Transformer graph model construction method and apparatus, computer device, and storage medium

Abstract

The present application relates to a transformer graph model construction method and apparatus, a computer device, and a storage medium. The method comprises: first extracting entity points and relation edges from device data of transformers; constructing initial graph models of the transformers; clustering attribute data of entity points in the initial graph models; and processing the initial graph models on the basis of target clusters obtained by the clustering to obtain target graph models. By means of the method, when users use transformer graph models, there is no need to learn each transformer graph model in advance, thereby greatly improving the convenience of using transformer graph models.

Description

Transformer diagram model construction method, device, computer equipment and storage medium Technical Field

[0001] The present application relates to the field of automation technology, and in particular to a method, device, computer equipment, and storage medium for constructing a transformer diagram model. Background Art

[0002] The graphical model of the transformer plays a vital role in the application field of the transformer. Using the graphical model of the transformer, all information related to the transformer can be easily queried.

[0003] However, with the development of society, the coverage of power grids has become increasingly extensive, requiring more and more transformers. However, different transformers may have different names, and the names of transformer-related components and units are also inconsistent. As a result, the current use of transformer diagram models has gradually become more limited. For example, the names of different transformer diagram models cannot be standardized, making it difficult for users to use other transformer diagram models smoothly. In order to master the use of other transformer diagram models, users need to learn in advance, which greatly reduces work efficiency.

[0004] Summary of the Invention

[0005] Based on this, it is necessary to provide a method, device, computer equipment and storage medium for constructing a transformer diagram model that can improve the convenience of using the transformer diagram model in order to address the above technical problems.

[0006] In a first aspect, the present application provides a method for constructing a transformer graph model. The method comprises:

[0007] Extract entity points and relationship edges from the equipment data of each transformer;

[0008] Based on the extracted entity points and relationship edges, the initial graph model corresponding to each transformer is constructed;

[0009] Cluster the attribute data of entity points in each initial graph model to obtain multiple target clusters;

[0010] According to multiple target clusters, each initial graph model is processed to obtain a target graph model corresponding to each transformer.

[0011] In one embodiment, the attribute data includes name data, and the attribute data of the entity points in each initial graph model are clustered to obtain multiple target clusters, including:

[0012] Determine the word vectors of the name data of the entity points in each initial graph model;

[0013] According to the word vectors of each name data, each name data is clustered to obtain multiple target clusters.

[0014] In one embodiment, each name data is clustered according to its word vector to obtain multiple target clusters, including:

[0015] Based on at least two initial cluster centers, cluster the name data according to the word vectors of the name data to obtain initial clusters corresponding to the initial cluster centers;

[0016] Determine the target cluster center corresponding to each initial cluster based on the word vectors of the name data contained in each initial cluster;

[0017] Determine whether the initial cluster center of any initial cluster is inconsistent with the target cluster center;

[0018] If so, each target cluster center is used as the new initial cluster center, and the operation of clustering each name data based on the word vector of each name data based on at least two initial cluster centers is returned to obtain the initial cluster corresponding to each initial cluster center;

[0019] If not, each initial cluster is used as the target cluster for the name data.

[0020] In one embodiment, each initial graph model is processed according to multiple target clusters to obtain a target graph model corresponding to each transformer, including:

[0021] Extract keywords from each target cluster;

[0022] The extracted keywords are used to replace the name data of the entity points in each initial graph model to obtain the target graph model corresponding to each transformer.

[0023] In one embodiment, the extracted keywords are used to replace the name data of the entity points in each initial graph model to obtain the target graph model corresponding to each transformer, including:

[0024] Obtaining at least two optional semantic information of the extracted keyword;

[0025] determining target semantic information from each optional semantic information according to the scores of each optional semantic information;

[0026] The extracted keywords are used to replace the name data of the entity points in each initial graph model to obtain the intermediate graph model of each transformer;

[0027] The target graph model corresponding to each transformer is obtained by removing all optional semantic information except the target semantic information corresponding to the keywords in each intermediate graph model.

[0028] In one embodiment, extracting keywords from each target cluster includes:

[0029] The name data corresponding to the target cluster center of each target cluster is used as the keyword of each target cluster; or

[0030] The name data closest to the center of each target cluster is used as the keyword of each target cluster.

[0031] In one embodiment, the method further comprises:

[0032] Display each target graph model to the operation and maintenance terminal;

[0033] Optimize each target graph model based on the editing operation of each target graph model

[0034] In a second aspect, the present application also provides a device for constructing a transformer graph model. The device includes:

[0035] A data extraction module is used to extract entity points and relationship edges from the equipment data of each transformer;

[0036] The model construction module is used to construct the initial graph model corresponding to each transformer based on the extracted entity points and relationship edges;

[0037] The data clustering module is used to cluster the attribute data of entity points in each initial graph model to obtain multiple target clusters;

[0038] The target determination module is used to process each initial graph model according to multiple target clusters to obtain a target graph model corresponding to each transformer.

