A digital twin model construction method, device and terminal equipment

Abstract

The application is suitable for the field of digital twin technology, and provides a digital twin model construction method, device and terminal equipment, which comprises the following steps: receiving device information of a plurality of devices sent by physical sensors in a power grid system, wherein the device information comprises device position information and device service information; clustering the device position information of the plurality of devices to obtain a plurality of position clustering clusters, each position clustering cluster corresponds to a device cluster, and each position clustering cluster comprises the device position information of each device in the device cluster; fusing the device service information of the plurality of devices by using an automatic encoder to obtain service association information between different devices; constructing a multi-scale model objective function based on the position clustering clusters and the service association information; and mapping the objective function to a digital twin system to obtain a digital twin model of the power grid system. The application can provide a guidance basis for the power grid operation at a future time, thereby improving the power grid operation efficiency.

Description

Technical Field

[0001] This application belongs to the field of digital twin technology, and in particular relates to a method, apparatus and terminal equipment for constructing a digital twin model. Background Technology

[0002] Limited by the imperfections of mapping and information technologies, traditional power grid systems primarily function as physical entities, leading to low efficiency and high energy losses in the specific operations of power companies. In recent years, the application of digital twin technology in power grids has gradually improved, enabling real-time interaction between power grid entities in physical and information spaces, as well as the creation of digital twin simulations of photovoltaic cells and their surrounding environment, and the prediction of final outcomes.

[0003] However, existing research on digital twin technology is limited to data processing at a single level when dealing with hierarchical power grids. It does not focus on the business interactions, data information, and constraints between this level and adjacent or other levels, resulting in limited improvement in power grid operation efficiency.

[0004] Therefore, there is an urgent need for a digital twin approach that can provide comprehensive consideration to guide power grid operation and improve its efficiency. Summary of the Invention

[0005] To overcome the problems existing in related technologies, this application provides a method, apparatus and terminal equipment for constructing a digital twin model, which can provide guidance for power grid operation and thus improve the efficiency of power grid operation.

[0006] This application is achieved through the following technical solution:

[0007] In a first aspect, embodiments of this application provide a method for constructing a digital twin model, including:

[0008] The system receives equipment information from multiple devices in a power grid system from physical sensors. This equipment information includes device location information and device service information. The system then clusters the device location information of multiple devices to obtain multiple location clusters. Each location cluster corresponds to a device cluster, and each location cluster includes the device location information of each device within that cluster. An autoencoder is used to fuse the device service information of multiple devices to obtain service association information between different devices. Based on the location clusters and service association information, a multi-scale model objective function is constructed. Finally, the objective function is mapped to a digital twin system to obtain a digital twin model of the power grid system.

[0009] In one possible implementation of the first aspect, an autoencoder is used to fuse the device business information of multiple devices to obtain business association information between different devices, including: constructing a data reconstruction model to reconstruct the device business information of multiple devices into first data; encoding association rules for the device business information of multiple devices, wherein the association rules follow a first loss function model; and decoding the business association information between different devices based on the first data and the association rules, wherein the business association information follows a second loss function model.

[0010] In one possible implementation of the first aspect, a data reconstruction model is constructed to reconstruct the device service information of multiple devices into first data, including: inputting the service information of multiple devices into an automatic encoder; and constructing a data reconstruction model, the expression of which is:

[0011]

[0012] In the formula, x i This represents the first data, which includes the device service information of the i-th device and the device service information of devices adjacent to the i-th device. i Let v represent the i-th device. j Let v represent the j-th device. j ∈N(v i () indicates that the j-th device is adjacent to the i-th device. This represents the device service information of the i-th device at time t. ω represents the device service information of the j-th device at time t. ij ω represents the association weight between the device service information of device i and device service information of device j. ij >0 indicates a weighted network, ω ij =1 indicates an unauthorized network.

[0013] In one possible implementation of the first aspect, encoding association rules for device service information of multiple devices includes: constructing an encoding layer model at the encoding layer to formulate association rules for device service information of multiple devices, wherein the constructed encoding layer model includes:

[0014] The expression for the first K-2 layers of the coding layer is:

[0015]

[0016]

[0017] The expression for the (K-1)th layer of the coding layer is:

[0018]

[0019]

[0020] The expression for the Kth layer of the coding layer is:

[0021]

[0022] In the formula, σ(·) represents the activation function, and ω k Let b represent the transition matrix. k This represents the bias vector, where K represents the number of coding and decoding layers, and y represents the bias vector. u and y σ Let E represent the vector output obtained by the (K-1)th layer through learning, and let E represent the distribution function, where E ~ (0, 1) follows a standard normal distribution. This represents the vector output obtained by the Kth layer through learning. This refers to association rules.

