property data matrix to obtain multiple sub-matrices.

[0047] (d) Construct multiple sub-matrices based on ACA's HODLR to compress the multiple sub-matrices to obtain the compressed data to be processed - multiple compressed sub-matrices.

[0048] (e) Solve the compressed data to be processed - multiple compressed sub-matrices to obtain a processing result.

[0049] (f) Determine whether the processing result needs to be corrected. If so, use the fast iterative method to correct it to obtain a corrected processing result. Otherwise, end the data processing flow.

[0050] Based on the above data processing methods and related embodiments, an embodiment of the present application provides a data processing method for a metal medium. Taking the metal medium mixed target in electromagnetic parasitic parameter extraction as an example, its data processing process using the present application can be described as follows: including: (1) Geometric discretization and basis function establishment: Perform mesh discretization processing on the metal and dielectric surfaces and establish a corresponding basis function system. Assume the number of basis functions on the metal surface is N, and the number of basis functions on the dielectric surface is M, to obtain N + M electrical property data.

[0051] (2) Basis function reordering optimization: Based on geometric coordinate information, spatially reorder the basis functions (N+M electrical data) on the metal and dielectric surfaces respectively, so that the indices of adjacent basis functions are as close as possible, and obtain the data to be processed after sorting, and finally merge and process the sorting results of the two types of basis functions.

[0052] (3) Matrix block modeling: The dimension of the system electrical data matrix is (N+M)×(N+M). Considering that the integral equations satisfied by the metal and dielectric surfaces have different characteristics, the matrix is divided into four sub-blocks: Diagonal block: A (N×N, metal-metal interaction) Diagonal block: B (M×M, dielectric-dielectric interaction) Off-diagonal block: C (N×M, metal-dielectric coupling) Off-diagonal block: D (M×N, dielectric-metal coupling) (4) HODLR matrix construction: Adopt a hierarchical matrix (HODLR) structure: ① Top-level processing: The off-diagonal blocks C and D are used as the 0th-level off-diagonal blocks and compressed by adaptive cross approximation (ACA) with a threshold of a.

[0053] ② Recursive decomposition: The diagonal blocks A and B are further decomposed hierarchically. To control error propagation, the sub-layer compression threshold decreases with the increase of the hierarchy.

[0054] ③ Symmetry utilization: If the integral equation has symmetry, the corresponding matrix block and its sub-blocks can adopt a symmetric filling strategy.

[0055] (5) Select a decomposition method according to the characteristics of the matrix block: For symmetric positive definite blocks: Adopt Cholesky (LLT) decomposition.

[0056] For non-symmetric / non-positive definite blocks: Adopt LU decomposition.

[0057] (6) Compress the multiple sub-matrices obtained by decomposition. Among them, for symmetric positive definite blocks: The sub-matrices obtained by Cholesky (LLT) decomposition are compressed by a dimensionality reduction algorithm. For non-symmetric / non-positive definite blocks: The sub-matrices obtained by LU decomposition adopt an adaptive cross approximation (ACA) compression strategy - ACA with an adaptive compression threshold for low-rank compression.

[0058] (7) Solve the compressed data to be processed to obtain the processing result.

[0059] (8) Adopt a fast iterative algorithm to correct the processing result - the initial solution to obtain the corrected processing result.

[0060] Large-scale problem optimization: When dealing with large-scale problems or when the low-rank property of non-diagonal blocks is poor, a relatively loose compression threshold is adopted to obtain an initial approximate solution, and the initial solution is substituted into a fast iterative algorithm (such as GMRES) for correction. The solution accuracy can be significantly improved through a small number of iterations while maintaining the computational efficiency.