[0039] In a third aspect, the present application further provides a computer device. The computer device includes a memory and a processor. The memory stores a computer program. When the processor executes the computer program, the following steps are performed:

[0040] Extract entity points and relationship edges from the equipment data of each transformer;

[0041] Based on the extracted entity points and relationship edges, the initial graph model corresponding to each transformer is constructed;

[0042] Cluster the attribute data of entity points in each initial graph model to obtain multiple target clusters;

[0043] According to multiple target clusters, each initial graph model is processed to obtain a target graph model corresponding to each transformer.

[0044] In a fourth aspect, the present application further provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the following steps:

[0045] Extract entity points and relationship edges from the equipment data of each transformer;

[0046] Based on the extracted entity points and relationship edges, the initial graph model corresponding to each transformer is constructed;

[0047] Cluster the attribute data of entity points in each initial graph model to obtain multiple target clusters;

[0048] According to multiple target clusters, each initial graph model is processed to obtain a target graph model corresponding to each transformer.

[0049] In a fifth aspect, the present application further provides a computer program product. The computer program product includes a computer program that, when executed by a processor, implements the following steps:

[0050] Extract entity points and relationship edges from the equipment data of each transformer;

[0051] Based on the extracted entity points and relationship edges, the initial graph model corresponding to each transformer is constructed;

[0052] Cluster the attribute data of entity points in each initial graph model to obtain multiple target clusters;

[0053] According to multiple target clusters, each initial graph model is processed to obtain a target graph model corresponding to each transformer.

[0054] The above-mentioned transformer graph model construction method, device, computer equipment and storage medium. First, based on the entity points and relationship edges extracted from the equipment data of each transformer, an initial graph model of each transformer is constructed. Then, the attribute data of the entity points in the initial graph model are clustered. Each initial graph model is processed according to the target cluster obtained after clustering to obtain a target graph model. The target graph model obtained in the above method is obtained after clustering the initial graph model. The clustering process can unify the attribute data of the entity points in the initial graph model. In this way, when using the transformer graph model, the user does not need to pre-learn each transformer graph model, which greatly increases the convenience of using the transformer graph model. BRIEF DESCRIPTION OF THE DRAWINGS

[0055] FIG1 is an application environment diagram of a method for constructing a transformer graph model in one embodiment;

[0056] FIG2 is a schematic flow chart of a method for constructing a transformer graph model in one embodiment;

[0057] FIG3 is a schematic diagram of a process for obtaining a target cluster in one embodiment;

[0058] FIG4 is a schematic diagram of a process for obtaining a target graph model in one embodiment;

[0059] FIG5 is a schematic diagram of a process for optimizing a target graph model in one embodiment;

[0060] FIG6 is a flow chart of a method for constructing a transformer graph model in another embodiment;

[0061] FIG7 is a structural block diagram of a device for constructing a transformer graphical model in one embodiment;

[0062] FIG8 is a structural block diagram of a device for constructing a transformer graphical model in another embodiment;

[0063] FIG9 is a structural block diagram of a device for constructing a transformer graphical model in yet another embodiment;

[0064] FIG10 is a structural block diagram of a device for constructing a transformer graphical model in yet another embodiment;

[0065] FIG11 is a diagram showing the internal structure of a computer device in one embodiment. DETAILED DESCRIPTION

[0066] In order to make the purpose, technical solutions and advantages of this application more clear, the following further describes this application in detail with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain this application and are not intended to limit this application.

[0067] The method for constructing a transformer graph model provided in an embodiment of the present application can be applied to an application environment as shown in Figure 1. In one embodiment, a computer device is provided, which can be a server, and its internal structure diagram can be shown in Figure 1. The computer device includes a processor, a memory, and a network interface connected via a system bus. The processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program, and a database. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The database of the computer device is used to store data required for related processing. The network interface of the computer device is used to communicate with an external terminal via a network connection. When the computer program is executed by the processor, it implements the method for constructing a transformer graph model shown in any of the following embodiments.

[0068] In one embodiment, as shown in FIG2 , a method for constructing a transformer graph model is provided. The method is described by applying the method to the computer device in FIG1 as an example, and includes the following steps:

[0069] S201: extract entity points and relationship edges from the equipment data of each transformer.

[0070] Among them, the equipment data of the transformer includes the attribute data of the transformer equipment, which may specifically include but is not limited to transformer equipment ledger data, transformer equipment supplier management data and transformer equipment topology relationship data; the transformer graph model is a graph used to intuitively describe transformer-related information; the entity point represents a random variable in the graph model, and the relationship edge represents the dependency relationship between these random variables; for example, the transformer and the manufacturer can be used as the entity points, and the relationship between the transformer manufacturer and the transformer can be used as the relationship edge.

[0071] Optionally, the transformer device data can be input into the extraction model to extract the transformer device's entity points and relationship edges. Alternatively, the entity point and relationship edge extraction requirements can be obtained from the server's storage system. Based on the extraction requirements, the entity points and relationship edges can be extracted from the device data of each transformer.

