Power grid data correlation analysis method, device, terminal device and storage medium

Abstract

The present invention discloses a method, device, terminal device and storage medium for analyzing the correlation of power grid data. The method includes: performing normalization processing on the data to be measured according to preset conditions, where the data to be measured includes power grid measurement data and simulation data; calculating the similarity of the processed data to be measured under different metrics to generate a variety of similarity calculation results; calculating the corresponding support degrees according to the variety of similarity calculation results, and constructing different weight calculation models by using the variety of similarity calculation results and support degrees; constructing a weighted integrated model by using the weight calculation models and support degrees, and analyzing the correlation between the data to be measured by using the weighted integrated model. The present invention reduces the time consumption and human resources for reviewing power data information, standardizes power data management, makes the calculation results of weights more reasonable and effective, and thus makes power-related work more efficient, more accurate and reliable, which is beneficial to maintaining the operation of the power grid system.

Description

Technical Field

[0001] The present invention relates to the technical field of power grid data analysis, and particularly to a method, device, terminal device and storage medium for analyzing the correlation of power grid data. Background Art

[0002] With the stable growth of the national economy, in order to enable the cyber-physical deeply integrated smart grid system to effectively perceive potential threats, detect abnormal behaviors and design active defense strategies, it is necessary to clearly understand its function, structural vulnerability characteristics and their correlations, so as to be able to use the vulnerability correlation technology to achieve batch and rapid vulnerability mining effects with known vulnerability information. Therefore, taking the effective calculation of weights as the key technical breakthrough point for discovering the correlation relationship between power grid measurement data and simulation data is very necessary in the field of power grid data application. However, the existing weight calculation method is determined according to the standards formulated by power grid experts. Once the indicators change, it is necessary to reorganize human and material resources to recalculate the weights. This subjective calculation method is both time-consuming and laborious, and has low efficiency. At the same time, since each indicator is relatively isolated during the calculation, it often causes unreasonable power grid data information analysis and even invalid analysis results. Therefore, in order to meet the development requirements of power grid operation and further standardize the management of power grid equipment, there is an urgent need for a method for analyzing the correlation of power grid data, which can discover the correlation relationship between power grid measurement data and simulation data to cope with the increasingly complex situation and ensure the more accurate and efficient development of power grid work. Summary of the Invention

[0003] The purpose of the present invention is to provide a method, device, terminal device and storage medium for analyzing the correlation of power grid data, so as to solve the problems of long time consumption, low efficiency and unreasonable results existing in the existing weight calculation method of power grid data.

[0004] To achieve the above purpose, the present invention provides a method for analyzing the correlation of power grid data, including:

[0005] Normalize the data to be measured according to preset conditions, where the data to be measured includes power grid measurement data and simulation data;

[0006] Calculate the similarity of the processed data to be measured under different indicators to generate a variety of similarity calculation results;

[0007] Calculate the corresponding support degrees according to the variety of similarity calculation results, and construct different weight calculation models by using the variety of similarity calculation results and the support degrees;

[0008] Construct a weight weighted integration model by using the weight calculation model and the support degrees, and analyze the correlation between the data to be measured by using the weight weighted integration model.

[0009] Further, preferably, the normalization processing of the data to be measured according to the preset conditions includes calculating using the following formula:

[0010]

[0011] In the formula, X ij represents the value of the power grid measurement data and the simulation data index, represents the normalized value of the power grid measurement data and the simulation data index, represents the sum of the values of the power grid measurement data and the simulation data index, and N(0,1) represents that the value is normalized within the range of [0 to 1].

[0012] Further, preferably, calculating the similarity of the processed data to be measured under different indexes includes:

[0013] Calculating the distance-based similarity, cosine-based similarity, Pearson correlation coefficient-based similarity, and mutual information-based similarity of the processed data to be measured.

[0014] Further, preferably, calculating the corresponding support degree according to the calculation results of the multiple similarities includes calculating using the following formula:

[0015] Sup(α i , α j ) = (Sim(α i , α j )) k , k>0;

[0016] Sup(α i , α j ) = exp(-k(2 - Sim(α i , α j )), k>0;

[0017] In the formula, Sim(α i , α j ) represents the similarity between indexes α i and α j , and k is a constant.

