Method for generating security detection model, security detection method and device

Abstract

The application discloses a security detection model generation method, a security detection method and device. The generation method comprises the following steps: acquiring state operation data of a power grid device, wherein the state operation data comprises multiple index data; determining an index subspace corresponding to each index data based on a first similarity between the index data; calculating a target weight of each index data in the corresponding index subspace; and generating a security detection model based on the multiple index data and the target weight. Compared with determining the weight of a certain index data in all index data, the embodiment is more reasonable in determining the target weight of a certain index data in the same index data, and also makes the detection result of the security detection model generated based on the index data and the target weight more accurate.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of power grid safety, and particularly relates to a safety detection model generation method, a safety detection method and device. BACKGROUND

[0002] The smart grid is a typical cyber-physical system (CPS), which is easy to collect operation state data of power grid equipment, and according to the state operation data, the safety of the power grid equipment is comprehensively quantitatively evaluated, and the power grid equipment with safety risks is accurately positioned, which provides beneficial help for the safety detection and protection of the smart grid.

[0003] At present, the safety of the power grid equipment is quantitatively evaluated based on an artificially generated index system. However, the index system generated artificially has strong subjectivity, and it is difficult to coordinate the influence of a large number of indexes on the safety of the equipment, resulting in low accuracy of the safety detection result. Therefore, how to detect the safety of the power grid equipment according to the state operation data has become a problem to be solved. SUMMARY

[0004] The present application provides a safety detection model generation method, a safety detection method and device, to solve the problem of low accuracy of the safety detection result of the smart grid equipment.

[0005] In order to solve the above technical problem, in a first aspect, the present application embodiment provides a safety detection model generation method, comprising:

[0006] Obtaining state operation data of power grid equipment, the state operation data comprising a plurality of index data;

[0007] Based on the first similarity between the index data, determining an index subspace corresponding to each index data;

[0008] Performing data analysis on the index data in each index subspace to obtain a target weight of each index data in the corresponding index subspace;

[0009] Generating a safety detection model based on the plurality of index data and the target weight.

[0010] The embodiment determines the index subspace corresponding to each index data through the first similarity between a plurality of index data in state operation data, realizes classification and attribution of the plurality of index data, and analyzes the target weight of each index data in the corresponding index subspace on the basis of the classified and attributed index data. Compared with determining the weight of a certain index data in all index data, the embodiment determines the target weight of a certain index data in the same index data more reasonably, and also makes the detection result of the security detection model generated based on the index data and the target weight more accurate.

[0011] In an embodiment, determining the index subspace corresponding to each index data based on the first similarity between the index data comprises:

[0012] Selecting K index data from the plurality of index data as initial index data;

[0013] Calculating the first similarity between each index data and the initial index data;

[0014] Clustering the plurality of first similarities, and dividing the index data corresponding to the first similarities of the same cluster into the same index subspace to obtain a plurality of index subspaces.

[0015] The embodiment takes the K index data as the initial index data to reduce the operation amount of the first similarity operation process, and uses the clustering algorithm to cluster based on the first similarity between the index data and the initial index data, to realize classification and attribution of the plurality of index data.

[0016] In an embodiment, calculating the target weight of each index data in the corresponding index subspace comprises:

[0017] Determining the information entropy weight of each index data in the corresponding index subspace;

[0018] Determining the similarity weight of each index data in the corresponding index subspace;

[0019] Calculating the final weight of the index data based on the information entropy weight and the similarity weight.

[0020] The embodiment calculates the final weight of the index data through the information entropy weight with monotonicity, non-negativity and accumulation, and the similarity weight representing the degree of similarity, so that the weight determination result of the index data is more accurate and objective.

[0021] In an embodiment, determining the information entropy weight of each index data in the corresponding index subspace comprises:

[0022] For each index data, calculating the information entropy of the index data in the corresponding index subspace;

[0023] The information entropy of the index data is compared with the information entropy of the target index data to obtain an information entropy comparison matrix, and the target index data is other index data in an index subspace corresponding to the index data;

[0024] According to the information entropy comparison matrix, the information entropy weight of the index data is determined.

[0025] In this embodiment, the information entropy comparison matrix is generated by information entropy, and the weight of the information entropy of the index data in the entire comparison matrix is determined, so that the information entropy weight is obtained, and the determination result of the information entropy weight is more objective and accurate.

