Transformer substation three-dimensional business model dynamic construction algorithm based on GIM model optimization

Abstract

The invention relates to the technical field of transformer substations, in particular to a transformer substation three-dimensional business model dynamic construction algorithm based on GIM model optimization. According to the algorithm, firstly, voltage, current, temperature and vibration operation parameter data of equipment in a transformer substation and spatial position information and engineering parameter data of the equipment are obtained, and the data are processed in real time; then, carrying out equipment attribute analysis and feature extraction on the processed data, and constructing an equipment feature matrix; thirdly, dynamically optimizing and adjusting the GIM model based on the equipment characteristic matrix, and simplifying the model through a QEM algorithm to realize the lightweight of the model; and finally, realizing real-time linkage of the optimized three-dimensional model and the actual equipment operation state through a digital twin platform. According to the method, the problem that a traditional static model is difficult to adapt to complex dynamic businesses is solved through technologies such as multi-dimensional data acquisition and fusion, feature matrix dynamic construction and model adaptive optimization, and the flexibility and response efficiency of the three-dimensional model of the transformer substation are remarkably improved.

Description

Technical Field

[0001] The present invention relates to the technical field of substations, and in particular to a dynamic construction algorithm of a substation three-dimensional business model based on GIM model optimization. Background Art

[0002] With the development of digitalization and intelligence of power grids, substation 3D modeling technology has gradually become a core tool in operation and maintenance. 3D models can intuitively display the physical structure, spatial distribution and operating status of equipment, providing strong auxiliary support for operation and maintenance personnel. Traditional 3D modeling methods mainly adopt static construction methods, and cannot respond to dynamic business needs in real time after model generation. This method shows obvious lag and limitations when facing frequent changes in equipment operating status and complex business processes. In addition, traditional models mostly use manual design or semi-automatic modeling based on static data, which is inefficient and inflexible, and it is difficult to meet the requirements of modern power grids for efficient, accurate and real-time operation and maintenance. In practical applications, the lag of static models is particularly prominent, and its lack of support for dynamic operation and maintenance needs seriously restricts the efficiency of operation and maintenance. At the same time, traditional methods are weak in multi-source data fusion, and it is difficult to effectively integrate the physical properties, engineering parameters and spatial location information of equipment, resulting in the lack of adaptability and practicality of the generated models.

[0003] Existing research has made some progress in improving the efficiency of substation 3D modeling and optimizing model performance. The 3D lightweight display method of substation based on GIM optimization transforms the traditional GIM model into a lighter 3D display model through hierarchical division, parameter modeling and model assembly. This method can effectively reduce the amount of model data and improve the efficiency of the model in visualization scenarios. It has significant advantages in data standardization and model simplification. However, the limitation of this method is that it is mainly oriented to the static display scene of the model, lacks the ability to respond to dynamic business needs in real time, and cannot adjust the model construction strategy according to the operation status of the equipment, so the actual application scope in operation and maintenance is limited. The fast modeling method based on the improved SSD neural network and ICP point cloud registration algorithm greatly improves the 3D modeling speed of substation equipment by combining target detection and point cloud registration technology. This method uses deep learning algorithm for target recognition and uses ICP algorithm for high-precision matching of point cloud models, especially showing excellent performance in the initial generation stage of the model. However, this method also has limitations. Its recognition accuracy is restricted by the number of point cloud samples and scanning conditions, and it pays more attention to static modeling tasks, and its adaptive adjustment ability to dynamic business needs is weak. In addition, the method itself lacks support for model optimization and updating, and cannot achieve continuous dynamic adjustment, which limits its application value in complex operation and maintenance scenarios. Summary of the invention

[0004] In view of the problems existing in the prior art, the present invention is proposed.

[0005] Therefore, the problem to be solved by the present invention is how to solve the data requirements in the substation operation and maintenance process by real-time analysis, and dynamically adjust the model structure in combination with the adaptive optimization algorithm to ensure that the model can quickly respond to complex business changes. At the same time, the physical properties, engineering parameters and spatial location information of the equipment are integrated, and the accuracy and practicality of the model are further improved through multi-source data fusion. The dynamically generated three-dimensional model can be efficiently linked with the digital twin system to provide intuitive and real-time auxiliary decision support for operation and maintenance personnel.

[0006] In order to solve the above technical problems, the present invention provides the following technical solutions:

[0007] In the first aspect, the embodiment of the present invention provides a dynamic construction algorithm of a three-dimensional business model of a substation based on GIM model optimization, which includes obtaining voltage, current, temperature, vibration operation parameter data of equipment in the substation, as well as spatial location information and engineering parameter data of the equipment, and performing real-time processing on the obtained data;

[0008] Perform device attribute analysis and feature extraction on the processed data to build a device feature matrix;

[0009] Dynamically optimizing and adjusting the GIM model based on the equipment feature matrix;

[0010] The digital twin platform enables real-time linkage between the optimized three-dimensional model and the actual equipment operating status.

[0011] As an optimal solution for the algorithm for dynamically constructing the three-dimensional business model of a substation based on GIM model optimization described in the present invention, the spatial position information of the equipment is obtained through the Beidou high-precision positioning system, and the engineering parameters include the equipment load capacity and the estimated value of the remaining life.

[0012] As a preferred solution of the algorithm for dynamically constructing the three-dimensional business model of a substation based on GIM model optimization described in the present invention, the equipment attribute analysis includes: quantitative description of the spatial posture of the equipment; normalized quantification of the equipment operating status; and generation of an equipment operating attribute matrix.

[0013] As a preferred solution of the algorithm for dynamically constructing the three-dimensional business model of a substation based on GIM model optimization described in the present invention, the feature extraction includes: generating feature vectors for the physical properties and operating data of the equipment; combining the feature vectors of all equipment to form an equipment feature matrix; and performing weight calculation on the feature matrix.

[0014] As a preferred solution of the algorithm for dynamically constructing the three-dimensional business model of a substation based on GIM model optimization described in the present invention, the dynamic optimization of the GIM model includes: performing redundant data analysis on the components in the initial model; optimizing the triangular faces and material textures of the components using the QEM algorithm; and calculating the model adjustment amount based on the change in equipment characteristics.

