GIL state integrated monitoring system based on multi-source information collaborative research and judgment

Abstract

The invention discloses a GIL state integrated monitoring system based on multi-source information collaborative research and judgment, and belongs to the technical field of electrical equipment monitoring. A GIL state integrated monitoring system based on multi-source information collaborative study and judgment comprises a multi-dimensional sensing unit and an intelligent analysis unit. According to the method, the problem that the real state of the GI L equipment is difficult to comprehensively and accurately reflect in the prior art is solved, the accuracy and efficiency of data monitoring are improved through optimized deployment of the sensors, the integrity and reliability of data are ensured through real-time monitoring of the data transmission state, and a data fusion algorithm and an artificial intelligence algorithm are adopted, so that the real-time monitoring of the GI L equipment is realized. According to the method, the multi-source data is deeply mined and analyzed, the abnormal mode and the fault trend of the G-IL equipment are effectively recognized, early warning is achieved, and workers can rapidly understand and respond by visually displaying the collected multi-source data, the analysis result and the fault warning.

Description

Technical Field

[0001] The present invention relates to the technical field of electrical equipment monitoring, and particularly to an integrated monitoring system for the state of Gas Insulated Metal Enclosed Transmission Lines (GIL) based on collaborative judgment of multi-source information. Background Technique

[0002] Gas Insulated Metal Enclosed Transmission Lines (GIL), as an efficient and reliable power transmission equipment, are widely used in high-voltage power transmission systems. However, during the operation of GIL equipment, it may be affected by various factors such as temperature, pressure, vibration, etc., resulting in abnormal equipment status or even failures. However, traditional GIL state monitoring methods usually rely on single sensors or simple data processing methods, which are difficult to comprehensively and accurately reflect the true state of GIL equipment, and there are problems of false alarms and missed alarms; therefore, they do not meet the existing requirements, and for this reason, we propose an integrated monitoring system for the state of GIL based on collaborative judgment of multi-source information. Summary of the Invention

[0003] The purpose of the present invention is to provide an integrated monitoring system for the state of GIL based on collaborative judgment of multi-source information. By optimizing the deployment of sensors, the accuracy and efficiency of data monitoring are improved. By real-time monitoring the data transmission status, the integrity and reliability of data are ensured. And by using data fusion algorithms and artificial intelligence algorithms, in-depth mining and analysis of multi-source data are carried out to effectively identify abnormal patterns and fault trends of GIL equipment, realizing early warning. By visually displaying the collected multi-source data, analysis results and fault warnings, it is convenient for staff to quickly understand and respond, solving the problems raised in the above background technique.

[0004] To achieve the above purpose, the present invention provides the following technical solution: An integrated monitoring system for the state of GIL based on collaborative judgment of multi-source information, the system includes a multi-dimensional perception unit and an intelligent analysis unit. The multi-dimensional perception unit is configured to collect in real time the data monitored by various sensors, including an equipment deployment module, a data collection module, and a data transmission module;

[0005] The equipment deployment module is configured to deploy different types of sensors and optimize the deployment positions of sensors based on the method of grid division, and monitor GIL equipment through the deployed multiple sensors, including multi-source data of temperature, pressure, strain, and vibration;

[0006] The data collection module is configured to collect and process in real time the multi-source data monitored by different types of sensors, and perform fusion processing on the collected multi-source data through a data fusion algorithm;

[0007] The data transmission module is configured to transmit the multi-source data after fusion processing to the intelligent analysis unit, and in the process of data transmission, real-time monitor the data transmission status;

[0008] The intelligent analysis unit is configured to conduct collaborative research and judgment on source information based on the received multi-source data, analyze the multi-source data using artificial intelligence algorithms, and achieve early warning of GIL equipment failures.

[0009] Further, the device deployment module includes:

[0010] The device selection module is configured to select appropriate sensor types according to the monitoring requirements of GIL equipment, including:

[0011] Temperature sensor: used to monitor the operating temperature of GIL equipment;

[0012] Pressure sensor: used to monitor the pressure of SF6 gas;

[0013] Strain sensor: used to monitor the mechanical strain of GIL equipment and determine whether deformation occurs;

[0014] Vibration sensor: used to detect the vibration of GIL equipment and determine whether there are mechanical failures;

[0015] Partial discharge sensor: used to monitor partial discharge faults;

[0016] The deployment optimization module is configured to optimize the deployment positions of sensors based on the grid division method and generate an optimized deployment plan;

[0017] The device layout module is configured to deploy the selected sensors according to the generated optimized deployment plan.

[0018] Further, the deployment optimization module specifically executes the following processes:

[0019] Grid division: Determine the size of the grid according to the area, shape of the deployment area, and the coverage radius of various sensors. The grid size should ensure that at least one sensor can be accommodated in each grid, and there should be some room to optimize the deployment position;

[0020] Coverage requirement assessment: Collect the coverage requirement information of each grid in the deployment area, including the type, quantity, and distribution of monitoring targets, and map the collected coverage requirements to the divided grids to form a coverage requirement distribution map;

[0021] Sensor characteristic analysis: Analyze the characteristics of different sensors, including their coverage radius, monitoring accuracy, and power consumption, and recommend suitable grids according to the sensor characteristics;

[0022] Based on grid division, coverage requirements, and sensor characteristics, generate an optimized deployment plan, including the number and location information of sensors in each grid.