[0026] In an embodiment, the similarity weight of each index data in the corresponding index subspace is determined, including:

[0027] For each index data, a second similarity between the index data and the target index data is calculated, and the target index data is other index data in an index subspace corresponding to the index data;

[0028] The second similarities are summed to obtain a comprehensive similarity of the index data;

[0029] The comprehensive similarity is normalized to obtain the similarity weight of the index data.

[0030] In an embodiment, a security detection model is generated based on the plurality of index data and the target weight, including:

[0031] According to the target weight, the index data of each index subspace is linearly combined to obtain upper layer index data of each index subspace;

[0032] The upper layer index data of each index subspace is iteratively linearly combined until the iteration number reaches a preset number, and the iteration is stopped to obtain the security detection model.

[0033] In this embodiment, the index data is linearly combined to fuse the lower layer index to obtain the upper layer index, and the security detection model is generated based on all index data by iteration of the upper layer index, which is more objective and accurate than establishing an index system artificially.

[0034] In a second aspect, the embodiments of the present application provide a security detection method of a power grid device, including:

[0035] Obtaining state operation data of the power grid device;

[0036] According to the security detection model, data analysis is performed on the state operation data to obtain a security quantization value of the power grid device, and the security quantization is used to evaluate the security degree of the power grid device, and the security detection model is generated based on the generation method of the first aspect.

[0037] In a third aspect, an embodiment of the present application provides a security detection model generation device, comprising:

[0038] A first acquisition module is configured to acquire state operation data of the power grid equipment, the state operation data comprising a plurality of index data;

[0039] A determination module is configured to determine, based on a first similarity between the index data, an index subspace corresponding to each index data;

[0040] A first analysis module is configured to perform data analysis on the index data in each index subspace, to obtain a target weight of each index data in the corresponding index subspace;

[0041] A generation module is configured to generate, based on the plurality of index data and the target weight, a security detection model.

[0042] In a fourth aspect, an embodiment of the present application provides a security detection device for power grid equipment, comprising:

[0043] A second acquisition module is configured to acquire state operation data of the power grid equipment;

[0044] A second analysis module is configured to perform data analysis on the state operation data according to the security detection model, to obtain a security quantization value of the power grid equipment, the security quantization value being used to evaluate the security level of the power grid equipment, and the security detection model being generated based on the generation method of the first aspect.

[0045] In a fifth aspect, an embodiment of the present application provides a computer device, comprising a memory and a processor, the memory storing computer readable instructions, and the processor executing the computer readable instructions to implement the steps of the security detection model generation method of the first aspect or the steps of the security detection method of the power grid equipment of the second aspect.

[0046] It should be noted that the beneficial effects of the second aspect to the fifth aspect are described in the first aspect, and will not be described here. BRIEF DESCRIPTION OF DRAWINGS

[0047] Figure 1 A flowchart of the security detection model generation method provided by an embodiment of the present application;

[0048] Figure 2 A flowchart of the security detection method provided by an embodiment of the present application;

[0049] Figure 3 A structure diagram of the security detection model generation device provided by an embodiment of the present application;

[0050] Figure 4A structural schematic diagram of a security detection device provided by an embodiment of the present application is shown.

[0051] Figure 5 A structural schematic diagram of a computer device provided by an embodiment of the present application is shown. DETAILED DESCRIPTION

[0052] The technical solutions in the embodiments of the present application will be described clearly and completely below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only some of the embodiments of the present application, but not all the embodiments of the present application. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative work fall within the scope of protection of the present application.

[0053] At present, an index system is artificially generated to quantitatively evaluate the security of power grid equipment. However, the index system artificially generated has strong subjectivity, and it is difficult to coordinate the influence of numerous indexes on the security of equipment, resulting in low accuracy of the security detection result. Therefore, how to detect the security of power grid equipment according to the state operation data has become a problem to be solved.

[0054] Therefore, an embodiment of the present application provides a security detection model generation method, a security detection method and device, which determines the index subspace corresponding to each index data through the first similarity between the plurality of index data in the state operation data, realizes classification and attribution of the plurality of index data, and analyzes the target weight of each index data in the corresponding index subspace on the basis of the classified and attributed index data. Compared with determining the weight of a certain index data in all index data, the embodiment only determines the target weight of a certain index data in the same index data, which is more reasonable, and also makes the detection result of the security detection model generated based on the index data and the target weight more accurate.