[0015] As a preferred scheme of the algorithm for dynamically constructing the three-dimensional business model of the substation based on GIM model optimization described in the present invention, the calculation of the model adjustment amount based on the equipment characteristic change includes: real-time monitoring of the changes in the equipment characteristic matrix; calculating the corresponding model adjustment parameters according to the characteristic changes; and real-time updating of the model status according to the preset update frequency.

[0016] As a preferred solution of the algorithm for dynamically constructing the three-dimensional business model of a substation based on GIM model optimization described in the present invention, the linkage achieved through the digital twin platform includes: establishing a dynamic association between the optimized three-dimensional model and the actual equipment operation status; generating a list of equipment that needs to be monitored based on equipment status data analysis; integrating all update results to form a digital twin system with dynamic response capabilities.

[0017] As a preferred solution for the dynamic construction algorithm of the substation three-dimensional business model based on GIM model optimization described in the present invention, the digital twin system has the following response functions: warning of potential faults by analyzing equipment operation trends; scheduling maintenance resources according to equipment operating status and priority; and providing real-time model support for multi-department collaborative operations.

[0018] As a preferred solution of the algorithm for dynamically constructing the three-dimensional business model of a substation based on GIM model optimization described in the present invention, the real-time processing of the acquired data includes: transmitting the collected data to a central data processing platform through a Beidou communication system; distributing and storing the data; and realizing efficient query and use of the data.

[0019] As a preferred solution of the algorithm for dynamically constructing the three-dimensional business model of a substation based on GIM model optimization described in the present invention, the optimized three-dimensional model is used to: intuitively display the physical structure and spatial distribution of the equipment; display the operating status of the equipment in real time; and provide maintenance and operation and maintenance decision support for operation and maintenance personnel.

[0020] In a second aspect, an embodiment of the present invention provides a computer device, comprising a memory and a processor, wherein the memory stores a computer program, wherein: when the computer program instructions are executed by the processor, the steps of the dynamic construction algorithm of the substation three-dimensional business model based on GIM model optimization as described in the first aspect of the present invention are implemented.

[0021] In a third aspect, an embodiment of the present invention provides a computer-readable storage medium having a computer program stored thereon, wherein: when the computer program instructions are executed by a processor, the steps of the algorithm for dynamically constructing a three-dimensional business model of a substation based on GIM model optimization as described in the first aspect of the present invention are implemented.

[0022] The beneficial effects of the present invention are as follows: the present invention can significantly improve the model flexibility and response efficiency in the substation operation and maintenance scenario. By real-time analysis of equipment operation status and business demand changes, the present invention realizes dynamic adjustment of the three-dimensional model, and solves the limitation that the traditional static model is difficult to adapt to complex dynamic business. At the same time, the present invention integrates the physical properties, operation status and spatial position information of the equipment, constructs a unified feature matrix as the basis for model optimization, and combines the QEM geometric optimization algorithm to effectively reduce redundant data while maintaining the accuracy of the model, thereby realizing the lightweight and precision of the model. The present invention also ensures that the dynamically adjusted three-dimensional model can synchronize the equipment status in real time and provide accurate business support through efficient linkage with the digital twin platform. Especially in multi-condition testing, the present invention shows strong adaptability to complex scenarios (such as high load, extreme temperature) and dynamic business (such as sudden maintenance needs), which not only significantly improves the operation and maintenance efficiency, but also provides a reliable basis for equipment fault diagnosis, status assessment and maintenance planning.

[0023] The technical points to be protected by the present invention include real-time dynamic adjustment mechanism, multi-source data fusion method, QEM-based model lightweight optimization technology, and dynamic linkage solution with the digital twin platform. These key innovations provide strong technical support for the intelligent and efficient operation and maintenance of substations, and also open up new directions for the widespread application of digital twin technology. BRIEF DESCRIPTION OF THE DRAWINGS

[0024] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0025] Figure 1 Flowchart of the algorithm for dynamically building a three-dimensional business model of a substation based on GIM model optimization;

[0026] Figure 2 Computer equipment diagram for dynamic construction algorithm of substation three-dimensional business model based on GIM model optimization;

[0027] Figure 3 A feature extraction flow chart for the dynamic construction algorithm of the substation 3D business model based on GIM model optimization;

[0028] Figure 4 Schematic diagram of the QEM algorithm for dynamically building the three-dimensional business model of a substation based on GIM model optimization. DETAILED DESCRIPTION

[0029] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the specific implementation methods of the present invention are described in detail below in conjunction with the accompanying drawings.

[0030] In the following description, many specific details are set forth to facilitate a full understanding of the present invention, but the present invention may also be implemented in other ways different from those described herein, and those skilled in the art may make similar generalizations without violating the connotation of the present invention. Therefore, the present invention is not limited to the specific embodiments disclosed below.

[0031] Secondly, the term "one embodiment" or "embodiment" as used herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The term "in one embodiment" that appears in different places in this specification does not necessarily refer to the same embodiment, nor is it a separate or selective embodiment that is mutually exclusive with other embodiments.

[0032] Example 1

[0033] Reference Figure 1-2 , which is the first embodiment of the present invention, and provides a dynamic construction algorithm of a substation three-dimensional business model based on GIM model optimization, including:

[0034] S100: Obtain voltage, current, temperature, vibration operation parameter data of equipment in the substation, as well as spatial location information and engineering parameter data of the equipment, and process the acquired data in real time;

[0035] In the embodiment of the present application, the collected data includes a variety of equipment operating parameters. Among them, voltage data is collected in real time by a high-precision voltage sensor, and the sampling frequency can reach 10KHz, ensuring that the rapid changes in voltage can be captured; current data is collected by a Hall current sensor, with a measurement range of 0-1000A and an accuracy of 0.5% FS; temperature data is measured non-contact by an infrared thermometer, with a temperature measurement range of -20℃~400℃ and a resolution of 0.1℃; vibration data is collected by a three-axis acceleration sensor, with a frequency response range of 0.5Hz-5KHz.

