A substation operation and maintenance system and method based on digital twinning

Abstract

This application discloses a substation operation and maintenance system and method based on digital twins, belonging to the field of operation and maintenance management technology. It includes: constructing a substation digital twin based on substation operating status data to generate a substation digital twin model; performing coupled simulation analysis on the substation digital twin model to confirm the substation coupled data; performing anomaly detection and comprehensive analysis of substation equipment operating status based on the substation coupled data to obtain comprehensive indicator data of equipment operating anomalies and equipment fault trend prediction data; evaluating the operation and maintenance priorities of the substation digital twin model to generate a substation equipment operation and maintenance priority sequence; generating substation operation and maintenance strategies based on the substation equipment operation and maintenance priority sequence to obtain a substation operation and maintenance strategy set; and distributing the substation operation and maintenance strategy set to the control terminal to execute intelligent substation operation and maintenance operations. This application has the effect of improving substation operation and maintenance efficiency.

Description

Technical Field

[0001] This application relates to the field of operation and maintenance management technology, and in particular to a substation operation and maintenance system and method based on digital twins. Background Technology

[0002] As a critical hub in the power grid, the safe and stable operation of substations is of paramount importance. With the development of sensor and communication technologies, digital operation and maintenance methods employing data acquisition and monitoring control systems and online monitoring devices have emerged. These methods enable remote acquisition and monitoring of certain status parameters of key equipment such as transformers and circuit breakers, improving the timeliness of operation and maintenance to some extent. However, such systems are typically limited to two-dimensional visualization and threshold alarms. Data from each subsystem is isolated, lacking deep integration and intelligent analysis capabilities. This makes it difficult to construct a high-fidelity virtual model that comprehensively maps the physical state of entities, accurately traces the fault evolution process, and supports forward-looking decision-making, thus limiting the evolution of operation and maintenance management towards intelligence, precision, and predictability. Digital twin technology, as a core means of achieving deep integration of cyber-physical systems, provides a novel solution for building a new generation of intelligent substation operation and maintenance systems.

[0003] Among related technologies, existing substation operation and maintenance systems have significant shortcomings:

[0004] First, the substation operation and maintenance monitoring data are mostly scattered and isolated, lacking in-depth integration and collaborative analysis of multi-source heterogeneous data, making it impossible to form a unified and panoramic understanding of the overall operation status of the substation, resulting in data silos.

[0005] Second, substation operation and maintenance mainly rely on real-time monitoring and historical queries, depending on simple over-threshold alarm mechanisms. They lack in-depth analysis and coupled simulation capabilities of equipment operation trends, making it difficult to achieve early warning of potential faults, resulting in insufficient early warning capabilities and a high false alarm rate.

[0006] Third, existing visualization methods are mostly two-dimensional panels or simple three-dimensional models, which fail to build a digital twin model that is linked and mapped to physical entities in real time. This results in a lack of intuitive and accurate digital model support for operation and maintenance decisions, and the decision-making process still relies on human experience, with limited intelligence.

[0007] Fourth, the substation operation and maintenance system failed to dynamically link and optimize the early warning information with operation and maintenance resources and maintenance strategies, thus failing to generate scientific and efficient adaptive operation and maintenance strategies, thereby reducing the efficiency of substation operation and maintenance management, and there are areas for improvement. Summary of the Invention

[0008] To address the shortcomings of existing technologies, this application provides a substation operation and maintenance system and method based on digital twins.

[0009] Firstly, this application provides a substation operation and maintenance method based on digital twins, comprising the following steps:

[0010] Step S1: Real-time data acquisition is performed through a multi-source sensor network deployed in the substation to obtain substation operation status data, and a digital twin of the substation is constructed based on the substation operation status data to generate a digital twin model of the substation.

[0011] Step S2: Perform coupled simulation analysis on the digital twin model of the substation, confirm the coupled data of the substation based on the results of the coupled simulation analysis, perform abnormal detection and comprehensive analysis of the substation equipment operation status based on the coupled data of the substation, obtain the comprehensive index data of equipment operation abnormality, and perform fault trend prediction on the comprehensive index data of equipment operation abnormality to confirm the equipment fault trend prediction data.

[0012] Step S3: Based on the comprehensive index data of equipment operation anomalies and the prediction data of equipment failure trends, perform operation and maintenance priority assessment on the digital twin model of the substation, and then generate a substation equipment operation and maintenance priority sequence based on the results of the operation and maintenance priority assessment.

[0013] Step S4: Generate substation operation and maintenance strategies based on the substation equipment operation and maintenance priority sequence, obtain a substation operation and maintenance strategy set, and send the substation operation and maintenance strategy set to the control terminal to execute the substation intelligent operation and maintenance operation.

[0014] Preferably, step S1 includes the following steps:

[0015] Step S11: Collect electrical parameter data, mechanical parameter data, and environmental parameter data through a multi-source sensor network deployed in the substation, and then integrate the substation operating status data based on the electrical parameter data, mechanical parameter data, and environmental parameter data;

[0016] Step S12: Perform spatiotemporal alignment processing on the substation operating status data, and then confirm the substation spatiotemporal synchronization data based on the result of the spatiotemporal alignment processing.

[0017] Step S13: Based on the spatiotemporal synchronization data of the substation, reconstruct the three-dimensional point cloud of the substation equipment, and then generate a point cloud model of the substation equipment based on the reconstruction result of the three-dimensional point cloud.

[0018] Step S14: Perform parameter mapping on the point cloud model of the substation equipment to construct a digital twin model of the substation.

[0019] Preferably, step S2 includes the following steps:

[0020] Step S21: Perform multiphysics coupling simulation analysis on the digital twin model of the substation to obtain the substation coupling data;

[0021] Step S22: Based on the substation coupling data, perform baseline modeling of the substation equipment operating status, and then generate a baseline curve for the normal operating status of the equipment;

[0022] Step S23: Anomaly detection is performed by analyzing the deviation between the substation operation status data and the equipment normal operation status benchmark curve. Based on the anomaly detection results, equipment operation anomaly index data is generated, and the equipment operation anomaly index data is comprehensively analyzed to confirm the comprehensive equipment operation anomaly index data.

[0023] Step S24: Perform time series prediction analysis on the comprehensive index data of equipment operation anomalies, and then confirm the equipment failure trend prediction data.

[0024] Preferably, step S23 includes the following steps:

[0025] Step S231: Extract features from the abnormal equipment operation index data, and obtain the equipment operation feature vector based on the feature extraction results;

[0026] Step S232: Based on the device operation feature vector, perform abnormal pattern identification, and then confirm the fault mode based on the result of abnormal pattern identification;

[0027] Step S233: Assess the severity of the fault mode to generate fault mode severity level data;

[0028] Step S234: Based on the severity level data of the fault mode, perform abnormal index fusion calculation to confirm the comprehensive index data of equipment operation abnormality.

[0029] Preferably, step S24 includes the following steps:

[0030] Step S241: Calculate the fault development rate based on the comprehensive index data of the equipment operation anomaly, and then obtain the fault development rate data based on the calculation results;

[0031] Step S242: Based on the fault development rate data, predict the remaining service life to generate equipment remaining service life prediction data;

[0032] Step S243: Perform confidence analysis on the equipment failure trend prediction data, and then confirm the failure prediction confidence index based on the results of the confidence analysis;

[0033] Step S244: Based on the remaining service life prediction data of the equipment and the failure prediction confidence index, a comprehensive evaluation of the failure trend is performed to obtain the equipment failure trend prediction data.