[0055] Reference Figure 1 , Figure 1 A flowchart of a security detection model generation method provided by an embodiment of the present application is shown. The generation method described below in the embodiment of the present application can be applied to a computer device, which includes but is not limited to a smart phone, a tablet computer, a desktop computer, a supercomputer, a personal digital assistant, a physical server and a cloud server, and the like. The generation method of the embodiment of the present application includes steps S101 to S104, which are described in detail as follows:

[0056] In step S101, state operation data of power grid equipment is obtained, and the state operation data includes a plurality of index data.

[0057] In this step, the status operation data refers to the status data of the power grid equipment during safe operation. It can be the status operation data of multiple power grid equipment or multiple sets of status operation data of a single power grid equipment.

[0058] For example, the dataset collected by N power grid devices is represented as X, C = {I1, I2, ..., I...} n} represents the set of metrics in dataset X, I i and I j This represents two different indicator data in C, α i ={x 1i ,x 2i ,…,x Ni} T isI i The corresponding data vector, α j ={x 1j ,x 2j ,…,x Nj} T isI j The corresponding data vector.

[0059] Step S102: Based on the first similarity between the indicator data, determine the indicator subspace corresponding to each indicator data.

[0060] In this step, the first similarity between the indicator data is calculated based on a similarity formula. Optionally, the similarity formula can be cosine similarity, Euclidean distance, etc. For example, the similarity between any two indicator data can be expressed as:

[0061] Sim(I i I j )=(1-d(α i α j )) q q > 0;

[0062] Among them, Sim(I i ,I j ) represents I i and I j The first similarity between them This represents the normalized Euclidean distance calculated based on data vectors αi and αj.

[0063] Furthermore, based on the above similarity formula, the first similarity between each pair of indicator data is calculated, and all first similarities are clustered using a clustering algorithm. The indicator data corresponding to the first similarities in the same cluster are classified into the same category, thereby achieving the purpose of intelligent classification.

[0064] Step S103, data analysis is performed on the index data in each index subspace to obtain a target weight of each index data in the corresponding index subspace.

[0065] In this step, the same type of index data is attributed to the same index subspace, and a target weight of each index data in the corresponding index subspace is calculated. Optionally, the target weight is determined by the similarity between each index data and other index data. It can be understood that the higher the similarity between index data A and other index data in the index subspace, the greater the target weight of index data A in the index subspace.

[0066] Step S104, generating a security detection model based on the plurality of index data and the target weight.

[0067] In this step, since each index data corresponds to a target weight, the index data and the target weight are weighted and summed to obtain a security detection model. Exemplarily, the security detection model is: security quantization value = index data I1x target weight 1 + index data I2x target weight 2 + … + index data INx target weight N. N

[0068] In an embodiment, based on the embodiment shown in Figure 1 The step S102 described above, specifically includes:

[0069] Selecting K index data from the plurality of index data as initial index data;

[0070] Calculating a first similarity between each index data and the initial index data;

[0071] Clustering the plurality of first similarities, and dividing the index data corresponding to the first similarities in the same cluster into the same index subspace to obtain a plurality of index subspaces.

[0072] In this embodiment, K initial index data is selected from the index set C, denoted as and the first similarity between all indexes and each initial representative index is calculated:

[0073]

[0074] wherein denotes the first similarity between the data index I i and the initial index data , a i is the data vector corresponding to the index I i , and a is the data vector corresponding to the index . ​

[0075] Thus, the similarity vector between each index data and the initial index data is obtained, denoted as {D i (k)|k∈{1,2,…,K}},i∈{1,2,…,n}。Further, the index similarity vectors are clustered based on the spectral clustering algorithm, the index data corresponding to the first similarity of the same cluster are divided into the same index subspace, a plurality of index subspaces are obtained, and the division of the index subspace is realized.

[0076] The embodiment takes K index data as initial index data, without calculating the similarity between each two index data, reduces the operation amount of the first similarity operation process, and classifies and attributes a plurality of index data based on the first similarity of the index data and the initial index data and the clustering algorithm.

[0077] In an embodiment, in Figure 1 Based on the embodiment shown, the step S103 specifically includes:

[0078] Determining the information entropy weight of each index data in the corresponding index subspace;

[0079] Determining the similarity weight of each index data in the corresponding index subspace;

[0080] Calculating the final weight of the index data based on the information entropy weight and the similarity weight.

