Method and system for evaluating digital twinborn integrity of power equipment

Abstract

The invention discloses an evaluation method and system for digital twinborn integrity of power equipment, and belongs to the technical field of digital twinborn of power grid equipment. The evaluation method provided by the invention comprises the following steps: constructing a maturity subjective rating criterion for the digital twinborn body of the power equipment; obtaining the structure data of the twin of the power equipment data, and extracting the feature data in the structure data; based on the feature data, training to obtain an evaluation model; and based on the evaluation model, according to the feature data of the digital twinborn of the target power equipment, evaluating the integrity of the digital twinborn of the target power equipment. The method is high in calculation efficiency, and meets the application requirements of the digital twin system.

Description

Technical Field

[0001] The present invention relates to the technical field of digital twins of power grid equipment, and more specifically, to a method and system for evaluating the integrity of digital twins of power equipment. Background Art

[0002] Visual quality assessment technology:

[0003] Visual intelligent quality assessment technology is an important research direction in the field of image processing and computer vision. Its goal is to provide an objective and accurate image quality assessment method. The quality of an image directly affects people's visual experience and the effective transmission of information. Therefore, effective evaluation of image quality has important practical application value.

[0004] The following is a brief description of the evaluation algorithm for visual intelligence quality:

[0005] Subjective evaluation method:

[0006] The subjective evaluation method is based on the perception of 3D content by the human visual system. This method usually requires a group of observers to score the 3D content according to a given evaluation standard. The advantage of the subjective evaluation method is that it can truly reflect the visual perception of the human eye to 3D content, but its disadvantages are also obvious, such as time-consuming, high cost, and affected by individual differences among observers.

[0007] Intelligent objective evaluation method:

[0008] Objective evaluation methods use mathematical models and algorithms to quantitatively evaluate the quality of 3D content. This type of method does not require human intervention and can quickly and automatically evaluate a large amount of 3D content. Objective evaluation methods can be divided into two categories: image / video quality-based methods and depth map-based methods.

[0009] (1) Methods based on image / video quality:

[0010] This type of method mainly focuses on the 2D image or video quality of 3D content. They use existing 2D image / video quality evaluation methods, such as peak signal-to-noise ratio (PSNR) and structural similarity index (SSIM), to evaluate the quality of left and right views or synthetic views of 3D content. However, this type of method ignores the depth information and stereoscopic vision characteristics unique to 3D content, so the accuracy of its evaluation results needs to be further improved.

[0011] (2) Depth map based method:

[0012] Depth map is an important component of 3D content, which contains the distance information of objects in the scene. Depth map-based methods evaluate the overall quality of 3D content by analyzing the quality of depth maps. Such methods usually focus on aspects such as clarity, smoothness, and noise of depth maps. However, due to the relatively complex acquisition and processing of depth maps, there are relatively few 3D quality evaluation methods based on depth maps.

[0013] As a key means of measuring the quality of 3D content, 3D quality evaluation has important practical application value. At present, research on 3D quality evaluation has made some progress, but there are still many challenges and problems. Future research will focus more on the exploration and application of multi-dimensional evaluation, combining human visual characteristics and deep learning, so as to promote the development and application of 3D quality evaluation technology.

[0014] Digital Twin Maturity Assessment Technology:

[0015] With the rapid development of information technology and the deepening of digital transformation, digital twin technology, as a bridge connecting the physical world and the digital world, is increasingly widely used in various fields. Digital twin refers to the comprehensive and accurate modeling and simulation of physical systems through digital technology, realizing the digital replication and virtual operation of physical systems. Digital twin maturity assessment technology is an important means to evaluate the integrity, accuracy, effectiveness and other aspects of digital twin systems.

[0016] Establishment of evaluation index system:

[0017] The primary task of digital twin maturity assessment technology is to establish a scientific and reasonable evaluation index system. At present, researchers have proposed some representative evaluation index systems, such as the digital twin model evaluation index system proposed by Professor Tao Fei's team based on 8 performance indicators of effectiveness, versatility, efficiency, intuitiveness, connectivity, integrity, flexibility and intelligence. These index systems cover many aspects of the digital twin system and provide a comprehensive reference for maturity assessment.