[0018] The present invention also provides a power grid data correlation analysis device, including:

[0019] A preprocessing module for normalizing the data to be measured according to preset conditions, where the data to be measured includes power grid measurement data and simulation data;

[0020] A similarity calculation module for calculating the similarity of the processed data to be measured under different indexes and generating multiple similarity calculation results;

[0021] A support degree calculation module, configured to calculate corresponding support degrees according to the multiple similarity calculation results, and construct different weight calculation models by using the multiple similarity calculation results and the support degrees;

[0022] A relevance analysis module, configured to construct a weighted integration model by using the weight calculation model and the support degrees, and analyze the relevance between the data to be measured by using the weighted integration model.

[0023] Further, preferably, the preprocessing module is further configured to perform normalization processing by using the following formula:

[0024]

[0025] In the formula, X ij represents the index value of the power grid measurement data and the simulation data, represents the normalized index value of the power grid measurement data and the simulation data, represents the sum of the index values of the power grid measurement data and the simulation data, and N(0,1) represents that the numerical specification is within the range of [0 to 1].

[0026] Further, preferably, the similarity calculation module is further configured to:

[0027] Calculate the distance-based similarity, cosine-based similarity, Pearson correlation coefficient-based similarity, and mutual information-based similarity of the processed data to be measured.

[0028] Further, preferably, the support degree calculation module is further configured to calculate by using the following formula:

[0029] Sup(α i , α j ) = (Sim(α i , α j )) k , k>0;

[0030] Sup(α i , α j ) = exp(-k(2 - Sim(α i , α j )), k>0;

[0031] In the formula, Sim(α i , α j ) represents the similarity between the indexes α i and α j , and k is a constant.

[0032] The present invention also provides a terminal device, including:

[0033] One or more processors;

[0034] A memory coupled to the processor for storing one or more programs;

[0035] When the one or more programs are executed by the one or more processors, the one or more processors implement the power grid data correlation analysis method as described in any one of the above.

[0036] The present invention also provides a computer-readable storage medium having a computer program stored thereon, and when the computer program is executed by a processor, the power grid data correlation analysis method as described in any one of the above is implemented.

[0037] Compared with the prior art, the beneficial effects of the present invention are as follows:

[0038] The present invention discloses a power grid data correlation analysis method, apparatus, terminal device, and storage medium. The method includes: normalizing the data to be measured according to preset conditions, where the data to be measured includes power grid measurement data and simulation data; calculating the similarity of the processed data to be measured under different indicators to generate a variety of similarity calculation results; calculating the corresponding support degrees according to the variety of similarity calculation results, and constructing different weight calculation models using the similarity calculation results and support degrees; constructing a weight weighted integration model using the weight calculation model and support degrees, and analyzing the correlation between the data to be measured using the weight weighted integration model. The present invention reduces the time consumption and human resources for reviewing power data information, standardizes power data management, makes the calculation results of weights more reasonable and effective, and thus makes power-related work more efficient, more accurate and reliable, which is beneficial to maintaining the operation of the power grid system. BRIEF DESCRIPTION OF THE DRAWINGS

[0039] In order to more clearly illustrate the technical solutions of the present invention, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings in the following description are only some embodiments of the present invention, and those of ordinary skill in the art can also obtain other drawings based on these drawings without creative efforts.

[0040] Figure 1 is a flowchart of the power grid data correlation analysis method provided by an embodiment of the present invention;

[0041] Figure 2 is Figure 1 a flowchart of the sub-steps of step S40 in

[0042] Figure 3 is a structural diagram of the power grid data correlation analysis apparatus provided by an embodiment of the present invention;

[0043] Figure 4 It is a schematic structural diagram of a terminal device provided by an embodiment of the present invention. Specific embodiments

[0044] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0045] It should be understood that the step numbers used in the text are only for convenient description and do not limit the execution order of the steps.