[0081] In the embodiment, the final weight of the index data is calculated by the information entropy weight with monotonicity, non-negativity and accumulation, and the similarity weight representing the degree of similarity, so that the weight determination result of the index data is more accurate and objective.

[0082] Optionally, determining the information entropy weight of each index data in the corresponding index subspace includes:

[0083] For each index data, calculating the information entropy of the index data in the corresponding index subspace;

[0084] Comparing the information entropy of the index data with the information entropy of the target index data to obtain an information entropy comparison matrix, the target index data being other index data in the index subspace corresponding to the index data;

[0085] Determining the information entropy weight of the index data according to the information entropy comparison matrix.

[0086] In the embodiment, the set of index subspaces divided is denoted as Wherein represents a subspace in and calculates the index data in the index subspace P l In the embodiment, the set of index subspaces divided is denoted as information entropy of corresponding data vector and the information entropy of the index data of P l The information entropy of each index data is compared with the information entropy of other index data in P

[0087]

[0088] According to the matrix A Entropy , the information entropy weight of the index data l in P is obtained.

[0089]

[0090] It can be understood that the information entropy weight of the index data in other subspaces of P can also be obtained in the above manner. In this embodiment, the comparison matrix is generated by information entropy, and the weight of the information entropy of the index data in the entire comparison matrix is determined, so that the information entropy weight is obtained, and the determination result of the information entropy weight is more objective and accurate.

[0091] Optionally, the similarity weight of each index data in the corresponding index subspace is determined, including:

[0092] For each index data, a second similarity between the index data and target index data is calculated, and the target index data is other index data in the index subspace corresponding to the index data;

[0093] The sum of all second similarities is summed to obtain a comprehensive similarity of the index data;

[0094] The comprehensive similarity is normalized to obtain the similarity weight of the index data.

[0095] In this embodiment, for each index subspace P l , the second similarity between the index data and other index data (wherein s, t ∈ {1, 2, …, m l , t≠s) is calculated:

[0096]

[0097] wherein, are the data vectors corresponding to the index data and respectively.

[0098] The sum of is summed to obtain the comprehensive similarity of the index According to the comprehensive similarity of the indexes, the sum normalization method is used to calculate P l Index data Similarity-based weight:

[0099]

[0100] It can be understood that the index data in other subspaces of P can also obtain corresponding similarity weights in the above manner.

[0101] Further, for the index subspace P l , the information entropy weight of the index data is multiplied by the similarity weight, and the sum normalization is performed to obtain the final weight of the index data :

[0102]

[0103] It can be understood that the index data in other subspaces of P can also obtain corresponding final weights in the above manner.

[0104] In an embodiment, on the basis of the embodiment shown in Figure 1 , the above step S103 specifically includes:

[0105] According to the target weight, the index data of each index subspace is linearly combined to obtain upper-level index data of each index subspace.

[0106] The upper-level index data of each index subspace is iteratively linearly combined until the iteration number reaches a preset number, and the iteration is stopped to obtain a security detection model.

[0107] In this embodiment, for each index subspace of P , according to the final weight of the calculated index data, the data vectors corresponding to the index data are linearly combined to obtain a data vector corresponding to an upper-level index of the index subspace. The above linear combination process is repeated for the index subspace upper-level index to obtain a higher-level index. This process is iterated until the requirements are met (for example, three iterations can meet the requirements), and further a index system for power grid equipment security quantitative evaluation is generated. Compared with the artificial establishment of the index system, this embodiment is more objective and accurate.

[0108] Referring to Figure 2 , Figure 2A flowchart of a security detection method of a power grid device is shown. The security detection method of the power grid device described in the embodiment of the present application can be applied to a computer device, including but not limited to a smart phone, a tablet computer, a desktop computer, a supercomputer, a personal digital assistant, a physical server, and a cloud server, and the like. It should be understood that the computer device implementing the above generation method and the computer device implementing the security detection method can be the same or different. The generation method of the embodiment of the present application includes steps S201 to S202, which are described in detail as follows:

[0109] In step S201, state operation data of the power grid device is obtained.

[0110] In step S202, the state operation data is analyzed according to a security detection model to obtain a security quantization value of the power grid device. The security quantization is used to evaluate the security degree of the power grid device, and the security detection model is based on Figure 1 The generation method of the embodiment shown in the figure is generated.

