Digital twin simulation optimization system for smart grid operation and maintenance scenarios

Abstract

The present application provides a digital twin simulation optimization system for smart grid operation and maintenance scenarios, which relates to the field of power grid operation and maintenance technology. The system includes: a data acquisition module for obtaining a multidimensional attribute characteristic data set of a power grid scenario; a digital twin modeling module for generating a digital twin model of a power grid operation and maintenance scenario; a simulation analysis module for obtaining a set of abnormal operation parameters of a partitioned architecture power grid; and an optimization decision module for operation and maintenance execution management and control. This application can solve the technical problem that traditional operation and maintenance methods are difficult to achieve comprehensive collection and classification labeling of multidimensional data, resulting in inaccurate operation and maintenance decisions, and achieve accurate simulation modeling of power grid operation and maintenance scenarios based on multidimensional attribute characteristic data sets and digital twin technology, ensuring the accuracy of operation and maintenance decisions and the effectiveness of optimization strategies, and improving the technical effect of power grid operation and maintenance efficiency.

Description

Technical Field

[0001] The present application relates to the field of power grid operation and maintenance technology, and in particular to a digital twin simulation optimization system for smart grid operation and maintenance scenarios. Background Art

[0002] With the rapid development of smart grids and the continuous increase in electricity demand, power grid operation and maintenance technology has been widely used in many fields. However, with the continuous expansion of the scale of power grids and the complexity of operation and maintenance tasks, the monitoring and management of power grid operation and maintenance processes are also facing increasing challenges. Traditional power grid operation and maintenance methods are difficult to achieve comprehensive and real-time collection and analysis of multi-dimensional data from all aspects of the power grid, resulting in incomplete data and unable to fully reflect the operation status of the power grid. Moreover, due to the complex power grid operation environment, traditional modeling methods are difficult to accurately describe the characteristics of each part of the power grid and their relationship with each other, resulting in insufficient accuracy of the simulation model, unable to effectively support operation and maintenance decisions, and unable to dynamically adjust and optimize the operation and maintenance strategy based on the real-time operation data of the power grid, resulting in rigid operation and maintenance strategies and difficulty in adapting to the complex and changing power grid operation environment. Summary of the invention

[0003] The purpose of this application is to provide a digital twin simulation optimization system for smart grid operation and maintenance scenarios, in order to solve the technical problem that traditional operation and maintenance methods are difficult to achieve comprehensive collection and classification and labeling of multi-dimensional data, resulting in inaccurate operation and maintenance decisions.

[0004] In view of the above problems, the present application provides a digital twin simulation optimization system for smart grid operation and maintenance scenarios.

[0005] The present application provides a digital twin simulation optimization system for smart grid operation and maintenance scenarios, the system comprising: a data acquisition module, used to perform multi-dimensional data acquisition on a target smart grid, obtain a grid scenario characteristic data set, classify and label the grid scenario characteristic data set, and obtain a grid scenario multi-dimensional attribute characteristic data set; a digital twin modeling module, used to determine a grid operation and maintenance functional architecture according to application requirements of the grid operation and maintenance scenario, perform association mapping and matching based on the grid operation and maintenance functional architecture and the grid scenario multi-dimensional attribute characteristic data set, obtain a grid operation and maintenance functional characteristic data set, and use digital twin technology to classify and label the grid scenario characteristic data set to obtain a grid operation and maintenance functional characteristic data set. The energy characteristic data set is simulated and modeled to generate a digital twin model of the power grid operation and maintenance scenario; the simulation analysis module is used to perform demand analysis on the power grid operation and maintenance functional architecture, determine the set of demand factors for the operation and maintenance function of the partition architecture, and simulate and analyze the set of demand factors for the operation and maintenance function of the partition architecture based on the digital twin model of the power grid operation and maintenance scenario to obtain the set of abnormal operation parameters of the partition architecture power grid; the optimization decision module is used to perform operation and maintenance strategy optimization analysis on the set of abnormal operation parameters of the partition architecture power grid, determine the power grid operation and maintenance optimization strategy parameters, and perform operation and maintenance execution control on the target smart grid based on the power grid operation and maintenance optimization strategy parameters.

[0006] The technical solution provided in this application has at least the following technical effects or advantages:

[0007] The digital twin simulation optimization system for the smart grid operation and maintenance scenario described above collects multi-dimensional data from the target smart grid, obtains the characteristic data of the grid scenario, and classifies and labels the data to form a multi-dimensional attribute characteristic data set. Then, according to the specific grid operation and maintenance requirements, the functional architecture of the grid operation and maintenance is determined, and the association mapping and matching between this architecture and the multi-dimensional attribute characteristic data set are performed to obtain a grid operation and maintenance functional characteristic data set. Subsequently, the digital twin technology is used to simulate and model this functional characteristic data set to generate a digital twin model of the grid operation and maintenance scenario. On this basis, the requirements of the grid operation and maintenance functional architecture are analyzed to determine the set of operation and maintenance functional requirement factors of the partition architecture. Afterwards, the digital twin model of the grid operation and maintenance scenario is used to simulate and analyze these requirement factors to obtain a set of abnormal grid operation parameters under the partition architecture. Then, the operation and maintenance strategy of these abnormal parameters is optimized and analyzed to determine the optimal grid operation and maintenance optimization strategy parameters, and the target smart grid is operated and maintained based on these optimization parameters to ensure the accuracy of the operation and maintenance decision and the effectiveness of the optimization strategy, and improve the efficiency of grid operation and maintenance.