[0018] Research on evaluation methods:

[0019] In terms of evaluation methods, researchers have proposed a variety of evaluation methods for different aspects of digital twin maturity. For example, evaluation methods based on completeness, standardization, data interface, integrity, and flexibility can be used to determine the maturity level of digital twin systems through quantitative analysis of these aspects. In addition, there are evaluation methods based on data richness, compatibility, accessibility, and quality. These methods focus on the evaluation of digital twin data, thereby reflecting the maturity of digital twin systems.

[0020] Development of assessment tools and platforms:

[0021] In order to better support the digital twin maturity assessment work, researchers have also developed a series of assessment tools and platforms. These tools and platforms provide a convenient assessment process, rich assessment functions and visual assessment results display, making the assessment work more efficient, accurate and intuitive.

[0022] As an important means to measure the performance and application effect of digital twin systems, digital twin maturity assessment technology has achieved certain research results. However, with the continuous development of digital twin technology and the expansion of its application fields, digital twin maturity assessment technology still faces many challenges and problems. Future research will focus more on the improvement and optimization of the evaluation index system, the innovation and expansion of evaluation methods, and the intelligentization and integration of evaluation tools and platforms, so as to promote the continuous advancement and application of digital twin maturity assessment technology.

[0023] Meta-learning techniques:

[0024] Meta-learning, as a special machine learning method, has attracted extensive attention and research in the field of artificial intelligence in recent years. Its core idea is to improve the generalization ability and learning efficiency of the model on new tasks by learning "how to learn".

[0025] Model-based meta-learning algorithms:

[0026] This type of algorithm usually improves learning efficiency by building a model that can quickly adapt to new tasks. For example, the MAML (Model-Agnostic Meta-Learning) algorithm is a typical model-based meta-learning algorithm. Its core idea is to find a set of good model initialization parameters during the training process so that the model can achieve good performance on new tasks with a small number of iterations. MAML finds such initialization parameters by optimizing the performance on multiple tasks, thereby improving the generalization ability of the model.

[0027] Optimization-based meta-learning algorithms:

[0028] This type of algorithm focuses on how to perform gradient descent or other optimization algorithms more effectively on new tasks. They accelerate the convergence of the model and improve performance by learning a general optimization strategy. This type of algorithm usually involves learning and improving the optimizer itself, so that the model can quickly find the optimal solution on different tasks and data distributions.

[0029] Data-based Meta-Learning Algorithms:

[0030] These algorithms improve learning by generating or selecting data suitable for new tasks. They may learn a data enhancement strategy to generate more training samples from the original data, or filter out the most valuable information for new tasks from a large amount of data. These algorithms can make full use of limited data resources and improve the training efficiency and performance of the model. Summary of the invention

[0031] In order to evaluate the integrity of a digital twin of an electric power device, the present invention proposes an evaluation method for the integrity of a digital twin of an electric power device, comprising:

[0032] Establishing subjective rating criteria for maturity of digital twins of power equipment;

[0033] Acquire the structural data of the power equipment data twin, and extract feature data from the structural data;

[0034] Based on the feature data, an evaluation model is trained;

[0035] Based on the evaluation model, the integrity of the digital twin of the target power equipment is evaluated according to the characteristic data of the digital twin of the target power equipment.

[0036] Optionally, subjective rating criteria for the maturity of the power equipment digital twin are constructed, including:

[0037] Establishing an assessment standard for the maturity of the digital twin of the power equipment;

[0038] Obtain digital twin models of power equipment from different manufacturers and in different scenarios;

[0039] Based on the evaluation criteria, standardize the scoring of the digital twin models of power equipment from different manufacturers and in different scenarios;

[0040] Based on the scoring results of the standardized scores, a subjective rating criterion for maturity is constructed.

[0041] Optionally, an evaluation standard for the maturity of the digital twin of the power equipment is established, including:

[0042] Formulate a maturity assessment indicator system for the digital twin of the power equipment;

[0043] Based on the maturity assessment indicator system, formulate assessment standards.