[0072] S202: Construct an initial graph model corresponding to each transformer based on the extracted entity points and relationship edges.

[0073] Specifically, by connecting each entity point according to the extracted relationship edges, the construction of the initial graph model corresponding to each transformer is completed.

[0074] Optionally, a graphical model construction model can be obtained in advance based on the equipment data of historical transformers and the historical transformer graphical model training. The entity points and transformation relationships extracted from the equipment data of the transformers can be input into the graphical model construction model to obtain the initial graphical model corresponding to each transformer.

[0075] S203: Cluster the attribute data of the entity points in each initial graph model to obtain multiple target clusters.

[0076] Clustering involves classifying the attribute data of the entities in each initial graph model. Specifically, clustering criteria uploaded by the user can be obtained, and the attribute data of the entities in each initial graph model can be clustered. The attribute data of each clustered entity point is then used as a target cluster, resulting in multiple target clusters.

[0077] S204 : Processing each initial graph model according to the multiple target clusters to obtain a target graph model corresponding to each transformer.

[0078] Specifically, disambiguation processing is performed on each target cluster to eliminate the ambiguous parts in each target cluster. Since each target cluster is constructed based on the attribute data of the entity points in the initial graph model, after the target cluster is disambiguated, the corresponding initial graph model is adaptively adjusted, which is equivalent to disambiguating the initial graph model. The initial graph model after disambiguation is the target graph model corresponding to each transformer.

[0079] In the above embodiment, the entity points and relationship edges are first extracted from the equipment data of each transformer to construct the initial graph model of each transformer, and then the attribute data of the entity points in the initial graph model are clustered. According to the target cluster obtained after clustering, each initial graph model is processed to obtain a target graph model. The target graph model obtained in the above method is obtained after clustering the initial graph model, and the clustering process can unify the entity point attribute data in the initial graph model. In this way, the user does not need to pre-learn each transformer graph model when using the transformer graph model, which greatly increases the convenience of using the transformer graph model.

[0080] The above embodiment generally introduces a method for constructing a target model based on transformer device data. The process of obtaining multiple target clusters is an important step in constructing a target graph model. Therefore, in this embodiment, as shown in FIG3 , the detailed process of obtaining multiple target clusters is described in detail. The specific method includes:

[0081] S301, determining the word vector of the name data of the entity point in each initial graph model.

[0082] Among them, the word vector is a vector converted from words in natural language. The more complex the semantics of a word in natural language, the higher the dimension of its corresponding word vector.

[0083] Specifically, a word vector training model is obtained according to word vector algorithm training, the name data of the entity points in each initial graph model is input into the word vector training model, and the word vector of the name data of the entity points in each initial graph model is output.

[0084] S302 , clustering each name data according to the word vector of each name data to obtain multiple target clusters.

[0085] Specifically, we can first count the dimensions of each word vector, and group the name data corresponding to word vectors with the same dimension or word vectors with similar dimensions. Each group of name data corresponds to a target cluster.

[0086] Optionally, after calculating the word vector distances between each word vector, the word vectors are sorted according to their distances, and the name data corresponding to the word vectors whose word vector distances are lower than a preset distance threshold are grouped together, with each group of name data corresponding to a target cluster.

[0087] Optionally, each name data may be clustered based on at least two initial cluster centers according to the word vectors of each name data to obtain initial clusters corresponding to each initial cluster center; the target cluster center corresponding to each initial cluster may be determined according to the word vectors of the name data contained in each initial cluster; it may be determined whether the initial cluster center of any initial cluster is inconsistent with the target cluster center; if so, each target cluster center is used as the new initial cluster center, and the operation of clustering each name data based on at least two initial cluster centers according to the word vectors of each name data to obtain initial clusters corresponding to each initial cluster center is returned; if not, each initial cluster is used as the target cluster for the name data.

[0088] Specifically, according to the number of name data types (for example, k types, for the sake of convenience, this implementation takes k types as an example, where k is not less than 2), randomly select k word vectors of name data as k initial cluster centers; for each word vector, calculate its distance to the k initial clusters to obtain k distance values, compare the obtained k distance values, and classify the word vector into the initial cluster where the initial cluster center corresponding to the minimum distance is located. In the same way, classify all word vectors to obtain k initial clusters; according to the word vectors of each name data in the initial cluster, connect each word vector in pairs to form a polygon, and use parsing to obtain the word vectors. The center point of the polygon is geometrically calculated, which is the center point of each initial cluster, and is used as the target cluster center of the initial cluster. It is also determined whether the initial cluster center of any initial cluster is inconsistent with the target cluster center. If not, k initial clusters are used as target clusters. If so, the target cluster center is used as the initial cluster center, and the operation of clustering each name data based on at least two initial cluster centers and the word vector of each name data is returned to obtain the initial cluster corresponding to each initial cluster center until all target cluster centers coincide with the position of the initial cluster center. The above process of obtaining the initial cluster corresponding to each initial cluster center has been explained in detail in the previous article and will not be repeated here.