[0034] Preferably, step S3 includes the following steps:

[0035] Step S31: Normalize the comprehensive index data of equipment operation anomalies to obtain normalized data of equipment operation anomalies;

[0036] Step S32: Perform weighted processing on the equipment failure trend prediction data to obtain weighted equipment failure trend data;

[0037] Step S33: Calculate the equipment risk index based on the normalized data of equipment operation anomalies and the weighted data of equipment failure trends, and generate equipment risk index data;

[0038] Step S34: Based on the equipment risk index data, sort the substation equipment in the substation digital twin model by maintenance priority, and generate a substation equipment maintenance priority sequence.

[0039] Preferably, step S34 includes the following steps:

[0040] Step S341: Perform equipment correlation analysis based on the equipment risk index data, and generate an equipment correlation network based on the results of the equipment correlation analysis;

[0041] Step S342: Perform fault propagation path analysis on the device association network to obtain potential fault propagation path data;

[0042] Step S343: Conduct a risk assessment based on the potential fault propagation path data to obtain a risk impact index;

[0043] Step S344: Based on the equipment risk index data and the risk impact index, a comprehensive priority assessment is performed to generate a substation equipment operation and maintenance priority sequence.

[0044] Preferably, step S4 includes the following steps:

[0045] Step S41: Optimize the allocation of operation and maintenance resources based on the substation equipment operation and maintenance priority sequence, and generate an operation and maintenance resource allocation scheme;

[0046] Step S42: Based on the operation and maintenance resource allocation scheme, perform operation and maintenance task scheduling and planning, and generate a detailed operation and maintenance task plan;

[0047] Step S43: Generate operation and maintenance instructions based on the detailed operation and maintenance task plan to obtain a substation operation and maintenance strategy set;

[0048] Step S44: The substation operation and maintenance strategy set is sent to the control terminal through a secure communication protocol, and the intelligent operation and maintenance operation of the substation is executed.

[0049] Secondly, this application provides a substation operation and maintenance system based on digital twins, including:

[0050] The model building module is used to collect real-time data through a multi-source sensor network deployed in the substation, obtain substation operation status data, and build a digital twin of the substation based on the substation operation status data, thereby generating a digital twin model of the substation.

[0051] The prediction module is used to perform coupled simulation analysis on the digital twin model of the substation, confirm the coupled data of the substation based on the results of the coupled simulation analysis, perform abnormal detection and comprehensive analysis of the substation equipment operation status based on the coupled data of the substation, obtain comprehensive index data of equipment operation abnormality, and perform fault trend prediction on the comprehensive index data of equipment operation abnormality to confirm the equipment fault trend prediction data.

[0052] The priority generation module is used to evaluate the operation and maintenance priority of the substation digital twin model based on the comprehensive index data of equipment operation anomalies and the prediction data of equipment failure trends, and then generate a substation equipment operation and maintenance priority sequence based on the results of the operation and maintenance priority evaluation.

[0053] The operation and maintenance control module is used to generate substation operation and maintenance strategies based on the substation equipment operation and maintenance priority sequence, obtain a substation operation and maintenance strategy set, and send the substation operation and maintenance strategy set to the control terminal to execute intelligent operation and maintenance operations of the substation.

[0054] Thirdly, this application provides a computer-readable storage medium storing instructions that, when executed on a computer, cause the computer to perform any of the above-described digital twin-based substation operation and maintenance methods.

[0055] In summary, this application includes the following beneficial technical effects:

[0056] This application provides a substation operation and maintenance method based on digital twins. By performing coupled simulation analysis on the substation digital twin model, coupled substation data is identified. Based on this coupled data, abnormal operation status detection and comprehensive analysis of substation equipment are performed to obtain comprehensive indicators of equipment operation anomalies. Fault trend prediction is then performed on these comprehensive indicators to identify predicted equipment fault trends. Operation and maintenance priorities are evaluated on the substation digital twin model to generate a substation equipment operation and maintenance priority sequence. Based on this priority sequence, substation operation and maintenance strategies are generated, resulting in a substation operation and maintenance strategy set. This set is then distributed to the control terminal to execute intelligent substation operation and maintenance operations. This effectively reduces the occurrence of insufficient early warning capabilities and high false alarm rates due to a lack of in-depth analysis and coupled simulation capabilities of equipment operation trends. Furthermore, the method dynamically correlates and optimizes early warning information with operation and maintenance resources and maintenance strategies, thereby effectively improving the efficiency of substation operation and maintenance management. Attached Figure Description

[0057] To more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0058] Figure 1 This is a flowchart of a substation operation and maintenance method based on digital twins, according to an embodiment of this application.

[0059] Figure 2 This is a schematic diagram of a substation operation and maintenance system based on digital twins, according to an embodiment of this application. Detailed Implementation

[0060] The following is in conjunction with the appendix Figure 1-2 This application will be described in further detail.

[0061] Example 1:

[0062] This application discloses a substation operation and maintenance method based on digital twins.

[0063] Reference Figure 1 A substation operation and maintenance method based on digital twins includes the following steps:

[0064] Step S1: Real-time data acquisition is performed through a multi-source sensor network deployed in the substation to obtain substation operation status data, and a digital twin of the substation is constructed based on the substation operation status data to generate a digital twin model of the substation.

[0065] Step S2: Perform coupled simulation analysis on the digital twin model of the substation, confirm the coupled data of the substation based on the results of the coupled simulation analysis, perform abnormal detection and comprehensive analysis of the substation equipment operation status based on the coupled data of the substation, obtain the comprehensive index data of equipment operation abnormality, and perform fault trend prediction on the comprehensive index data of equipment operation abnormality to confirm the equipment fault trend prediction data.

[0066] Step S3: Based on the comprehensive index data of equipment operation anomalies and the prediction data of equipment failure trends, perform operation and maintenance priority assessment on the digital twin model of the substation, and then generate a substation equipment operation and maintenance priority sequence based on the results of the operation and maintenance priority assessment.

[0067] Step S4: Generate substation operation and maintenance strategies based on the substation equipment operation and maintenance priority sequence, obtain a substation operation and maintenance strategy set, and send the substation operation and maintenance strategy set to the control terminal to execute the substation intelligent operation and maintenance operation.

[0068] Specifically, in this embodiment, a multi-source sensor network is deployed in the key equipment and surrounding environment inside the substation. The key equipment includes, but is not limited to, main transformers, circuit breakers, and disconnectors. The multi-source sensor network encompasses electrical quantity sensors, mechanical quantity sensors, and environmental sensors. Real-time data collection and integration generate substation operating status data. Based on this data, digital twin modeling technology, such as BIM, Unity, or specialized industrial modeling software, is used to construct a 1:1 digital twin of the physical substation. This ultimately generates a digital twin model of the substation that reflects its real-time operating status. The digital twin modeling technology can be set to BIM, Unity, or specialized industrial modeling software. Next, multiphysics simulation tools are used to perform coupled simulation analysis on the digital twin model, extracting substation coupling data from the simulation analysis results. Based on this substation coupling data, the real-time collected substation operating status data is compared with normal status data. The system compares and contrasts equipment operation status anomalies, comprehensively analyzes abnormal information such as voltage fluctuations, excessive temperature, and abnormal vibration, and calculates comprehensive equipment operation anomaly index data. Then, using time series analysis and machine learning algorithms, it predicts fault trends based on this comprehensive equipment operation anomaly index data, determining equipment fault trend prediction data. Next, combining the comprehensive equipment operation anomaly index data and the equipment fault trend prediction data, it quantifies the current and potential risks of the equipment, prioritizes all equipment in the substation's digital twin model, and generates a substation equipment operation and maintenance priority sequence. Finally, considering the substation's operation and maintenance resources, it formulates a set of substation operation and maintenance strategies, including operation and maintenance tasks, execution times, operation procedures, and resource allocation. This set of strategies is then distributed to the mobile control terminals of operation and maintenance personnel or the substation's automatic control terminals via an encrypted and secure communication protocol, guiding personnel to conduct targeted operation and maintenance operations or triggering automatic equipment operation and maintenance actions.