[0044] Optionally, obtaining the structural data of the power equipment data twin and extracting the characteristic data in the structural data includes:

[0045] Acquire different types of structural data of the power equipment data twin, preprocess the structural data to obtain target data, determine the importance of different attributes of the target data, determine key attributes based on the importance, and set static weights of the key attributes;

[0046] Based on the set static weight, the dynamic weight of the key attribute is adjusted and the encoding length is dynamically adjusted to represent the characteristics of the target data and generate characteristic data.

[0047] Optionally, representing the characteristics of the target data includes:

[0048] The target data is mapped into a low-dimensional space using a hash learning method to obtain feature representation.

[0049] Optionally, based on the feature data, an evaluation model is trained, including:

[0050] Divide the characteristic data into different tasks according to device type or function.

[0051] Each task corresponds to a specific device type or function, and the feature data is divided into training and validation sets accordingly;

[0052] A base model is trained on each task using the corresponding training set and validation set.

[0053] Parameters or features of the base model are extracted and represented as inputs of a meta-learner to construct a meta-learner, train the meta-learner, and generate an evaluation model.

[0054] Optionally, after the evaluation model is trained, the accuracy and recall of the evaluation model are evaluated based on the validation set of the new task, and based on the accuracy and recall obtained by the evaluation, the model parameters of the evaluation model are adjusted until the evaluated accuracy and recall meet the requirements.

[0055] On the other hand, the present invention also proposes an evaluation system for the integrity of a digital twin of a power device, comprising:

[0056] A criteria building unit, used to build maturity subjective rating criteria for power equipment digital twins;

[0057] A data acquisition unit, used to obtain the structural data of the power equipment data twin and extract characteristic data from the structural data;

[0058] A training unit, used for training an evaluation model based on the feature data;

[0059] An evaluation unit is used to evaluate the integrity of the digital twin of the target power equipment based on the evaluation model and according to the characteristic data of the digital twin of the target power equipment.

[0060] Optionally, subjective rating criteria for the maturity of the power equipment digital twin are constructed, including:

[0061] Establishing an assessment standard for the maturity of the digital twin of the power equipment;

[0062] Obtain digital twin models of power equipment from different manufacturers and in different scenarios;

[0063] Based on the evaluation criteria, standardize the scoring of the digital twin models of power equipment from different manufacturers and in different scenarios;

[0064] Based on the scoring results of the standardized scores, a subjective rating criterion for maturity is constructed.

[0065] Optionally, an evaluation standard for the maturity of the digital twin of the power equipment is established, including:

[0066] Formulate a maturity assessment indicator system for the digital twin of the power equipment;

[0067] Based on the maturity assessment indicator system, formulate assessment standards.

[0068] Optionally, obtaining the structural data of the power equipment data twin and extracting the characteristic data in the structural data includes:

[0069] Acquire different types of structural data of the power equipment data twin, preprocess the structural data to obtain target data, determine the importance of different attributes of the target data, determine key attributes based on the importance, and set static weights of the key attributes;

[0070] Based on the set static weight, the dynamic weight of the key attribute is adjusted and the encoding length is dynamically adjusted to represent the characteristics of the target data and generate characteristic data.

[0071] Optionally, representing the characteristics of the target data includes:

[0072] The target data is mapped into a low-dimensional space using a hash learning method to obtain feature representation.

[0073] Optionally, based on the feature data, an evaluation model is trained, including:

[0074] Divide the characteristic data into different tasks according to device type or function.

[0075] Each task corresponds to a specific device type or function, and the feature data is divided into training and validation sets accordingly;

[0076] A base model is trained on each task using the corresponding training set and validation set.

[0077] Parameters or features of the base model are extracted and represented as inputs of a meta-learner to construct a meta-learner, train the meta-learner, and generate an evaluation model.

[0078] Optionally, after the evaluation model is trained, the accuracy and recall of the evaluation model are evaluated based on the validation set of the new task, and based on the accuracy and recall obtained by the evaluation, the model parameters of the evaluation model are adjusted until the evaluated accuracy and recall meet the requirements.