[0061] Based on the same concept, an embodiment of the present application provides a data processing device related to an electrical device. Figure 3 As shown in the schematic diagram of a data processing device related to an electrical device provided by an embodiment of the present application, Figure 3 it includes: An acquisition unit 301, configured to acquire data to be processed, where the data to be processed includes a plurality of electrical data, and the electrical data is used to characterize a certain physical property of the electrical device; A sorting unit 302, configured to re-sort the plurality of electrical data based on a preset sorting method to obtain the sorted data to be processed, and the preset sorting algorithm is used to perform proximity sorting on the electrical data with adjacent positions; A first processing unit 303, configured to perform compression processing on the sorted data to be processed based on a preset compression method to obtain the compressed data to be processed; A second processing unit 304, configured to obtain a processing result according to the compressed data to be processed, and the processing result is used to characterize the target physical property of the electrical device.

[0062] Optionally, the preset sorting method is KD-Tree. Based on the preset sorting method, the sorting unit 302 is specifically configured to, perform index processing on the plurality of electrical data, and construct a KD-Tree according to the coordinate data of each of the plurality of electrical data; Based on the KD-Tree, re-sort the plurality of electrical data to obtain the sorted data to be processed.

[0063] Optionally, the plurality of electrical data in the sorted data to be processed is an electrical data matrix, and the preset compression method is a low-rank matrix method; the first processing unit 303 is specifically configured to, adopt a low-rank matrix method to perform hierarchical decomposition processing on the electrical data matrix to obtain a plurality of sub-matrices; For each sub-matrix, perform compression processing using a corresponding adaptive compression threshold to obtain a corresponding compressed sub-matrix until the compressed data to be processed formed by the compressed sub-matrices corresponding to each of the plurality of sub-matrices is obtained.

[0064] Optionally, the first processing unit 303 is specifically configured to, If the submatrix has symmetric characteristics, an adaptive compression threshold is used to compress the submatrix with symmetric characteristics to obtain a compressed submatrix of the submatrix with symmetric characteristics; If the submatrix has asymmetric characteristics, based on the adaptive cross approximation method, an adaptive compression threshold is used to compress the submatrix with asymmetric characteristics to obtain a compressed submatrix of the submatrix without symmetric characteristics.

[0065] Optionally, the first processing unit 303 is specifically configured to, if the electrical data matrix is a symmetric matrix, hierarchically decompose the electrical data matrix by using the Cholesky decomposition method based on the low-rank matrix method to obtain the multiple submatrices; If the electrical data matrix is an asymmetric matrix, hierarchically decompose the electrical data matrix by using the LU decomposition method based on the low-rank matrix method to obtain the multiple submatrices.

[0066] Optionally, the second processing unit 304 is further configured to correct the processing result by using a preset iterative algorithm to obtain a corrected processing result.

[0067] Those skilled in the art should understand that the embodiments of the present application can be provided as a method, a system, or a computer program product. Therefore, the present application can adopt the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware aspects. Moreover, the present application can adopt 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.

[0068] The present application is described with reference to the flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to the present application. It should be understood that each flow and / or block in the flowcharts and / or block diagrams, and the combination of flows and / or blocks in the flowcharts and / or block diagrams can be implemented by computer program instructions. These computer program instructions can be provided to the processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing devices to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing devices generate means for implementing the functions specified in Figure 1 one process or multiple processes and / or blocks Figure 1 one block or multiple blocks.

[0069] These computer program instructions can also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to operate in a particular manner, such that the instructions stored in the computer-readable memory produce a manufacture including an instruction device that implements the functions specified in one or more of the flows Figure 1 one or more flows and / or boxes Figure 1 specified in the boxes.

[0070] These computer program instructions can also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer-implemented process, whereby the instructions executed on the computer or other programmable apparatus provide steps for implementing the functions specified in one or more of the flows Figure 1 one or more flows and / or boxes Figure 1 specified in the boxes.

[0071] Obviously, those skilled in the art can make various changes and modifications to this application without departing from the spirit and scope of this application. Thus, if these modifications and variations of this application fall within the scope of the claims of this application and their equivalent technologies, this application is also intended to include these changes and modifications.