[0008] The above description is only an overview of the technical solution of the present application. In order to more clearly understand the technical means of the present application, it can be implemented according to the contents of the specification, and in order to make the above and other purposes, features and advantages of the present application more obvious and easy to understand, the specific implementation methods of the present application are specifically cited below. It should be understood that the content described in this section is not intended to identify the key or important features of the embodiments of the present application, nor is it intended to limit the scope of the present application. Other features of the present application will become easy to understand through the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0009] In order to more clearly illustrate the technical solutions in the present application or the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings in the following description are only exemplary, and for ordinary technicians in this field, other drawings can be obtained based on the provided drawings without paying any creative work.

[0010] Figure 1 This is a schematic diagram of the structure of the digital twin simulation optimization system used in the smart grid operation and maintenance scenario in this application;

[0011] Figure 2 A schematic diagram of the process of obtaining a multi-dimensional attribute characteristic data set of a power grid scenario for a digital twin simulation optimization system used in smart grid operation and maintenance scenarios in this application.

[0012] Description of reference numerals:

[0013] Data acquisition module 1, digital twin modeling module 2, simulation analysis module 3, optimization decision module 4. DETAILED DESCRIPTION

[0014] This application provides a digital twin simulation optimization system for smart grid operation and maintenance scenarios, which solves the technical problem that traditional operation and maintenance methods are difficult to achieve comprehensive collection and classification labeling of multi-dimensional data, resulting in inaccurate operation and maintenance decisions. It achieves accurate power grid operation and maintenance scenario simulation modeling based on multi-dimensional attribute characteristic data sets and digital twin technology, ensures the accuracy of operation and maintenance decisions and the effectiveness of optimization strategies, and improves the technical effect of power grid operation and maintenance efficiency.

[0015] Below, the technical solutions in the present application will be clearly and completely described with reference to the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all of the embodiments of the present application. It should be understood that the present application is not limited to the example embodiments described herein. Based on the embodiments of the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present application. It should also be noted that, for the convenience of description, only the parts related to the present application are shown in the accompanying drawings, rather than all of them.

[0016] For examples, please see the attached Figure 1 , the present application provides a digital twin simulation optimization system for smart grid operation and maintenance scenarios, the system specifically includes the following modules:

[0017] The data acquisition module 1 is used to collect multi-dimensional data of the target smart grid, obtain a grid scenario characteristic data set, classify and label the grid scenario characteristic data set, and obtain a grid scenario multi-dimensional attribute characteristic data set.

[0018] Specifically, in the data acquisition module 1, the first step for the system terminal to optimize the operation and maintenance of the target smart grid is to conduct comprehensive multi-dimensional data acquisition. This process aims to capture detailed data of all areas covered by the smart grid in real time, obtain the grid scenario characteristic data set, and ensure the comprehensiveness and real-time nature of the information. The grid scenario characteristic data set includes but is not limited to grid distribution data (line layout, substation location, etc.), equipment status information (equipment operation status, fault alarm, etc.), power information (power generation, power consumption, load changes, etc.) and environmental parameters (temperature, humidity, wind speed, etc.). After obtaining the grid scenario characteristic data set, the system terminal classifies and labels the grid scenario characteristic data set according to the pre-built grid data attribute classifier to obtain the grid scenario multi-dimensional attribute characteristic data set. Through this processing process, the final grid scenario multi-dimensional attribute characteristic data set not only contains rich grid operation information, but also has a clear hierarchical structure and clear attribute identification, providing a solid data foundation for subsequent operation and maintenance analysis and decision-making.

[0019] Further, if Figure 2 As shown, the present application provides the method for obtaining a multi-dimensional attribute characteristic data set of a power grid scenario, including:

[0020] Acquire a power grid operation and maintenance scenario knowledge base, extract knowledge attributes from the power grid operation and maintenance scenario knowledge base, and obtain a power grid operation and maintenance scenario knowledge attribute set; perform attribute classification training on the power grid operation and maintenance scenario knowledge base based on the power grid operation and maintenance scenario knowledge attribute set, and construct a power grid data attribute classifier.

[0021] In a preferred embodiment, in order to improve the intelligence level of power grid operation and maintenance, the system terminal first obtains a pre-built power grid operation and maintenance scenario knowledge base. This knowledge base contains a large amount of knowledge and historical operation and maintenance data about power grid operation and maintenance, and is an important basis for subsequent analysis and decision-making. Subsequently, the system terminal determines the key attributes that can describe and distinguish different operation and maintenance scenarios based on the needs and goals of power grid operation and maintenance. These attributes involve multiple dimensions such as line layout, equipment type, equipment status, fault type, load change, environmental conditions, operation and maintenance time, and processing results. Then define a clear description and scope for each attribute to ensure the independence and distinguishability between the attributes. After that, extract the historical operation and maintenance data related to the defined attributes from the power grid operation and maintenance scenario knowledge base. This historical operation and maintenance data includes historical collection data and historical marking data, and these data are summarized to obtain a set of power grid operation and maintenance scenario knowledge attributes. This set of power grid operation and maintenance scenario knowledge attributes can reflect the characteristics and differences of different operation and maintenance scenarios. After obtaining the power grid operation and maintenance scenario knowledge attribute set, the system terminal divides the power grid operation and maintenance scenario knowledge attribute set into a training set and a test set, and then sets the parameters of the random forest model according to the actual data situation and the requirements of the classification task, including the number of trees, the maximum depth, the minimum number of samples required for splitting, the minimum number of samples required for leaf nodes, the maximum number of features, etc. Then, the same number of samples as the power grid operation and maintenance scenario knowledge attribute set are randomly selected from the training set with replacement as the training data of the current decision tree. When each node splits, the system terminal randomly selects a part of the features from all the features according to the set maximum number of features, and finds the best split point based on these features. The node is split into two child nodes according to the best split point found until the set maximum depth is reached. When the node no longer splits, the category of the leaf node is marked according to the label distribution of the samples in the node. Repeat the above steps of building a decision tree until the specified number of decision trees are generated, and these decision trees are integrated into the random forest model. Finally, the system terminal uses the test set to evaluate the trained random forest model and calculate performance indicators such as accuracy and recall. If the calculated performance index does not meet the performance index threshold, the parameters of the random forest model are tuned, including the number of trees, maximum depth, minimum number of samples required for splitting, and other parameters, in order to obtain better classification performance. Otherwise, the current random forest model is output as a power grid data attribute classifier. This power grid data attribute classifier can quickly and accurately identify the key attributes in the power grid operation and maintenance data and classify and organize them. In this way, in the subsequent operation and maintenance process, these data can be used more efficiently to support decision-making and actions, thereby improving the efficiency and accuracy of operation and maintenance.