[0079] In yet another aspect, the present invention further provides a computing device, comprising: one or more processors;

[0080] a processor for executing one or more programs;

[0081] When the one or more programs are executed by the one or more processors, the above-described method is implemented.

[0082] In yet another aspect, the present invention further provides a computer-readable storage medium having a computer program stored thereon, wherein when the computer program is executed, the method described above is implemented.

[0083] Compared with the prior art, the present invention has the following beneficial effects:

[0084] The present invention provides an evaluation method for the integrity of a digital twin of an electric power device, comprising: constructing a maturity subjective rating criterion for a digital twin of an electric power device; obtaining structural data of the digital twin of the electric power device and extracting characteristic data from the structural data; training an evaluation model based on the characteristic data; and evaluating the integrity of the digital twin of the target electric power device based on the characteristic data of the digital twin of the target electric power device based on the evaluation model. The present invention has high computational efficiency and meets the application requirements of the digital twin system. BRIEF DESCRIPTION OF THE DRAWINGS

[0085] Figure 1 is a flow chart of the method of the present invention;

[0086] Figure 2 It is a structural diagram of the system of the present invention. DETAILED DESCRIPTION

[0087] Now, exemplary embodiments of the present invention are described with reference to the accompanying drawings. However, the present invention can be implemented in many different forms and is not limited to the embodiments described herein. These embodiments are provided to disclose the present invention in detail and completely and to fully convey the scope of the present invention to those skilled in the art. The terms used in the exemplary embodiments shown in the accompanying drawings are not intended to limit the present invention. In the accompanying drawings, the same units / elements are marked with the same reference numerals.

[0088] Unless otherwise specified, the terms (including technical terms) used herein have the commonly understood meanings to those skilled in the art. In addition, it is understood that the terms defined in commonly used dictionaries should be understood to have the same meanings as those in the context of the relevant fields, and should not be understood as idealized or overly formal meanings.

[0089] Embodiment 1:

[0090] The present invention proposes a method for evaluating the integrity of a digital twin of a power device, such as Figure 1 As shown, including:

[0091] Step 1: Establish a subjective maturity rating criterion for digital twins of power equipment;

[0092] Step 2: Obtain the structural data of the power equipment data twin, and extract feature data from the structural data;

[0093] Step 3: Based on the feature data, an evaluation model is trained;

[0094] Step 4: Based on the evaluation model, the integrity of the digital twin of the target power equipment is evaluated according to the characteristic data of the digital twin of the target power equipment.

[0095] Among them, subjective rating criteria for building maturity of digital twins of power equipment include:

[0096] Establishing an assessment standard for the maturity of the digital twin of the power equipment;

[0097] Obtain digital twin models of power equipment from different manufacturers and in different scenarios;

[0098] Based on the evaluation criteria, standardize the scoring of the digital twin models of power equipment from different manufacturers and in different scenarios;

[0099] Based on the scoring results of the standardized scores, a subjective rating criterion for maturity is constructed.

[0100] Among them, the evaluation criteria for the maturity of the digital twin of the power equipment are established, including:

[0101] Formulate a maturity assessment indicator system for the digital twin of the power equipment;

[0102] Based on the maturity assessment indicator system, formulate assessment standards.

[0103] Wherein, obtaining the structural data of the power equipment data twin and extracting the characteristic data in the structural data includes:

[0104] Acquire different types of structural data of the power equipment data twin, preprocess the structural data to obtain target data, determine the importance of different attributes of the target data, determine key attributes based on the importance, and set static weights of the key attributes;

[0105] Based on the set static weight, the dynamic weight of the key attribute is adjusted and the encoding length is dynamically adjusted to represent the characteristics of the target data and generate characteristic data.

[0106] The characteristics of the target data are represented, including:

[0107] The target data is mapped into a low-dimensional space using a hash learning method to obtain feature representation.

[0108] Wherein, based on the feature data, an evaluation model is trained, including:

[0109] Divide the characteristic data into different tasks according to device type or function.

[0110] Each task corresponds to a specific device type or function, and the feature data is divided into training and validation sets accordingly;

[0111] A base model is trained on each task using the corresponding training set and validation set.