[0036] In an optional embodiment, the spatial location information of the device is obtained by combining the Beidou high-precision positioning system with UWB positioning technology. The Beidou positioning system provides centimeter-level positioning accuracy, while UWB positioning can provide sub-meter-level positioning services in indoor environments. The integration of the two positioning technologies can not only realize all-weather and full-coverage location information collection, but also complement and verify each other, significantly improving the reliability of positioning.

[0037] In an optional embodiment, other types of operating data can be collected. For example, the collection of equipment noise data (i.e., the fifth type of data) can be added, and the noise characteristics of the equipment during operation can be collected through an acoustic sensor array to evaluate the operating status of the equipment. In addition, oil chromatography analysis data can be introduced to diagnose potential equipment failures by online monitoring of the content of dissolved gas in transformer oil.

[0038] It should be noted that the multi-dimensional data collection solution adopted by the present invention can fully reflect the operating status of substation equipment. Voltage and current data reflect the electrical performance of the equipment, temperature and vibration data reflect the mechanical characteristics, and spatial position information is used to build a geometric model of the equipment. This fusion of multi-source heterogeneous data not only provides a full range of equipment status information, but also lays a solid data foundation for subsequent model construction and optimization.

[0039] In the embodiment of the present application, the real-time processing of the acquired data mainly includes the following steps: First, the raw data is subjected to noise reduction processing using an anti-interference filtering algorithm to remove random errors and systematic errors in the measurement process. Then, data of different dimensions are mapped to a unified value interval through data normalization processing. Finally, a distributed storage architecture is used to store and manage the processed data.

[0040] Specifically, data processing adopts a layered architecture design. At the edge layer, high-performance industrial computers are deployed to perform preliminary filtering and compression of data to reduce transmission bandwidth requirements. At the cloud layer, a distributed computing framework is used to perform parallel processing of large-scale data to improve processing efficiency. Data storage uses a time series database (such as InfluxDB) to store high-frequency sampled equipment operation data, and a relational database (such as PostgreSQL) to store static attribute information of equipment.

[0041] It should be noted that the real-time data processing scheme of the present invention fully considers the characteristics of substation data. First, edge computing reduces the amount of data transmission and improves the system response speed. Second, the use of a distributed architecture improves the scalability and reliability of the system. Finally, different types of databases are selected in a targeted manner to optimize the storage and query efficiency of data.

[0042] S101: The spatial location information of the equipment is obtained through the Beidou high-precision positioning system. The engineering parameters include the equipment load capacity and the estimated remaining life.

[0043] In the embodiment of this application, the Beidou positioning system adopts RTK differential positioning technology, equipped with high-precision multi-band antennas and high-performance receivers. The system can simultaneously receive signals from the three major satellite navigation systems of Beidou, GPS and GLONASS. In an open environment, the horizontal positioning accuracy can reach ±1cm+1ppm, and the vertical positioning accuracy can reach ±2cm+1ppm, which fully meets the needs of substation equipment positioning.

[0044] In an optional embodiment, in order to improve the positioning accuracy indoors and in obstructed areas, the system also integrates an indoor positioning system based on UWB technology. The UWB system operates in the 3.1GHz-10.6GHz frequency band and uses a time differential positioning algorithm to achieve a positioning accuracy of less than 30cm inside the substation. The system deploys UWB base stations at key locations to build an indoor positioning network and achieve continuous tracking of device locations.

[0045] Specifically, load capacity assessment uses a combination of real-time monitoring and historical data analysis. By collecting real-time operating data of the equipment, such as current, temperature and other parameters, combined with the rated parameters of the equipment, the current load level is calculated. At the same time, the system maintains a load history database to record the operating performance of the equipment under different working conditions, providing data support for load capacity assessment.

[0046] It should be noted that the remaining life estimation adopts a life prediction model based on deep learning. The model input includes multi-dimensional data such as equipment operating parameters, maintenance records, and failure history. The performance degradation trend of the equipment is modeled through the long short-term memory network (LSTM) to achieve dynamic prediction of the remaining life. The reliability of the prediction results is verified and corrected through regular actual test data.

[0047] S200: Perform device attribute analysis and feature extraction on the processed data to construct a device feature matrix;

[0048] In the embodiment of this application, the equipment attribute analysis mainly includes two aspects: physical property analysis and operation status analysis. Physical property analysis mainly focuses on static characteristics such as geometric dimensions, material properties, and installation location of the equipment; operation status analysis focuses on dynamic properties such as electrical parameters, thermal properties, and vibration characteristics. The system adopts a modular analysis framework and designs special analysis algorithms for different types of attributes.

[0049] In an optional embodiment, for large transformer equipment, the physical property analysis also includes special analyses such as oil level analysis and cooling efficiency evaluation. The oil level analysis tracks the changes in the transformer oil level in real time through the online monitoring system, performs compensation calculations based on the ambient temperature, and evaluates the sealing of the oil system. The cooling efficiency evaluation calculates the actual heat dissipation capacity by measuring the inlet and outlet temperature difference, flow rate and other parameters of the cooling device.

[0050] Specifically, feature extraction uses a multi-level feature engineering method. The first is the extraction of basic features, including time domain features (mean, standard deviation, peak, etc.) and frequency domain features (spectral analysis, harmonic analysis, etc.). Then comes the construction of combined features, generating new feature dimensions through feature combination and crossover operations. Finally, deep feature learning is carried out, using deep learning models such as autoencoders to automatically extract high-order features from the data.

[0051] In the embodiment of the present application, the construction of the feature matrix adopts intelligent dimensionality reduction and feature selection technology. The system first uses principal component analysis (PCA) to reduce the dimension of the original features to remove redundancy and noise. Then, through feature importance evaluation, the feature subset that contributes the most to the model performance is selected. The feature matrix finally formed not only maintains the main information of the data, but also has good computational efficiency.