[0046] It should be understood that the terms used in the specification of the present invention are only for the purpose of describing specific embodiments and are not intended to limit the present invention. As used in the specification of the present invention and the appended claims, unless otherwise clearly specified in the context, the singular forms "a", "an" and "the" are intended to include the plural forms.

[0047] The terms "comprising" and "including" indicate the presence of the described features, wholes, steps, operations, elements and / or components, but do not exclude the presence or addition of one or more other features, wholes, steps, operations, elements, components and / or their combinations.

[0048] The term "and / or" refers to any combination and all possible combinations of one or more of the associated listed items, and includes these combinations.

[0049] Please refer to Figure 1 , an embodiment of the present invention provides a method for analyzing the relevance of power grid data. As Figure 1 shown, the method for analyzing the relevance of power grid data includes steps S10 to S40. The specific steps are as follows:

[0050] S10. Normalize the data to be measured according to preset conditions, where the data to be measured includes power grid measurement data and simulation data.

[0051] In this step, the simulation data usually includes data such as power load and load. Among them, the preset condition means that the value of the data to be measured is normalized within the range of [0-1]. Specifically, the calculation formula for normalization processing is:

[0052]

[0053] In the formula, X ijRepresents the index values of power grid measurement data and simulation data Represents the normalized index values of power grid measurement data and simulation data Represents the sum of the index values of power grid measurement data and simulation data, and N(0,1) indicates that the numerical specification is within the range of [0 to 1].

[0054] S20. Calculate the similarity of the data to be measured after calculation and processing under different indexes, and generate multiple similarity calculation results.

[0055] It should be noted that calculating the similarity under different indexes in this step means: calculating the distance-based similarity, cosine-based similarity, Pearson correlation coefficient-based similarity, and mutual information-based similarity of the data to be measured after calculation and processing.

[0056] Specifically, the calculation formula is as follows:

[0057] 2.1) Distance-based similarity:

[0058] Sim Distance (α i ,α j )=(1 - d N (α i ,α j )) (2)

[0059] In the formula, d N (α i ,α j ) represents the distance between two normalized indexes. Sim represents the similarity between two indexes, Distance (α i ,α j ) indicates that this similarity is calculated based on distance, and α i and α j represent two different indexes.

[0060] 2.2) Cosine-based similarity:

[0061]

[0062] In the formula, Cosine (α i ,α j ) indicates that this similarity is calculated based on cosine, and α i and α j represent two different indexes.

[0063] 2.3) Pearson correlation coefficient-based similarity:

[0064]

[0065] In the formula,pearson (α i ,α j ) indicates that this similarity is calculated based on the Pearson correlation coefficient. α i and α j represent two different metrics. Cov(α i ,α j ) is the covariance, is the variance.

[0066] 2.4) Similarity based on mutual information:

[0067] Sim MutualInformation (α i ,α j ) = I(α i ) - I(α i |α j ) (5)

[0068] In the formula, MutualInformation (α i ,α j ) indicates that this similarity is calculated based on mutual information. α i and α j represent two different metrics. I(α i ) represents the mutual information of α i . I(α i |α j ) represents the mutual information calculated with α i and α j as the conditional probability.

[0069] S30. Calculate the corresponding support degrees according to the multiple similarity calculation results, and construct different weight calculation models by using the similarity calculation results and the support degrees.

[0070] In this step, based on the similarity between feature vectors, calculate the support degrees between features. The common calculations of three types of feature support degrees are as follows:

[0071] Sup(α i ,α j ) = (Sim(α i ,α j )) k , k > 0 (6)

[0072] Sup(α i ,α j ) = exp(-k(2 - Sim(α i ,α j , k > 0 (8)

[0073] In the formula, Sim(αi , α j ) represents the index α i and α j The similarity between them, where k is a constant. As a preferred embodiment, k is taken as 2 in this embodiment.

[0074] Furthermore, in this step, the calculation of feature support degree and the calculation of similarity between features are combined to construct a variety of weight calculation models. For example, by combining formula (6) and (2), a corresponding weight calculation model can be constructed.