[0111] In steps S201 and S202, state operation data related to security in the power grid device is collected, the bottom layer index data in the state operation data is input into the security detection model, and the linear weighting method is used in combination with the index system to sum up layer by layer to calculate the security quantization evaluation value of the power grid device.

[0112] In order to execute the generation method of the security detection model corresponding to the above method embodiment to realize the corresponding functions and technical effects. Referring to Figure 3 , Figure 3 A structure block diagram of a security detection model generation device provided by an embodiment of the present application is shown. For ease of illustration, only the part related to the present embodiment is shown. The security detection model generation device provided by the embodiment of the present application includes:

[0113] The first acquisition module 301 is configured to obtain state operation data of the power grid device, and the state operation data includes a plurality of index data.

[0114] The determination module 302 is configured to determine an index subspace corresponding to each index data based on a first similarity between the index data.

[0115] The first analysis module 303 is configured to analyze the index data in each index subspace to obtain a target weight of each index data in the corresponding index subspace.

[0116] The generation module 304 is configured to generate a security detection model based on the plurality of index data and the target weight.

[0117] In an embodiment, the determining module 302 comprises:

[0118] The selecting unit is configured to select K index data from the plurality of index data as initial index data.

[0119] The first calculating unit is configured to calculate a first similarity between each index data and the initial index data.

[0120] The clustering unit is configured to cluster the plurality of first similarities, and divide index data corresponding to first similarities in a same cluster into a same index subspace, to obtain a plurality of index subspaces.

[0121] In an embodiment, the first analyzing module 303 comprises:

[0122] The first determining unit is configured to determine an information entropy weight of each index data in a corresponding index subspace.

[0123] The second determining unit is configured to calculate a similarity weight of each index data in a corresponding index subspace.

[0124] The second calculating unit is configured to calculate a final weight of the index data based on the information entropy weight and the similarity weight.

[0125] In an embodiment, the first determining unit comprises:

[0126] The calculating sub-unit is configured to calculate, for each index data, an information entropy of the index data in a corresponding index subspace.

[0127] The comparing sub-unit is configured to compare the information entropy of the index data with information entropies of target index data to obtain an information entropy comparison matrix, the target index data being other index data in the index subspace corresponding to the index data.

[0128] The determining sub-unit is configured to determine the information entropy weight of the index data according to the information entropy comparison matrix.

[0129] In an embodiment, the second determining unit comprises:

[0130] The calculating sub-unit is configured to calculate, for each index data, a second similarity between the index data and target index data, the target index data being other index data in the index subspace corresponding to the index data.

[0131] The summing sub-unit is configured to sum all the second similarities to obtain a comprehensive similarity of the index data.

[0132] The normalization sub-unit is configured to normalize the comprehensive similarity to obtain the similarity weight of the index data.

[0133] In one embodiment, the generation module 304 includes:

[0134] The combination unit is used to linearly combine the indicator data of each indicator subspace according to the final weight value to obtain the upper-level indicator data of each indicator subspace.

[0135] The iterative unit is used to iteratively combine the upper-level indicator data of each indicator subspace until the number of iterations reaches the preset number, at which point the iteration stops and the security detection model is obtained.

[0136] The aforementioned security detection model generation apparatus can implement the security detection model generation method corresponding to the above method embodiments. The options in the above method embodiments are also applicable to this embodiment, and will not be detailed here. The remaining content of this application embodiment can be referred to the content of the above method embodiments, and will not be repeated in this embodiment.

[0137] To implement the safety detection method for power grid equipment corresponding to the above method embodiments, and to achieve the corresponding functions and technical effects. See also Figure 4 , Figure 4 This diagram illustrates a structural block diagram of a safety detection device for power grid equipment according to an embodiment of this application. For ease of explanation, only the parts relevant to this embodiment are shown. The safety detection device for power grid equipment provided in this embodiment includes:

[0138] The second acquisition module 401 is used to acquire the status operation data of power grid equipment;

[0139] The second analysis module 402 is used to perform data analysis on the state operation data according to the safety detection model to obtain the safety quantification value of the power grid equipment. The safety quantification is used to evaluate the safety level of the power grid equipment. The safety detection model is based on... Figure 1 The method described in the embodiment is used to generate the product.