[0052] It should be noted that in order to improve the accuracy and robustness of feature extraction, the system implements an adaptive feature update mechanism. When a significant change in the device operation mode is detected, the system automatically triggers the feature reconstruction process to ensure that the feature matrix can reflect the changes in the device status in a timely manner. At the same time, the system will regularly evaluate the effectiveness of the features, eliminate invalid features, and add new valid features.

[0053] S201: Equipment attribute analysis includes: quantitatively describing the equipment spatial posture; normalizing and quantifying the equipment operation status; and generating an equipment operation attribute matrix.

[0054] S202: Feature extraction includes: generating feature vectors for physical properties and operation data of the equipment; combining feature vectors of all equipment to form a device feature matrix; and performing weight calculation on the feature matrix.

[0055] S300: Dynamically optimize and adjust the GIM model based on the equipment feature matrix;

[0056] In the embodiment of the present application, the dynamic optimization of the GIM model adopts a hierarchical optimization strategy, and optimizes the geometric model, texture mapping and operating status respectively. The geometric model optimization mainly simplifies the triangular facets through the QEM algorithm, reducing the amount of data while maintaining the visual effect of the model; the texture mapping optimization adopts multi-level texture compression technology to dynamically adjust the texture accuracy according to the observation angle; the operating status optimization is based on real-time monitoring data to dynamically update the state parameters of the model.

[0057] In an optional embodiment, the system also introduces a hierarchical simplification mechanism of the model based on the viewpoint. For the model area far from the viewpoint, a higher degree of simplification strategy is adopted; for the focus area near the viewpoint, a higher model accuracy is maintained. This hierarchical simplification strategy not only improves the rendering efficiency of the model, but also ensures the display quality of important areas. Specifically, the system divides the observation space into three levels: near, medium, and far, and adopts 100%, 75%, and 50% of the patch retention rates respectively.

[0058] Specifically, the optimization process of the QEM algorithm includes the following steps: first, the quadratic error matrix of each vertex is calculated, which reflects the contribution of the vertex to the model shape; then, all possible edge folding operations are evaluated, and the operation that causes the smallest geometric deformation is selected for execution; finally, the error matrix of the affected area is updated, and the next optimal folding operation is continued to be found. The whole process is repeated until the preset simplification goal is reached.

[0059] It should be noted that in order to avoid model distortion caused by over-simplification, the system sets multiple constraints. These include geometric feature preservation constraints to ensure that important geometric features (such as corners and contours) are not destroyed; topology preservation constraints to avoid self-intersection or holes in the model; and texture consistency constraints to ensure the continuity of texture mapping. The setting of these constraints allows the model simplification process to achieve the optimization goal without compromising the usability of the model.

[0060] S301: Dynamic optimization of the GIM model includes: performing redundant data analysis on the components in the initial model; optimizing the triangular facets and material textures of the components using the QEM algorithm; and calculating the model adjustment amount based on the change in device characteristics.

[0061] In the embodiment of the present application, redundant data analysis adopts an entropy-based information redundancy evaluation method. The system first calculates the geometric entropy of each region of the model, which reflects the geometric complexity of the region; then analyzes the information redundancy between adjacent regions and identifies regions that can be merged or simplified; finally, a differentiated optimization strategy is formulated based on the evaluation results.

[0062] In an optional embodiment, the system also implements an intelligent redundancy determination mechanism. By analyzing the frequency of use, observation angle, and importance of the model, different optimization weights are assigned to different areas. For example, for equipment areas in the substation that often need to be repaired, a high model accuracy will be maintained even if there is a certain redundancy; while for areas that are not frequently used, a more radical simplification strategy can be adopted.

[0063] Specifically, material texture optimization uses multi-level texture mapping technology. The system generates mipmap levels of different resolutions for each texture image, and selects the appropriate texture level according to the viewing distance and angle during rendering. At the same time, for large areas of repeated textures, texture atlas technology is used to reduce texture memory usage. Texture compression uses the BC7 format, which reduces the amount of texture data to 1 / 4 to 1 / 8 of the original while ensuring visual quality.

[0064] In the embodiment of the present application, the calculation of the model adjustment amount adopts an adaptive weight method. The system dynamically adjusts the weight coefficients of different features according to the importance and urgency of the changes in the equipment features. For key features that affect the safe operation of the equipment, a higher adjustment weight is given; for secondary features, a lower adjustment weight is used. This differentiated adjustment strategy ensures the efficiency and accuracy of the model update.

[0065] It should be noted that in order to ensure the real-time and reliability of model optimization, the system has established a three-level cache mechanism. The first level is the memory cache, which is used to store the most frequently accessed model data; the second level is the video memory cache, which is used to store the model data required for the current view; the third level is the disk cache, which is used to store the complete model data. This multi-level cache strategy significantly improves the efficiency of model access and update.

[0066] S302: Calculating the model adjustment amount based on the equipment feature change includes: real-time monitoring of the change of the equipment feature matrix; calculating the corresponding model adjustment parameters according to the feature change; and updating the model state in real time according to a preset update frequency.

[0067] In the embodiment of the present application, the monitoring of the characteristic matrix change adopts a multi-threshold detection mechanism. The system sets different change thresholds for different types of characteristics. When the change of the characteristic value exceeds the corresponding threshold, the model adjustment process is triggered. Among them, the change threshold of electrical parameters (such as voltage and current) is set to ±5% of the rated value, the temperature change threshold is ±10°C, and the change threshold of vibration characteristics is ±20% of the reference value.

[0068] In an optional embodiment, the system implements an intelligent trend analysis function. By performing time series analysis on feature data, the system can identify gradual trends and sudden changes. For gradual trends, the system adopts a smooth adjustment strategy to avoid frequent model updates; for sudden changes, the system immediately triggers an emergency adjustment process to ensure that the model can reflect the drastic changes in the device status in a timely manner.