[0069] The above technical solutions achieve the following: First, they enable a shift in substation operation and maintenance from passive response to proactive prediction. Through real-time mapping and coupled simulation of the substation's digital twin model, potential faults that are difficult to detect through traditional inspections can be identified in advance, avoiding large-scale power outages caused by sudden failures. Second, the generation of operation and maintenance priority sequences makes resource allocation more efficient. Traditional operation and maintenance often leads to delays in the maintenance of high-risk equipment due to a lack of scientific prioritization. However, the above method, by quantifying risk prioritization, can prioritize manpower and spare parts to critical equipment that is currently severely abnormal or has a high risk of future failures, reducing resource waste. Third, the issuance and execution of substation operation and maintenance strategies improve operation and maintenance efficiency. Operation and maintenance personnel no longer need to rely on experience to judge operating procedures; they only need to execute standardized instructions from the substation strategy set, reducing human error. At the same time, the linkage of automatic control terminals can achieve rapid self-healing of some faults, shortening fault handling time and ensuring the long-term stable operation of the substation.

[0070] It should be noted that step S1 includes the following steps:

[0071] Step S11: Collect electrical parameter data, mechanical parameter data, and environmental parameter data through a multi-source sensor network deployed in the substation, and then integrate the substation operating status data based on the electrical parameter data, mechanical parameter data, and environmental parameter data;

[0072] Step S12: Perform spatiotemporal alignment processing on the substation operating status data, and then confirm the substation spatiotemporal synchronization data based on the result of the spatiotemporal alignment processing.

[0073] Step S13: Based on the spatiotemporal synchronization data of the substation, reconstruct the three-dimensional point cloud of the substation equipment, and then generate a point cloud model of the substation equipment based on the reconstruction result of the three-dimensional point cloud.

[0074] Step S14: Perform parameter mapping on the point cloud model of the substation equipment to construct a digital twin model of the substation.

[0075] Specifically, multi-source sensors are deployed inside the substation. Adaptive sensors are selected based on the characteristics of different equipment. For example, electrical quantity sensors are installed at the main transformer windings to collect data such as voltage, current, and power factor; mechanical quantity sensors are installed at the transformer core and casing to collect data such as vibration frequency, displacement, and noise levels; and environmental sensors are installed both inside and outside the substation to collect data such as temperature, humidity, atmospheric pressure, dust concentration, and harmful gas content. Through the real-time data transmission function of the multi-source sensor network, the above-mentioned scattered data are aggregated and preprocessed using data cleaning algorithms to form substation operating status data. Secondly, due to differences in the acquisition frequency and installation location of different sensors, the acquired data may have discrepancies in timestamps and spatial coordinates. Therefore, a spatiotemporal alignment algorithm is required. In the time dimension, the timestamp of the high-frequency acquired data is used as a reference, and interpolation processing is performed on the low-frequency data to achieve time synchronization. In the spatial dimension, the installation location of the sensors is considered. Coordinates are used to assign precise spatial attributes to each data point, ultimately confirming the substation's spatiotemporal synchronization data. Then, laser scanning equipment is used to perform a comprehensive scan of the substation's physical equipment and surrounding environment, acquiring 3D point cloud data of the substation equipment. The spatiotemporal synchronization data is correlated with the 3D point cloud data, and through point cloud denoising, registration, and fusion processing techniques, a 3D point cloud model of the substation equipment, consistent with the size and structure of the physical equipment, is reconstructed. This 3D point cloud model not only visually presents the equipment's appearance but also correlates with its real-time operating data. Finally, a parameter mapping relationship is established between the substation's spatiotemporal synchronization data and the 3D point cloud model, mapping electrical, mechanical, and environmental parameters to the corresponding equipment components in the 3D point cloud model. This allows the 3D point cloud model to reflect real-time changes in the physical equipment's operating parameters. Simultaneously, the physical characteristics of the equipment are embedded to participate in the operating rules, ultimately constructing a digital twin model of the substation.

[0076] By adopting the above technical solution, firstly, the substation operation status data collected by the multi-source sensor network avoids the one-sidedness of traditional data collection. Traditional operation and maintenance mainly focuses on electrical parameters, while the above steps simultaneously collect mechanical and environmental parameters, which can more comprehensively reflect the equipment status. Secondly, the spatiotemporal alignment processing solves the problems of data asynchrony and difficulty in correlation, ensuring that different types of data can be mutually verified in the same spatiotemporal dimension in subsequent simulation analysis, avoiding simulation errors caused by data deviation. Thirdly, 3D point cloud reconstruction and parameter mapping make the substation digital twin model more realistic and usable. The intuitive substation digital twin model makes it easy for operation and maintenance personnel to quickly locate equipment positions, and real-time parameter mapping makes the substation digital twin model a digital mirror of the physical equipment. Subsequent coupled simulation and anomaly detection can be carried out based on the substation digital twin model, which greatly improves the accuracy of subsequent operation and maintenance analysis and lays a solid foundation for the effective implementation of the entire operation and maintenance method.

[0077] It should be noted that step S2 includes the following steps:

[0078] Step S21: Perform multiphysics coupling simulation analysis on the digital twin model of the substation to obtain the substation coupling data;

[0079] Step S22: Based on the substation coupling data, perform baseline modeling of the substation equipment operating status, and then generate a baseline curve for the normal operating status of the equipment;

[0080] Step S23: Anomaly detection is performed by analyzing the deviation between the substation operation status data and the equipment normal operation status benchmark curve. Based on the anomaly detection results, equipment operation anomaly index data is generated, and the equipment operation anomaly index data is comprehensively analyzed to confirm the comprehensive equipment operation anomaly index data.

[0081] Step S24: Perform time series prediction analysis on the comprehensive index data of equipment operation anomalies, and then confirm the equipment failure trend prediction data.