[0023] Furthermore, the data acquisition module includes:

[0024] A real-time acquisition module configured to formulate a data acquisition strategy according to the characteristics of various sensors and monitoring requirements, including the acquisition frequency and data accuracy, and to perform real-time acquisition of multi-source data monitored by different types of sensors according to the data acquisition strategy;

[0025] A data processing module configured to preprocess the multi-source data acquired in real time, including data cleaning, denoising, and normalization;

[0026] A data fusion module configured to perform fusion processing on the preprocessed multi-source data using a data fusion algorithm, where the data fusion algorithm includes weighted average, Kalman filtering, and Bayesian network.

[0027] Furthermore, the data transmission module includes:

[0028] A real-time transmission module configured to transmit multi-source data in a wired network or wireless network manner;

[0029] A transmission monitoring module configured to monitor various status indicators during the transmission process of multi-source data in real time, including transmission rate, packet loss rate, and latency, etc., and to judge whether there is an abnormality or fault in the transmission according to the various status indicators;

[0030] An exception handling module configured to design an exception handling mechanism. When the transmission monitoring module detects a transmission abnormality or fault, corresponding measures are immediately taken, including caching data, retransmitting data, and switching the transmission channel, and at the same time, logging the exception or fault event;

[0031] Among them, caching data: Design a data caching mechanism to cope with network fluctuations or temporary unavailability of the transmission channel;

[0032] Retransmitting data: When a transmission abnormality or fault causes the data transmission to fail, start a retry mechanism, and retransmit the multi-source data according to a preset retry strategy. The retry mechanism includes the number of retries and the retry interval;

[0033] Switching the transmission channel: If there is an abnormality or fault in the transmission channel used by the multi-source data, switch to another transmission channel for data transmission.

[0034] Furthermore, the intelligent analysis unit includes:

[0035] A real-time analysis module configured to construct, train, and optimize a fault warning model using artificial intelligence algorithms, and to perform real-time analysis on multi-source data through the fault warning model to achieve early warning of GIL equipment faults. Among them, the artificial intelligence algorithms include deep learning and machine learning;

[0036] The fault warning module is configured to preset warning thresholds and warning levels. According to the model analysis results of the real-time analysis module, when the abnormal patterns in the multi-source data reach or exceed the warning threshold, it timely issues fault warning information.

[0037] Furthermore, the real-time analysis module specifically executes the following processes:

[0038] Algorithm selection: According to the characteristics of the GIL device and the requirements of fault warning, select appropriate artificial intelligence algorithms, including deep learning and machine learning, to construct and optimize the fault warning model.

[0039] Model training: Use historical fault data and normal operation data to train the selected algorithm, construct a fault warning model for the GIL device, and through continuous iteration and optimization, improve the prediction accuracy and generalization ability of the fault warning model;

[0040] Real-time analysis: Input the multi-source data into the trained fault warning model for real-time analysis. Automatically identify abnormal patterns and trends in the multi-source data through the fault warning model, and predict the fault type and time of the GIL device.

[0041] Furthermore, the system further includes a visualization unit configured to visually display the multi-source data collected by the data acquisition module, the analysis results of the intelligent analysis unit, and the fault warning. Specifically:

[0042] Multi-source data display: Use charts and dashboards to display the real-time values of the multi-source data, and provide a data comparison function to support comparing data differences at different time periods, different devices, or different parameters;

[0043] Analysis result display: Display in a graphical manner, including showing the correlation and abnormal patterns between data through heat maps and clustering maps;

[0044] Fault warning display: Design a prominent warning display interface. When the intelligent analysis unit issues a fault warning, use flashing and color-changing methods to remind the staff to pay attention to the warning.

[0045] Furthermore, the system further includes:

[0046] A digital twin model construction module for importing the GIL design drawing, combining with the parameter analysis of the GIL, constructing a geometric model of the GIL, and associating the geometric model with the corresponding material properties and electrical parameters to form a digital twin model;

[0047] A data synchronization module for mapping and binding the collected multi-source data with the digital twin model, and realizing real-time synchronization between the GIL physical entity and the virtual digital twin model through data interfaces and middleware technologies;

[0048] A virtual simulation fault prediction module is used to perform virtual simulation based on a digital twin model through physical simulation and data analysis. By simulating the operating conditions under different working conditions, it predicts the fault risk. Using the finite element analysis method, under extreme conditions of high temperature and high voltage, it simulates the stress distribution and deformation of the GIL. If the prediction result indicates a structural hidden danger, an alarm message is sent.

[0049] A decision support module is used for AI data analysis algorithms to provide fault diagnosis suggestions, maintenance plans, and decision support information to help formulate a scientific and reasonable GIL maintenance strategy.

[0050] Furthermore, the virtual simulation fault prediction module includes:

[0051] A temperature field model construction sub-module is used to simulate the finite element thermal state of the GIL based on the finite element analysis, combined with the thermal conductivity of the GIL physical object, and based on the heat conduction equation. According to the finite element thermal state of the GIL and its changes, a temperature field model of the GIL physical object is constructed.