[0069] Specifically, the calculation of model adjustment parameters is based on the device status assessment model. The system first normalizes the feature changes, and then calculates the health index of the device through the status assessment model. According to the degree of change in the health index, the system automatically adjusts the display properties, alarm levels, and interactive response parameters of each component. For example, when the temperature of a device rises abnormally, the system automatically adjusts the color display and transparency of the device to highlight the abnormal area.

[0070] In the embodiment of the present application, the model state update adopts a multi-threaded parallel processing mechanism. The main thread is responsible for user interaction and scene rendering, and the separate update thread is responsible for the calculation and update of the model state. This asynchronous update mechanism ensures the smoothness of the system interface and the real-time nature of the model update. The default update frequency of the system is 10Hz, which can be adjusted within the range of 1Hz-30Hz according to actual needs.

[0071] It should be noted that in order to optimize system performance and reduce resource consumption, the system implements an intelligent update strategy. Through a comprehensive analysis of model importance, change frequency, and user attention, the system assigns different update priorities to different components. High-priority components get more frequent update opportunities, while low-priority components are updated less frequently. This differentiated update strategy ensures the real-time nature of key information while avoiding unnecessary waste of computing resources.

[0072] S400: Realize real-time linkage between the optimized 3D model and the actual equipment operating status through the digital twin platform.

[0073] In the embodiment of the present application, the digital twin platform adopts a microservice architecture design, which mainly includes four modules: data acquisition service, model management service, state synchronization service and visualization service. The platform is deployed based on Docker container technology and uses Kubernetes for container orchestration and management to ensure high availability and scalability of the system.

[0074] In an optional embodiment, the platform implements a data stream processing mechanism based on a message queue. Using RabbitMQ as the message middleware, three message channels are established for device data, model update, and state synchronization. Through the asynchronous processing and flow control functions of the message queue, the speed mismatch problem between data acquisition and model update is effectively solved. Each message channel is configured with an independent consumer group, and the number of consumers is dynamically adjusted according to data priority and processing capacity.

[0075] Specifically, the state synchronization adopts an incremental update strategy. The system maintains a state cache pool and only transmits the changed state data. For the static attributes of the device (such as device number, model, etc.), a periodic full synchronization method is adopted; for the dynamic state (such as operating parameters, alarm information, etc.), real-time incremental synchronization is adopted. This differentiated synchronization strategy significantly reduces the amount of data transmission and improves the system response speed.

[0076] S401: Realizing linkage through the digital twin platform includes: establishing a dynamic association between the optimized 3D model and the actual equipment operation status; generating a list of equipment that needs to be monitored based on equipment status data analysis; and integrating all update results to form a digital twin system with dynamic response capabilities.

[0077] S402: The digital twin system has the following response functions: warning of potential failures by analyzing equipment operation trends; scheduling maintenance resources based on equipment operating status and priority; and providing real-time model support for multi-department collaborative operations.

[0078] S403: Real-time processing of the acquired data includes: transmitting the collected data to the central data processing platform through the Beidou communication system; distributing and storing the data; and realizing efficient query and use of the data.

[0079] S404: The optimized three-dimensional model is used to: intuitively display the physical structure and spatial distribution of the equipment; display the equipment operation status in real time; and provide maintenance and operation decision support for operation and maintenance personnel.

[0080] In summary, by introducing a multi-dimensional data collection mechanism (including voltage, current, temperature, vibration, and spatial position information), combined with the Beidou high-precision positioning system and UWB indoor positioning technology, all-round and high-precision data collection of substation equipment is achieved. This multi-source heterogeneous data fusion solution not only solves the problem of single data collection and insufficient accuracy in traditional methods, but also provides more comprehensive data support for subsequent model optimization, enabling the model to more accurately reflect the actual status of the equipment.

[0081] By adopting a modular equipment attribute analysis framework and a multi-level feature engineering method, the physical characteristics and operating status of the equipment are converted into a structured feature matrix. This hierarchical analysis method not only improves the accuracy of feature extraction, but also enables the system to respond to changes in equipment status in a timely manner through an adaptive feature update mechanism, effectively solving the problem of incomplete feature extraction and untimely update in traditional methods.

[0082] By introducing the geometric optimization based on the QEM algorithm and the hierarchical simplification mechanism based on the viewpoint, the adaptive optimization of the model is realized. This optimization strategy not only significantly reduces the amount of model data, but also ensures the retention of key features in the process of model simplification through the setting of multiple constraints, solves the problem of large model volume and slow loading in traditional methods, and ensures the accuracy and availability of the model.

[0083] By designing an adaptive adjustment mechanism based on feature changes, combined with multi-threshold detection and intelligent change trend analysis, the real-time dynamic update of the model is achieved. This differentiated adjustment strategy not only ensures the real-time nature of model updates, but also avoids unnecessary resource consumption by optimizing the update frequency, effectively solving the problem that traditional static models cannot respond to dynamic business needs.

[0084] By adopting a microservice architecture and a data stream processing mechanism based on a message queue, efficient linkage between physical devices and digital models is achieved. This design not only improves the scalability and reliability of the system, but also significantly reduces the amount of data transmission through an incremental update strategy, solving the problems of low data synchronization efficiency and delayed response in traditional methods.

[0085] Example 2

[0086] Reference Figure 2 - Figure 4 , which is the second embodiment of the present invention, and this embodiment provides a dynamic construction algorithm of a substation three-dimensional business model based on GIM model optimization.

[0087] Step 1: Data collection and classification processing. In the dynamic construction of the substation 3D business model, data collection and classification processing are the basic links of the entire technical process. Accurate and efficient data collection and classification can provide reliable support for subsequent model optimization and adjustment. This step aims to build a complete data foundation for the dynamic adjustment of the 3D model through real-time data collection, normalization processing and multi-source data fusion.

[0088] (1) Data collection and sources. In order to build a dynamic 3D business model, data collection needs to cover multiple data sources of the substation, including equipment operating status, spatial location information, and environmental parameters. The collected data and description are shown in Table 1.

[0089] Table 1 Collection data categories and detailed description

[0090]

[0091] Let the collected data set be:

[0092] D = {d1, d2, ..., d n},

[0093] Among them, d i Represents the i-th category of data.