[0022] Based on the power grid data attribute classifier, the power grid scenario characteristic data set is attribute classified and marked to obtain a power grid scenario multidimensional attribute classification data set; the power grid scenario multidimensional attribute classification data set is sequenced and integrated according to the data occurrence time to obtain the power grid scenario multidimensional attribute characteristic data set.

[0023] In a preferred embodiment, the system terminal inputs the obtained power grid scenario characteristic data set into the constructed power grid data attribute classifier, and the power grid data attribute classifier performs attribute classification marking on each data in the power grid scenario characteristic data set based on the learned classification knowledge to generate a power grid scenario multidimensional attribute classification data set. Subsequently, the power grid scenario multidimensional attribute classification data set is sequenced and integrated according to the time sequence of the data occurrence. The purpose of this step is to organize data within the same time period or with time correlation together, so as to better analyze the characteristics of the power grid scenario that change over time. Through sequence integration, not only can the static characteristics of the power grid scenario at a certain moment, such as equipment status, load conditions, etc., be understood, but also the changing trends of these characteristics over time and the correlation between them can be understood. After the sequence integration is completed, the system terminal obtains the power grid scenario multidimensional attribute characteristic data set, which provides strong data support for subsequent power grid operation and maintenance decisions.

[0024] The digital twin modeling module 2 is used to determine the power grid operation and maintenance functional architecture according to the application requirements of the power grid operation and maintenance scenario, perform association mapping and matching based on the power grid operation and maintenance functional architecture and the multi-dimensional attribute characteristic data set of the power grid scenario, obtain the power grid operation and maintenance functional characteristic data set, use the digital twin technology to simulate and model the power grid operation and maintenance functional characteristic data set, and generate a digital twin model of the power grid operation and maintenance scenario.

[0025] Specifically, in the digital twin modeling module 2, in the complex application scenarios of power grid operation and maintenance, in order to meet the operation and maintenance requirements of different links, the system terminal first determines the functional architecture of power grid operation and maintenance. This architecture is divided based on the key links of power grid operation, such as the detailed architecture of operation and maintenance scenarios such as the power generation side, the power grid side (including transmission and distribution), and the power consumption side. Each part bears specific functions and operation and maintenance challenges, and requires corresponding operation and maintenance strategies and technical support. After determining the power grid operation and maintenance functional architecture, the system terminal divides the power grid operation and maintenance functional architecture into node areas according to the functional architecture attribute description set, and then associates and maps the division results with the power grid scenario multidimensional attribute characteristic data set to extract the characteristic data set closely related to each operation and maintenance function, that is, the power grid operation and maintenance functional characteristic data set. Subsequently, the system terminal constructs a power grid scenario distribution panoramic model and a power grid structure functional attribute model set according to the division results of the power grid operation and maintenance functional architecture and the obtained power grid operation and maintenance functional characteristic data set. After that, the constructed power grid scenario distribution panoramic model and the power grid structure functional attribute model set are simulated and fused modeled using digital twin technology to generate a power grid operation and maintenance scenario digital twin model. This model not only includes the physical structure and equipment information of the power grid, but also incorporates the characteristics of the operation and maintenance functions and the dynamic changes in the operation and maintenance process. By simulating and analyzing this model, we can have a deeper understanding and prediction of the power grid operation and maintenance process, providing strong support for operation and maintenance decisions.

[0026] Furthermore, the present application provides the power grid operation and maintenance function characteristic data set, including:

[0027] Attribute labels are extracted from the multidimensional attribute characteristic data set of the power grid scenario to obtain a set of attribute labels for the power grid scenario data; attributes are described for the power grid operation and maintenance functional architecture based on architecture attribute factors to obtain a set of functional architecture attribute descriptions, wherein the set of functional architecture attribute descriptions includes functional types and functional levels.