[0052] A key position determination sub-module is used to obtain historical temperature data and the monitored temperatures of each position of the GIL physical object under the corresponding temperature conditions. Conduct a comparative analysis of the monitored temperatures of each position for the same GIL physical object, sort the position points according to the change of their monitored temperatures from large to small, and determine the top several position points as the key positions to be concerned.

[0053] A field model training sub-module is used to use the historical temperature data and the monitored temperatures of each position of the GIL physical object under the corresponding temperature conditions as the learning and training data of the temperature field model, implement model optimization, and obtain a trained temperature field model through training. Record the changes in the monitored temperatures of each key position during the training process.

[0054] A model boundary determination sub-module is used to obtain real-time temperature data, input the trained temperature field model according to the current GIL physical object parameters, and after being processed by the temperature field model, output the predicted temperature data of each key position of the GIL physical object.

[0055] An alarm sub-module is used to compare the predicted temperature data of the key positions to be concerned with the preset temperature range threshold. If the predicted temperature data deviates from the preset temperature range threshold, it indicates that there is a structural hidden danger in the GIL physical object, and then an alarm message is sent.

[0056] Compared with the prior art, the beneficial effects of the present invention are:

[0057] 1. The multi-dimensional perception unit of the present invention realizes the precise deployment of sensors through grid division, ensuring the comprehensiveness and accuracy of data monitoring, while reducing the deployment cost. Then, it collects the multi-source data monitored by different types of sensors in real time and transmits it to the intelligent analysis unit for real-time analysis. During the data transmission process, it monitors the transmission rate, packet loss rate, and latency in real time, and designs cache, retransmission, and transmission channel switching mechanisms to improve data transmission efficiency and flexibility.

[0058] 2. Before data transmission, the multi-dimensional perception unit of the present invention integrates multi-source data using a data fusion algorithm, improving the availability and reliability of the data. After receiving the multi-source data, the intelligent analysis unit constructs a fault warning model based on artificial intelligence algorithms, which can automatically identify abnormal patterns, realize early warning of GIL equipment faults, and improve the efficiency of fault prevention and handling. BRIEF DESCRIPTION OF THE DRAWINGS

[0059] Figure 1 It is a schematic diagram of the overall structure of the GIL status integrated monitoring system based on multi-source information collaborative judgment of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0060] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.

[0061] In order to solve the technical problems that the existing GIL status monitoring methods usually rely on single sensors or simple data processing methods, it is difficult to comprehensively and accurately reflect the true state of GIL equipment, and there are false alarms and missed alarms. Please refer to Figure 1 , the following technical solutions are provided in this embodiment:

[0062] A GIL status integrated monitoring system based on multi-source information collaborative judgment, the system includes a multi-dimensional perception unit and an intelligent analysis unit. The multi-dimensional perception unit is configured to collect the data monitored by various sensors in real time, including a device deployment module, a data collection module, and a data transmission module;

[0063] The device deployment module is configured to deploy different types of sensors and optimize the deployment positions of the sensors based on the grid division method. The GIL equipment is monitored through the deployed multiple sensors, including multi-source data such as temperature, pressure, strain, and vibration.

[0064] The data acquisition module is configured to collect and process multi-source data monitored by different types of sensors in real time, and fuse the collected multi-source data through a data fusion algorithm;

[0065] The data transmission module is configured to transmit the multi-source data after fusion processing to the intelligent analysis unit, and during the data transmission process, monitor the data transmission status in real time;

[0066] The intelligent analysis unit is configured to conduct collaborative research and judgment on source information based on the received multi-source data, and analyze the multi-source data using artificial intelligence algorithms to achieve early warning of GIL equipment failures.

[0067] The technical effects of the above content are as follows: By deploying multiple sensors through the multi-dimensional perception unit, it is possible to collect multi-source data such as the temperature, pressure, strain, and vibration of the GIL equipment in real time. Using a multi-dimensional monitoring method can comprehensively reflect the operating state of the equipment and avoid the limitations of single-dimensional data.

[0068] In the multi-dimensional perception unit, the equipment deployment module optimizes the deployment positions of sensors based on the grid division method, enabling precise monitoring of the GIL equipment. By optimizing the deployment method, it can maximize the coverage of key parts of the equipment, improve the comprehensiveness and accuracy of monitoring. The data transmission module can monitor the data transmission status in real time when transmitting the multi-source data after fusion processing, ensuring the stability and reliability of data transmission, avoiding data loss or transmission delay, and guaranteeing the real-time performance of the system.

[0069] The intelligent analysis unit uses artificial intelligence algorithms to conduct collaborative research and judgment on multi-source data, enabling early warning of GIL equipment failures, thus providing sufficient time for maintenance personnel to take preventive measures.

[0070] In summary, the GIL status integrated monitoring system based on collaborative research and judgment of multi-source information significantly improves the monitoring accuracy and operation and maintenance efficiency of GIL equipment through multi-dimensional monitoring, intelligent analysis, as well as technical means such as efficient data transmission and optimized deployment, providing strong guarantee for the safe operation of the equipment.