[0023] The encoding layer model follows a first loss function model, the expression of which is:

[0024]

[0025] In one possible implementation of the first aspect, decoding, based on the first data and association rules, obtains business association information between different devices, including:

[0026] In the decoding layer, a decoding layer model is constructed, wherein the number of layers and parameters of the constructed decoding layer model are the same as those of the encoding layer model;

[0027] Based on the first data x i and association rules The result obtained is the output X of the decoding layer. i Output result X i This refers to the business association information between different devices;

[0028] The output result X i The service information of the i-th device and the service information of the devices adjacent to the i-th device x i The comparison is performed, and the result follows the second loss function model, which is expressed as follows:

[0029]

[0030] In one possible implementation of the first aspect, a multi-scale model objective function is constructed based on location clustering and business association information, including: the expression of the multi-scale model objective function is:

[0031] L=αL RE +βL KL +γdis

[0032] In the formula, α represents the hyperparameter guarantee for parameter balance between the encoder and the Skip-gram, and L RE This represents the second loss function model, where β indicates that the association rules must follow a predetermined functional distribution, and L... KL Let represent the first loss function model, γ represent the hyperparameters of the bias matrix and weight matrix, and dis represent the device location information.

[0033] In one possible implementation of the first aspect, the device information expression is: This represents the device location information of the i-th device at time t. This represents the device service information of the i-th device at time t, where the device location information includes the device's three-dimensional coordinates. Let X represent the coordinate of the i-th device on the X-axis at time t, and Y represent the coordinate of the i-th device on the X-axis at time t. t i This represents the Y-axis coordinate of the i-th device at time t. This represents the Z-axis coordinate of the i-th device at time t. Multiple device location information is clustered to obtain multiple location clusters, including: using the DBSCAN method to cluster the device location information of multiple devices to obtain multiple location clusters.

[0034] Secondly, embodiments of this application provide a digital twin model construction apparatus, comprising: a receiving module for receiving device information from multiple devices in a power grid system via physical sensors, wherein the device information includes device location information and device service information; a clustering module for clustering the device location information of multiple devices to obtain multiple location clusters, each location cluster corresponding to a device cluster, each location cluster including the device location information of each device in the device cluster; a neural network module for fusing the device service information of multiple devices using an autoencoder to obtain service association information between different devices, and constructing a multi-scale model objective function based on the location clusters and service association information; and a mapping module for mapping the objective function to a digital twin system to obtain a digital twin model.

[0035] Thirdly, embodiments of this application provide a terminal device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement a digital twin model construction method as described in any of the first aspects.

[0036] Fourthly, embodiments of this application provide a computer-readable storage medium storing a computer program that, when executed by a processor, implements a digital twin model construction method as described in any of the first aspects.

[0037] Fifthly, embodiments of this application provide a computer program product that, when run on a terminal device, causes the terminal device to execute a digital twin model construction method according to any one of the first aspects described above.

[0038] It is understood that the beneficial effects of the second to fifth aspects mentioned above can be found in the relevant descriptions in the first aspect mentioned above, and will not be repeated here.

[0039] The beneficial effects of the embodiments in this application compared with the prior art are:

[0040] This application's embodiment receives device information from physical sensors in the power grid system, clusters the location information of multiple devices into location clusters, and fuses business information between different devices to form business association information between devices. Then, based on the device location clusters and business association information, a multi-scale model objective function is constructed. Finally, the objective function is mapped to a digital twin system to obtain a digital twin model of the power grid system. This digital twin construction method can provide guidance for future power grid operation based on existing power grid data, thereby improving power grid operating efficiency.

[0041] It should be understood that the above general description and the following detailed description are exemplary and explanatory only, and are not intended to limit this specification. Attached Figure Description

[0042] To more clearly illustrate the technical solutions in the embodiments of this application, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0043] Figure 1 This is a schematic diagram of a scenario based on digital twin technology provided in an embodiment of this application;

[0044] Figure 2 This is a schematic diagram of multi-scale modeling provided in an embodiment of this application;

[0045] Figure 3 This is a schematic diagram of a multi-scale modeling process provided in an embodiment of this application;

[0046] Figure 4 This is a schematic diagram of the process for constructing a digital twin model according to an embodiment of this application;

[0047] Figure 5 This is a schematic diagram of the process for fusing device service information according to an embodiment of this application;

[0048] Figure 6 This is a structural block diagram of the digital twin model construction device provided in the embodiments of this application;

[0049] Figure 7 This is a schematic diagram of the structure of the terminal device provided in the embodiments of this application. Detailed Implementation

[0050] In the following description, specific details such as particular system architectures and techniques are set forth for illustrative purposes and not for limitation, in order to provide a thorough understanding of the embodiments of this application. However, those skilled in the art will understand that this application may also be implemented in other embodiments without these specific details. In other instances, detailed descriptions of well-known systems, apparatuses, circuits, and methods have been omitted so as not to obscure the description of this application with unnecessary detail.