[0028] In a preferred embodiment, in the scenario of power grid operation and maintenance, in order to more effectively manage and utilize multidimensional attribute characteristic data sets, the system terminal extracts key information such as equipment type, operating status, fault mode, etc. from the complex power grid scenario multidimensional attribute characteristic data sets to form a set of power grid scenario data attribute labels. These labels provide clear and identifiable identification for subsequent data analysis and application. In order to regionalize the power grid operation and maintenance functional architecture, the system terminal clearly defines the attribute factors that affect the power grid operation and maintenance functional architecture. These factors include function type, function level, relationship between functions, physical location of functions, business objectives of functions, etc. Subsequently, according to the actual needs of power grid operation and maintenance, key functional modules are identified, such as power generation management, transmission scheduling, distribution operation and maintenance, power consumption services, etc. The identified functional modules are then classified and summarized according to their business attributes and technical characteristics to form different functional types. For example, they can be divided into monitoring, control, management, service, etc. Afterwards, the functional level division criteria are determined according to the scope level that the function can affect in power grid operation and maintenance. High-level functions have a wider influence and more complex coordination tasks, while low-level functions are more specific and local. Then, according to the determined classification criteria, each function type is divided into different levels. For example, it can be divided into the first level (such as power grid planning, technology selection), the second level (such as transmission scheduling, distribution management), the third level (such as equipment inspection, data collection), etc. Then, for each function type and level, the system terminal records its detailed attribute description. The attribute description should include information such as the name, type, level, and main function of the function. Then, the attribute descriptions of all function types are sorted out to form the construction of the functional architecture attribute description set.

[0029] The power grid operation and maintenance functional architecture is divided into node areas based on the functional architecture attribute description set to obtain a power grid functional architecture partition node set; the power grid scenario data attribute label set is associated and mapped with each node area in the power grid functional architecture partition node set to obtain the power grid operation and maintenance functional characteristic data set.

[0030] In a preferred embodiment, due to the complexity and hugeness of the power grid operation and maintenance functional architecture, in order to more effectively manage and utilize these functions, the system terminal uses the key attributes such as the function type and level in the functional architecture attribute description set as the basis for division, and divides the power grid functional architecture into multiple node areas to form a power grid functional architecture partition node set. Each area in this power grid functional architecture partition node set contains a group of functional modules with similar functions and levels. This partitioning method not only simplifies the complexity of the functional architecture, but also facilitates subsequent data management and analysis. After forming the power grid functional architecture partition node set, the system terminal associates and maps the power grid scene data attribute label set with these node areas, that is, each label in the power grid scene data attribute label set is compared with each node area in the power grid functional architecture partition node set one by one, and determines which node area each label matches best. Then the matching results are recorded to form an initial power grid operation and maintenance functional characteristic data set. Subsequently, the initial power grid operation and maintenance functional characteristic data set is optimized to obtain a power grid operation and maintenance functional characteristic data set. Optimization refers to removing redundant labels and node areas to ensure the simplicity and efficiency of the data set.

[0031] Furthermore, the present application provides the generation of a digital twin model of a power grid operation and maintenance scenario, including:

[0032] According to the grid function architecture partition node set, distributed data extraction is performed on the grid operation and maintenance function characteristic data set to obtain the architecture partition node distribution characteristic data set; three-dimensional modeling and topological connection are performed based on the architecture partition node distribution characteristic data set to generate a grid scene distribution panoramic model; based on the functional architecture attribute description set, functional training is performed on the grid scene multidimensional attribute characteristic data set to obtain a grid structure function attribute model set; digital twin technology is used to simulate and fuse the grid scene distribution panoramic model with the grid structure function attribute model set to generate a grid operation and maintenance scene digital twin model.

[0033] In an optional implementation, the system terminal performs targeted data extraction on the power grid operation and maintenance function characteristic data set according to the power grid function architecture partition node set. This process aims to extract data closely related to each functional node area from a huge data set to form an architecture partition node distribution characteristic data set. These data sets reflect the operating status and characteristics of the power grid in different functional node areas, providing a basis for subsequent three-dimensional modeling and topological connection. Subsequently, based on the architecture partition node distribution characteristic data set, the system terminal performs three-dimensional modeling and topological connection to generate a panoramic model of the power grid scene distribution. This model is a physical model of the overall distribution of the power grid, which uses three-dimensional modeling technology to show the geographical distribution, equipment layout and connection relationship between each functional node area of ​​the power grid. The panoramic model of the power grid scene distribution not only provides an intuitive visual display, but also provides an important reference for subsequent simulation analysis and operation and maintenance decisions. Afterwards, the system terminal uses the functional architecture attribute description set to perform model framework analysis, determine the power grid function architecture model training framework set, and then perform functional training on the power grid function architecture model training framework set according to the power grid scene multi-dimensional attribute characteristic data set to obtain a power grid structure function attribute model set. This grid structure functional attribute model set includes multiple functional models such as power generation side equipment status analysis, power generation prediction, and transmission efficiency evaluation, which can simulate and predict the operation status and performance of the power grid in different scenarios. Then, the system terminal designs the overall framework of the digital twin model according to the actual needs of power grid operation and maintenance. This framework covers the physical space, virtual space of the power grid and the interaction mechanism between them. After obtaining the overall framework of the digital twin model, the system terminal maps the three-dimensional geographical distribution, equipment layout and other information of the grid scene distribution panorama model to the digital twin model through three-dimensional modeling software to ensure that the digital twin model is visually consistent with the actual power grid. Then, the functional models in the grid structure functional attribute model set, such as power generation side equipment status analysis, power generation prediction, etc., are integrated into the digital twin model to provide intelligent analysis and prediction capabilities for the digital twin model. Finally, the system terminal outputs the digital twin model that maps and integrates the grid scene distribution panorama model and the grid structure functional attribute model set to form a grid operation and maintenance scenario digital twin model.

[0034] Furthermore, the present application provides the generation of a panoramic model of power grid scenario distribution, including:

[0035] The three-dimensional modeling technology is used to perform visual modeling on the architecture partition node distribution characteristic data set to generate a power grid node area model set; the power grid function architecture partition node set is spatially analyzed to obtain the node area spatial position sequence relationship.