[0089] In the above embodiment, the word vectors of the name data of the entity points in each initial graph model are first determined, and then the name data are clustered according to the word vectors to obtain multiple target clusters. This method converts the abstract name data into specific word vectors when determining the target cluster, which greatly increases the accuracy of determining the target cluster.

[0090] The above embodiment describes how to obtain a target cluster. In this embodiment, as shown in FIG4 , it describes how to obtain a target graph model based on the target cluster. The specific method includes:

[0091] S401: extract keywords from each target cluster.

[0092] Optionally, the name data corresponding to the target cluster center of each target cluster may be used as the keyword of each target cluster; or the name data closest to the target cluster center in each target cluster may be used as the keyword of each target cluster.

[0093] Specifically, if the cluster center of the target cluster coincides with the word vector of any name data, the name data will be used as the keyword of the target cluster; if the cluster center of the target cluster does not coincide with the word vector of any name data, the name data in each target cluster that is closest to the target cluster center will be used as the keyword of each target cluster. It is necessary to calculate the distance between the word vector of all name data in each target cluster and the target cluster center. The specific calculation method is to calculate the Euclidean distance and / or cosine distance between the word vector of each name data and the target cluster center according to the mathematical formula. The Euclidean distance can be used as the distance between the word vector and the target cluster center; the cosine distance can also be used as the distance between the word vector and the target cluster center; the comprehensive distance can also be obtained according to the preset calculation method based on the Euclidean distance and cosine distance, and the comprehensive distance is used as the distance between the word vector and the target cluster center; the name data corresponding to the word vector closest to the target cluster center is selected as the keyword of the target cluster.

[0094] S402 , using the extracted keywords to replace the name data of the entity points in each initial graph model, to obtain the target graph model corresponding to each transformer.

[0095] Specifically, all name data except keywords are deleted, and the extracted keywords are used as the name data of the entity points in each initial graph model to obtain the target graph model corresponding to each transformer.

[0096] Optionally, at least two optional semantic information of the extracted keywords can be obtained first; based on the score of each optional semantic information, the target semantic information can be determined from each optional semantic information; the name data of the entity points in each initial graph model can be replaced with the extracted keywords to obtain the intermediate graph model of each transformer; finally, the optional semantic information other than the target semantic information corresponding to the keywords in each intermediate graph model can be eliminated to obtain the target graph model corresponding to each transformer.

[0097] Specifically, first parse out all the semantic information of the keyword, score the parsed semantic information according to the preset scoring rules, sort all the semantic information of the keyword according to the score, use the semantic information with the highest score as the target semantic information of the keyword, use the extracted keyword as the name data of the entity point in each initial graph model, and you can get the intermediate graph model. Delete other semantic information with lower scores (that is, other semantic information except the target semantic information) from the intermediate graph model to get the target graph model corresponding to each transformer.

[0098] In the above embodiment, the keywords of the target cluster are first extracted, and then the keywords are used to replace the name data of the entity points in each initial graph model, so as to obtain the target graph model corresponding to each transformer. The target graph models constructed using this method all use keywords as name data, so that the target graph models corresponding to each transformer can unify the name data.

[0099] Based on the above embodiment, as shown in FIG5 , in order to make the target graph model more in line with the user's needs, the target graph model may be optimized according to the user's needs. The specific method includes:

[0100] S501: Display each target graph model to the operation and maintenance terminal.

[0101] Specifically, each target graph model is displayed on the display screen of the operation and maintenance terminal. Alternatively, the name of each target graph model is displayed on the display screen of the operation and maintenance terminal, and after the user clicks, the target graph model clicked by the user is displayed.

[0102] S502: Optimize each target graph model according to the editing operation on each target graph model.

[0103] Specifically, the entity points and relationship edges of the target graph model may be optimized according to the user's editing operation on the target graph model on the display screen, thereby achieving optimization of each target graph model.

[0104] Optionally, the user's editing operations on each target graph model can also be made into optimization requirements. After the server obtains the optimization requirements, it parses the optimization requirements for the target graph model entity points and the optimization requirements for the target graph model relationship edges in the optimization requirements, and optimizes the entity points and relationship edges of the target graph model based on the optimization requirements for the target graph model entity points and the optimization requirements for the target graph model relationship edges, thereby completing the optimization of the target graph model.

[0105] In the above embodiment, the target graph model is first displayed, and then each target graph model is optimized according to the editing operation on each target graph model. This method can make the optimized target transformer graph model more in line with user needs.

[0106] In order to more comprehensively demonstrate this solution, this embodiment provides an optional method for constructing a transformer graph model, as shown in FIG6 :

[0107] S601: Extract entity points and relationship edges from the equipment data of each transformer.

[0108] S602: Construct an initial graph model corresponding to each transformer based on the extracted entity points and relationship edges.