[0051] It should be understood that, when used in this application specification and the appended claims, the term "comprising" indicates the presence of the described features, integrals, steps, operations, elements and / or components, but does not exclude the presence or addition of one or more other features, integrals, steps, operations, elements, components and / or a collection thereof.

[0052] It should also be understood that the term “and / or” as used in this application specification and the appended claims means any combination of one or more of the associated listed items and all possible combinations, and includes such combinations.

[0053] As used in this application specification and the appended claims, the term "if" may be interpreted, depending on the context, as "when," "once," "in response to determination," or "in response to detection." Similarly, the phrase "if determined" or "if detected [the described condition or event]" may be interpreted, depending on the context, as meaning "once determined," "in response to determination," "once detected [the described condition or event]," or "in response to detection [the described condition or event]."

[0054] Furthermore, in the description of this application and the appended claims, the terms "first," "second," "third," etc., are used only to distinguish descriptions and should not be construed as indicating or implying relative importance.

[0055] References to "one embodiment" or "some embodiments" as described in this specification mean that one or more embodiments of this application include a specific feature, structure, or characteristic described in connection with that embodiment. Therefore, the phrases "in one embodiment," "in some embodiments," "in other embodiments," "in still other embodiments," etc., appearing in different parts of this specification do not necessarily refer to the same embodiment, but rather mean "one or more, but not all, embodiments," unless otherwise specifically emphasized. The terms "comprising," "including," "having," and variations thereof mean "including but not limited to," unless otherwise specifically emphasized.

[0056] Digital twins are simulation processes that fully utilize information such as physical models, sensor updates, and historical operational data, integrating multiple disciplines, physical quantities, scales, and probabilities. They are mapped in virtual space to reflect the entire lifecycle of the corresponding physical equipment.

[0057] In power grid systems, as the scale of power grid data continues to expand, digital twins built using traditional methods suffer from high latency in the massive node sensing and data transmission, making it difficult to predict the state of the power grid in the future.

[0058] Based on the above problems, this application provides a digital twin model construction method that uses sensors on the physical end of the digital twin to collect data such as the attributes of devices in the power grid and the business information executed by the devices. It also uses digital twin technology to perform multi-scale modeling, calculates based on the attributes and business information of devices in the actual power grid, and finally maps it to the power grid to predict the state of the power grid at future times, providing guidance for the efficient operation of the power grid.

[0059] For example, embodiments of this application can be applied to, for example... Figure 1 In the exemplary scenario shown, the physical domain and grid principles of the power grid system are completely transferred to the digital twin. The digital twin receives information from the power grid system through physical sensors and feeds back the data calculated by the digital twin model to the power grid system.

[0060] For example, Figure 2 This is a schematic diagram of multi-scale modeling provided in an embodiment of this application, with reference to... Figure 2 Let P represent a cluster in the power grid system. Cluster P is now decomposed into four layers: s, t, i, and j, corresponding to... Figure 1The system involves complex layers such as the system layer and the unit layer. In existing technologies, calculating the Euclidean distance between different clusters doesn't consider the business-driven relationship between the two clusters, leading to deviations in the virtual mapping. In the diagram, Mps represents the mapping between the problem node granularity and the processing solution granularity at a certain level. Msc represents the mapping between the processing solution and the final computational granularity. Mpc represents the mapping between the problem node granularity and the final computational granularity. Mpc represents the feedback from the current optimal computational granularity solution to the problem node, achieving a closed-loop feedback in digital twins.

[0061] For example, Figure 3 This is a schematic diagram of a multi-scale modeling process provided in an embodiment of this application. (Refer to...) Figure 3 The main steps of the multi-scale modeling process are as follows: First, information such as the attributes of each device and the business data executed are collected through the physical sensors of the power grid system; then, the relationship between devices in the power grid is converted into a granular structure guided by node attributes and business information, and clustering is performed in the scale space of each layer according to the three-dimensional Euclidean distance between device clusters; then, the business relationship is calculated according to the topology between device clusters and combined with historical data and current time data, and the objective function is solved by a neural network algorithm so that the objective function meets the mapping requirements; finally, the calculated objective function is received by the digital twin, the modeling scheme is displayed, and feedback is given to the power grid system.

[0062] The following combination Figure 3 This application provides a detailed description of the digital twin modeling method.

[0063] Figure 4 This is a flowchart illustrating a digital twin model construction method according to an embodiment of this application. (Refer to...) Figure 4 The method is described in detail below:

[0064] In step 101, device information from multiple devices transmitted by physical sensors in the power grid system is received.