[0036] In an optional implementation, the system terminal uses a three-dimensional modeling technology to create a corresponding basic three-dimensional model for each node in the architecture partition node distribution characteristic data set according to its geographical location and device type. Subsequently, details such as texture, color, and logo on the surface of the device are added to the corresponding basic three-dimensional model to enhance the realism and recognition of the basic three-dimensional model. Then, the attribute information of the node, such as load conditions, operating status, etc., is associated with the corresponding basic three-dimensional model. After that, the three-dimensional model of a single node is integrated according to the grid functional architecture partition to form a node area model, and the relative position and angle between each node model are adjusted to ensure that the connection relationship and spatial position sequence relationship between them are accurate. Then, the above modeling process is repeated for all node areas in the power grid until a complete set of grid node area models is generated. After obtaining the grid node area model set, the system terminal performs spatial position analysis on the grid functional architecture partition node set. The purpose of this step is to clarify the spatial position of each node in the power grid, as well as the relative relationship and connection path between them. Through spatial position analysis, the system terminal obtains the spatial position sequence relationship of the node area, that is, the relationship between the nodes arranged in a certain order and path. This relationship is of great significance for understanding the overall layout, operation logic and troubleshooting of the power grid.

[0037] The spatial position sequence relationship of the node area is used as a coding element, and a node area coding system is designed. Based on the node area coding system, the grid functional architecture partition node set is encoded to generate a grid node area coding set; based on the grid node area coding set, the grid node area model set is topologically connected to generate a panoramic model of the grid scene distribution.

[0038] In an optional implementation, after obtaining the spatial position sequence relationship of the node area, the system terminal uses the spatial position sequence relationship of the node area as a coding element to design a set of node area coding systems. This node area coding system is intended to identify each node area in the power grid by a unique code, and at the same time reflect the spatial position sequence relationship between them. The node area coding system contains multiple levels, such as area code, partition code, node code, etc., to ensure the uniqueness and traceability of the coding. Subsequently, the system terminal encodes each node in the partition node set of the power grid functional architecture according to the designed node area coding system. This coding process converts information such as the spatial position and functional attributes of the node into a series of numbers and letters to form a power grid node area coding set. Afterwards, the system terminal establishes a mapping relationship between the power grid node area code and the corresponding three-dimensional model to ensure that each code can quickly locate its corresponding three-dimensional model. Then, the power grid node area code is analyzed to identify the spatial position code contained therein. Then, the identified spatial position code is used to retrieve other node area codes related to the current node area and their connection information. This connection information includes connection type, connection strength, connection direction, etc. Then, according to the retrieved topological relationship, the corresponding node area models are connected, which involves adding lines, pipes or other virtual elements to the model to represent the actual physical connection. Finally, the system terminal integrates all the node area models with topological connections into a unified three-dimensional scene to generate a panoramic model of the power grid scene distribution.

[0039] Furthermore, the present application provides the method of obtaining a set of power grid structure function attribute models, including:

[0040] Based on the functional architecture attribute description set, a model framework analysis is performed to determine the power grid functional architecture model training framework set; the power grid scenario multidimensional attribute characteristic data set is functionally identified through the functional architecture attribute description set to obtain a power grid scenario functional architecture attribute sample set.

[0041] In an optional implementation, the system terminal performs a model framework analysis based on the functional architecture attribute description set of the power grid. This step aims to understand how the various components of the power grid work together to achieve specific functional goals and how these functional goals are characterized by machine learning algorithms. Based on the model framework analysis, the system terminal selects multiple appropriate modeling methods according to the nature of the problem and the characteristics of the data to construct a set of power grid functional architecture model training frameworks, which includes recurrent neural networks (RNNs), support vector machines (SVMs), etc. Each algorithm has its own unique advantages and applicable scenarios. For example, recurrent neural networks are suitable for processing sequence data and can capture time dependencies and long-term dependencies in data. Therefore, they are very effective in processing load, power generation and other data that change over time in the power grid. Support vector machines are suitable for processing classification and regression problems, especially in finding optimal classification boundaries in high-dimensional space. It can be used for tasks such as power grid fault classification and state prediction. Subsequently, the system terminal uses the functional architecture attribute description set to functionally identify the multidimensional attribute characteristic data set of the power grid scenario. The purpose of this step is to extract features directly related to the power grid functional architecture from the multidimensional attribute characteristic data set of the power grid scenario and label these features with corresponding functional labels. Through this process, the system terminal generates a sample set of grid scenario functional architecture attributes. This sample set not only contains various multi-dimensional attribute data of the grid scenario, but also clarifies the relationship between these data and the grid functional architecture through functional identification. In this way, the sample set can provide rich and targeted data support for subsequent model training.

[0042] The power grid scenario function architecture attribute sample set is iteratively trained through the power grid function architecture model training framework set to obtain a power grid scenario function model set; performance evaluation and optimization are performed on the power grid scenario function model set to obtain the power grid structure function attribute model set.

[0043] In an optional implementation, the system terminal uses the determined grid functional architecture model training framework set and combines the obtained grid scenario functional architecture attribute sample set to perform multiple iterative training. Each iteration will perform forward propagation, loss calculation, back propagation and other processes, and adjust the parameters of the model according to the difference between the prediction results of the current model and the actual sample set to optimize the performance of the model. This process is designed to enable the model to better learn and understand the functional architecture attributes in the grid scenario, so as to accurately simulate and predict the actual operating state of the grid. After the iterative training is completed, the system terminal performs a performance evaluation on the trained grid scenario functional model set. This includes using the grid scenario functional architecture attribute sample set that is not used in the training process to verify the accuracy, generalization ability, stability and other key indicators of the corresponding model. The purpose of performance evaluation is to ensure that the model can perform well in actual applications and can adapt to different grid scenarios and needs. Subsequently, according to the results of the performance evaluation, the corresponding model is tuned, which includes adjusting the parameters of the model, optimizing the structure of the model, etc. The purpose of tuning is to further improve the performance of the model so that it is more in line with actual needs. After iterative training, performance evaluation and tuning, the system terminal obtains a set of grid structure function attribute models with excellent performance and suitable for different grid scenarios and needs. These models can accurately reflect the functional architecture attributes of the grid and provide strong support for the planning, design, operation and maintenance of the grid.