[0075] S40. Use the weight calculation model and the support degree to construct a weighted integrated model, and use the weighted integrated model to analyze the correlation between the data to be measured.

[0076] It should be noted that based on step S30, multiple weight calculation models can be obtained in this step, and then by combining the support degrees between them, an adaptive weighted integrated model can be obtained to calculate the final feature weights and analyze the correlation between the data to be measured according to the feature weights.

[0077] As Figure 2 shown, this step S40 further includes the following steps:

[0078] 4.1) The feature I i weights obtained by t feature weight calculation methods are expressed as:

[0079]

[0080] In the formula, M j represents a certain feature weight calculation method.

[0081] 4.2) Calculate the support degree between two different weights of the same feature:

[0082]

[0083] 4.3) Based on the formula in 4.2), obtain the support degree of a certain feature weight:

[0084]

[0085] 4.4) Normalize the support degree of the above feature weight to obtain a weight value of a certain feature weight:

[0086]

[0087] 4.5) Finally, use linear weighted averaging to fuse multiple feature weights to obtain the final weight of feature I i :

[0088]

[0089] The power grid data correlation analysis method provided by the embodiments of the present invention can reduce the time consumption and human resources for reviewing power data information, standardize power data management, make the calculation results of weights more reasonable and effective, and thus make power-related work more efficient, accurate and reliable, which is beneficial to maintaining the operation of the power grid system.

[0090] To help understand the solution provided by the present invention, the following will be described in conjunction with specific examples:

[0091] Please refer to Table 1, the index value table of power measurement data and simulation data. The first row is the index type, and the following 8 rows are 6 values of each index. Among them, the power grid measurement data are the first three indexes, and the simulation data are the last three. The existing method is only to sum the index data first and then perform normalization processing, and use the normalization result as the index weight:

[0092] Table 1 Index value table of power measurement data and simulation data

[0093]

[0094] Furthermore, the result of the normalization processing is shown in Table 2, the existing method power network data index value weight table. By observing the change of weights, the correlation between power grid data is not considered, which is likely to cause unreasonable or even invalid results.

[0095] Table 2 Existing method power measurement data and simulation data index value weight table

[0096]

[0097] Furthermore, adopting the method for the correlation relationship between the power grid measurement data and the simulation data of the present invention, first perform normalization processing on the power grid data index values, as shown in Table 3, the power grid index value normalization table.

[0098] Table 3 Power measurement data and simulation data index value normalization table

[0099]

[0100] Secondly, for the normalized power measurement data and simulation data index data, various similarities between pairwise data are calculated through formulas (1), (2), (3), and (4).

[0101] Taking formula (1) as an example: calculate the similarity between pairwise indexes. The calculation results of formula (2), formula (3), and formula (4) are similar to that of formula (1), as shown in Table 4.

[0102] Further combine the support degree, calculate the support degree result, and construct various similarity calculation results. Only one support degree calculation method is listed here, and the other two methods are similar, as shown in Table 5.

[0103] Table 4 Example of (one kind of) similarity calculation result between indicators

[0104]

[0105]

[0106] Table 5 Example of (one kind of) support degree calculation result data

[0107]

[0108] First, normalize the similarity calculation obtained based on the support degree to obtain various weight calculation methods. For the sake of brief description, only one weight calculation is listed in Table 6, and the calculation of the remaining methods is similar.

[0109] Table 6 Result of combining (one kind of) support degree and (one kind of) similarity

[0110]

[0111] A certain weight represented by formula (9) is shown in Table 7;

[0112] Table 7 Example of (a certain) calculated weight

[0113] Summation 1 1.000078 0.619444 0.967405 0.980032 0.802303 Weight value 0.186245 0.18626 0.115369 0.180175 0.182526 0.149425

[0114] Then, combine the support degrees between each other, propose an adaptive weighted integration method to obtain the calculated adaptive weights. The calculation of this method is from formula (10) to formula (13). Integrate the weights obtained by various weight calculation methods to obtain the final feature weights.