[0082] Specifically, multiphysics simulation software is used to perform multiphysics coupling simulation on the digital twin model of the substation. For example, for the core physical processes of the substation equipment, such as the electric field, temperature field, and mechanical field of the main transformer, and the arc field and magnetic field of the circuit breaker, the interaction and influence between various physical fields under different operating conditions are simulated. Through simulation calculations, parameter change data of each component of the equipment under different operating conditions are obtained. This data collectively constitutes substation coupling data that reflects the interaction of multiple factors in the equipment. Next, operating parameter data of the equipment under rated operating conditions and normal environment are filtered from the substation coupling data. Statistical analysis algorithms are used to process the operating parameter data to generate a reference curve for the normal operating state of the equipment. This reference curve includes, but is not limited to, a reference curve for transformer winding temperature changing with load and a reference curve for circuit breaker opening time changing with ambient temperature. The reference curve for the normal operating state is used to determine whether the equipment is abnormal. Then, the real-time operating status data of the substation is compared with the normal operating status data of the equipment. The system compares each parameter of the state baseline curve to calculate the deviation between the real-time operating status data of the substation and the baseline curve. By setting a deviation threshold, when the deviation of a parameter exceeds the threshold, the parameter is determined to be abnormal, and equipment operation abnormality index data is generated. For example, transformer winding temperature abnormality: deviation of 5℃, circuit breaker tripping time abnormality: deviation of 0.2 seconds. The above single abnormality index data are then comprehensively analyzed. For example, it is determined whether there is a correlation between temperature abnormality and tripping time abnormality. Finally, the comprehensive equipment operation abnormality index data is confirmed. Finally, the historical comprehensive operation abnormality index data and the real-time comprehensive operation abnormality index data are collected. The time series prediction algorithm is used to analyze the above data to explore the changing trend of abnormal indicators and predict the comprehensive equipment abnormality index data in the future. This confirms the equipment failure trend prediction data that can reflect the development direction and degree of equipment failure. For example, in the next month, the transformer operation abnormality index will increase by 15%, which poses a risk of accelerated insulation aging.

[0083] The above-mentioned technical solutions significantly improve the accuracy of equipment anomaly detection and the foresight of fault prediction. Firstly, multi-physics coupled simulation overcomes the limitations of traditional single-physics analysis. Traditional maintenance often analyzes electrical or temperature parameters in isolation, while the above steps consider the interaction of multiple physics fields, enabling the discovery of potential problems that are difficult to detect with single-parameter analysis. Secondly, normal operation baseline modeling provides a scientific basis for anomaly detection, avoiding the subjectivity of traditional maintenance relying on experience to judge anomalies. The quantitative standard of the equipment's normal operation status benchmark curve makes anomaly judgment more objective and consistent. Thirdly, deviation analysis can accurately locate abnormal parameters, helping maintenance personnel quickly find the source of the problem. Fourthly, time series prediction upgrades maintenance from anomaly detection to fault prediction, allowing for advance knowledge of fault trends and providing maintenance personnel with sufficient time to develop response plans, avoiding losses caused by sudden fault occurrences and significantly improving the initiative of substation maintenance.

[0084] Furthermore, step S23 includes the following steps:

[0085] Step S231: Extract features from the abnormal equipment operation index data, and obtain the equipment operation feature vector based on the feature extraction results;

[0086] Step S232: Based on the device operation feature vector, perform abnormal pattern identification, and then confirm the fault mode based on the result of abnormal pattern identification;

[0087] Step S233: Assess the severity of the fault mode to generate fault mode severity level data;

[0088] Step S234: Based on the severity level data of the fault mode, perform abnormal index fusion calculation to confirm the comprehensive index data of equipment operation abnormality.

[0089] Specifically, for abnormal equipment operation data, such as abnormal temperature, vibration, and current fluctuation data of transformers, and abnormal opening time and arc-extinguishing chamber pressure data of circuit breakers, feature extraction algorithms are used to process each abnormal operation indicator. For example, features such as temperature rise rate, peak temperature, and temperature fluctuation frequency are extracted from abnormal temperature data; features such as vibration frequency, vibration amplitude, and vibration duration are extracted from abnormal vibration data. Through feature extraction of the above abnormal indicator data, the original abnormal data is transformed into key feature information that reflects the essence of the abnormality, and then integrated to form equipment operation feature maps. For example, the equipment operating characteristic vector is: temperature rise rate 2℃ / h, peak temperature 95℃, vibration frequency 50Hz, vibration amplitude 0.5mm, and current fluctuation amplitude 5%. Next, a database of common fault modes for substation equipment is constructed. This database contains characteristic vectors corresponding to various faults from historical fault cases. For example, the characteristic vector for a "short circuit fault" is: temperature rise rate 5℃ / h, current fluctuation amplitude 20%, and vibration frequency 100Hz; the characteristic vector for an insulation aging fault is: temperature rise rate 1℃ / h, increased partial discharge, and decreased insulation resistance. The equipment operating characteristic vector is then compared with... The system performs similarity matching on the feature vectors of common fault modes in the substation equipment fault mode database. When the matching degree of a fault mode exceeds a set threshold, the fault type corresponding to the current equipment anomaly is determined, thereby confirming the equipment's fault mode. Then, a severity assessment index system is established for each fault mode. For example, assessment standards are set from three dimensions: fault impact range, fault development speed, and fault consequence severity. For instance, if a transformer insulation aging fault affects only the transformer itself, develops slowly, and results in a shortened equipment lifespan, it is classified as a mild severity level; if the impact range extends to the line where the main transformer is located, develops rapidly, and so on... If the consequence is a potential short circuit and power outage, it is classified as a severe level. Fault mode severity level data is generated through assessment (e.g., represented by a score of 1-5, where 1 is mild and 5 is severe). Finally, the weights of different abnormal indicators are determined based on the fault mode severity level data. Higher severity levels correspond to higher weights for abnormal indicators with greater impact on equipment operation, while lower weights are assigned to abnormal indicators with less impact. A weighted summation algorithm is used to multiply the deviation values ​​of each abnormal indicator data by their corresponding weights and sum them to obtain comprehensive equipment operation anomaly index data. For example, a comprehensive index value of 8.5 out of 10 indicates a relatively severe current abnormal condition of the equipment.

[0090] The above technical solutions address the problems of vague qualitative analysis and one-sided quantitative analysis in traditional anomaly analysis, providing a precise basis for subsequent fault trend prediction and maintenance priority assessment: First, feature extraction refines key information from anomaly data, avoiding redundancy and interference in the original data, making subsequent pattern recognition more accurate; second, anomaly pattern recognition enables rapid fault characterization. In traditional maintenance, maintenance personnel need to conduct numerous tests to determine the fault type, while the above steps, through feature vector matching, can quickly locate fault patterns, significantly shortening fault diagnosis time; third, severity assessment distinguishes the urgency of faults, avoiding the misconception of treating all anomalies equally in traditional maintenance, allowing maintenance personnel to prioritize serious faults; fourth, the comprehensive index formed by the fusion calculation of anomaly indicators transforms multi-dimensional and fragmented anomaly information into a single quantitative value, facilitating rapid comparison of the anomaly severity of different devices and providing a clear and unified quantitative standard for maintenance priority ranking.

[0091] Furthermore, step S24 includes the following steps:

[0092] Step S241: Calculate the fault development rate based on the comprehensive index data of the equipment operation anomaly, and then obtain the fault development rate data based on the calculation results;

[0093] Step S242: Based on the fault development rate data, predict the remaining service life to generate equipment remaining service life prediction data;

[0094] Step S243: Perform confidence analysis on the equipment failure trend prediction data, and then confirm the failure prediction confidence index based on the results of the confidence analysis;

[0095] Step S244: Based on the remaining service life prediction data of the equipment and the failure prediction confidence index, a comprehensive evaluation of the failure trend is performed to obtain the equipment failure trend prediction data.