[0044] The simulation analysis module 3 is used to analyze the requirements of the power grid operation and maintenance functional architecture, determine the set of required factors for the operation and maintenance functions of the partition architecture, simulate and analyze the set of required factors for the operation and maintenance functions of the partition architecture based on the digital twin model of the power grid operation and maintenance scenario, and obtain the set of abnormal parameters of the power grid operation of the partition architecture.

[0045] Specifically, in the simulation analysis module 3, the system terminal identifies the respective operation and maintenance function requirements for different partition architectures in the power grid operation and maintenance function architecture. For example, the power generation side may include equipment status monitoring, power generation forecasting, and identification of hidden dangers in the power generation environment; the transmission side may focus on line safety, load balancing, and rapid fault location; the distribution side may focus on user power quality and fault repair efficiency. Subsequently, the identified operation and maintenance function requirements are refined into specific operation and maintenance function requirement factors, and these operation and maintenance function requirement factors are sorted out to determine the partition architecture operation and maintenance function requirement factor set. For example, the equipment status monitoring on the power generation side may include multiple monitoring indicators such as equipment temperature, vibration, and oil level; power generation forecasting may involve multiple factors such as historical power generation data and market demand forecasting. After that, the system terminal uses the constructed power grid operation and maintenance scenario digital twin model for simulation analysis. By simulating the partition architecture operation and maintenance function requirement factor set in the model, simulating the power grid behavior under different operation and maintenance scenarios, and then calling the integrated power grid structure function attribute model set, the abnormal parameter set that may appear in the operation of the partition architecture power grid is identified, that is, the abnormal parameter set of the partition architecture power grid operation. The abnormal parameters of the grid operation in this partitioned architecture include voltage fluctuations, current overloads, equipment failure warning signals, environmental parameter exceeding the standard, etc. They directly reflect the challenges and risks that the grid may face under specific operation and maintenance requirements. Through the analysis and evaluation of the abnormal parameter set, the formulation of grid operation and maintenance strategies can be guided to ensure the safe, stable and efficient operation of the grid under the partitioned architecture.

[0046] The optimization decision module 4 is used to perform operation and maintenance strategy optimization analysis on the abnormal operation parameter set of the partitioned architecture power grid, determine the power grid operation and maintenance optimization strategy parameters, and perform operation and maintenance execution control on the target smart grid based on the power grid operation and maintenance optimization strategy parameters.

[0047] Specifically, in the optimization decision module 4, the system terminal performs an optimization analysis of the operation and maintenance strategy for the obtained partitioned architecture power grid operation abnormality parameter set to find the most optimized power grid operation and maintenance solution. This process aims to analyze and divide the power grid operation and maintenance strategy parameter space through the partitioned architecture power grid operation abnormality parameter set, and randomly generate multiple power grid strategy parameters. Subsequently, the multiple generated power grid strategy parameters are simulated and evaluated in combination with the power grid operation and maintenance scenario digital twin model, and the optimization direction of the strategy parameters is determined based on the evaluation results. Local optimization is performed based on the parameter optimization step size to determine the power grid operation and maintenance optimization strategy parameters. Afterwards, the system terminal uses the power grid operation and maintenance optimization strategy parameters to guide the operation and maintenance execution control of the target smart grid to ensure that the power grid can operate safely, stably and efficiently.

[0048] Furthermore, the present application provides the method of determining the grid operation and maintenance optimization strategy parameters, including:

[0049] Construct a power grid operation and maintenance strategy parameter space, analyze and divide the power grid operation and maintenance strategy parameter space based on the partition architecture power grid operation abnormal parameter set, and obtain a power grid operation and maintenance strategy parameter selection threshold; randomly generate multiple power grid strategy parameters within the power grid operation and maintenance strategy parameter selection threshold, and simulate and evaluate the multiple power grid strategy parameters through the power grid operation and maintenance scenario digital twin model to obtain multiple parameter operation and maintenance effects; perform parameter optimization analysis on the power grid operation and maintenance strategy parameter selection threshold based on the multiple parameter operation and maintenance effects, and determine the power grid operation and maintenance optimization strategy parameters.

[0050] In a preferred embodiment, in order to find the most suitable optimization strategy parameters for power grid operation and maintenance, the system terminal first constructs a parameter space containing various operation and maintenance strategy parameters, i.e., the power grid operation and maintenance strategy parameter space, based on historical operation and maintenance data and expert advice. Subsequently, according to the obtained partition architecture power grid operation abnormality parameter set, the power grid operation and maintenance strategy parameter space is divided and analyzed, and the range of each parameter in the partition architecture power grid operation abnormality parameter set is extracted from the power grid operation and maintenance strategy parameter space to obtain the power grid operation and maintenance strategy parameter selection threshold. Afterwards, within the power grid operation and maintenance strategy parameter selection threshold, the system terminal randomly generates multiple power grid strategy parameters, and then simulates these power grid strategy parameters using the power grid operation and maintenance scenario digital twin model, and evaluates the simulation results according to the internally integrated power grid structure function attribute model set to obtain multiple parameter operation and maintenance effects. If these parameter operation and maintenance effects do not meet the parameter operation and maintenance standards, the system terminal determines the optimization direction of the strategy parameters based on these parameter operation and maintenance effects, and locally optimizes the power grid operation and maintenance strategy parameter selection threshold in combination with the set parameter optimization step length. Then, multiple power grid strategy parameters are regenerated for simulation and evaluation. Otherwise, the current multiple power grid strategy parameters are output as power grid operation and maintenance optimization strategy parameters.