[0094] (2) Data normalization. Since the data collected come from different sources, with different units, dimensions, and value ranges, directly using the original data may lead to inconsistent model inputs and affect the accuracy of the analysis results. Therefore, it is necessary to normalize the data and map it to a unified standard range [0,1]. The normalization formula is:

[0095]

[0096] Among them, d i ′ is the normalized data value; d i is the original data value; a min , a max is the minimum and maximum value of the corresponding data type. The normalized data is transmitted to the central data processing platform through Beidou communication technology. The data is transmitted to the central data processing platform in real time through the Beidou communication system. The data transmission delay can be expressed as:

[0097]

[0098] Among them, L (unit: byte) is the size of a single data packet; R (unit: byte / second) is the communication bandwidth. The data transmitted to the central processing platform adopts a distributed storage architecture to ensure efficient query and use of data.

[0099] Step 2: Equipment attribute analysis and feature extraction. Equipment attribute analysis and feature extraction are key steps to achieve model optimization. The core goal is to extract the physical and operating characteristics related to the equipment from the normalized multi-source data, and provide effective support for the subsequent dynamic adjustment of the model. In the three-dimensional modeling of substations, the spatial position, physical structure and operating parameters of the equipment are important factors affecting the accuracy and practicality of the model. Therefore, this step not only requires the analysis of the basic attributes of the equipment, but also the characterization of these attributes to form a feature matrix that can be input into the model optimization process.

[0100] (1) Physical attribute analysis. Physical attribute analysis mainly involves quantification and modeling of the spatial position, geometric shape and state parameters of the device. The present invention uses a transformation matrix to describe the spatial position of the device, which facilitates the positioning and relationship construction of the device in three-dimensional space. Assuming that the original coordinates of the device in three-dimensional space are P = (x, y, z), its spatial posture can be represented by the following homogeneous transformation matrix T:

[0101]

[0102] Where R is a 3×3 rotation matrix representing the device orientation, which is defined as:

[0103]

[0104] θ is the rotation angle of the device; t is a 3×1 translation vector, which represents the spatial position offset of the device and is defined as:

[0105]

[0106] The transformed device coordinates p′ can be expressed as:

[0107]

[0108] Through this method, a unique posture description matrix in three-dimensional space can be generated for each device, which can be used for the accurate construction and optimization of subsequent models. Combined with the data collection results of step 1, the operating status of the device (such as current, voltage, temperature, vibration) is normalized and quantified to form the operating attribute matrix S = [UITV] of the device.

[0109] (2) Feature extraction: Feature extraction aims to convert the physical properties and operating data of the equipment into quantifiable feature vectors for subsequent model optimization. Figure 3 This is the general process of feature extraction.

[0110] First, the feature generation function g(d, P) is used to generate feature vectors for each device, and then the feature matrix is ​​constructed. The feature generation function g(d, P) represents the comprehensive relationship between device-specific attributes and location, and is defined as:

[0111] f i =g(d i , P) = w i ·d i +b

[0112] Among them, f i represents the i-th feature; d i is the input property (such as voltage, current, temperature, etc.); w iis the feature weight, reflecting the importance of the attribute to model optimization; b is the bias term, used to balance the deviation of different data. After feature extraction is completed, the feature vector F = {f1, f2…, f m}. Combine the feature vectors of all devices to form the device feature matrix F:

[0113]

[0114] Among them, f ij represents the jth feature of the i-th device; n is the number of devices, and m is the number of features. These features serve as inputs for subsequent model optimization and dynamic adjustment, directly affecting the real-time and accuracy of the model. In addition, the constructed feature matrix is ​​highly structured, easy to calculate and update, and can efficiently support the dynamic model construction algorithm.

[0115] Step 3: GIM model optimization and adaptive adjustment. The core task of this step is to optimize the structure of the GIM model in real time based on the equipment features and attribute data extracted in step 2 and combined with business needs to ensure that the model can dynamically respond to changes in operation and maintenance scenarios. Traditional static models are difficult to adapt to dynamic needs, and the present invention uses an adaptive optimization algorithm to achieve continuous adjustment and improvement of the GIM model, so that the three-dimensional model can be highly consistent with the equipment operating status and spatial changes.

[0116] (1) Construction and optimization of the initial model. In the GIM framework, the initial model consists of multiple components C = {C1, C2…, C k}, each component represents a specific device or structure in the substation. In order to improve the efficiency and availability of the initial model, it needs to be optimized. The optimization process aims to reduce redundancy and improve model accuracy. The specific optimization formula is:

[0117]

[0118] Where E represents the optimization error; C i is the i-th original component; C′ i is the optimized component of the i-th component. By comparing C i and C′ i The difference between i -C′ i ‖ 2 , find out redundant data or inefficient structures. Then use the QEM (Quadric Error Metrics) algorithm to optimize the triangles and material textures in the component. Figure 4 A schematic diagram of the QEM algorithm is shown.

[0119] In the 3D model of the substation, the geometric data of the equipment is usually stored in the form of triangular facets. Although high-precision 3D models can provide more intuitive display effects, too many triangular facets will significantly increase the storage and rendering burden of the model. Especially in real-time operation and maintenance scenarios, excessive geometric data may cause slow model loading and system response lag. Iterative simplification can reduce redundant data while maintaining the integrity and accuracy of the model's geometric structure.

[0120] (2) Dynamic adjustment strategy. In order to meet the real-time changing needs of substation operation and maintenance, the GIM model needs to dynamically adjust the structure and parameters of components according to real-time data. This paper proposes an adaptive adjustment strategy based on equipment feature changes. Let the feature matrix F = {f1, f2…, f m} represents the device feature set of the current model. When the feature of a device changes, its feature change △F is defined as:

[0121] ΔF=F new -F old

[0122] Among them, F new is the latest feature matrix collected; F old Represents the feature matrix of the previous moment. According to the feature change △F, the model adjustment △C can be calculated by the following formula:

[0123] △C=α·△F

[0124] Among them, △C represents the model adjustment amount; α is the adjustment coefficient, which is obtained through historical data training and reflects the sensitivity of the adjustment. For example, suppose the operating temperature of a transformer rises from 60℃ to 75℃, and its characteristic change is △T=15℃. The adjustment coefficient α=0.05, then the model adjustment amount is △C=0.05*15=0.75. According to the adjustment amount △C, the thermal properties or warning status of the transformer in the model can be dynamically changed to ensure that the model is consistent with the actual operating status.