[0096] Specifically, firstly, based on the comprehensive indicator data of equipment operation anomalies, historical change data of the comprehensive indicator data of equipment operation anomalies over a certain period of time is collected. For example, the comprehensive indicator data of equipment operation anomalies in week 1 is 4.2, week 2 is 5.5, week 3 is 6.8, and week 4 is 8.1. A linear regression method is used to fit and analyze the historical change data to calculate the rate of change of the comprehensive indicator of anomalies. For example, based on the historical change data over the aforementioned period of time, the rate of change of the comprehensive indicator of anomalies is confirmed to be 1.3. This rate of change of the comprehensive indicator of anomalies is then used as the fault development rate data. Simultaneously, the rate data is corrected in conjunction with fault modes. Fault modes include, but are not limited to, insulation aging faults and short-circuit faults. When the fault mode is insulation aging fault, the fault development rate data is relatively stable. When the fault mode is short-circuit fault, the fault development rate data may show abrupt changes, ensuring that the rate calculation accurately reflects the actual development of the fault. For example, if the comprehensive indicator data of operation anomalies in a certain week is detected to rise from 6.8 to... At 10:00, it is determined to be an abnormal fluctuation. The corrected fault development rate data is still taken as 1.3. Then, based on the rated service life of the equipment and the current years of use, combined with the fault development rate data, and referring to the remaining service life data of similar fault modes in the equipment's historical fault cases, the prediction results are calibrated, and finally the equipment's remaining service life prediction data is generated. For example, the critical value corresponding to the comprehensive index data of abnormal equipment operation is identified first. The critical value is set as the comprehensive index of abnormal equipment operation when the equipment cannot operate normally. In this embodiment, the critical value is set to 10. Based on the current comprehensive index data of abnormal equipment operation of 8.1 and the fault development rate data of 1.3, the time required from the current comprehensive index data of abnormal equipment operation to the critical value is calculated, which is (10.0-8.1) / 1.3≈1.46 weeks, about 10 days. At the same time, the remaining service life data of similar fault modes in the equipment's historical fault cases is referred. For example, the comprehensive index data of abnormal equipment operation corresponding to insulation aging fault in the equipment's historical fault cases starts from 8.The average time required to go from 1 to 10 is 12 days. Therefore, the time required to reach the critical value from the current equipment operation anomaly comprehensive index data (10 days) is calculated and averaged with the remaining life data under similar failure modes in historical equipment failure cases (12 days). Based on the average calculation result, the prediction results are calibrated, ultimately generating the equipment's remaining life prediction data (11 days). Next, the confidence level of the failure trend prediction process is evaluated from three dimensions: data quality, model accuracy, and operating condition stability. Data quality dimension: Analyzing the completeness and accuracy of historical equipment operation anomaly comprehensive index data; the more complete and accurate the data, the higher the confidence level. Model accuracy dimension: Using cross-validation to test the error of the equipment's remaining life prediction data. Operating condition stability dimension: Analyzing whether the current substation operating conditions are consistent with the operating conditions at the time of modeling; the more stable the operating conditions, the more reliable the prediction results, and the higher the confidence level. Through quantitative scoring of the above three dimensions, such as data quality 80 points, model accuracy 85 points, and operating condition stability 90 points, a weighted average method is used to calculate the overall confidence level. The confidence score is calculated as 80 × 0.3 + 85 × 0.4 + 90 × 0.3 = 85 points. This score represents the fault prediction confidence index, expressed as a percentage (85%). Finally, the remaining service life prediction data (e.g., 11 days) is comprehensively considered with the fault prediction confidence index. If the remaining service life is short (less than 15 days) and the confidence level is high (above 80%), the fault trend is judged as short-term high risk, requiring emergency maintenance. If the remaining service life is long (more than 3 months) but the confidence level is low (below 60%), the fault trend is judged as long-term uncertain risk, requiring enhanced monitoring. Simultaneously, the severity of the fault mode is considered for a final assessment of the fault trend. For example, if the equipment fault mode is a short-circuit fault, even with a remaining service life of 1 month, the fault risk level is higher than that of an insulation aging fault. This results in equipment fault trend prediction data that includes the remaining service life prediction data, the fault prediction confidence index, and the fault risk level. For example, transformer fault trend: remaining service life approximately 11 days, confidence level 85%, fault risk level: urgent.

[0097] The above technical solutions upgrade fault trend prediction from rough estimation to "precise and reliable," providing reliable support for operation and maintenance decisions: First, the fault development rate calculation quantifies the pace of fault development. Traditional operation and maintenance can only vaguely judge the development of a fault, while the above steps allow operation and maintenance personnel to clearly understand the urgency of the fault. Second, the remaining service life prediction provides a clear time node for operation and maintenance preparation. Operation and maintenance personnel can purchase spare parts and arrange operation and maintenance personnel in advance based on the predicted 11 days, avoiding failure to handle faults in a timely manner due to insufficient preparation. Third, confidence analysis avoids the risk of blindly trusting the prediction results. If the confidence level is low, operation and maintenance personnel will choose to strengthen monitoring rather than immediately shut down for maintenance, reducing unnecessary power outage losses. Fourth, the comprehensive fault trend data formed by the comprehensive evaluation makes operation and maintenance decisions more comprehensive, significantly reducing the error rate of operation and maintenance decisions and ensuring that substation equipment is handled in a timely and appropriate manner before a fault occurs.

[0098] It should be noted that step S3 includes the following steps:

[0099] Step S31: Normalize the comprehensive index data of equipment operation anomalies to obtain normalized data of equipment operation anomalies;

[0100] Step S32: Perform weighted processing on the equipment failure trend prediction data to obtain weighted equipment failure trend data;

[0101] Step S33: Calculate the equipment risk index based on the normalized data of equipment operation anomalies and the weighted data of equipment failure trends, and generate equipment risk index data;

[0102] Step S34: Based on the equipment risk index data, sort the substation equipment in the substation digital twin model by maintenance priority, and generate a substation equipment maintenance priority sequence.

[0103] Specifically, since the comprehensive indicators of operational anomalies for different equipment may have differences in units or numerical ranges—for example, the comprehensive indicator range for transformers is 0-10 points, while that for switchgear is 0-5 points—directly comparing a transformer's 8 points with a switchgear's 4 points would be misleading. Therefore, a normalization algorithm is needed to process the data. For min-max normalization, first determine the minimum value (min) and maximum value (max) of the comprehensive indicator for a certain type of equipment, and then normalize the data using the formula: Normalized Data = (Original Data - min) / (Max - Min), converting the original data into a value within the 0-1 range. For example, if the original value of the transformer's comprehensive indicator is 8 points, then min = 0, max=10, after normalization it is (8-0) / (10-0)=0.8; the original value of the switchgear abnormal comprehensive index is 4 points, min=0, max=5, after normalization it is (4-0) / (5-0)=0.8; through normalization, the difference in the dimensions and range of abnormal indicators of different equipment is eliminated, and normalized data of equipment operation abnormality is obtained. Then, according to the importance of different indicators in the fault trend prediction data, the weight is set. For example, the remaining service life has the highest weight (e.g., 0.5), the fault prediction confidence has the second highest weight (e.g., 0.3), and the fault risk level has the lowest weight (e.g., 0.2); each indicator is quantified; and then the fault trend of each equipment is calculated by weighted summation formula. Weighted data is used; weighting highlights key information in fault trend data, preventing secondary information from interfering with risk assessment. Then, a risk index is calculated by weighted summation of normalized abnormal equipment operation data and weighted fault trend data. For example, considering that current and future risks are equally important, both are weighted at 0.5. The formula is: Equipment Risk Index = Abnormalized Abnormal Data × 0.5 + Weighted Fault Trend Data × 0.5. If a transformer's abnormalized normalized data is 0.8 and its fault trend weighted data is 0.905, the equipment risk index = 0.8 × 0.5 + 0.905 × 0.5 = 0.8525. If a switchgear's abnormalized normalized data is 0.8 and its fault trend weighted data is 0.905, the equipment risk index = 0.8 × 0.5 + 0.905 × 0.5 = 0.8525. With a weighted average of 0.41, the equipment risk index is calculated as 0.8 × 0.5 + 0.41 × 0.5 = 0.605. This calculation yields an equipment risk index that comprehensively reflects the current and future risks of the equipment. Finally, the risk index data of all equipment in the substation digital twin model are sorted from highest to lowest: for example, transformers 0.8525, busbar circuit breakers 0.72, switchgear 0.605, and disconnectors 0.35. Simultaneously, the importance of the equipment is considered (e.g., the main transformer is a core piece of equipment in the substation; even if its risk index is similar to other equipment, its ranking can be appropriately increased). Fine-tuning is then performed to ultimately generate a substation equipment maintenance priority sequence that is comprehensively ranked by risk level and equipment importance.