[0065] In a digital twin, all devices in the power grid system have corresponding physical sensors, which are used to transmit relevant information about the devices to the digital twin system.

[0066] For example, equipment information may include various data such as equipment quantity, equipment number, equipment model, equipment status, equipment operation data, equipment location information, and equipment service information. For ease of understanding, this application does not further limit the equipment information, but only provides illustrative examples of the equipment information used in the embodiments of this application.

[0067] For example, the expression for device information can be: This represents the device location information of the i-th device at time t. This represents the device service information of the i-th device at time t.

[0068] Optionally, the device location information dis may include the device's two-dimensional or three-dimensional coordinates.

[0069] For example, two-dimensional coordinates can be represented as This represents the X-axis coordinate of the i-th device at time t. This represents the Y-axis coordinate of the i-th device at time t.

[0070] For example, three-dimensional coordinates are represented as in, This represents the X-axis coordinate of the i-th device at time t. This represents the Y-axis coordinate of the i-th device at time t. Let represent the coordinate of the i-th device along the Z-axis at time t.

[0071] In step 102, the device location information of multiple devices is clustered to obtain multiple location clusters.

[0072] In some embodiments, the DBSCAN (Density-Based Spatial Clustering of Applications with Noise) method is used to cluster multiple device location information to obtain multiple location clusters. Since each location cluster corresponds to a device cluster, each location cluster contains the device location information of each device in that device cluster.

[0073] For example, the specific process of clustering device location information using the DBSCAN method is as follows:

[0074] Step 1: Use the obtained device location information as clustering samples and preset the clustering radius E. ps and the preset minimum number of cluster samples MinP ts .

[0075] Step 2: Using any sample as the starting point for clustering, combined with the preset cluster radius E... ps and the preset minimum number of cluster samples MinP ts The algorithm calculates and determines whether the sample point is a kernel object. If the sample point is a kernel object, a cluster S is created.

[0076] For example, whether a sample is a kernel object can be determined by calculating the Euclidean distance. The expression for calculating the Euclidean distance is:

[0077]

[0078] In the formula, Let be the Euclidean distance between the positions of the i-th and j-th devices at time t, where n is the dimension and k∈[1,n]. and Let represent the position coordinates of the i-th and j-th devices in the k-th dimension at time t. If Less than or equal to the preset cluster radius E ps If the sample is included in cluster S, then the sample will be included in cluster S.

[0079] Step 3: Search for unvisited device location information within cluster S, and determine whether it is a kernel object by calculating Euclidean distance. If so, classify the samples in the kernel object's neighborhood into cluster S.

[0080] Step 4: Repeat Step 1 to Step 3 until all device location information has a corresponding cluster.

[0081] It should be noted that the preset cluster radius E ps and the preset minimum number of cluster samples MinP ts Based on actual circumstances and experience, this application will not impose further limitations.

[0082] In step 102, the preset clustering radius E is used. ps and the preset minimum number of cluster samples MinP ts This transforms devices of different dimensions into a unified dimension, preparing for the subsequent establishment of multi-scale models and the realization of mapping at different scales.

[0083] In step 103, an automatic encoder is used to fuse the device service information of multiple devices to obtain service association information between different devices.

[0084] In some embodiments, some devices are geographically distant but perform closely related or identical services. To achieve the fusion of device service information and find service correlation information between different devices, this application employs an automatic encoder to fuse the device service information.

[0085] An autoencoder (AE) is a type of artificial neural network used in semi-supervised and unsupervised learning. Its function is to learn representations of input information by using the input information as the learning target.

[0086] Figure 5 This is a schematic diagram of the device service information fusion process provided in an embodiment of this application, referring to... Figure 5 .

[0087] In step 1031, a data reconstruction model is constructed to reconstruct the device business information of multiple devices into the first data.

[0088] In some embodiments, business information from multiple devices is input into the automatic encoder.

[0089] Optionally, it will receive equipment service information sent by physical sensors in the power grid system. Decompose into a vector group at time t

[0090] Furthermore, a data reconstruction model is constructed to obtain the first data. The expression of the data reconstruction model is:

[0091]

[0092] In the formula, x i This represents the first data, which includes the service information of the i-th device and the service information of the devices adjacent to the i-th device. i Let v represent the i-th device. j Let v represent the j-th device. j ∈N(v i () indicates that the j-th device is adjacent to the i-th device. This represents the device service information of the i-th device at time t. ω represents the device service information of the j-th device at time t. ij ω represents the association weight between the device service information of device i and device service information of device j. ij >0 indicates a rightful network, ω ij =1 indicates an unauthorized network.