[0112] Parameters or features of the base model are extracted and represented as inputs of a meta-learner to construct a meta-learner, train the meta-learner, and generate an evaluation model.

[0113] After the evaluation model is trained, the accuracy and recall of the evaluation model are evaluated based on the validation set of the new task, and based on the accuracy and recall obtained by the evaluation, the model parameters of the evaluation model are adjusted until the evaluated accuracy and recall meet the requirements.

[0114] The present invention is further described below in conjunction with the implementation cases of the present invention:

[0115] The present invention aims at the complete digital twin of the appearance, state, and parameters of power equipment. It constructs an unsupervised multi-attribute feature learning method that integrates the state, parameters, and equipment appearance, and builds a mapping relationship between the descriptive features obtained by this method and the subjective feelings of the human body, so as to achieve the maturity level of the digital twin model of the equipment measured based on the subjective feelings of the human body.

[0116] The implementation steps of the present invention include:

[0117] Construction of subjective maturity rating criteria;

[0118] Step 1: Design the digital twin maturity assessment criteria for power equipment, as shown in Table 1:

[0119] (1) Analyze the characteristics and requirements of power equipment, combine the existing general digital twin maturity standards, and develop a digital twin maturity evaluation index system suitable for power equipment. The index system should cover multiple aspects such as model accuracy, real-time performance, stability, and scalability.

[0120] (2) Based on the evaluation indicator system, formulate specific evaluation methods and processes, including data collection, model construction, performance testing, expert review, etc.

[0121] Table 1. Maturity subjective rating criteria scoring

[0122]

[0123]

[0124] Step 2: Collect dynamic power equipment digital twin models built by different manufacturers and in various ways;

[0125] (1) Cooperate with major power equipment manufacturers to collect the digital twin models they construct, including model structure, parameters, data, etc.

[0126] (2) According to different types of power equipment and application scenarios, digital twin models with various construction methods are collected to enrich the content of the database.

[0127] Step 3: Invite industry experts to conduct standardized scoring

[0128] (1) Establish an expert review team, including experts and scholars in the power industry and engineers with rich practical experience.

[0129] (2) Organize experts to review the collected digital twin models, score the models according to the evaluation index system, and give specific evaluation opinions and suggestions.

[0130] Step 4: Form a standardized digital twin maturity assessment database

[0131] (1) Organize the experts’ scoring results and evaluation opinions, classify and sort the digital twin models, and form a standardized maturity assessment standard.

[0132] (2) Establish a database to store and manage the collected digital twin models, assessment results, evaluation opinions and other information for subsequent query and use.

[0133] Multi-attribute feature learning;

[0134] Step 1: Data preprocessing:

[0135] (1) Preprocess different types of data, including standardization and normalization, to facilitate subsequent processing.

[0136] (2) Set static weights for key attributes based on the importance of different attributes.

[0137] Step 2: Dynamic weight adjustment:

[0138] (1) Use dynamic weight adjustment methods to dynamically adjust weights according to the importance of different attribute data. This can be achieved by using some adaptive algorithms or feedback-based methods.

[0139] (2) Dynamic weight adjustment formula:

[0140] Among them, w i (t) is the weight of attribute i in the tth iteration, x i is the actual value of attribute i, is the expected value of attribute i, and α is the learning rate.

[0141] Step 3: Dynamic adjustment of encoding length:

[0142] (1) Dynamically adjust the encoding length according to the function and needs of the current model. This can be done through supervised learning or some adaptive algorithms.

[0143] (2) The formula for dynamic adjustment of the code length is: L(t+1)=L(t)+β×(loss(t)-loss(t-1))

[0144] Where L is the encoding length, loss is the model loss function, and β is the learning rate.

[0145] Step 4: Feature Representation:

[0146] (1) Use hash learning methods to map data into a low-dimensional space to obtain feature representation. Some classic hash learning methods are used, such as Locality Sensitive Hashing (LSH) and deep hash learning.

[0147]

[0148] Meta-learning training for multi-device completeness;

[0149] Step 1: Task definition and division;

[0150] (1) Divide the collected device data into different tasks according to device type or function.