[0109] S603: Determine the word vector of the name data of the entity point in each initial graph model.

[0110] S604 , clustering each name data based on the word vector of each name data based on at least two initial cluster centers to obtain initial clusters corresponding to each initial cluster center.

[0111] S605 , determining the target cluster center corresponding to each initial cluster according to the word vector of the name data contained in each initial cluster.

[0112] S606: Determine whether the initial cluster center of any initial cluster is inconsistent with the target cluster center. If so, execute S607; if not, execute S608.

[0113] S607: Each target cluster center is used as a new initial cluster center, and the operation of clustering each name data according to the word vector of each name data based on at least two initial cluster centers is returned to obtain the initial cluster corresponding to each initial cluster center.

[0114] S608: Each initial cluster is used as a target cluster for the name data.

[0115] S609 , using the name data corresponding to the target cluster center of each target cluster as the keyword of each target cluster; or using the name data closest to the target cluster center in each target cluster as the keyword of each target cluster.

[0116] S610: Obtain at least two optional semantic information of the extracted keyword.

[0117] S611 : Determine target semantic information from each optional semantic information according to the score of each optional semantic information.

[0118] S612: Using the extracted keywords, replace the name data of the entity points in each initial graph model to obtain the intermediate graph model of each transformer.

[0119] S613: Eliminate other optional semantic information except the target semantic information corresponding to the keywords in each intermediate graph model to obtain the target graph model corresponding to each transformer.

[0120] S614: Display each target graph model to the operation and maintenance terminal.

[0121] S615: Optimize each target graph model according to the editing operation on each target graph model.

[0122] The specific process of the above S601-S615 can be found in the description of the above method embodiment. The implementation principle and technical effects are similar and will not be repeated here.

[0123] It should be understood that, although the steps in the flowcharts of the above embodiments are shown in sequence as indicated by the arrows, these steps are not necessarily performed in the order indicated by the arrows. Unless otherwise specified herein, there is no strict order restriction on the execution of these steps, and these steps can be performed in other orders. Moreover, at least a portion of the steps in the flowcharts of the above embodiments may include multiple steps or multiple stages, and these steps or stages are not necessarily performed at the same time, but can be performed at different times. The execution order of these steps or stages is not necessarily to be performed in sequence, but can be performed in turn or alternately with other steps or at least a portion of steps or stages in other steps.

[0124] Based on the same inventive concept, embodiments of the present application also provide a device for constructing a transformer graphical model for implementing the aforementioned method for constructing a transformer graphical model. The solution provided by this device is similar to the solution described in the aforementioned method. Therefore, the specific limitations of the one or more embodiments of the device for constructing a transformer graphical model provided below can be found in the aforementioned definition of the method for constructing a transformer graphical model, and will not be further elaborated here.

[0125] In one embodiment, as shown in FIG7 , a device 7 for constructing a transformer graph model is provided, comprising: a data extraction module 70 , a model construction module 71 , a data clustering module 72 , and a target determination module 73 , wherein:

[0126] A data extraction module 70 is used to extract entity points and relationship edges from the equipment data of each transformer;

[0127] A model building module 71 is used to build an initial graph model corresponding to each transformer based on the extracted entity points and relationship edges;

[0128] A data clustering module 72 is used to cluster the attribute data of the entity points in each initial graph model to obtain multiple target clusters;

[0129] The target determination module 73 is used to process each initial graph model according to multiple target clusters to obtain a target graph model corresponding to each transformer.

[0130] In another embodiment, as shown in FIG8 , the data clustering module 72 in FIG7 includes:

[0131] A word vector determination unit 720 is used to determine the word vector of the name data of each entity point in the initial graph model;

[0132] The data clustering unit 721 is used to cluster each name data according to the word vector of each name data to obtain multiple target clusters.

[0133] In another embodiment, the data clustering unit 721 in FIG. 7 is specifically configured to:

[0134] Based on at least two initial cluster centers, each name data is clustered according to the word vector of each name data to obtain the initial cluster corresponding to each initial cluster center; based on the word vector of the name data contained in each initial cluster, the target cluster center corresponding to each initial cluster is determined; it is judged whether the initial cluster center of any initial cluster is inconsistent with the target cluster center; if so, each target cluster center is used as the new initial cluster center, and the operation of clustering each name data based on at least two initial cluster centers and according to the word vector of each name data to obtain the initial cluster corresponding to each initial cluster center is returned; if not, each initial cluster is used as the target cluster for the name data.

[0135] In another embodiment, as shown in FIG9 , the target determination module 73 in FIG7 includes:

[0136] A keyword acquisition unit 730 is used to extract keywords from each target cluster;

[0137] The data replacement unit 731 is used to replace the name data of the entity points in each initial graph model with the extracted keywords to obtain the target graph model corresponding to each transformer.