[0115] Table 8 A certain feature weight value λ

[0116]

[0117] Furthermore, based on Tables 7 and 8, the final weight is calculated by formula (13), as shown in Table 9:

[0118] Table 9 Weight calculation result

[0119]

[0120]

[0121] Compared with the existing weight calculation methods, it can be found that the weight differences of the method provided by the present invention are relatively mild. Since the correlation between grid measurement data and simulation data is considered, the weight results are reasonable and effective. In addition, compared with the existing methods, the present invention can independently design and add similarity calculations according to requirements, enabling the weight calculation to be adaptive. When the index system changes, it can also adapt extremely quickly to ensure the accuracy of the results. At the same time, compared with a single weight calculation that cannot overcome the defects existing in local information, the present invention integrates multiple weights through adaptive weighted integration, making the weight calculation more comprehensive and reasonable.

[0122] Please refer to Figure 3 , the present invention also provides a grid data correlation analysis device, including:

[0123] A preprocessing module 01, configured to perform normalization processing on the data to be measured according to preset conditions, where the data to be measured includes grid measurement data and simulation data;

[0124] A similarity calculation module 02, configured to calculate the similarity of the processed data to be measured under different indicators, and generate multiple similarity calculation results;

[0125] A support degree calculation module 03, configured to calculate the corresponding support degree according to the multiple similarity calculation results, and construct different weight calculation models by using the multiple similarity calculation results and the support degree;

[0126] A correlation analysis module 04, configured to construct a weight weighted integration model by using the weight calculation model and the support degree, and analyze the correlation between the data to be measured by using the weight weighted integration model.

[0127] In a specific embodiment, the preprocessing module 01 is further configured to perform normalization processing by using the following formula:

[0128]

[0129] In the formula, X ij represents the index value of the grid measurement data and the simulation data, represents the normalized index value of the grid measurement data and the simulation data, represents the sum of the index values of the grid measurement data and the simulation data, and N(0,1) represents that the numerical specification is within the range of [0 to 1].

[0130] In a specific embodiment, the similarity calculation module 02 is further configured to:

[0131] Calculate the distance-based similarity, cosine-based similarity, Pearson correlation coefficient-based similarity, and mutual information-based similarity of the processed data to be measured.

[0132] In a specific embodiment, the support calculation module 03 is further configured to calculate using the following formula:

[0133] Sup(α i ,α j ) = (Sim(α i ,α j )) k , k > 0;

[0134] Sup(α i ,α j ) = exp(-k(2 - Sim(α i ,α j )), k > 0;

[0135] In the formula, Sim(α i ,α j ) represents the similarity between the indicators α i and α j , and k is a constant.

[0136] It can be understood that the power grid data correlation analysis device provided by the embodiments of the present invention is used to execute the power grid data correlation analysis method described in any one of the above embodiments. This embodiment can reduce the time consumption and human resources for reviewing power data information, standardize power data management, make the calculation results of weights more reasonable and effective, and thus make power-related work more efficient, more accurate and reliable, which is beneficial to maintaining the operation of the power grid system.

[0137] Please refer to Figure 4 , an embodiment of the present invention provides a terminal device, including:

[0138] One or more processors;

[0139] A memory, coupled to the processor, for storing one or more programs;

[0140] When the one or more programs are executed by the one or more processors, the one or more processors implement the power grid data correlation analysis method as described above.

[0141] The processor is used to control the overall operation of the terminal device to complete all or part of the steps of the above-mentioned power grid data correlation analysis method. The memory is used to store various types of data to support the operation of the terminal device. Such data may include, for example, instructions for any application or method operating on the terminal device, as well as application-related data. The memory can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as Static Random Access Memory (SRAM), Electrically Erasable Programmable Read-Only Memory (EEPROM), Erasable Programmable Read-Only Memory (EPROM), Programmable Read-Only Memory (PROM), Read-Only Memory (ROM), magnetic memory, flash memory, magnetic disk or optical disc.