[0151] (2) Each task corresponds to a specific device type or function and contains corresponding training and validation sets.

[0152] Step 2: base model training;

[0153] (1) A base model is trained on each task to preliminarily learn the feature representation and evaluation logic of the task.

[0154] (2) Commonly used machine learning algorithms or deep learning models can be selected as the base model.

[0155] Step 3: Meta-learner training;

[0156] (1) Using the trained base models, extract their parameters or feature representations as the input of the meta-learner.

[0157] (2) Build a meta-learner to learn how to quickly find the optimal model parameters or structure based on the representation of the task.

[0158] (3) Meta-learners can be trained using algorithms such as meta-gradient descent and model-agnostic meta-learning (MAML).

[0159] New task adaptation and evaluation model fine-tuning;

[0160] Step 1: Data collection and processing of new tasks;

[0161] (1) When faced with a new equipment maturity assessment task, collect the corresponding data and perform preprocessing.

[0162] (2) Ensure that the data for the new task is consistent with the data for previous tasks in terms of format, feature space, etc.

[0163] Step 2: base model parameter loading;

[0164] (1) Load the parameters of the trained meta-learner into the base model as the initial parameters of the new task.

[0165] Step 3: Fine-tune the base model;

[0166] (1) Use the data of the new task to fine-tune the base model to better adapt it to the characteristics of the new task.

[0167] (2) You can choose an appropriate fine-tuning strategy and learning rate based on the requirements of the new task and the size of the data.

[0168] Step 4: Model evaluation and optimization;

[0169] (1) Use the validation set of the new task to evaluate the fine-tuned model and calculate evaluation metrics (such as accuracy, recall, etc.).

[0170] (2) Based on the evaluation results, the model is further optimized and adjusted to improve its performance on new tasks.

[0171] Embodiment 2:

[0172] The present invention also proposes an evaluation system 200 for the integrity of a digital twin of a power device, such as Figure 2 As shown, including:

[0173] A criterion construction unit 201 is used to construct a maturity subjective rating criterion for the digital twin of the power equipment;

[0174] The data acquisition unit 202 is used to obtain the structural data of the power equipment data twin and extract the characteristic data in the structural data;

[0175] A training unit 203, configured to train an evaluation model based on the feature data;

[0176] The evaluation unit 204 is used to evaluate the integrity of the digital twin of the target power equipment based on the evaluation model and according to the characteristic data of the digital twin of the target power equipment.

[0177] Among them, subjective rating criteria for building maturity of digital twins of power equipment include:

[0178] Establishing an assessment standard for the maturity of the digital twin of the power equipment;

[0179] Obtain digital twin models of power equipment from different manufacturers and in different scenarios;

[0180] Based on the evaluation criteria, standardize the scoring of the digital twin models of power equipment from different manufacturers and in different scenarios;

[0181] Based on the scoring results of the standardized scores, a subjective rating criterion for maturity is constructed.

[0182] Among them, the evaluation criteria for the maturity of the digital twin of the power equipment are established, including:

[0183] Formulate a maturity assessment indicator system for the digital twin of the power equipment;

[0184] Based on the maturity assessment indicator system, formulate assessment standards.

[0185] Wherein, obtaining the structural data of the power equipment data twin and extracting the characteristic data in the structural data includes:

[0186] Acquire different types of structural data of the power equipment data twin, preprocess the structural data to obtain target data, determine the importance of different attributes of the target data, determine key attributes based on the importance, and set static weights of the key attributes;

[0187] Based on the set static weight, the dynamic weight of the key attribute is adjusted and the encoding length is dynamically adjusted to represent the characteristics of the target data and generate characteristic data.

[0188] The characteristics of the target data are represented, including:

[0189] The target data is mapped into a low-dimensional space using a hash learning method to obtain feature representation.

[0190] Wherein, based on the feature data, an evaluation model is trained, including:

[0191] Divide the characteristic data into different tasks according to device type or function.

[0192] Each task corresponds to a specific device type or function, and the feature data is divided into training and validation sets accordingly;

[0193] A base model is trained on each task using the corresponding training set and validation set.