[0093] In step 1032, the association rules for device service information of multiple devices are encoded.

[0094] In some embodiments, an encoding layer model is constructed in the encoding layer of the autoencoder to formulate association rules for device service information of multiple devices. The constructed encoding layer model may be:

[0095] The expression for the first K-2 layers of the coding layer is:

[0096]

[0097]

[0098] The expression for the (K-1)th layer of the coding layer is:

[0099]

[0100]

[0101] The expression for the Kth layer of the coding layer is:

[0102]

[0103] In the formula, σ(·) represents the activation function, and ω k Let b represent the transition matrix. k This represents the bias vector, where K represents the number of coding and decoding layers, and y represents the bias vector. u and y σ Let E represent the vector output obtained by the (K-1)th layer through learning, and let E represent the distribution function, where E ~ (0, 1) follows a standard normal distribution. This represents the vector output obtained by the Kth layer through learning. This refers to association rules.

[0104] Furthermore, to ensure that data loss remains within a controllable range, the coding layer model must also conform to the first loss function model, the expression of which is:

[0105]

[0106] In step 1033, decoding is performed based on the first data and association rules to obtain business association information between different devices.

[0107] In some embodiments, a decoding layer model is constructed at the decoding layer, wherein the constructed decoding layer model has the same number of layers and parameters as the encoding layer model. Based on the first data x i and the association rules The result obtained is the output X of the decoding layer. i Output result X i This refers to the business association information between different devices.

[0108] Furthermore, to ensure that the loss of business-related information between devices is within a controllable range, the output result X will be... i The service information of the i-th device and the service information of the devices adjacent to the i-th device x i The comparison is performed, and the result follows a second loss function model, the expression of which is:

[0109]

[0110] In step 103, by reconstructing device service information and establishing association rules between devices, the device service information of different devices was merged, fully mining the hidden information between different device services, and finding the service association information between different devices. Furthermore, a first loss function model and a second loss function model were established to constrain the autoencoder, ensuring the accuracy of the acquired service association information.

[0111] Step 103 also fused the business information between adjacent devices, addressing the issue of business information not being fused between device entities at relatively distant geographical locations under multi-scale conditions. This prepares for the subsequent establishment of a multi-scale model.

[0112] It should be noted that there is no specific order between steps 102 and 103. Step 102 can be executed first, or step 103 can be executed first, or they can be performed simultaneously. This application will not impose any further restrictions.

[0113] In step 104, a multi-scale model objective function is constructed based on location clustering and business association information.

[0114] In some embodiments, a multi-scale model objective function is constructed based on the location clusters of devices and the service association information of device components. The expression of the multi-scale model objective function may be:

[0115] L=αL RE +βL KL +γdis

[0116] In the formula, α represents the hyperparameter guarantee for parameter balance between the encoder and the Skip-gram, and L RE This represents the second loss function model, where β indicates that the association rules must follow a predetermined functional distribution, and L... KL Let represent the first loss function model, γ represent the hyperparameters of the bias matrix and weight matrix, and dis represent the device location information.

[0117] This step overcomes the problem in existing technologies that, when dealing with hierarchical power grids, digital twin technology is limited to data processing at a single level and lacks attention to business interactions, data information, and constraints between that level and adjacent or other levels.

[0118] In step 105, the objective function is mapped to the digital twin system to obtain a digital twin model of the power grid system.

[0119] This step inputs the designed multi-scale model objective function into the digital twin system, which then feeds the solution back to the physical layer of the power grid system, providing guidance for future power grid operation.

[0120] It should be understood that the sequence number of each step in the above embodiments does not imply the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of this application.

[0121] Corresponding to the digital twin model construction method described in the above embodiments, Figure 6A structural block diagram of the digital twin model construction apparatus provided in the embodiments of this application is shown. For ease of explanation, only the parts related to the embodiments of this application are shown.

[0122] See Figure 6 The digital twin model construction device in this application embodiment may include a receiving module 201, a clustering module 202, a neural network module 203, and a mapping module 204.

[0123] The receiving module 201 is used to receive device information from multiple devices sent by physical sensors in the power grid system, wherein the device information includes device location information and device service information.

[0124] The device information expression is as follows: This represents the device location information of the i-th device at time t. This represents the device service information of the i-th device at time t, where the device location information includes the device's three-dimensional coordinates. Let X represent the coordinate of the i-th device on the X-axis at time t, and Y represent the coordinate of the i-th device on the X-axis at time t. t i This represents the Y-axis coordinate of the i-th device at time t. Let represent the coordinate of the i-th device along the Z-axis at time t.

[0125] Clustering module 202 is used to cluster the device location information of multiple devices to obtain multiple location clusters. Each location cluster corresponds to a device cluster, and each location cluster includes the device location information of each device in the device cluster.