[0125] (3) Real-time update rules. The dynamically adjusted model needs to be synchronized according to a certain update frequency to ensure the real-time performance of the model. In the present invention, the update frequency of the model is defined as fupdate, and the adjusted model state M′ can be expressed as:

[0126] M'=M+ΔC

[0127] Among them, M is the original model, △C is the adjustment amount calculated by the characteristic change amount, and the delay time of model update is calculated as

[0128] Model optimization and dynamic adjustment have significant synergistic effects. Optimization reduces model redundancy and improves operational efficiency, while dynamic adjustment enables the model to quickly respond to changes in equipment status and achieve efficient business adaptation. For example, during the cable laying process of a substation, by adjusting the spatial position information in the model, possible collision risks can be detected and corrected in a timely manner, providing real-time support for operation and maintenance personnel.

[0129] Step 4: Digital twin linkage and system response. The optimized three-dimensional business model needs to be synchronized with the actual equipment operation status and linked with substation operation and maintenance through the digital twin platform. This step aims to ensure the consistency of virtual models and physical equipment through real-time data docking and dynamic model updates, and provide support for intelligent operation and maintenance.

[0130] (1) Digital twin linkage mechanism. The core of digital twin linkage is to establish a dynamic association between real-time data streams and optimized three-dimensional models to reflect the operating status of the equipment in real time. Based on Beidou communication and distributed data processing architecture, this paper designs an efficient data interaction mechanism.

[0131] First, real-time data is collected from device sensors and transmitted to the twin platform through the Beidou communication system. The twin platform updates the properties of device components in the model based on data changes. Secondly, the twin platform analyzes the status data of the equipment and generates a list of equipment that needs to be monitored or repaired, helping operation and maintenance personnel to quickly locate problems. Finally, the platform integrates all update results into the three-dimensional model to form a digital twin system with dynamic response capabilities.

[0132] (2) System response and dynamic support. In addition to reflecting the status of equipment in real time, the digital twin system can also provide dynamic support in business processes. For common substation scenarios, the linkage effect is mainly reflected in the following aspects:

[0133] Fault prediction: By analyzing the operating trends of equipment, potential faults can be warned in advance to avoid system risks caused by sudden problems.

[0134] Resource optimization: Rationally dispatch maintenance resources according to the operating status and priority of the equipment to improve operation and maintenance efficiency.

[0135] Business collaboration: In complex business scenarios, it provides real-time updated model support to ensure the accuracy of multi-department collaborative work.

[0136] 3. Comparison of system implementation effects

[0137] In order to verify the superiority of the method of the present invention, a comparative experiment was designed to compare the method of the present invention with two existing methods. Taking a 500kV substation as the scenario, the experimental equipment includes multiple types of substation component models, combined with a real-time data acquisition system and a high-precision operation monitoring device for testing. The experiment is equipped with the following main experimental equipment:

[0138] (1) Real-time data acquisition system: used to collect operating status data of substation equipment, including key parameters such as voltage, current, temperature, and vibration.

[0139] (2) High-precision posture monitoring device: obtains the precise position and dynamic changes of the equipment in three-dimensional space through the Beidou system.

[0140] (3) Distributed data processing platform: used for real-time transmission, processing and storage of collected data, and supports dynamic modeling and linkage optimization.

[0141] The experiment was conducted in an actual substation operation and maintenance scenario, and three key performance indicators were compared. The indicators and their descriptions are shown in Table 2.

[0142] Table 2 Test indicators and their descriptions

[0143]

[0144] The experiment compares the method of the present invention with the existing evaluation methods in the literature, and the main indicators are shown in Table 2. For the experimental monitoring data of the same set of equipment, GIM lightweight optimization and deep learning and point cloud registration are used for testing. Table 3 is a comparison of the implementation effects of different methods.

[0145] Table 3 Comparison of indicators of three methods

[0146]

[0147] The experimental results show that the method of the present invention is superior to the other two methods in terms of model response time, accuracy and operation and maintenance efficiency. In terms of model response time, the average time consumed by the method of the present invention is 27 seconds, which is significantly shorter than the other two methods. This advantage is due to the dynamic optimization mechanism and real-time data analysis capabilities introduced by the present invention, which can quickly adapt to the complex business needs in the substation operation and maintenance scenario. In terms of model accuracy, the root mean square error of the method of the present invention is 1.8cm. The present invention retains the key features of the model through multi-source data fusion and geometric optimization algorithm, and significantly improves the restoration degree of the three-dimensional model to the actual equipment. At the same time, in terms of operation and maintenance efficiency improvement, the method of the present invention has achieved an efficiency improvement of 50%. This efficiency improvement is attributed to the technical breakthrough of the method of the present invention in the real-time performance of the model, which makes equipment maintenance and operation status adjustment faster and more accurate, and reduces manual intervention and operation time. In summary, the method of the present invention has shown significant technical advantages in many key indicators and can better meet the actual needs in the substation operation and maintenance scenario.

[0148] Example 3

[0149] This embodiment also provides a computer device, which is suitable for a dynamic construction algorithm of a substation three-dimensional business model based on GIM model optimization, including a memory and a processor; the memory is used to store computer executable instructions, and the processor is used to execute computer executable instructions to implement a forced oscillation detection and positioning method for a distribution network as proposed in the above embodiment.

[0150] This embodiment further provides a storage medium on which a computer program is stored. When the program is executed by a processor, a forced oscillation detection and positioning method for a distribution network is implemented as proposed in the above embodiment.