[0104] The above technical solutions achieve a scientific and standardized approach to prioritizing maintenance, overcoming the drawbacks of traditional experience-based prioritization: First, normalization eliminates obstacles to data comparison. In traditional maintenance, abnormal indicators of different devices cannot be directly compared due to varying ranges; normalization allows for fair comparison. Second, weighted processing highlights key information in fault trends, giving higher scores to devices with shorter remaining lifespans and higher confidence levels, ensuring these high-urgency devices receive priority attention. Third, the risk index combines current anomalies with future faults, avoiding the bias of traditional prioritization that only considers current anomalies or future trends. Fourth, the priority sequence provides a clear basis for resource allocation, allowing maintenance personnel to directly allocate manpower and spare parts according to the sequence, significantly improving resource utilization efficiency, ensuring high-risk equipment receives priority maintenance, and minimizing substation operational risks.

[0105] Furthermore, step S34 includes the following steps:

[0106] Step S341: Perform equipment correlation analysis based on the equipment risk index data, and generate an equipment correlation network based on the results of the equipment correlation analysis;

[0107] Step S342: Perform fault propagation path analysis on the device association network to obtain potential fault propagation path data;

[0108] Step S343: Conduct a risk assessment based on the potential fault propagation path data to obtain a risk impact index;

[0109] Step S344: Based on the equipment risk index data and the risk impact index, a comprehensive priority assessment is performed to generate a substation equipment operation and maintenance priority sequence.

[0110] Specifically, firstly, based on the substation's electrical wiring diagram, equipment topology, and historical fault cases, a database of equipment relationships is constructed to clarify the direct relationships between equipment. For example, the main transformer and the bus circuit breaker are directly connected, and a circuit breaker fault will affect the transformer's power supply. Secondly, indirect relationships are also established (e.g., the bus circuit breaker and the line disconnect switch are indirectly connected via the busbar, and a disconnect switch fault may indirectly lead to a circuit breaker overload). Graph theory algorithms are then used to quantify these relationships. For instance, an undirected graph is constructed where nodes represent equipment, edges represent relationships, and edge weights represent relationship strength. The relationship strength is set according to the degree of influence between the equipment. For example, the relationship strength between the main transformer and the bus circuit breaker is 0.9, and the relationship between the bus circuit breaker and the line disconnect switch is... The correlation strength is 0.6. By quantifying the correlation relationships of all equipment, an equipment correlation network is generated. For example, main transformer-bus circuit breaker (correlation strength 0.9)-line disconnector (correlation strength 0.6), main transformer-cooling system (correlation strength 0.8). Secondly, based on equipment risk index data, high-risk equipment with higher risk indices (such as main transformer 0.8525, bus circuit breaker 0.72) are selected as potential fault sources. Fault propagation analysis algorithms (such as breadth-first search and depth-first search) are used to simulate the propagation process of faults from potential fault sources in the equipment correlation network. For example, taking the main transformer as the fault source, the equipment that its fault may propagate to through correlation relationships is analyzed: first propagating to directly related equipment... The fault propagated from the main transformer to the busbar circuit breaker (association strength 0.9), causing the circuit breaker to overload. The propagation then spread from the busbar circuit breaker to the indirectly associated disconnector (association strength 0.6), causing the disconnector to overheat. Simultaneously, the propagation probability (80% probability of the main transformer fault propagating to the busbar circuit breaker, and 90% probability of propagating to the cooling system) and propagation time (3 hours from the main transformer fault to the circuit breaker overload, and 5 hours to the disconnector overheating) were analyzed. Through simulation analysis, the possible propagation paths, involved equipment, propagation probabilities, and propagation times were determined, obtaining potential fault propagation path data. For example, main transformer fault → busbar circuit breaker (80% propagation probability, 3 hours propagation time) → disconnector (60% propagation probability, 5 hours propagation time). (5-hour interval); Main transformer fault → Cooling system (propagation probability 90%, propagation time 2 hours). Then, for each potential fault propagation path, a risk assessment is conducted from three dimensions: propagation range (e.g., whether the path involves 3 or 5 devices), propagation speed (e.g., 2 hours or 10 hours), and propagation consequences (e.g., whether the propagation causes a power outage on the entire line or only a single device to stop). Assessment indicators and weights are set (propagation range weight 0.4, propagation speed weight 0.3, propagation consequences weight 0.3). The weights of propagation range, propagation speed, and propagation consequences can be obtained by fitting historical data. Each dimension is quantitatively scored (e.g., propagation range involving 5 devices is quantified as 1.0, and 3 devices is quantified as 0).6; the propagation speed is quantified as 1.0 for 2 hours and 0.2 for 10 hours; the propagation consequence of power outage is quantified as 1.0, and the downtime of a single device is quantified as 0.3. The risk impact index for each path is calculated using the weighted summation formula: Risk Impact Index = Propagation Range Quantification × 0.4 + Propagation Speed ​​Quantification × 0.3 + Propagation Consequence Quantification × 0.3 (e.g., if a path has a propagation range quantification of 0.8, a propagation speed quantification of 1.0, and a propagation consequence quantification of 1.0, the risk impact index = 0.8 × 0.4 + 1.0 × 0.3 + 1.0 × 0.3 = 0.32 + 0.3 + 0.3 = 0.92). Simultaneously, the risk impact indices of all devices involved in the path are summarized to obtain the risk impact index of each device due to fault propagation (e.g., the risk impact index of a busbar circuit breaker due to a main transformer fault propagation is 0.92, the device risk index is 0.72, and the total risk impact index is 0.72 + 0.92 = 1.64). Finally, the device's own risk index data is compared with the device risk impact index. The system combines numerical data, with the weighting coefficients corresponding to the equipment's own risk index and risk impact index data obtained by fitting historical data. The comprehensive priority score is calculated using the formula: Comprehensive Priority Score = Equipment Risk Index × 0.6 + Risk Impact Index × 0.4. For example, if the main transformer's risk index is 0.8525 and its risk impact index is 1.2, the comprehensive priority score is 0.8525 × 0.6 + 1.2 × 0.4 = 0.9915; if the busbar circuit breaker's risk index is 0.72 and its risk impact index is 1.64, the comprehensive priority score is 0.72 × 0.6 + 1.64 × 0.4 = 1.088. In this case, the comprehensive priority score of the busbar circuit breaker is higher than that of the main transformer, requiring a reassessment of its importance. Considering the main transformer is a core piece of equipment, it is still ranked first even with a slightly lower comprehensive score, but the busbar circuit breaker's ranking is significantly advanced. Finally, a more comprehensive substation equipment operation and maintenance priority sequence is generated based on the comprehensive priority score and equipment importance.