[0140] The aforementioned power grid equipment safety detection device can implement the power grid equipment safety detection method corresponding to the above method embodiments. The remaining content of this application embodiment can be referred to the content of the above method embodiments, and will not be repeated in this embodiment.

[0141] Figure 5 This is a schematic diagram of the structure of a computer device provided in an embodiment of this application. Figure 5 As shown, the computer device 5 of this embodiment includes: at least one processor 50 ( Figure 5 (Only one is shown) a processor, a memory 51, and a computer program 52 stored in the memory 51 and executable on the at least one processor 50, wherein the processor 50 executes the computer program 52 to implement the steps in any of the above method embodiments.

[0142] The computer device 5 can be a host computer, a tablet computer, a desktop computer, etc. The computer device can include but is not limited to a processor 50, a memory 51. Those skilled in the art can understand that, Figure 5 The computer device 5 is only an example and does not constitute a limitation on the computer device 5, and can include more or fewer components than shown, or combine certain components, or different components, for example, can also include an input / output device, a network access device, etc.

[0143] The processor 50 can be a central processing unit (CPU), and can also be other general-purpose processors, a digital signal processor (DSP), an application specific integrated circuit (ASIC), a field-programmable gate array (FPGA) or other programmable logic device, discrete gate or transistor logic, discrete hardware components, etc. The general-purpose processor can be a microprocessor or can also be any conventional processor.

[0144] The memory 51 can be an internal storage unit of the computer device 5 in some embodiments, for example, a hard disk or a memory of the computer device 5. The memory 51 can also be an external storage device of the computer device 5 in other embodiments, for example, a plug-in hard disk, a smart media card (SMC), a secure digital (SD) card, a flash card, etc. Further, the memory 51 can include both the internal storage unit and the external storage device of the computer device 5. The memory 51 is used to store an operating system, an application program, a boot loader, data, and other programs, for example, program codes of the computer program, etc. The memory 51 can also be used to temporarily store data that has been output or will be output.

[0145] In addition, the embodiment of the present application further provides a computer readable storage medium, the computer readable storage medium stores a computer program, and the computer program is executed by a processor to implement the steps in any method embodiment described above.

[0146] The embodiment of the present application provides a computer program product, when the computer program product runs on a computer device, so that the computer device executes the steps in each method embodiment described above.

[0147] In several embodiments provided by the present application, it can be understood that each block in the flowchart or block diagram can represent a module, a segment or a portion of code which includes one or more executable instructions for implementing the specified logic function. It should also be noted that in some alternative implementations, the functions noted in the blocks can occur out of the order noted in the figure. For example, two blocks noted in succession can in fact be executed substantially concurrently or in the reverse order, depending on the functionality involved.

[0148] If the functions are realized in the form of software function modules and sold or used as independent products, they can be stored in a computer readable storage medium. Based on such understanding, the technical solutions of the present application or the parts of the technical solutions which essentially contribute to the prior art or the parts of the technical solutions can be embodied in the form of a software product. The computer software product is stored in a storage medium and includes a plurality of instructions for causing a computer device to perform all or part of the steps of the methods described in the embodiments of the present application. The aforementioned storage medium includes: a U disk, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk, and various media capable of storing program codes.

[0149] The specific embodiments described above further illustrate the purposes, technical solutions and beneficial effects of the present application. It should be understood that the above description is only for specific embodiments of the present application and is not intended to limit the protection scope of the present application. It is particularly pointed out that any modification, equivalent replacement, improvement, etc. made by those skilled in the art within the spirit and principles of the present application shall be included in the protection scope of the present application.

Claims

1. A method for generating a security detection model, characterized in that, include: Acquire status operation data of power grid equipment, the status operation data including multiple indicator data; The status operation data refers to the status data of power grid equipment during safe operation, including voltage, current, and temperature. Based on the first similarity between the indicator data, determine the indicator subspace corresponding to each indicator data; Data analysis is performed on the indicator data in each indicator subspace to obtain the target weight of each indicator data in the corresponding indicator subspace; A security detection model is generated based on multiple indicator data and the target weights; The operational data is analyzed based on the security detection model to obtain a quantitative value for the security of the power grid equipment. The step of performing data analysis on the indicator data within each indicator subspace to obtain the target weight of each indicator data in the corresponding indicator subspace includes: Determine the information entropy weight of each of the indicator data in the corresponding indicator subspace; Determine the similarity weight of each indicator data in the corresponding indicator subspace; The final weight of the indicator data is obtained by multiplying the information entropy weight and similarity weight, summing and normalizing the results. The process of generating a security detection model based on multiple indicator data and the target weights includes: Based on the target weight, the indicator data of each indicator subspace are linearly combined to obtain the upper-level indicator data of each indicator subspace. The upper-level indicator data of each indicator subspace are iteratively linearly combined until the number of iterations reaches a preset number, at which point the iteration stops, and the security detection model is obtained.