[0071] The equipment deployment module includes:

[0072] The equipment selection module is configured to select appropriate sensor types according to the monitoring requirements of the GIL equipment, including:

[0073] Temperature sensor: used to monitor the operating temperature of the GIL equipment;

[0074] Pressure sensor: used to monitor the pressure of SF6 gas;

[0075] Strain sensor: used to monitor the mechanical strain of the GIL equipment and judge whether deformation occurs;

[0076] Vibration sensor: used to detect the vibration condition of the GIL device and judge whether there is a mechanical fault;

[0077] Partial discharge sensor: used to monitor partial discharge faults;

[0078] Deployment optimization module, configured to optimize the deployment positions of sensors based on the grid division method, generate an optimized deployment plan, and specifically execute the following processes:

[0079] Grid division: Determine the size of the grid according to the area and shape of the deployment area and the coverage radius of various sensors. The grid size should ensure that at least one sensor can be accommodated in each grid, and there should be some room to optimize the deployment position;

[0080] Coverage requirement assessment: Collect the coverage requirement information of each grid in the deployment area, including the type, quantity, and distribution of monitoring targets, and map the collected coverage requirements to the divided grids to form a coverage requirement distribution map;

[0081] Sensor characteristic analysis: Analyze the characteristics of different sensors, including their coverage radius, monitoring accuracy, and power consumption, and recommend suitable grids according to the characteristics of the sensors;

[0082] Based on grid division, coverage requirements, and sensor characteristics, generate an optimized deployment plan, including the quantity and location information of sensors in each grid;

[0083] Device layout module, configured to deploy the selected sensors according to the generated optimized deployment plan.

[0084] The technical effects of the above content are as follows: The device selection module can accurately select the appropriate type of sensor according to the specific monitoring requirements of the GIL device. Through targeted sensor selection, the accuracy and reliability of monitoring data can be ensured, meeting the monitoring requirements of different aspects of the GIL device. Multiple types of sensors can comprehensively monitor the operating status of the GIL device, from multiple dimensions such as temperature, pressure, mechanical strain, and partial discharge, avoiding monitoring blind spots. The deployment optimization module can generate an optimized deployment plan based on grid division, coverage requirements, and sensor characteristics, clarifying the quantity and location information of sensors in each grid. This optimized plan can maximize the monitoring efficiency, reduce the number of sensors, and lower the system cost. The device layout module accurately deploys the selected sensors to the specified positions according to the optimized deployment plan. Precise layout can ensure the monitoring effect of the sensors and avoid monitoring errors caused by unreasonable positions. It should be noted that the system supports flexible adjustment of the sensor deployment plan according to the actual operating conditions. For example, when the monitoring requirements change or the status of the GIL device is abnormal, the deployment plan can be quickly re-optimized and the sensor positions can be adjusted.

[0085] In summary, through the collaborative work of the device selection module, the deployment optimization module, and the device layout module, the precise selection and optimized deployment of sensors are achieved, significantly improving the accuracy and efficiency of GIL device monitoring, reducing the system cost and the workload of operation and maintenance, and providing a strong guarantee for the safe operation of GIL devices.

[0086] The data acquisition module includes:

[0087] The real-time acquisition module is configured to formulate a data acquisition strategy according to the characteristics of various sensors and the monitoring requirements, including the acquisition frequency and data accuracy, and to perform real-time acquisition of multi-source data monitored by different types of sensors according to the data acquisition strategy;

[0088] The data processing module is configured to preprocess the multi-source data acquired in real time, including data cleaning, denoising, and normalization;

[0089] The data fusion module is configured to perform fusion processing on the preprocessed multi-source data by using a data fusion algorithm. Among them, the data fusion algorithm includes weighted average, Kalman filtering, and Bayesian network;

[0090] The weighted average method: applicable to the case where each data source has the same type and the importance difference is not significant. By assigning different weights to different data sources and then performing weighted average calculation, a comprehensive result can be obtained;

[0091] The Kalman filtering method: a linear filtering method, especially suitable for real-time data processing and prediction. By fusing data at different time points, noise can be effectively eliminated, and the reliability and stability of the data can be improved;

[0092] The Bayesian network: based on Bayes' theorem in probability theory, applicable to data fusion scenarios with high uncertainty. By updating the prior probability to obtain the posterior probability, the effective integration of different data sources can be achieved.

[0093] The technical effects of the above content are as follows: The real-time acquisition module formulates flexible data acquisition strategies according to the characteristics of various sensors and monitoring requirements, including acquisition frequency and data accuracy. By formulating data acquisition strategies, it can ensure that the acquired data not only meets the monitoring requirements but also avoids data redundancy. The data processing module preprocesses the acquired multi-source data, which can improve the quality and usability of the data and also provides standardized data input for subsequent data fusion, improving the fusion effect. The data fusion module adopts a variety of data fusion algorithms and can select the most suitable fusion method according to different scenarios, improving the accuracy and reliability of the fusion results. Through the collaborative work of the real-time acquisition module, data processing module, and data fusion module, the efficient acquisition, processing, and fusion of multi-source data of GIL equipment are realized, significantly improving the data quality, fusion accuracy, real-time performance, and stability of the system, providing strong technical support for the intelligent monitoring and fault warning of GIL equipment.