[0111] The above technical solutions overcome the isolation limitations of relying solely on risk index ranking, fully considering the systemic interrelationships of substation equipment and making maintenance priorities more aligned with actual operating scenarios: First, equipment correlation analysis breaks the limitations of independent assessment of individual equipment. Traditional ranking only considers the risk of the equipment itself, while the above steps focus on the mutual influence between equipment, avoiding situations where maintaining equipment A might overlook the impact of a failure on equipment B. Second, fault propagation path analysis identifies the risk of fault spread in advance. Maintenance personnel not only know the current high-risk equipment but also which equipment might be affected by its failure, providing direction for preventative maintenance. Third, propagation risk assessment quantifies the impact of fault spread, allowing maintenance personnel to clearly understand the severity of the chain reaction caused by a failure of a certain equipment, thus prioritizing equipment with high propagation risk in the priority ranking. Fourth, the comprehensive priority sequence better meets the maintenance needs of the substation system, ensuring that not only individual high-risk equipment is addressed but also critical node equipment that may trigger a chain reaction of failures, minimizing the probability of substation system failures and improving overall operational stability.

[0112] It should be noted that step S4 includes the following steps:

[0113] Step S41: Optimize the allocation of operation and maintenance resources based on the substation equipment operation and maintenance priority sequence, and generate an operation and maintenance resource allocation scheme;

[0114] Step S42: Based on the operation and maintenance resource allocation scheme, perform operation and maintenance task scheduling and planning, and generate a detailed operation and maintenance task plan;

[0115] Step S43: Generate operation and maintenance instructions based on the detailed operation and maintenance task plan to obtain a substation operation and maintenance strategy set;

[0116] Step S44: The substation operation and maintenance strategy set is sent to the control terminal through a secure communication protocol, and the intelligent operation and maintenance operation of the substation is executed.

[0117] Specifically, based on the substation equipment maintenance priority sequence (e.g., 1. main transformer; 2. bus circuit breaker; 3. cooling system) and the substation's current maintenance resource inventory (personnel: 5 engineers, including 2 senior and 3 junior engineers; spare parts: 1 main transformer spare winding, 2 circuit breakers, 3 sets of cooling system components; tools: 2 sets of high-voltage testing equipment, 1 set of disassembly and assembly tools), resources are allocated according to the principle of allocating high-priority equipment with high-level resources: 2 senior engineers, 1 spare winding, 1 set of high-voltage testing equipment, and 1 set of disassembly and assembly tools are allocated to the main transformer (Sequence 1); 1 senior engineer + 1 junior engineer, 1 spare circuit breaker, and 1 set of high-voltage testing equipment are allocated to the bus circuit breaker (Sequence 2); and 1 junior engineer, 1 set of cooling system components, and 1 set of general tools are allocated to the cooling system (Sequence 3). This generates a maintenance resource allocation scheme that includes the correspondence between equipment, personnel, spare parts, and tools. Secondly, considering the substation's operating conditions (daytime is peak electricity consumption, making shutdown for maintenance unsuitable; early morning off-peak electricity consumption is suitable for main equipment maintenance), the required maintenance time (main transformer maintenance requires 8 hours, bus circuit breaker maintenance requires 4 hours), and available resources (engineers' free time), a detailed maintenance task plan is generated. For example, main transformer maintenance is scheduled for early morning during engineers' free time; bus circuit breaker maintenance is scheduled for the following day during engineers' free time. Then, based on the detailed maintenance task plan and in conjunction with the equipment's technical specifications and safe operating procedures, standardized maintenance operation instructions are generated. All equipment maintenance operation instructions are integrated according to task sequence to form a substation maintenance strategy. The substation operation and maintenance strategy set is encrypted using a communication protocol that conforms to power industry security standards to prevent instructions from being tampered with or stolen during transmission. The encrypted strategy set is then distributed to the corresponding control terminals, mobile terminals of maintenance personnel, and automatic control terminals of the substation via the substation's industrial control network. Maintenance personnel receive the strategy set via an app and execute the operations step-by-step according to the instructions, while simultaneously uploading the operation progress in real time through their terminals. The automatic control terminals, upon receiving the strategy set, can automatically execute some standardized operations. During the operation and maintenance process, the control center can monitor the operation progress and equipment status in real time via the network, ensuring that the operation and maintenance are executed safely and according to plan, ultimately completing the intelligent operation and maintenance of the substation.

[0118] Example 2:

[0119] This application also discloses a substation operation and maintenance system based on digital twins.

[0120] Reference Figure 2 A substation operation and maintenance system based on digital twins includes:

[0121] The model building module is used to collect real-time data through a multi-source sensor network deployed in the substation, obtain substation operation status data, and build a digital twin of the substation based on the substation operation status data, thereby generating a digital twin model of the substation.

[0122] The prediction module is used to perform coupled simulation analysis on the digital twin model of the substation, confirm the coupled data of the substation based on the results of the coupled simulation analysis, perform abnormal detection and comprehensive analysis of the substation equipment operation status based on the coupled data of the substation, obtain comprehensive index data of equipment operation abnormality, and perform fault trend prediction on the comprehensive index data of equipment operation abnormality to confirm the equipment fault trend prediction data.

[0123] The priority generation module is used to evaluate the operation and maintenance priority of the substation digital twin model based on the comprehensive index data of equipment operation anomalies and the prediction data of equipment failure trends, and then generate a substation equipment operation and maintenance priority sequence based on the results of the operation and maintenance priority evaluation.

[0124] The operation and maintenance control module is used to generate substation operation and maintenance strategies based on the substation equipment operation and maintenance priority sequence, obtain a substation operation and maintenance strategy set, and send the substation operation and maintenance strategy set to the control terminal to execute intelligent operation and maintenance operations of the substation.

[0125] The above content is merely an example and illustration of the concept of the present invention. Those skilled in the art can make various modifications or additions to the specific embodiments described or use similar methods to replace them, as long as they do not deviate from the concept of the invention, they should all fall within the protection scope of the present invention.

[0126] In the description of this specification, references to terms such as "an embodiment," "example," "specific example," etc., indicate that a specific feature, structure, material, or characteristic described in connection with that embodiment or example is included in at least one embodiment or example of the invention. In this specification, illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples.

[0127] The preferred embodiments of the present invention disclosed above are merely illustrative of the invention. These preferred embodiments do not exhaustively describe all details, nor do they limit the invention to any specific implementation. Clearly, many modifications and variations can be made based on the content of this specification. This specification selects and specifically describes these embodiments to better explain the principles and practical applications of the invention, thereby enabling those skilled in the art to better understand and utilize the invention.