[0151] The computer device may be a terminal, and the computer device includes a processor, a memory, a communication interface, a display screen and an input device connected via a system bus. The processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The communication interface of the computer device is used to communicate with an external terminal in a wired or wireless manner, and the wireless manner can be achieved through WIFI, an operator network, NFC (near field communication) or other technologies. The display screen of the computer device may be a liquid crystal display screen or an electronic ink display screen, and the input device of the computer device may be a touch layer covering the display screen, or a key, trackball or touchpad provided on the housing of the computer device, or an external keyboard, touchpad or mouse, etc.

[0152] If the function is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, or the part of the technical solution, can be embodied in the form of a software product. The computer software product is stored in a storage medium, including several instructions for a computer device (which can be a personal computer, server, or network device, etc.) to perform all or part of the steps of the methods of each embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk, etc. Various media that can store program codes.

[0153] The logic and / or steps represented in the flowchart or otherwise described herein, for example, can be considered as an ordered list of executable instructions for implementing logical functions, and can be embodied in any computer-readable medium for use by an instruction execution system, device or apparatus (such as a computer-based system, a system including a processor, or other system that can fetch instructions from an instruction execution system, device or apparatus and execute instructions), or in conjunction with such instruction execution systems, devices or apparatuses. For the purposes of this specification, "computer-readable medium" can be any device that can contain, store, communicate, propagate or transmit a program for use by an instruction execution system, device or apparatus, or in conjunction with such instruction execution systems, devices or apparatuses.

[0154] More specific examples of computer-readable media (a non-exhaustive list) include the following: an electrical connection with one or more wires (electronic device), a portable computer disk case (magnetic device), a random access memory (RAM), a read-only memory (ROM), an erasable and programmable read-only memory (EPROM or flash memory), an optical fiber device, and a portable compact disk read-only memory (CDROM). In addition, the computer-readable medium may even be a paper or other suitable medium on which the program is printed, since the program may be obtained electronically, for example, by optically scanning the paper or other medium, followed by editing, deciphering, or processing in another suitable manner as necessary, and then stored in a computer memory.

[0155] It should be understood that the various parts of the present invention can be implemented by hardware, software, firmware or a combination thereof. In the above embodiments, a plurality of steps or methods can be implemented by software or firmware stored in a memory and executed by a suitable instruction execution system. For example, if implemented by hardware, as in another embodiment, any one of the following technologies known in the art or a combination thereof can be used to implement: a discrete logic circuit having a logic gate circuit for implementing a logic function for a data signal, a dedicated integrated circuit having a suitable combination of logic gate circuits, a programmable gate array (PGA), a field programmable gate array (FPGA), etc.

[0156] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present invention, which should all be included in the scope of the claims of the present invention.

Claims

1. A dynamic construction algorithm of a substation three-dimensional business model based on GIM model optimization, characterized by: This includes obtaining the voltage, current, temperature, vibration operating parameter data of the equipment in the substation, as well as the spatial location information and engineering parameter data of the equipment, and processing the obtained data in real time; Perform device attribute analysis and feature extraction on the processed data to build a device feature matrix; Dynamically optimizing and adjusting the GIM model based on the equipment feature matrix; The digital twin platform enables real-time linkage between the optimized three-dimensional model and the actual equipment operating status.

2. The algorithm for dynamically constructing a three-dimensional business model of a substation based on GIM model optimization according to claim 1, characterized in that: The spatial position information of the equipment is obtained through the Beidou high-precision positioning system, and the engineering parameters include the equipment load capacity and the estimated value of remaining life.

3. The algorithm for dynamically constructing a three-dimensional business model of a substation based on GIM model optimization according to claim 2 is characterized in that: The device attribute analysis includes: quantitatively describing the device spatial posture; normalizing and quantifying the device operation status; and generating a device operation attribute matrix.

4. The algorithm for dynamically constructing a three-dimensional business model of a substation based on GIM model optimization according to claim 3 is characterized in that: The feature extraction includes: generating feature vectors for physical properties and operation data of equipment; combining feature vectors of all equipment to form a device feature matrix; and performing weight calculation on the feature matrix.

5. The algorithm for dynamically constructing a three-dimensional business model of a substation based on GIM model optimization according to claim 4, characterized in that: The dynamic optimization of the GIM model includes: performing redundant data analysis on components in the initial model; optimizing the triangular facets and material textures of the components using the QEM algorithm; and calculating the model adjustment amount based on the device feature change amount.

6. The algorithm for dynamically constructing a three-dimensional business model of a substation based on GIM model optimization according to claim 5, characterized in that: The method of calculating the model adjustment amount based on the equipment feature change amount includes: real-time monitoring of changes in the equipment feature matrix; calculating corresponding model adjustment parameters according to the feature change amount; and real-time updating of the model state according to a preset update frequency.

7. The algorithm for dynamically constructing a three-dimensional business model of a substation based on GIM model optimization according to claim 6, characterized in that: The linkage achieved through the digital twin platform includes: establishing a dynamic association between the optimized three-dimensional model and the actual equipment operation status; generating a list of equipment that needs to be monitored based on equipment status data analysis; and integrating all update results to form a digital twin system with dynamic response capabilities.

8. The algorithm for dynamically constructing a three-dimensional business model of a substation based on GIM model optimization according to claim 7, characterized in that: The digital twin system has the following response functions: early warning of potential failures by analyzing equipment operation trends; scheduling maintenance resources according to equipment operation status and priority; Provide real-time model support for multi-department collaboration.

9. The algorithm for dynamically constructing a three-dimensional business model of a substation based on GIM model optimization according to claim 8, characterized in that: The real-time processing of the acquired data includes: transmitting the collected data to a central data processing platform through a Beidou communication system; performing distributed storage on the data; and realizing efficient query and use of the data.

10. The algorithm for dynamically constructing a three-dimensional business model of a substation based on GIM model optimization according to claim 9, characterized in that: The optimized three-dimensional model is used to: intuitively display the physical structure and spatial distribution of the equipment; display the equipment operating status in real time; and provide maintenance and operation decision support for operation and maintenance personnel.

Images