[0126] Optionally, the device location information of multiple devices can be clustered to obtain multiple location clusters, including: using the DBSCAN method to cluster the device location information of multiple devices to obtain multiple location clusters.

[0127] The neural network module 203 is used to fuse the device business information of multiple devices using an autoencoder to obtain business association information between different devices.

[0128] The process involves using an autoencoder to fuse device business information from multiple devices to obtain business association information between different devices. This includes: constructing a data reconstruction model to reconstruct the device business information from multiple devices into first data; encoding association rules for the device business information from multiple devices, wherein the association rules follow a first loss function model; and decoding the business association information between different devices based on the first data and the association rules, wherein the business association information follows a second loss function model.

[0129] Optionally, a data reconstruction model is constructed to reconstruct the device business information of multiple devices into first data, including: inputting the business information of multiple devices into an automatic encoder; and constructing the data reconstruction model, the expression of which is:

[0130]

[0131] In the formula, x i This represents the first data, which includes the device service information of the i-th device and the device service information of devices adjacent to the i-th device. i Let v represent the i-th device. j Let v represent the j-th device. j ∈N(v i () indicates that the j-th device is adjacent to the i-th device. This represents the device service information of the i-th device at time t. ω represents the device service information of the j-th device at time t. ij ω represents the association weight between the device service information of device i and device service information of device j. ij >0 indicates a weighted network, ω ij =1 indicates an unauthorized network.

[0132] Optionally, the association rules for encoding device service information of multiple devices include: constructing an encoding layer model in the encoding layer to formulate association rules for device service information of multiple devices, wherein the constructed encoding layer model includes:

[0133] The expression for the first K-2 layers of the coding layer is:

[0134]

[0135]

[0136] The expression for the (K-1)th layer of the coding layer is:

[0137]

[0138]

[0139] The expression for the Kth layer of the coding layer is:

[0140]

[0141] In the formula, σ(·) represents the activation function, and ω k Let b represent the transition matrix. k This represents the bias vector, where K represents the number of coding and decoding layers, and y represents the bias vector. u and y σLet E represent the vector output obtained by the (K-1)th layer through learning, and let E represent the distribution function, where E ~ (0, 1) follows a standard normal distribution. This represents the vector output obtained by the Kth layer through learning. For association rules.

[0142] The encoding layer model follows a first loss function model, the expression of which is:

[0143]

[0144] Optionally, decoding, based on the first data and association rules, obtains business association information between different devices, including:

[0145] A decoding layer model is constructed in the decoding layer, wherein the number of layers and parameters of the constructed decoding layer model are the same as those of the encoding layer model; based on the first data x i and association rules The result obtained is the output X of the decoding layer. i Output result X i This refers to the business association information between different devices.

[0146] The output result X i The service information of the i-th device and the service information of the devices adjacent to the i-th device x i The comparison is performed, and the result follows the second loss function model. The expression for the second loss function model is:

[0147]

[0148] The neural network module 203 is also used to construct a multi-scale model objective function based on the location clusters and the business association information.

[0149] Optionally, based on location clustering and business association information, a multi-scale model objective function is constructed, including: the expression of the multi-scale model objective function is:

[0150] L=αL RE +βL KL +γdis

[0151] In the formula, α represents the hyperparameter guarantee for parameter balance between the encoder and the Skip-gram, and L RE This represents the second loss function model, where β indicates that the association rules must follow a predetermined functional distribution, and L... KL Let represent the first loss function model, γ represent the hyperparameters of the bias matrix and weight matrix, and dis represent the device location information.

[0152] The mapping module 204 is used to map the objective function to the digital twin system to obtain the digital twin model.

[0153] It should be noted that the information interaction and execution process between the above-mentioned devices / units are based on the same concept as the method embodiments of this application. For details on their specific functions and technical effects, please refer to the method embodiments section, and they will not be repeated here.

[0154] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the above-described division of functional units and modules is merely an example. In practical applications, the above functions can be assigned to different functional units and modules as needed, that is, the internal structure of the device can be divided into different functional units or modules to complete all or part of the functions described above. The functional units and modules in the embodiments can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit. Furthermore, the specific names of the functional units and modules are only for easy differentiation and are not intended to limit the scope of protection of this application. The specific working process of the units and modules in the above system can be referred to the corresponding process in the foregoing method embodiments, and will not be repeated here.

[0155] This application also provides a terminal device, see [link to relevant documentation] Figure 7 The terminal 300 may include: at least one processor 310, a memory 320, and a computer program 321 stored in the memory 320 and executable on the at least one processor 310. When the processor 310 executes the computer program 321, it implements the steps in any of the above-described method embodiments, for example... Figure 4 Steps 101 to 105 in the illustrated embodiment. Alternatively, when the processor 310 executes the computer program, it implements the functions of each module / unit in the above-described device embodiments, for example... Figure 6 The functions of modules 201 to 204 are shown.