[0094] The data transmission module includes:

[0095] The real-time transmission module is configured to transmit multi-source data by means of a wired network or a wireless network;

[0096] The transmission monitoring module is configured to monitor in real time various status indicators during the transmission process of multi-source data, including transmission rate, packet loss rate, and latency, etc., and judge whether there are any abnormalities or faults in the transmission according to the various status indicators;

[0097] The exception handling module is configured to design an exception handling mechanism. When the transmission monitoring module detects a transmission abnormality or fault, corresponding measures are immediately taken, including caching data, retransmitting data, and switching the transmission channel. At the same time, the exception or fault event is logged for post-event analysis and problem troubleshooting;

[0098] Among them, caching data: Design a data caching mechanism to cope with network fluctuations or temporary unavailability of the transmission channel. By temporarily storing the data that has not been successfully transmitted, it will be retransmitted after the network is restored;

[0099] Retransmitting data: When a transmission abnormality or fault causes the data transmission to fail, a retry mechanism is started, and the multi-source data is retransmitted according to the preset retry strategy to ensure that the data can be finally successfully transmitted. The retry mechanism includes the number of retries and the retry interval;

[0100] Switching the transmission channel: If there are abnormalities or faults in the transmission channel used by the multi-source data, switch to another transmission channel for data transmission to ensure the continuity of data transmission.

[0101] The technical effects of the above content are as follows: The real-time transmission module supports two transmission methods, wired network and wireless network. When performing data transmission, the optimal transmission method can be selected according to actual needs to ensure the efficiency and reliability of data transmission. During the data transmission process, the transmission monitoring module can monitor various status indicators in the multi-source data transmission process in real time. Through the monitored indicators, it can quickly determine whether there are abnormalities or faults in the transmission. Once an abnormality or fault occurs in the transmission, a warning is issued in a timely manner, which can reduce the risk of data loss and transmission interruption. When an abnormality or fault occurs in the transmission, the exception handling module can quickly respond to the transmission fault through the designed exception handling mechanism, reduce the system downtime, and improve the overall operation stability.

[0102] In summary, through the collaborative work of the real-time transmission module, the transmission monitoring module, and the exception handling module, the efficient transmission, real-time monitoring, and intelligent exception handling of multi-source data are realized, which can improve the reliability, stability, and system operation and maintenance efficiency of data transmission, and provide strong technical support for the intelligent monitoring and fault warning of GIL equipment.

[0103] The intelligent analysis unit includes:

[0104] The real-time analysis module is configured to construct, train, and optimize a fault warning model using artificial intelligence algorithms. Through the fault warning model, real-time analysis of multi-source data is performed to achieve early warning of GIL equipment faults. Among them, the artificial intelligence algorithms include deep learning and machine learning, and the following processes are specifically executed:

[0105] Algorithm selection: According to the characteristics of GIL equipment and the requirements of fault warning, select appropriate artificial intelligence algorithms, including deep learning and machine learning, to construct and optimize the fault warning model.

[0106] Model training: Use historical fault data and normal operation data to train the selected algorithm, construct a GIL equipment fault warning model, and continuously iterate and optimize to improve the prediction accuracy and generalization ability of the fault warning model;

[0107] Real-time analysis: Input multi-source data into the trained fault warning model for real-time analysis. Through the fault warning model, automatically identify abnormal patterns and trends in multi-source data, and predict the fault type and time of GIL equipment;

[0108] The fault warning module is configured to preset warning thresholds and warning levels. According to the model analysis results of the real-time analysis module, if the abnormal pattern in the multi-source data reaches or exceeds the warning threshold, a fault warning message is issued in a timely manner.

[0109] The technical effects of the above content are as follows: The real-time analysis module selects appropriate artificial intelligence algorithms (including deep learning and machine learning) for the construction, training, and optimization of the fault warning model according to the characteristics of GIL equipment and the requirements of fault warning. The flexibility of algorithm selection can ensure that the model has a high recognition ability for different types of fault modes. Real-time input of multi-source data into the fault warning model can provide more comprehensive equipment status information, avoiding the limitations of a single data source. Through the real-time analysis of the fault warning model, abnormal patterns and trends can be quickly identified, and the fault type and time can be predicted, so that early warnings can be issued in a timely manner before the fault occurs, providing sufficient time for maintenance personnel to take preventive measures. The fault warning module has pre-set warning thresholds and warning levels, and can judge whether the abnormal pattern reaches or exceeds the warning threshold according to the real-time analysis results. Through the hierarchical warning mechanism, different response strategies can be provided according to the severity of the fault, avoiding over-warning or missed warnings. When the abnormal pattern in the multi-source data reaches the warning threshold, the system can promptly issue a fault warning message to notify the maintenance personnel to take corresponding measures. This timeliness can significantly reduce the impact of faults on equipment operation, reduce downtime and maintenance costs.

[0110] In summary, through the collaborative work of the real-time analysis module and the fault warning module, using artificial intelligence algorithms to efficiently analyze and warn multi-source data, the accuracy, real-time performance, and reliability of GIL equipment fault warning have been significantly improved, providing strong technical support for the safe operation and intelligent maintenance of GIL equipment.