[0051] Furthermore, the present application provides the method of determining the power grid operation and maintenance optimization strategy parameters, including:

[0052] Based on the operation and maintenance effects of the multiple parameters, an optimization direction is selected for the threshold value for selecting the parameters of the power grid operation and maintenance strategy, and the optimization direction of the strategy parameters is determined; a parameter optimization step length is set, and the power grid operation and maintenance strategy parameter selection threshold value is locally optimized according to the strategy parameter optimization direction and the parameter optimization step length, and the power grid operation and maintenance optimization strategy parameters are determined.

[0053] In an optional implementation, after determining the operation and maintenance effects of multiple power grid strategy parameters, the system terminal subdivides the optimization direction of the power grid operation and maintenance strategy parameter selection threshold based on the operation and maintenance effect data. By analyzing which parameter combinations have brought significant positive effects, the optimization direction of the power grid operation and maintenance strategy parameter selection threshold is selected. These directions point to parameter areas that can bring significant improvements in operation and maintenance effects. Subsequently, the system terminal sets the parameter optimization step length according to the characteristics of the power grid operation and maintenance strategy parameters and the actual operation and maintenance needs, and based on the parameter optimization step length, takes the current multiple power grid strategy parameters as the initial point, and regenerates multiple power grid strategy parameters according to the strategy parameter optimization direction for simulation evaluation. If the evaluation results meet the parameter operation and maintenance standards, the system terminal uses the current multiple power grid strategy parameters as the power grid operation and maintenance optimization strategy parameters. Otherwise, repeat the above process until the requirements of the parameter operation and maintenance standards are met.

[0054] In summary, the digital twin simulation optimization system for smart grid operation and maintenance scenarios provided by this application has the following technical effects:

[0055] This application uses the data acquisition module 1 to perform multi-dimensional data collection on the target smart grid, obtain the characteristic data of the power grid scene, and classify and mark these data to form a multi-dimensional attribute characteristic data set. The digital twin modeling module 2 determines the functional architecture of the power grid operation and maintenance according to the specific needs of the power grid operation and maintenance, and performs association mapping based on this functional architecture and the multi-dimensional attribute characteristic data set to generate a power grid operation and maintenance functional characteristic data set. The digital twin technology is used to simulate and model this data set, and finally a digital twin model of the power grid operation and maintenance scene is generated. The simulation analysis module 3 performs demand analysis on the power grid operation and maintenance functional architecture, determines the operation and maintenance function demand factors of each partition, and simulates and analyzes these demand factors based on the digital twin model to identify the set of abnormal operation parameters in the partition architecture. The optimization decision module 4 performs strategy optimization analysis on these abnormal parameters, determines the optimization strategy parameters of the power grid operation and maintenance, and performs operation and maintenance execution and control on the target smart grid based on these parameters to achieve efficient and accurate power grid operation and maintenance. In addition, by constructing the grid operation and maintenance strategy parameter space, analyzing the abnormal parameters of the partition architecture, generating multiple strategy parameters for simulation evaluation, and performing parameter optimization analysis based on the evaluation results, the optimization strategy parameters of the grid operation and maintenance are finally determined to ensure the scientificity and effectiveness of the operation and maintenance strategy. These technical effects jointly solve the technical problem that traditional operation and maintenance methods are difficult to achieve comprehensive collection and classification and labeling of multi-dimensional data, resulting in inaccurate operation and maintenance decisions, and achieve accurate grid operation and maintenance scenario simulation modeling based on multi-dimensional attribute characteristic data sets and digital twin technology, ensuring the accuracy of operation and maintenance decisions and the effectiveness of optimization strategies, and improving the efficiency of grid operation and maintenance.

[0056] The above description of the disclosed embodiments enables those skilled in the art to implement or use the present application. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present application. Therefore, the present application will not be limited to the embodiments shown herein, but will conform to the widest scope consistent with the principles and novel features disclosed herein.

[0057] Obviously, those skilled in the art can make various changes and modifications to the present application without departing from the spirit and scope of the present application. Thus, if these modifications and variations of the present application belong to the scope of the present application and its equivalent technology, the present application is also intended to include these modifications and variations.