[0138] In another embodiment, the keyword acquisition unit 730 in FIG. 9 is specifically configured to:

[0139] The name data corresponding to the target cluster center of each target cluster is used as the keyword of each target cluster; or the name data closest to the target cluster center in each target cluster is used as the keyword of each target cluster.

[0140] In another embodiment, the data replacement unit 731 in FIG. 9 is specifically configured to:

[0141] Obtain at least two optional semantic information of the extracted keywords; determine the target semantic information from each optional semantic information based on the score of each optional semantic information; use the extracted keywords to replace the name data of the entity points in each initial graph model to obtain the intermediate graph model of each transformer; eliminate the other optional semantic information except the target semantic information corresponding to the keywords in each intermediate graph model to obtain the target graph model corresponding to each transformer.

[0142] In another embodiment, as shown in FIG10 , the device 7 for constructing the transformer graphical model in FIG7 further includes:

[0143] Model display module 74, used to display each target graph model to the operation and maintenance terminal;

[0144] The model optimization module 75 is used to optimize each target graph model according to the editing operation on each target graph model.

[0145] Each module in the aforementioned transformer graphical model construction device can be implemented in whole or in part through software, hardware, or a combination thereof. Each module can be embedded in or independent of a processor in a computer device in the form of hardware, or can be stored in a memory in the computer device in the form of software, so that the processor can call and execute the corresponding operations of each module.

[0146] In one embodiment, a computer device is provided, which may be a terminal. Its internal structure diagram may be as shown in Figure 11. The computer device includes a processor, memory, an input / output interface, a communication interface, a display unit, and an input device. The processor, memory, and input / output interface are connected via a system bus, and the communication interface, display unit, and input device are connected to the system bus via the input / output interface. The processor of the computer device provides computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operating system and computer program stored in the non-volatile storage medium. The input / output interface of the computer device is used to exchange information between the processor and external devices. The communication interface of the computer device is used to communicate with external terminals via wired or wireless communication, which may be achieved via Wi-Fi, a mobile cellular network, NFC (near-field communication), or other technologies. When executed by the processor, the computer program implements a method for constructing a transformer diagram model. The display unit of the computer device is used to produce a visually visible image and may be a display screen, a projection device, or a virtual reality imaging device. The display screen can be a liquid crystal display screen or an electronic ink display screen, and the input device of the computer device can be a touch layer covering the display screen, or a button, trackball or touchpad set on the computer device casing, or an external keyboard, touchpad or mouse.

[0147] Those skilled in the art will understand that the structure shown in FIG11 is merely a block diagram of a portion of the structure related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may include more or fewer components than shown in the figure, or combine certain components, or have a different arrangement of components.

[0148] In one embodiment, a computer device is provided, including a memory and a processor, wherein a computer program is stored in the memory, and when the processor executes the computer program, the following steps are implemented:

[0149] Extract entity points and relationship edges from the equipment data of each transformer;

[0150] Based on the extracted entity points and relationship edges, the initial graph model corresponding to each transformer is constructed;

[0151] Cluster the attribute data of entity points in each initial graph model to obtain multiple target clusters;

[0152] According to multiple target clusters, each initial graph model is processed to obtain a target graph model corresponding to each transformer.

[0153] In one embodiment, when the processor executes the computer program, the processor further implements the following steps:

[0154] Determine the word vectors of the name data of the entity points in each initial graph model; cluster the name data according to the word vectors to obtain multiple target clusters.

[0155] In one embodiment, when the processor executes the computer program, the processor further implements the following steps:

[0156] Based on at least two initial cluster centers, each name data is clustered according to the word vector of each name data to obtain the initial cluster corresponding to each initial cluster center; based on the word vector of the name data contained in each initial cluster, the target cluster center corresponding to each initial cluster is determined; it is judged whether the initial cluster center of any initial cluster is inconsistent with the target cluster center; if so, each target cluster center is used as the new initial cluster center, and the operation of clustering each name data based on at least two initial cluster centers and according to the word vector of each name data to obtain the initial cluster corresponding to each initial cluster center is returned; if not, each initial cluster is used as the target cluster for the name data.

[0157] In one embodiment, when the processor executes the computer program, the processor further implements the following steps:

[0158] Keywords are extracted from each target cluster; the name data of the entity points in each initial graph model are replaced with the extracted keywords to obtain the target graph model corresponding to each transformer.

[0159] In one embodiment, when the processor executes the computer program, the processor further implements the following steps:

[0160] Obtain at least two optional semantic information of the extracted keywords; determine the target semantic information from each optional semantic information based on the score of each optional semantic information; use the extracted keywords to replace the name data of the entity points in each initial graph model to obtain the intermediate graph model of each transformer; eliminate the other optional semantic information except the target semantic information corresponding to the keywords in each intermediate graph model to obtain the target graph model corresponding to each transformer.

[0161] In one embodiment, when the processor executes the computer program, the processor further implements the following steps:

[0162] The name data corresponding to the target cluster center of each target cluster is used as the keyword of each target cluster; or the name data closest to the target cluster center in each target cluster is used as the keyword of each target cluster.