[0194] Parameters or features of the base model are extracted and represented as inputs of a meta-learner to construct a meta-learner, train the meta-learner, and generate an evaluation model.

[0195] After the evaluation model is trained, the accuracy and recall of the evaluation model are evaluated based on the validation set of the new task, and based on the accuracy and recall obtained by the evaluation, the model parameters of the evaluation model are adjusted until the evaluated accuracy and recall meet the requirements.

[0196] The present invention realizes the integrity maturity analysis between the entity and the twin, and has high computational efficiency, meeting the application requirements of the digital twin system.

[0197] Embodiment 3:

[0198] Based on the same inventive concept, the present invention also provides a computer device, which includes a processor and a memory, wherein the memory is used to store a computer program, the computer program includes program instructions, and the processor is used to execute the program instructions stored in the computer storage medium. The processor may be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gates or transistor logic devices, discrete hardware components, etc. It is the computing core and control core of the terminal, which is suitable for implementing one or more instructions, and is specifically suitable for loading and executing one or more instructions in the computer storage medium to implement the corresponding method flow or corresponding functions, so as to implement the steps of the method in the above embodiment.

[0199] Embodiment 4:

[0200] Based on the same inventive concept, the present invention also provides a storage medium, specifically a computer-readable storage medium (Memory), which is a memory device in a computer device for storing programs and data. It is understandable that the computer-readable storage medium here can include both a built-in storage medium in a computer device and an extended storage medium supported by the computer device. The computer-readable storage medium provides a storage space, which stores the operating system of the terminal. In addition, one or more instructions suitable for being loaded and executed by a processor are also stored in the storage space, and these instructions can be one or more computer programs (including program codes). It should be noted that the computer-readable storage medium here can be a high-speed RAM memory or a non-volatile memory, such as at least one disk memory. The processor can load and execute one or more instructions stored in the computer-readable storage medium to implement the steps of the method in the above embodiment.

[0201] It will be appreciated by those skilled in the art that embodiments of the present invention may be provided as methods, systems, or computer program products. Therefore, the present invention may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Moreover, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program codes. The schemes in the embodiments of the present invention may be implemented in various computer languages, for example, object-oriented programming language Java and literal scripting language JavaScript, etc.

[0202] The present invention is described with reference to flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0203] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0204] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 A step that specifies a function in one or more boxes.

[0205] Although the preferred embodiments of the present invention have been described, those skilled in the art may make other changes and modifications to these embodiments once they have learned the basic creative concept. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments and all changes and modifications that fall within the scope of the present invention.

[0206] Obviously, those skilled in the art can make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if these modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalents, the present invention is also intended to include these modifications and variations.

Claims

1. A method for evaluating the integrity of a digital twin of an electric power device, characterized in that: include: Establishing subjective rating criteria for maturity of digital twins of power equipment; Acquire the structural data of the power equipment data twin, and extract feature data from the structural data; Based on the feature data, an evaluation model is trained; Based on the evaluation model, the integrity of the digital twin of the target power equipment is evaluated according to the characteristic data of the digital twin of the target power equipment.

2. The evaluation method according to claim 1, characterized in that: The subjective rating criteria for the maturity of building digital twins of power equipment include: Establishing an assessment standard for the maturity of the digital twin of the power equipment; Obtain digital twin models of power equipment from different manufacturers and in different scenarios; Based on the evaluation criteria, standardize the scoring of the digital twin models of power equipment from different manufacturers and in different scenarios; Based on the scoring results of the standardized scores, a subjective rating criterion for maturity is constructed.

3. The evaluation method according to claim 2, characterized in that: The establishment of the evaluation criteria for the maturity of the digital twin of the power equipment includes: Formulate a maturity assessment indicator system for the digital twin of the power equipment; Based on the maturity assessment indicator system, formulate assessment standards.

4. The evaluation method according to claim 1, characterized in that: The obtaining of the structural data of the power equipment data twin and extracting the characteristic data in the structural data includes: Acquire different types of structural data of the power equipment data twin, preprocess the structural data to obtain target data, determine the importance of different attributes of the target data, determine key attributes based on the importance, and set static weights of the key attributes; Based on the set static weight, the dynamic weight of the key attribute is adjusted and the encoding length is dynamically adjusted to represent the characteristics of the target data and generate characteristic data.