Claims

1. A digital twin simulation optimization system for smart grid operation and maintenance scenarios, characterized by: The system comprises: A data acquisition module is used to collect multi-dimensional data of the target smart grid, obtain a grid scenario characteristic data set, classify and mark the grid scenario characteristic data set by attributes, and obtain a grid scenario multi-dimensional attribute characteristic data set; A digital twin modeling module is used to determine the power grid operation and maintenance functional architecture according to the application requirements of the power grid operation and maintenance scenario, perform association mapping and matching based on the power grid operation and maintenance functional architecture and the multi-dimensional attribute characteristic data set of the power grid scenario, obtain the power grid operation and maintenance functional characteristic data set, use the digital twin technology to simulate and model the power grid operation and maintenance functional characteristic data set, and generate a digital twin model of the power grid operation and maintenance scenario; A simulation analysis module is used to perform demand analysis on the power grid operation and maintenance functional architecture, determine a set of demand factors for the operation and maintenance functions of the partition architecture, perform simulation analysis on the set of demand factors for the operation and maintenance functions of the partition architecture based on the digital twin model of the power grid operation and maintenance scenario, and obtain a set of abnormal operation parameters of the partition architecture power grid; An optimization decision module is used to perform an operation and maintenance strategy optimization analysis on the abnormal operation parameter set of the partitioned architecture power grid, determine the power grid operation and maintenance optimization strategy parameters, and perform operation and maintenance execution control on the target smart grid based on the power grid operation and maintenance optimization strategy parameters; The step of obtaining a multi-dimensional attribute characteristic data set of a power grid scenario includes: Acquire a power grid operation and maintenance scenario knowledge base, extract knowledge attributes from the power grid operation and maintenance scenario knowledge base, and obtain a power grid operation and maintenance scenario knowledge attribute set; Based on the power grid operation and maintenance scenario knowledge attribute set, attribute classification training is performed on the power grid operation and maintenance scenario knowledge base to construct a power grid data attribute classifier; Based on the power grid data attribute classifier, attribute classification marking is performed on the power grid scenario characteristic data set to obtain a power grid scenario multidimensional attribute classification data set; Sequentially integrating the power grid scenario multidimensional attribute classification data set according to the data generation time to obtain the power grid scenario multidimensional attribute characteristic data set; The power grid operation and maintenance function characteristic data set is obtained, including: Extracting attribute labels from the multi-dimensional attribute characteristic data set of the power grid scenario to obtain a power grid scenario data attribute label set; Based on the architecture attribute factors, the power grid operation and maintenance function architecture is described by attributes to obtain a function architecture attribute description set, wherein the function architecture attribute description set includes a function type and a function level; Divide the power grid operation and maintenance functional architecture into node regions based on the functional architecture attribute description set to obtain a power grid functional architecture partition node set; Associating and mapping the power grid scenario data attribute label set with each node area in the power grid function architecture partition node set to obtain the power grid operation and maintenance function characteristic data set; The generating of the digital twin model of the power grid operation and maintenance scenario includes: Extracting distributed data from the power grid operation and maintenance function characteristic data set according to the power grid function architecture partition node set to obtain an architecture partition node distribution characteristic data set; Perform three-dimensional modeling and topological connection based on the architecture partition node distribution characteristic data set to generate a panoramic model of power grid scene distribution; Based on the functional architecture attribute description set, function training is performed on the multidimensional attribute characteristic data set of the power grid scenario to obtain a power grid structure functional attribute model set; The digital twin technology is used to simulate and fuse the power grid scenario distribution panoramic model with the power grid structure functional attribute model set to generate a digital twin model of the power grid operation and maintenance scenario.

2. The digital twin simulation optimization system for smart grid operation and maintenance scenarios according to claim 1, characterized in that: The generating of the grid scenario distribution panoramic model comprises: Using three-dimensional modeling technology to perform visual modeling on the architecture partition node distribution characteristic data set to generate a set of power grid node area models; Performing spatial position analysis on the partition node set of the power grid functional architecture to obtain a spatial position sequence relationship of the node area; Taking the spatial position sequence relationship of the node area as a coding element, designing a node area coding system, encoding the power grid functional architecture partition node set based on the node area coding system, and generating a power grid node area coding set; The power grid node area model set is topologically connected based on the power grid node area code set to generate the power grid scene distribution panoramic model.

3. The digital twin simulation optimization system for smart grid operation and maintenance scenarios according to claim 1, characterized in that: The obtaining of a set of power grid structure function attribute models comprises: Performing a model framework analysis based on the functional architecture attribute description set to determine a power grid functional architecture model training framework set; Functionally identifying the power grid scenario multidimensional attribute characteristic data set through the functional architecture attribute description set to obtain a power grid scenario functional architecture attribute sample set; Iteratively training the power grid scenario function architecture attribute sample set respectively through the power grid function architecture model training framework set to obtain a power grid scenario function model set; The power grid scenario function model set is subjected to performance evaluation and tuning to obtain the power grid structure function attribute model set.

4. The digital twin simulation optimization system for smart grid operation and maintenance scenarios according to claim 1, characterized in that: Determining the power grid operation and maintenance optimization strategy parameters includes: Constructing a power grid operation and maintenance strategy parameter space, parsing and dividing the power grid operation and maintenance strategy parameter space based on the partition architecture power grid operation abnormality parameter set, and obtaining a power grid operation and maintenance strategy parameter selection threshold; Randomly generate multiple power grid strategy parameters within the power grid operation and maintenance strategy parameter selection threshold, and simulate and evaluate the multiple power grid strategy parameters through the power grid operation and maintenance scenario digital twin model to obtain multiple parameter operation and maintenance effects; Based on the operation and maintenance effects of the multiple parameters, parameter optimization analysis is performed on the thresholds for selecting the parameters of the power grid operation and maintenance strategy to determine the power grid operation and maintenance optimization strategy parameters.

5. The digital twin simulation optimization system for smart grid operation and maintenance scenarios according to claim 4, characterized in that: Determining the power grid operation and maintenance optimization strategy parameters includes: Based on the operation and maintenance effects of the multiple parameters, an optimization direction is selected for the selection threshold of the power grid operation and maintenance strategy parameters, and an optimization direction of the strategy parameters is determined; The parameter optimization step length is set, and the power grid operation and maintenance strategy parameter selection threshold is locally optimized according to the strategy parameter optimization direction and the parameter optimization step length to determine the power grid operation and maintenance optimization strategy parameters.

Images