Claims

1. A data processing method related to an electrical device, characterized in that, The method includes: Obtaining data to be processed, where the data to be processed contains multiple electrical property data, and the electrical property data is used to characterize a certain physical property of an electrical device; Based on a preset sorting method, re-sorting the multiple electrical property data to obtain the sorted data to be processed. The preset sorting method is used to perform proximity sorting on the electrical property data with adjacent positions, and the multiple electrical property data in the sorted data to be processed form an electrical property data matrix; Using a low-rank matrix method, performing hierarchical decomposition processing on the electrical property data matrix to obtain multiple sub-matrices; For each sub-matrix, performing compression processing using a corresponding adaptive compression threshold to obtain a corresponding compressed sub-matrix until the compressed data to be processed formed by the compressed sub-matrices corresponding to the multiple sub-matrices is obtained; According to the compressed data to be processed, obtaining a processing result, and the processing result is used to characterize the target physical property of the electrical device.

2. The method according to claim 1, wherein The preset sorting method is KD-Tree. Based on the preset sorting method, re-sorting the multiple electrical property data to obtain the sorted data to be processed includes: Performing index processing on the multiple electrical property data, and constructing a KD-Tree according to the coordinate data of the multiple electrical property data; Based on the KD-Tree, re-sorting the multiple electrical property data to obtain the sorted data to be processed.

3. The method according to claim 1, characterized in that, For each sub-matrix, performing compression processing using the corresponding adaptive compression threshold to obtain a corresponding compressed sub-matrix includes: If the sub-matrix has a symmetric feature, performing compression processing on the sub-matrix with symmetric characteristics using the corresponding adaptive compression threshold to obtain a compressed sub-matrix of the sub-matrix with symmetric characteristics; If the sub-matrix has an asymmetric feature, based on the adaptive cross approximation method, performing compression processing on the sub-matrix with asymmetric characteristics using the corresponding adaptive compression threshold to obtain a compressed sub-matrix of the sub-matrix without symmetric characteristics.

4. The method according to claim 1 or 3, characterized in that, Using the low-rank matrix method to perform hierarchical decomposition processing on the electrical property data matrix to obtain multiple sub-matrices includes: If the electrical property data matrix is a symmetric matrix, based on the low-rank matrix method, using the Cholesky decomposition method to perform hierarchical decomposition processing on the electrical property data matrix to obtain the multiple sub-matrices; If the electrical property data matrix is an asymmetric matrix, based on the low-rank matrix method, using the LU decomposition method to perform hierarchical decomposition processing on the electrical property data matrix to obtain the multiple sub-matrices.

5. The method according to claim 1, characterized in that, After obtaining the processing result according to the compressed data to be processed, it further includes: Using a preset iterative algorithm to perform correction processing on the processing result to obtain a corrected processing result.

6. A data processing device related to an electrical device, characterized in that, The device includes: An acquisition unit for acquiring data to be processed, where the data to be processed contains multiple electrical property data, and the electrical property data is used to characterize a certain physical property of an electrical device; A sorting unit, configured to re-sort the multiple electrical property data based on a preset sorting method to obtain the to-be-processed data after sorting, where the preset sorting method is used to perform near-neighbor sorting on the electrical property data with adjacent positions, and the multiple electrical property data in the to-be-processed data after sorting form an electrical property data matrix; A first processing unit, configured to perform hierarchical decomposition processing on the electrical property data matrix by using a low-rank matrix method to obtain multiple sub-matrices; for each sub-matrix, perform compression processing by using a corresponding adaptive compression threshold to obtain a corresponding compressed sub-matrix until the compressed sub-matrices corresponding to the multiple sub-matrices respectively are obtained to form the to-be-processed data after compression; A second processing unit, configured to obtain a processing result according to the to-be-processed data after compression, where the processing result is used to characterize the target physical property of the electrical device.

7. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a program, and when the program runs on a computer, the computer is caused to implement the method according to any one of claims 1 to 5.

8. A computer device, characterized in that, Comprising: A memory, configured to store a computer program; A processor, configured to call the computer program stored in the memory and execute the method according to any one of claims 1 to 5 according to the obtained program.

9. A computer program product, characterized in that, When a computer reads and executes the computer program product, the computer is caused to execute the method according to any one of claims 1 to 5.

Images