Claims

1. A digital-twin-based substation operation and maintenance method, characterized in that, Includes the following steps: Step S1: Real-time data acquisition is performed through a multi-source sensor network deployed in the substation to obtain substation operation status data, and a digital twin of the substation is constructed based on the substation operation status data to generate a digital twin model of the substation. Step S2: Perform coupled simulation analysis on the digital twin model of the substation, confirm the coupled data of the substation based on the results of the coupled simulation analysis, perform abnormal detection and comprehensive analysis of the substation equipment operation status based on the coupled data of the substation, obtain the comprehensive index data of equipment operation abnormality, and perform fault trend prediction on the comprehensive index data of equipment operation abnormality to confirm the equipment fault trend prediction data. Step S3: Based on the comprehensive index data of equipment operation anomalies and the prediction data of equipment failure trends, perform operation and maintenance priority assessment on the digital twin model of the substation, and then generate a substation equipment operation and maintenance priority sequence based on the results of the operation and maintenance priority assessment. The comprehensive index data for abnormal equipment operation includes: Feature extraction is performed on the abnormal equipment operation index data to obtain the equipment operation feature vector; Based on the device's operating feature vector, anomaly pattern recognition is performed to identify the fault mode; the severity of the fault mode is assessed to generate fault mode severity level data; Based on the severity level data of the aforementioned fault modes, anomaly index fusion calculations are performed to identify comprehensive equipment operation anomaly index data. The generation of substation equipment operation and maintenance priority sequences includes: The comprehensive index data of the equipment operation anomaly is normalized; the equipment failure trend prediction data is weighted. Equipment risk index is calculated based on normalized and weighted data; Based on the equipment risk index data, an equipment correlation analysis is performed to generate an equipment correlation network; Fault propagation path analysis is performed on the device relationship network to obtain potential fault propagation path data; risk assessment is performed based on the potential fault propagation path data to obtain a risk impact index; A comprehensive priority assessment is conducted based on equipment risk index data and risk impact index to generate a substation equipment operation and maintenance priority sequence. Step S4: Generate a substation operation and maintenance strategy based on the substation equipment operation and maintenance priority sequence, obtain a substation operation and maintenance strategy set, and send the substation operation and maintenance strategy set to the control terminal to execute intelligent operation and maintenance operations of the substation.

2. The power substation operation and maintenance method based on digital twinning according to claim 1, characterized in that, Step S1 includes the following steps: Step S11: Collect electrical parameter data, mechanical parameter data, and environmental parameter data through a multi-source sensor network deployed in the substation, and then integrate the substation operating status data based on the electrical parameter data, mechanical parameter data, and environmental parameter data; Step S12: Perform spatiotemporal alignment processing on the substation operating status data, and then confirm the substation spatiotemporal synchronization data based on the result of the spatiotemporal alignment processing. Step S13: Based on the spatiotemporal synchronization data of the substation, reconstruct the three-dimensional point cloud of the substation equipment, and then generate a point cloud model of the substation equipment based on the reconstruction result of the three-dimensional point cloud. Step S14: Perform parameter mapping on the point cloud model of the substation equipment to construct a digital twin model of the substation.

3. The substation operation and maintenance method based on digital twin according to claim 1, characterized in that, Step S2 includes the following steps: Step S21: Perform multiphysics coupling simulation analysis on the digital twin model of the substation to obtain the substation coupling data; Step S22: Based on the substation coupling data, perform baseline modeling of the substation equipment operating status, and then generate a baseline curve for the normal operating status of the equipment; Step S23: Anomaly detection is performed by analyzing the deviation between the substation operation status data and the equipment normal operation status benchmark curve. Based on the anomaly detection results, equipment operation anomaly index data is generated, and the equipment operation anomaly index data is comprehensively analyzed to confirm the comprehensive equipment operation anomaly index data. Step S24: Perform time series prediction analysis on the comprehensive index data of equipment operation anomalies, and then confirm the equipment failure trend prediction data.

4. The power substation operation and maintenance method based on digital twinning of claim 3, wherein, Step S24 includes the following steps: Step S241: Calculate the fault development rate based on the comprehensive index data of the equipment operation anomaly, and then obtain the fault development rate data based on the calculation results; Step S242: Based on the fault development rate data, predict the remaining service life to generate equipment remaining service life prediction data; Step S243: Perform confidence analysis on the equipment failure trend prediction data, and then confirm the failure prediction confidence index based on the results of the confidence analysis; Step S244: Based on the remaining service life prediction data of the equipment and the failure prediction confidence index, a comprehensive evaluation of the failure trend is performed to obtain the equipment failure trend prediction data.

5. The power substation operation and maintenance method based on digital twinning of claim 1, wherein, Step S4 includes the following steps: Step S41: Optimize the allocation of operation and maintenance resources based on the substation equipment operation and maintenance priority sequence, and generate an operation and maintenance resource allocation scheme; Step S42: Based on the operation and maintenance resource allocation scheme, perform operation and maintenance task scheduling and planning, and generate a detailed operation and maintenance task plan; Step S43: Generate operation and maintenance instructions based on the detailed operation and maintenance task plan to obtain the substation operation and maintenance strategy set; Step S44: The substation operation and maintenance strategy set is sent to the control terminal through a secure communication protocol to execute intelligent operation and maintenance operations of the substation.

6. A digital-twin-based substation operation and maintenance system, applied to the digital-twin-based substation operation and maintenance method of any one of claims 1-5, characterized in that, include: The model building module is used to collect real-time data through a multi-source sensor network deployed in the substation, obtain substation operation status data, and build a digital twin of the substation based on the substation operation status data, thereby generating a digital twin model of the substation. The prediction module is used to perform coupled simulation analysis on the digital twin model of the substation, confirm the coupled data of the substation based on the results of the coupled simulation analysis, perform abnormal detection and comprehensive analysis of the substation equipment operation status based on the coupled data of the substation, obtain comprehensive index data of equipment operation abnormality, and perform fault trend prediction on the comprehensive index data of equipment operation abnormality to confirm the equipment fault trend prediction data. The priority generation module is used to evaluate the operation and maintenance priority of the substation digital twin model based on the comprehensive index data of equipment operation anomalies and the prediction data of equipment failure trends, and then generate a substation equipment operation and maintenance priority sequence based on the results of the operation and maintenance priority evaluation. The comprehensive index data for abnormal equipment operation includes: Feature extraction is performed on the abnormal equipment operation index data to obtain the equipment operation feature vector; Based on the device's operating feature vector, anomaly pattern recognition is performed to identify the fault mode; the severity of the fault mode is assessed to generate fault mode severity level data; Based on the severity level data of the aforementioned fault modes, anomaly index fusion calculations are performed to identify comprehensive equipment operation anomaly index data. The generation of substation equipment operation and maintenance priority sequences includes: The comprehensive index data of the equipment operation anomaly is normalized; the equipment failure trend prediction data is weighted. Equipment risk index is calculated based on normalized and weighted data; Based on the equipment risk index data, an equipment correlation analysis is performed to generate an equipment correlation network; Fault propagation path analysis is performed on the device relationship network to obtain potential fault propagation path data; risk assessment is performed based on the potential fault propagation path data to obtain a risk impact index; A comprehensive priority assessment is conducted based on equipment risk index data and risk impact index to generate a substation equipment operation and maintenance priority sequence. The operation and maintenance control module is used to generate substation operation and maintenance strategies based on the substation equipment operation and maintenance priority sequence, obtain a substation operation and maintenance strategy set, and send the substation operation and maintenance strategy set to the control terminal to execute intelligent operation and maintenance operations of the substation.

7. A computer-readable storage medium, characterized in that: The system stores instructions that, when executed on a computer, cause the computer to perform a substation operation and maintenance method based on digital twins as described in any one of claims 1 to 5.

Images