[0163] In one embodiment, when the processor executes the computer program, the processor further implements the following steps:

[0164] Display each target graph model to the operation and maintenance terminal; optimize each target graph model based on the editing operation on each target graph model.

[0165] In one embodiment, a computer-readable storage medium is provided, on which a computer program is stored. When the computer program is executed by a processor, the following steps are implemented:

[0166] Extract entity points and relationship edges from the equipment data of each transformer;

[0167] Based on the extracted entity points and relationship edges, the initial graph model corresponding to each transformer is constructed;

[0168] Cluster the attribute data of entity points in each initial graph model to obtain multiple target clusters;

[0169] According to multiple target clusters, each initial graph model is processed to obtain a target graph model corresponding to each transformer.

[0170] In one embodiment, when the computer program is executed by a processor, the following steps are further implemented:

[0171] Determine the word vectors of the name data of the entity points in each initial graph model; cluster the name data according to the word vectors to obtain multiple target clusters.

[0172] In one embodiment, when the computer program is executed by a processor, the following steps are further implemented:

[0173] Based on at least two initial cluster centers, each name data is clustered according to the word vector of each name data to obtain the initial cluster corresponding to each initial cluster center; based on the word vector of the name data contained in each initial cluster, the target cluster center corresponding to each initial cluster is determined; it is judged whether the initial cluster center of any initial cluster is inconsistent with the target cluster center; if so, each target cluster center is used as the new initial cluster center, and the operation of clustering each name data based on at least two initial cluster centers and according to the word vector of each name data to obtain the initial cluster corresponding to each initial cluster center is returned; if not, each initial cluster is used as the target cluster for the name data.

[0174] In one embodiment, when the computer program is executed by a processor, the following steps are further implemented:

[0175] Keywords are extracted from each target cluster; the name data of the entity points in each initial graph model are replaced with the extracted keywords to obtain the target graph model corresponding to each transformer.

[0176] In one embodiment, when the computer program is executed by a processor, the following steps are further implemented:

[0177] Obtain at least two optional semantic information of the extracted keywords; determine the target semantic information from each optional semantic information based on the score of each optional semantic information; use the extracted keywords to replace the name data of the entity points in each initial graph model to obtain the intermediate graph model of each transformer; eliminate the other optional semantic information except the target semantic information corresponding to the keywords in each intermediate graph model to obtain the target graph model corresponding to each transformer.

[0178] In one embodiment, when the computer program is executed by a processor, the following steps are further implemented:

[0179] The name data corresponding to the target cluster center of each target cluster is used as the keyword of each target cluster; or the name data closest to the target cluster center in each target cluster is used as the keyword of each target cluster.

[0180] In one embodiment, when the computer program is executed by a processor, the following steps are further implemented:

[0181] Display each target graph model to the operation and maintenance terminal; optimize each target graph model based on the editing operation on each target graph model.

[0182] In one embodiment, a computer program product is provided, comprising a computer program, which, when executed by a processor, implements the following steps:

[0183] Extract entity points and relationship edges from the equipment data of each transformer;

[0184] Based on the extracted entity points and relationship edges, the initial graph model corresponding to each transformer is constructed;

[0185] Cluster the attribute data of entity points in each initial graph model to obtain multiple target clusters;

[0186] According to multiple target clusters, each initial graph model is processed to obtain a target graph model corresponding to each transformer.

[0187] In one embodiment, when the computer program is executed by a processor, the following steps are further implemented:

[0188] Determine the word vectors of the name data of the entity points in each initial graph model; cluster the name data according to the word vectors to obtain multiple target clusters.

[0189] In one embodiment, when the computer program is executed by a processor, the following steps are further implemented:

[0190] Based on at least two initial cluster centers, each name data is clustered according to the word vector of each name data to obtain the initial cluster corresponding to each initial cluster center; based on the word vector of the name data contained in each initial cluster, the target cluster center corresponding to each initial cluster is determined; it is judged whether the initial cluster center of any initial cluster is inconsistent with the target cluster center; if so, each target cluster center is used as the new initial cluster center, and the operation of clustering each name data based on at least two initial cluster centers and according to the word vector of each name data to obtain the initial cluster corresponding to each initial cluster center is returned; if not, each initial cluster is used as the target cluster for the name data.

[0191] In one embodiment, when the computer program is executed by a processor, the following steps are further implemented:

[0192] Keywords are extracted from each target cluster; the name data of the entity points in each initial graph model are replaced with the extracted keywords to obtain the target graph model corresponding to each transformer.

[0193] In one embodiment, when the computer program is executed by a processor, the following steps are further implemented:

[0194] Obtain at least two optional semantic information of the extracted keywords; determine the target semantic information from each optional semantic information based on the score of each optional semantic information; use the extracted keywords to replace the name data of the entity points in each initial graph model to obtain the intermediate graph model of each transformer; eliminate the other optional semantic information except the target semantic information corresponding to the keywords in each intermediate graph model to obtain the target graph model corresponding to each transformer.