[0111] The system also includes a visualization unit configured to visually display the multi-source data collected by the data acquisition module, the analysis results of the intelligent analysis unit, and the fault warning. Specifically:

[0112] Multi-source data display: Display the real-time values of multi-source data in the form of charts and dashboards. Maintenance personnel can quickly understand the operating status of GIL equipment, including key parameters such as temperature, pressure, strain, vibration, and partial discharge. The intuitive display method can help maintenance personnel quickly discover abnormal fluctuations in the data;

[0113] At the same time, it supports comparing the data differences of different time periods, different devices, or different parameters, enabling maintenance personnel to analyze the change trends of equipment status and identify potential problems. For example, by comparing the data of different time periods, it can be found whether there is long-term performance degradation of the equipment. By comparing the data of different devices, the problem device can be quickly located;

[0114] Analysis result display: The use of heat maps and clustering maps to display the correlation and abnormal patterns between data can help operation and maintenance personnel quickly identify key information in the data. By graphically displaying the analysis results, operation and maintenance personnel can more intuitively understand the operating status and potential problems of the equipment, thereby improving the efficiency and accuracy of fault diagnosis;

[0115] Fault warning display: Design a prominent warning display interface. When the intelligent analysis unit issues a fault warning, it uses flashing and color-changing methods to alert the staff to ensure that the warning information can be noticed immediately. The prominent display method can significantly improve the response speed of operation and maintenance personnel to faults.

[0116] The technical effects of the above content are as follows: Through the intuitive visualization display of the visualization unit, operation and maintenance personnel can quickly obtain equipment status information, reduce the time for fault troubleshooting, and improve operation and maintenance efficiency. The visualization display of multi-source data and analysis results can help operation and maintenance personnel verify the accuracy of fault warnings, reduce false alarms and missed alarms, and the intuitive charts and graphical displays provide scientific decision-making basis for operation and maintenance personnel, enabling them to better formulate maintenance plans and response measures.

[0117] Working principle: By deploying a variety of sensors, multi-source data of GIL equipment can be collected in real time, thus avoiding the limitations of single-dimensional data. And based on the grid division method to optimize the deployment location of sensors to improve the comprehensiveness and accuracy of monitoring. During the transmission of the multi-source data monitored by the sensors, the data transmission status will be monitored in real time to ensure the stability and reliability of data transmission, so as to avoid data loss or transmission delay. By using data fusion algorithms and artificial intelligence algorithms to deeply mine and analyze the multi-source data, early warning of GIL equipment faults can be realized, thus providing sufficient time for operation and maintenance personnel to take preventive measures.

[0118] On the basis of the foregoing embodiments, the system further includes:

[0119] Digital twin model construction module, which is used to import the GIL design drawing, construct the geometric model of GIL in combination with the parameter analysis of GIL, and associate the geometric model with the corresponding material properties and electrical parameters to form a digital twin model;

[0120] Data synchronization module, which is used to map and bind the collected multi-source data with the digital twin model, and realize the real-time synchronization of the GIL physical object and the virtual digital twin model through data interfaces and middleware technologies;

[0121] The virtual simulation fault prediction module is used to perform virtual simulation based on the digital twin model through physical simulation and data analysis, predict fault risks by simulating the operating conditions under different working conditions; use the finite element analysis method to simulate the stress distribution and deformation of GIL under extreme conditions of high temperature and high voltage, and send an alarm message if the prediction result indicates a structural hidden danger;

[0122] The decision support module is used for AI data analysis algorithms to provide fault diagnosis suggestions, maintenance plans and decision support information to help formulate a scientific and reasonable GIL maintenance strategy.

[0123] The working principle and technical effects of the above content are as follows: By importing the GIL design drawing, based on parsing the parameters obtained from the GIL design drawing, a geometric model of GIL is constructed, and the geometric model is associated with the corresponding material properties and electrical parameters to form a digital twin model; then, through multi-source data mapping and binding, the real-time synchronization between the GIL physical object and the virtual digital twin model is realized; in the case of data synchronization, virtual simulation is carried out based on the digital twin model to predict fault risks, and according to the predicted risk situation, it is determined whether to send an alarm message; in addition, AI data analysis algorithms can be used to provide fault diagnosis suggestions, maintenance plans and decision support information to help formulate a scientific and reasonable GIL maintenance strategy in real time; adopting this solution, on the one hand, it can effectively control the risk of temperature imbalance, and timely remind to eliminate the risk when it is discovered; on the other hand, by carrying out virtual simulation analysis before implementation, costs can be saved.

[0124] On the basis of the foregoing embodiments, the virtual simulation fault prediction module includes:

[0125] The temperature field model construction sub-module is used to simulate the finite element thermal state of GIL based on the finite element analysis, combined with the heat conduction performance of the GIL physical object, and construct the temperature field model of the GIL physical object according to the finite element thermal state and change of GIL; among them, the heat conduction equation adopts the following three-dimensional unsteady heat conduction equation:

[0126]

[0127] where T is the temperature at the position (x, y, z) at time t; γ is the thermal diffusivity of the GIL physical object; ρ is the density of the GIL physical object material; c is the specific heat capacity of the GIL physical object material; (x, y, z) is the position coordinate of the GIL physical object point; q is the heat source rate per unit volume of the GL physical object;

[0128] The key position determination sub-module is used to obtain historical temperature data and the monitored temperatures at various positions of the GIL physical object under corresponding temperature conditions, conduct a comparative analysis of the monitored temperatures at various positions for the same GIL physical object, sort the positions according to the change of their monitored temperatures from large to small, and determine the positions of the top several before sorting as the key positions of concern;

[0129] The field model training sub-module is used to use the historical temperature data and the monitored temperatures at various positions of the GIL physical object under corresponding temperature conditions as the learning and training data of the temperature field model, implement model optimization, and obtain the trained temperature field model through training, and record the change of the monitored temperature at each key position of concern during the training process;

[0130] The model boundary determination sub-module is used to obtain real-time temperature data, input the trained temperature field model according to the current GIL physical object parameters, and output the predicted temperature data at each key position of concern of the GIL physical object after being processed by the temperature field model;

[0131] The alarm sub-module is used to compare the predicted temperature data at the key positions of concern with the preset temperature range threshold. If the predicted temperature data deviates from the preset temperature range threshold, it indicates that there are structural hidden dangers in the GIL physical object, and then an alarm message is sent.