[0156] For example, a computer program may be divided into one or more modules / units, one or more of which are stored in memory 320 and executed by processor 310 to complete this application. The one or more modules / units may be a series of computer program segments capable of performing specific functions, which describe the execution process of the computer program in terminal device 300.

[0157] Those skilled in the art will understand that Figure 7 This is merely an example of a terminal device and does not constitute a limitation on the terminal device. It may include more or fewer components than shown, or combine certain components, or different components, such as input / output devices, network access devices, buses, etc.

[0158] The processor 310 can be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor can be a microprocessor or any conventional processor.

[0159] The memory 320 can be an internal storage unit of the terminal device or an external storage device, such as a plug-in hard drive, a smart media card (SMC), a secure digital card (SD), or a flash card. The memory 320 is used to store the computer program and other programs and data required by the terminal device. The memory 320 can also be used to temporarily store data that has been output or will be output.

[0160] The bus can be an Industry Standard Architecture (ISA) bus, a Peripheral Component Interconnect (PCI) bus, or an Extended Industry Standard Architecture (EISA) bus, etc. Buses can be categorized as address buses, data buses, control buses, etc. For ease of illustration, the buses shown in the accompanying drawings are not limited to a single bus or a single type of bus.

[0161] The digital twin model construction method provided in this application can be applied to terminal devices such as computers, wearable devices, in-vehicle devices, tablet computers, laptops, netbooks, personal digital assistants (PDAs), augmented reality (AR) / virtual reality (VR) devices, and mobile phones. This application does not impose any restrictions on the specific type of terminal device.

[0162] This application also provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the steps described in the various embodiments of the digital twin model construction method.

[0163] This application provides a computer program product that, when run on a mobile terminal, enables the mobile terminal to implement the steps described in the various embodiments of the digital twin model construction method.

[0164] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, all or part of the processes in the methods of the above embodiments of this application can be implemented by a computer program instructing related hardware. The computer program can be stored in a computer-readable storage medium, and when executed by a processor, it can implement the steps of the various method embodiments described above. The computer program includes computer program code, which can be in the form of source code, object code, executable files, or certain intermediate forms. The computer-readable medium can include at least: any entity or device capable of carrying computer program code to a photographing device / terminal device, a recording medium, a computer memory, a read-only memory (ROM), a random access memory (RAM), an electrical carrier signal, a telecommunication signal, and a software distribution medium. Examples include USB flash drives, portable hard drives, magnetic disks, or optical disks. In some jurisdictions, according to legislation and patent practice, computer-readable media cannot be electrical carrier signals or telecommunication signals.

[0165] In the above embodiments, the descriptions of each embodiment have different focuses. For parts that are not described in detail or recorded in a certain embodiment, please refer to the relevant descriptions of other embodiments.

[0166] Those skilled in the art will recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0167] In the embodiments provided in this application, it should be understood that the disclosed apparatus / network devices and methods can be implemented in other ways. For example, the apparatus / network device embodiments described above are merely illustrative. For instance, the division of modules or units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection between devices or units may be electrical, mechanical, or other forms.

[0168] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0169] The above-described embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit them. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of this application, and should all be included within the protection scope of this application.

Claims

1. A method for constructing a digital twin model, applied to a power grid system, characterized in that, include: Receive device information from multiple devices transmitted by physical sensors in the power grid system, wherein the device information includes device location information and device service information; The device location information of the multiple devices is clustered to obtain multiple location clusters. Each location cluster corresponds to a device cluster, and each location cluster includes the device location information of each device in the device cluster. An autoencoder is used to fuse the device service information of the multiple devices to obtain service association information between different devices. This includes: constructing a data reconstruction model to reconstruct the device service information of the multiple devices into first data; encoding association rules for the device service information of the multiple devices, wherein the association rules follow a first loss function model; and decoding the service association information between different devices based on the first data and the association rules, wherein the service association information follows a second loss function model. The expression for the first loss function model is: The expression for the second loss function model is: ; Indicates the first K The mean vector obtained by learning from layer -1 Indicates the first K The standard deviation vector learned by layer -1 This represents the first data, which includes the first data. i The device's business information and its relationship with the first device. i Equipment service information of adjacent devices. This is the output of the decoding layer; Based on the location clusters and the business association information, a multi-scale model objective function is constructed; the expression of the multi-scale model objective function is: In the formula, This indicates that the hyperparameters ensure parameter balance between the encoder and the Skip-gram. This represents the second loss function model. This indicates that the association rules are guaranteed to follow a defined function distribution. This represents the first loss function model. The hyperparameters representing the paranoia matrix and the weight matrix, This indicates the location information of the device; The objective function is mapped to a digital twin system to obtain a digital twin model of the power grid system.