5. The evaluation method according to claim 4, characterized in that: The representing the characteristics of the target data includes: The target data is mapped into a low-dimensional space using a hash learning method to obtain feature representation.

6. The evaluation method according to claim 1, characterized in that: The step of training an evaluation model based on the feature data comprises: Divide the characteristic data into different tasks according to device type or function. Each task corresponds to a specific device type or function, and the feature data is divided into training and validation sets accordingly; A base model is trained on each task using the corresponding training set and validation set. Parameters or features of the base model are extracted and represented as inputs of a meta-learner to construct a meta-learner, train the meta-learner, and generate an evaluation model.

7. The evaluation method according to claim 6, characterized in that: After the evaluation model is obtained through training, the accuracy and recall of the evaluation model are evaluated based on the validation set of the new task, and the model parameters of the evaluation model are adjusted based on the accuracy and recall obtained by the evaluation until the evaluated accuracy and recall meet the requirements.

8. A system for evaluating the integrity of a digital twin of an electric power device, characterized in that: include: A criteria building unit, used to build maturity subjective rating criteria for power equipment digital twins; A data acquisition unit, used to obtain the structural data of the power equipment data twin and extract characteristic data from the structural data; A training unit, used for training an evaluation model based on the feature data; An evaluation unit is used to evaluate the integrity of the digital twin of the target power equipment based on the evaluation model and according to the characteristic data of the digital twin of the target power equipment.

9. The evaluation system according to claim 8, characterized in that The subjective rating criteria for the maturity of building digital twins of power equipment include: Establishing an assessment standard for the maturity of the digital twin of the power equipment; Obtain digital twin models of power equipment from different manufacturers and in different scenarios; Based on the evaluation criteria, standardize the scoring of the digital twin models of power equipment from different manufacturers and in different scenarios; Based on the scoring results of the standardized scores, a subjective rating criterion for maturity is constructed.

10. The evaluation system according to claim 9, characterized in that The establishment of the evaluation criteria for the maturity of the digital twin of the power equipment includes: Formulate a maturity assessment indicator system for the digital twin of the power equipment; Based on the maturity assessment indicator system, formulate assessment standards.

11. The evaluation system according to claim 8, characterized in that The obtaining of the structural data of the power equipment data twin and extracting the characteristic data in the structural data includes: Acquire different types of structural data of the power equipment data twin, preprocess the structural data to obtain target data, determine the importance of different attributes of the target data, determine key attributes based on the importance, and set static weights of the key attributes; Based on the set static weight, the dynamic weight of the key attribute is adjusted and the encoding length is dynamically adjusted to represent the characteristics of the target data and generate characteristic data.

12. The evaluation system according to claim 11, characterized in that The representing the characteristics of the target data includes: The target data is mapped into a low-dimensional space using a hash learning method to obtain feature representation.

13. The evaluation system according to claim 8, characterized in that The step of training an evaluation model based on the feature data comprises: Divide the characteristic data into different tasks according to device type or function. Each task corresponds to a specific device type or function, and the feature data is divided into training and validation sets accordingly; A base model is trained on each task using the corresponding training set and validation set. Parameters or features of the base model are extracted and represented as inputs of a meta-learner to construct a meta-learner, train the meta-learner, and generate an evaluation model.

14. The evaluation system according to claim 13, characterized in that: After the evaluation model is obtained through training, the accuracy and recall of the evaluation model are evaluated based on the validation set of the new task, and the model parameters of the evaluation model are adjusted based on the accuracy and recall obtained by the evaluation until the evaluated accuracy and recall meet the requirements.

15. A computer device, characterized in that: include: one or more processors; a processor for executing one or more programs; When the one or more programs are executed by the one or more processors, the method according to any one of claims 1 to 7 is implemented.

16. A computer-readable storage medium, characterized in that: A computer program is stored thereon, and when the computer program is executed, the method according to any one of claims 1 to 7 is implemented.

Images