[0132] The working principle and technical effect of the above content are as follows: By using finite element analysis, the GIL physical object is divided into each grid position point, the finite element thermal state of the GIL is simulated based on the heat conduction equation, and according to the finite element thermal state and change of the GIL, a temperature field model of the GIL physical object is constructed. After being trained by historical data, the temperature field model is used for the temperature prediction of the GIL physical object. According to the prediction result, compare it with the preset temperature range threshold. If a deviation is found, an alarm message is sent; By adopting this solution, with the use of the trained temperature field model, while ensuring temperature monitoring, the number of temperature sensors set for the GIL physical object can be saved, and the cost can be saved.

[0133] It should be noted that in this article, relational terms such as first and second are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the term "including", "comprising" or any other variant thereof is intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements not only includes those elements, but also includes other elements not expressly listed, or also includes elements inherent to such process, method, article or device.

[0134] Although embodiments of the present invention have been shown and described, it will be understood by those of ordinary skill in the art that various changes, modifications, substitutions and variations can be made to these embodiments without departing from the principles and spirit of the present invention.

Claims

1. The GIL status integrated monitoring system based on multi-source information collaborative analysis is characterized by: The system includes a multi-dimensional perception unit and an intelligent analysis unit, wherein the multi-dimensional perception unit includes a device deployment module, a data acquisition module and a data transmission module; The equipment deployment module is configured to optimize the deployment location of sensors based on a grid partitioning method and monitor the GIL equipment through the deployed different types of sensors; The data acquisition module is configured to perform real-time acquisition, processing and fusion processing of multi-source data monitored by different types of sensors; The data transmission module is configured to transmit the fused multi-source data to the intelligent analysis unit and monitor the data transmission status in real time; The intelligent analysis unit is configured to perform collaborative analysis of multi-source information based on multi-source data, and to achieve early warning of GIL equipment failure by analyzing multi-source data through artificial intelligence algorithms.

2. The GIL status integrated monitoring system based on multi-source information collaborative analysis according to claim 1 is characterized by: The device deployment module includes: The device selection module is configured to select the appropriate sensor type based on the monitoring requirements of the GIL device, including: Temperature sensor: used to monitor the operating temperature of GIL equipment; Pressure sensor: used to monitor the pressure of SF6 gas; Strain sensor: used to monitor the mechanical strain of the GIL equipment and determine whether deformation occurs; Vibration sensor: used to detect the vibration of GIL equipment and determine whether there is a mechanical failure; Partial discharge sensor: used to monitor partial discharge faults; A deployment optimization module, configured to optimize the deployment position of the sensor based on a grid partitioning method and generate an optimized deployment plan; The device deployment module is configured to deploy the selected sensors according to the generated optimized deployment plan.

3. The GIL status integrated monitoring system based on multi-source information collaborative analysis according to claim 2 is characterized by: The deployment optimization module specifically performs the following process: Grid division: Determine the size of the grid based on the area and shape of the deployment area and the coverage radius of various sensors. The grid size should ensure that each grid can accommodate at least one sensor and leave some room to optimize the deployment location. Coverage demand assessment: Collect coverage demand information of each grid in the deployment area, including the type, quantity and distribution of monitoring targets, and map the collected coverage demand to the divided grids to form a coverage demand distribution map; Sensor characteristic analysis: Analyze the characteristics of different sensors, including their coverage radius, monitoring accuracy, and power consumption, and recommend appropriate grids based on the characteristics of the sensors; Based on grid division, coverage requirements and sensor characteristics, an optimized deployment plan is generated, including the number and location information of sensors in each grid.

4. The GIL status integrated monitoring system based on multi-source information collaborative analysis according to claim 1 is characterized by: The data acquisition module comprises: The real-time acquisition module is configured to formulate data acquisition strategies, including acquisition frequency and data accuracy, according to the characteristics and monitoring requirements of various sensors, and to collect multi-source data monitored by different types of sensors in real time according to the data acquisition strategies; A data processing module configured to pre-process the multi-source data collected in real time, including data cleaning, denoising and normalization; The data fusion module is configured to use a data fusion algorithm to perform fusion processing on the pre-processed multi-source data, wherein the data fusion algorithm includes weighted averaging, Kalman filtering and Bayesian network.