2. The generation method as described in claim 1, characterized in that, The step of determining the indicator subspace corresponding to each indicator data based on the first similarity between the indicator data includes: From the multiple sets of indicator data, select K indicator data as initial indicator data; Calculate the first similarity between each of the indicator data and the initial indicator data; Clustering is performed on multiple first similarities, and the index data corresponding to the first similarities in the same cluster are divided into the same index subspace to obtain multiple index subspaces.

3. The generation method as described in claim 1, characterized in that, Determining the information entropy weight of each indicator data in the corresponding indicator subspace includes: For each of the aforementioned indicator data, calculate the information entropy of the indicator data in the corresponding indicator subspace; The information entropy of the indicator data is compared with the information entropy of the target indicator data to obtain an information entropy comparison matrix. The target indicator data is other indicator data in the indicator subspace corresponding to the indicator data. The information entropy weights of the indicator data are determined based on the information entropy comparison matrix.

4. The generation method as described in claim 1, characterized in that, Determining the similarity weight of each indicator data in the corresponding indicator subspace includes: For each of the indicator data, a second similarity is calculated between the indicator data and the target indicator data, where the target indicator data is other indicator data in the indicator subspace corresponding to the indicator data. Summing all the second similarities yields the comprehensive similarity of the index data; The overall similarity is normalized to obtain the similarity weight of the indicator data.

5. A method for safety testing of power grid equipment, characterized in that, include: Acquire status and operational data of power grid equipment; According to the security detection model, the state operation data is analyzed to obtain the security quantification value of the power grid equipment. The security quantification is used to evaluate the security level of the power grid equipment. The security detection model is generated based on the generation method described in any one of claims 1 to 4.

6. A device for generating a security detection model, characterized in that, include: The first acquisition module is used to acquire the status operation data of power grid equipment, the status operation data including multiple indicator data; The status operation data refers to the status data of power grid equipment during safe operation, including voltage, current, and temperature. The determination module is used to determine the indicator subspace corresponding to each indicator data based on the first similarity between the indicator data; The first analysis module is used to perform data analysis on the indicator data in each indicator subspace to obtain the target weight of each indicator data in the corresponding indicator subspace. The generation module is used to generate a security detection model based on multiple indicator data and the target weight; and to analyze the state operation data according to the security detection model to obtain the security quantification value of the power grid equipment. The first analysis module is used to perform data analysis on the indicator data within each indicator subspace to obtain the target weight of each indicator data in the corresponding indicator subspace, including: Determine the information entropy weight of each of the indicator data in the corresponding indicator subspace; Determine the similarity weight of each indicator data in the corresponding indicator subspace; The final weight of the indicator data is obtained by multiplying the information entropy weight and similarity weight, summing and normalizing the results. The generation module is used to generate a security detection model based on multiple indicator data and the target weight, including: Based on the target weight, the indicator data of each indicator subspace are linearly combined to obtain the upper-level indicator data of each indicator subspace. The upper-level indicator data of each indicator subspace are iteratively linearly combined until the number of iterations reaches a preset number, at which point the iteration stops, and the security detection model is obtained.

7. A safety detection device for power grid equipment, characterized in that, include: The second acquisition module is used to acquire the status and operation data of power grid equipment; The second analysis module is used to perform data analysis on the state operation data according to the security detection model to obtain the security quantification value of the power grid equipment. The security quantification is used to evaluate the security level of the power grid equipment. The security detection model is generated based on the generation method according to any one of claims 1 to 4.

8. A computer device, characterized in that, It includes a memory and a processor, wherein the memory stores computer-readable instructions, and the processor executes the computer-readable instructions to implement the steps of the method for generating a security detection model as described in any one of claims 1 to 4, or the steps of the method for security detection of power grid equipment as described in claim 5.

Images