[0195] In one embodiment, when the computer program is executed by a processor, the following steps are further implemented:

[0196] The name data corresponding to the target cluster center of each target cluster is used as the keyword of each target cluster; or the name data closest to the target cluster center in each target cluster is used as the keyword of each target cluster.

[0197] In one embodiment, when the computer program is executed by a processor, the following steps are further implemented:

[0198] Display each target graph model to the operation and maintenance terminal; optimize each target graph model based on the editing operation on each target graph model.

[0199] Those skilled in the art will appreciate that all or part of the processes in the above-mentioned embodiments can be implemented by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to memory, database or other media used in the embodiments provided in this application may include at least one of non-volatile and volatile memory. Non-volatile memory may include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory may include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM). The database involved in the various embodiments provided herein may include at least one of a relational database and a non-relational database. Non-relational databases may include, but are not limited to, distributed databases based on blockchains. The processor involved in the various embodiments provided herein may be, but are not limited to, a general-purpose processor, a central processing unit, a graphics processing unit, a digital signal processor, a programmable logic unit, a data processing logic unit based on quantum computing, and the like.

[0200] The technical features of the above embodiments can be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0201] The above embodiments merely illustrate several implementation methods of the present application. While the descriptions are relatively specific and detailed, they should not be construed as limiting the scope of the present invention. It should be noted that a person skilled in the art may make various modifications and improvements without departing from the spirit of the present invention, all of which fall within the scope of protection of the present application. Therefore, the scope of protection of the present application shall be determined by the appended claims.

Claims

1. A method for constructing a transformer graph model, characterized in that: The method comprises: Extracting entity points and relationship edges from the equipment data of each transformer; wherein the entity points represent random variables in the transformer graph model, and the relationship edges represent dependency relationships between different entity points; According to the extracted entity points and relationship edges, the initial graph model corresponding to each transformer is constructed; Cluster the attribute data of entity points in each initial graph model to obtain multiple target clusters; According to the multiple target clusters, each initial graph model is processed to obtain a target graph model corresponding to each transformer.

2. The method according to claim 1, characterized in that The attribute data includes name data, and the attribute data of the entity points in each initial graph model are clustered to obtain multiple target clusters, including: Determine the word vector of the name data of the entity point in each initial graph model; According to the word vectors of each name data, each name data is clustered to obtain multiple target clusters.

3. The method according to claim 2, characterized in that The method clusters each name data according to the word vector of each name data to obtain multiple target clusters, including: Based on at least two initial cluster centers, clustering each name data according to the word vector of each name data to obtain an initial cluster corresponding to each initial cluster center; Determine the target cluster center corresponding to each initial cluster according to the word vectors of the name data contained in each initial cluster; Determine whether the initial cluster center of any initial cluster is inconsistent with the target cluster center; If yes, each target cluster center is used as a new initial cluster center, and the operation of clustering each name data based on at least two initial cluster centers and the word vector of each name data is returned to obtain the initial cluster corresponding to each initial cluster center; If not, each initial cluster is used as the target cluster for the name data.

4. The method according to claim 3, characterized in that The method of processing each initial graph model according to multiple target clusters to obtain a target graph model corresponding to each transformer includes: Extract keywords from each target cluster; The extracted keywords are used to replace the name data of the entity points in each initial graph model to obtain the target graph model corresponding to each transformer.

5. The method according to claim 4, characterized in that The extracted keywords are used to replace the name data of the entity points in each initial graph model to obtain the target graph model corresponding to each transformer, including: Obtaining at least two optional semantic information of the extracted keywords; According to the scores of each optional semantic information, the target semantic information is determined from each optional semantic information; The extracted keywords are used to replace the name data of the entity points in each initial graph model to obtain the intermediate graph model of each transformer; The target graph model corresponding to each transformer is obtained by eliminating other optional semantic information except the target semantic information corresponding to the keyword in each intermediate graph model.

6. The method according to claim 4, characterized in that The extracting keywords from each target cluster includes: The name data corresponding to the target cluster center of each target cluster is used as the keyword of each target cluster; or, The name data closest to the center of each target cluster is used as the keyword of each target cluster.

7. The method according to claim 1, characterized in that The method further comprises: Display each target graph model to the operation and maintenance terminal; According to the editing operation on each target graph model, each target graph model is optimized.

8. A device for constructing a transformer graphical model, characterized in that: The device comprises: A data extraction module is used to extract entity points and relationship edges from the equipment data of each transformer; A model building module is used to build an initial graph model corresponding to each transformer based on the extracted entity points and relationship edges; The data clustering module is used to cluster the attribute data of entity points in each initial graph model to obtain multiple target clusters; The target determination module is used to process each initial graph model according to multiple target clusters to obtain a target graph model corresponding to each transformer.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 7 are implemented.

Images