5. The GIL status integrated monitoring system based on multi-source information collaborative analysis according to claim 1 is characterized by: The data transmission module comprises: A real-time transmission module is configured to transmit multi-source data using a wired network or a wireless network; The transmission monitoring module is configured to monitor various status indicators in the multi-source data transmission process in real time, including transmission rate, packet loss rate and delay, and judge whether there is abnormality or failure in the transmission according to various status indicators; The exception handling module is configured to design an exception handling mechanism. When the transmission monitoring module detects a transmission anomaly or failure, it immediately takes corresponding measures, including caching data, retransmitting data, and switching transmission channels, and logs the anomaly or failure event. Among them, cache data: design a data cache mechanism to cope with network fluctuations or temporary unavailability of the transmission channel; Retransmit data: When transmission is abnormal or fails, resulting in data transmission failure, the retry mechanism is activated to retransmit multi-source data according to the preset retry strategy. The retry mechanism includes the number of retries and the retry interval. Switch transmission channel: If there is an abnormality or failure in the transmission channel used by the multi-source data, switch to another transmission channel for data transmission.

6. The GIL status integrated monitoring system based on multi-source information collaborative analysis according to claim 1 is characterized by: The intelligent analysis unit comprises: A real-time analysis module is configured to use artificial intelligence algorithms to build, train and optimize fault warning models, and to perform real-time analysis of multi-source data through the fault warning models to achieve early warning of GIL equipment failures, wherein the artificial intelligence algorithms include deep learning and machine learning; The fault warning module is configured to pre-set the warning threshold and warning level. According to the model analysis results of the real-time analysis module, if the abnormal pattern in the multi-source data reaches or exceeds the warning threshold, the fault warning information is issued in time.

7. The GIL status integrated monitoring system based on multi-source information collaborative analysis according to claim 6 is characterized by: The real-time analysis module specifically performs the following process: Algorithm selection: According to the characteristics of GIL equipment and the needs of fault warning, select appropriate artificial intelligence algorithms, including deep learning and machine learning, to build and optimize the fault warning model. Model training: Use historical fault data and normal operation data to train the selected algorithm, build a GIL equipment fault warning model, and improve the prediction accuracy and generalization ability of the fault warning model through continuous iteration and optimization; Real-time analysis: Multi-source data is input into the trained fault warning model for real-time analysis. The fault warning model automatically identifies abnormal patterns and trends in multi-source data and predicts the type and time of failure of GIL equipment.

8. The GIL status integrated monitoring system based on multi-source information collaborative analysis according to claim 1 is characterized by: The system further includes a visualization unit configured to visualize the multi-source data collected by the data collection module and the analysis results and fault warnings of the intelligent analysis unit, specifically: Multi-source data display: Displays the real-time values ​​of multi-source data in the form of charts and dashboards, and provides data comparison functions to support comparison of data differences in different time periods, different devices or different parameters; Analysis results presentation: Graphical presentation, including heat maps and cluster diagrams to show the correlation and abnormal patterns between data; Fault warning display: Design an eye-catching warning display interface. When the intelligent analysis unit issues a fault warning, it uses flashing and color changing methods to remind staff to pay attention to the warning.

9. The GIL status integrated monitoring system based on multi-source information collaborative analysis according to claim 1 is characterized by: Also includes: The digital twin model construction module is used to import the GIL design drawing, combine the GIL parameter analysis, build the GIL geometric model, and associate the geometric model with the corresponding material properties and electrical parameters to form a digital twin model; The data synchronization module is used to map and bind the collected multi-source data to the digital twin model, and to achieve real-time synchronization between the physical and virtual digital twin models of the GIL through data interfaces and middleware technology; The virtual simulation fault prediction module is used to implement virtual simulation based on the digital twin model through physical simulation and data analysis, and predict the risk of faults by simulating the operation under different working conditions. It uses the finite element analysis method to simulate the stress distribution and deformation of the GIL under extreme conditions of high temperature and high voltage. If the prediction results show that there are structural hazards, an alarm message will be issued. The decision support module is used for AI data analysis algorithms to provide fault diagnosis suggestions, maintenance plans and decision support information to help formulate scientific and reasonable GIL maintenance strategies.

10. The GIL status integrated monitoring system based on multi-source information collaborative analysis according to claim 9 is characterized in that: The virtual simulation fault prediction module comprises: The temperature field model construction submodule is used to simulate the GIL finite element thermal state based on the heat conduction equation and combine the thermal conductivity of the GIL object on the basis of finite element analysis, and to construct the temperature field model of the GIL object according to the GIL finite element thermal state and change; The key position determination submodule is used to obtain historical temperature data and the monitored temperature of each position of the GIL object under the corresponding temperature conditions, conduct comparative analysis of the monitored temperature of each position of the same GIL object, sort each position point from large to small according to its monitored temperature change, and determine the top several position points as key focus positions; The field model training submodule is used to use the historical temperature data and the monitored temperatures of various locations of the GIL physical object under the corresponding temperature conditions as the learning and training data of the temperature field model, implement model optimization, obtain the trained temperature field model through training, and record the monitored temperature changes of each key location during the training process; The model boundary determination submodule is used to obtain real-time temperature data, input the trained temperature field model according to the current GIL object parameters, and output the predicted temperature data of each key focus position of the GIL object after being processed by the temperature field model; The alarm submodule is used to compare the predicted temperature data of the key focus location with the preset temperature range threshold. If the predicted temperature data deviates from the preset temperature range threshold, it indicates that there are structural risks in the GIL object, and an alarm message is issued.

Images