2. The method as described in claim 1, characterized in that, The construction of the data reconstruction model reconstructs the device service information of the multiple devices into first data, including: The business information of the multiple devices is input into the automatic encoder; Construct a data reconstruction model to obtain the first data. The expression of the data reconstruction model is: In the formula, This represents the first data, which includes the first data. i The device's business information and its relationship with the first device. i Equipment service information of adjacent devices. Indicates the first i One device, Indicates the first j One device, Indicates the first j The device and the first i The devices are adjacent. Indicates the first i The device is t Real-time device service information, Indicates the first j The device is t Real-time device service information, Indicates the first i The equipment business information of the device and the first j The correlation weight between the device business information of each device Indicates the right to the network, This indicates that the network is not authorized.

3. The method as described in claim 2, characterized in that, The association rules for encoding the device service information of the multiple devices include: A coding layer model is constructed at the coding layer to formulate association rules for the device service information of the multiple devices. The constructed coding layer model includes: Before the coding layer K The expression for level -2 is: Coding layer K The expression for level -1 is: Coding layer K The expression for the layer is: In the formula, (·) represents the activation function. Represents the transition matrix. This represents the bias vector. K This indicates the number of encoding and decoding layers. and Indicates the first K The -1 layer outputs the vectors learned from it. E Represents the distribution function. Follows a standard normal distribution. Indicates the first K The layer obtains a vector output through learning, and the vector output... For the association rule; The coding layer model follows a first loss function model.

4. The method as described in claim 3, characterized in that, The decoding, based on the first data and the association rules, obtains business association information between different devices, including: A decoding layer model is constructed in the decoding layer, wherein the constructed decoding layer model has the same number of layers and parameters as the encoding layer model; Based on the first data and the association rules The result obtained is the output of the decoding layer. The output result This refers to the business association information between different devices; The output results will be and the i The device's business information and its connection with the first i Device service information of adjacent devices The comparison is performed, and the comparison result follows the second loss function model.

5. The method as described in claim 1, characterized in that, The device information expression is D= , Indicates the first i The device is t Real-time device location information, Indicates the first i The device is t The device service information at any given time, wherein the device location information includes the device's three-dimensional coordinates. , Indicates the first i The device is t The coordinate of the X-axis at time [time]. Indicates the first i The device is t The coordinate of the Y-axis at time [time]. Indicates the first i The device is t The coordinate of the Z-axis at that moment; The process of clustering the device location information of the multiple devices to obtain multiple location clusters includes: using the DBSCAN method to cluster the device location information of the multiple devices to obtain multiple location clusters.

6. A digital twin model construction device, applied to a power grid system, characterized in that, include: A receiving module is used to receive device information from multiple devices transmitted by physical sensors in the power grid system, wherein the device information includes device location information and device service information; The clustering module is used to cluster the device location information of the multiple devices to obtain multiple location clusters. Each location cluster corresponds to a device cluster, and each location cluster includes the device location information of each device in the device cluster. A neural network module is used to fuse device service information from multiple devices using an autoencoder to obtain service association information between different devices. This includes: constructing a data reconstruction model to reconstruct the device service information from the multiple devices into first data; encoding association rules for the device service information from the multiple devices, wherein the association rules follow a first loss function model; and decoding the service association information between different devices based on the first data and the association rules, wherein the service association information follows a second loss function model. The expression for the first loss function model is: The expression for the second loss function model is: ; Indicates the first K The mean vector obtained by learning from layer -1 Indicates the first K The standard deviation vector learned by layer -1 This represents the first data, which includes the first data. i The device's business information and its relationship with the first device. i Equipment service information of adjacent devices. The output of the decoding layer is as follows: Based on the location clusters and the business association information, a multi-scale model objective function is constructed; the expression of the multi-scale model objective function is: In the formula, This indicates that the hyperparameters ensure parameter balance between the encoder and the Skip-gram. This represents the second loss function model. This indicates that the association rules are guaranteed to follow a defined function distribution. This represents the first loss function model. The hyperparameters representing the paranoia matrix and the weight matrix, This indicates the location information of the device; The mapping module is used to map the objective function to the digital twin system to obtain a digital twin model.

7. A terminal device, comprising a memory and a processor, wherein the memory stores a computer program executable on the processor, characterized in that, When the processor executes the computer program, it implements the method as described in any one of claims 1 to 5.

8. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed by a processor, it implements the method as described in any one of claims 1 to 5.

Images