[0142] In an exemplary embodiment, the terminal device can be implemented by one or more Application Specific Integrated Circuits (ASICs), Digital Signal Processors (DSPs), Digital Signal Processing Devices (DSPDs), Programmable Logic Devices (PLDs), Field Programmable Gate Arrays (FPGAs), controllers, microcontrollers, microprocessors or other electronic components, and is used to execute the power grid data correlation analysis method described in any of the above embodiments, and achieve the same technical effects as the above method.

[0143] In another exemplary embodiment, a computer-readable storage medium including a computer program is further provided. When the computer program is executed by a processor, it implements the steps of the power grid data correlation analysis method described in any of the above embodiments. For example, the computer-readable storage medium can be the above-mentioned memory including the computer program, and the above computer program can be executed by the processor of the terminal device to complete the power grid data correlation analysis method described in any of the above embodiments, and achieve the same technical effects as the above method.

[0144] The above are the preferred embodiments of the present invention. It should be noted that for those of ordinary skill in the art, without departing from the principle of the present invention, several improvements and refinements can be made, and these improvements and refinements are also regarded as the protection scope of the present invention.

Claims

1. A method for analyzing the correlation of power grid data, characterized in that, Including: Normalize the data to be measured according to preset conditions, where the data to be measured includes power grid measurement data and simulation data; Including using the following formula for calculation: where X ij represents the values of power grid measurement data and simulation data indicators, represents the normalized values of power grid measurement data and simulation data indicators, represents the sum of the values of power grid measurement data and simulation data indicators, and N(0,1) represents that the numerical specification is within the range of [0 to 1]; Calculate the similarity of the processed data to be measured under different metrics, generating multiple similarity calculation results; including: calculating the distance-based similarity, cosine-based similarity, Pearson correlation coefficient-based similarity, and mutual information-based similarity of the processed data to be measured; Calculate the corresponding support degrees according to the multiple similarity calculation results, and construct different weight calculation models using the multiple similarity calculation results and the support degrees; including using the following formula for calculation: Sup(α i ,α j ) = (Sim(α i ,α j )) k , k > 0; Sup(α i ,α j ) = exp(-k(2 - Sim(α i ,α j )), k > 0; where Sim(α i , α j ) represents the similarity between the indices αi and αj, and k is a constant; Construct a weighted integrated model using the weight calculation model and the support degrees, and analyze the correlation between the data to be measured using the weighted integrated model.

2. A device for analyzing the correlation of power grid data, characterized in that, Including: A preprocessing module for normalizing the data to be measured according to preset conditions, where the data to be measured includes power grid measurement data and simulation data; Also used for normalizing using the following formula: Wherein, X ij represents the value of the power grid measurement data and the simulation data index, represents the normalized value of the power grid measurement data and the simulation data index, represents the sum of the power grid measurement data and the index values of the simulation data, and N(0,1) represents that the numerical specification is in the range of [0 to 1]; A similarity calculation module for calculating the similarity of the processed data to be measured under different metrics, generating multiple similarity calculation results; Also used for: calculating the distance-based similarity, cosine-based similarity, Pearson correlation coefficient-based similarity, and mutual information-based similarity of the processed data to be measured; A support degree calculation module for calculating the corresponding support degrees according to the multiple similarity calculation results, and constructing different weight calculation models using the multiple similarity calculation results and the support degrees; Also used for using the following formula for calculation: Sup(α i ,α j ) = (Sim(α i ,α j )) k , k > 0; Sup(α i ,α j ) = exp(-k(2 - Sim(α i ,α j )), k > 0; where Sim(α i , α j ) represents the similarity between indices α i and α j , and k is a constant; A correlation analysis module for constructing a weighted integrated model using the weight calculation model and the support degrees, and analyzing the correlation between the data to be measured using the weighted integrated model.

3. A terminal device, characterized in that, Including: One or more processors; A memory coupled to the processor 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 power grid data correlation analysis method as claimed in claim 1.

4. A computer-readable storage medium, on which a computer program is stored, characterized in that, The computer program, when executed by the processor, implements the power grid data correlation analysis method as claimed in claim 1.

Images