GIS bus duct telescopic joint deformation on-line monitoring system based on multi-source data fusion

Abstract

The application discloses a GIS bus chamber expansion joint deformation online monitoring system based on multi-source data fusion and relates to the technical field of GIS equipment monitoring.The multi-source sensor type calibration and precision trend monitoring structure of the application solves the problem of lack of prediction in traditional calibration, guarantees reliable monitoring data, eliminates the limitations of single sensors by relying on finite element analysis models and multi-source data fusion technology, deeply excavates the correlation between parameters and deformation, improves the accuracy of deformation risk identification by combining three-dimensional model labeling and double determination mechanism, realizes data specification storage, risk early prediction and hidden danger accurate positioning by virtue of anti-electromagnetic interference transmission, dynamic deformation prediction and multi-terminal visualization design.The overall breakthrough of the traditional monitoring passivity, inaccurate determination and low-efficiency operation provides a scientific decision basis for the operation and maintenance of the GIS bus chamber expansion joint, effectively guarantees the long-term safe and stable operation of the GIS equipment, and significantly improves the pertinence and efficiency of monitoring and operation and maintenance.

Description

Technical Field

[0001] This invention relates to the field of GIS equipment monitoring technology, and in particular to an online monitoring system for deformation of expansion joints in GIS busbar compartments based on multi-source data fusion. Background Technology

[0002] GIS (Gas Insulated Switchgear), as a high-voltage power distribution device with insulating medium, consists of components such as busbars, circuit breakers, disconnectors, and expansion joints sealed in a metal cylinder. It boasts significant advantages such as small footprint, high operational reliability, strong resistance to environmental interference, and low maintenance workload, making it a core component of high-voltage and ultra-high-voltage power grids. However, GIS equipment operates in complex environments such as the field, coastal areas, and suburbs, and must withstand the effects of temperature field changes, foundation settlement, equipment vibration, and electromagnetic coupling. As service time increases, various potential faults gradually emerge, with SF6 gas leakage and mechanical structure failure being the most prominent. In particular, displacement and deformation of the busbar expansion joints and insulation faults are also common problems. SF6 gas leakage is especially prevalent in extreme weather conditions, not only damaging the insulation and arc-extinguishing performance of GIS equipment and leading to a decrease in the overall operational stability of the substation, but also posing serious hazards to the surrounding environment and human health due to the strong greenhouse effect and toxic decomposition products of SF6 gas.

[0003] Current GIS equipment monitoring technologies mainly focus on electrical parameters such as SF6 gas parameters, partial discharge, and heating of conductive connections. Examples include monitoring SF6 pressure using gas density relays, detecting partial discharge using ultrasonic sensors, and monitoring temperature rise in conductive parts using infrared thermography. While domestic and international scholars have conducted some research on the causes of GIS failures, such studies largely focus on post-failure structural analysis or offline detection, neglecting online monitoring of busbar expansion joint displacement and deformation. Furthermore, existing monitoring methods have significant limitations.

[0004] 1. It is impossible to capture the three-dimensional displacement of the expansion joint and the overall deformation of the busbar compartment in real time, making it difficult to predict faults such as weld cracking and fatigue damage of the expansion joint caused by deformation accumulation.

[0005] 2. Existing monitoring relies heavily on data from a single sensor, without considering the coupled effects of multiple parameters such as temperature, vibration, and displacement. This fails to eliminate the limitations and redundancy of single data, resulting in insufficient accuracy in determining the operating status of GIS equipment. The cause can only be deduced by tracing the traces after a fault occurs, which fails to achieve the maintenance goal of early warning and early prevention.

[0006] 3. The complex structure of GIS equipment, the sealing characteristics of the metal shell, and the multi-field coupling result in a complex displacement deformation mechanism. Traditional offline detection is not only inefficient and costly, but also prone to insufficient data reference due to limitations of measuring tools, reading errors, and differences in measurement positions. It may even exacerbate equipment damage due to misjudgment of displacement status. There is a lack of quantitative judgment standards for the causes of deformation. Summary of the Invention

[0007] The purpose of this invention is to provide an online monitoring system for deformation of expansion joints in GIS busbar compartments based on multi-source data fusion. By using a multi-source sensor classification calibration and accuracy trend monitoring structure, relying on finite element analysis models and multi-source data fusion technology, and combining three-dimensional model annotation and dual judgment mechanisms, the system improves the accuracy of deformation risk identification, effectively ensures the long-term safe and stable operation of GIS equipment, and significantly enhances the pertinence and efficiency of monitoring and maintenance, thereby solving the problems mentioned in the background art.

[0008] To achieve the above objectives, the present invention provides the following technical solution:

[0009] The GIS busbar expansion joint deformation online monitoring system based on multi-source data fusion includes:

[0010] The monitoring parameter acquisition module is used to collect multiple types of monitoring parameters in real time based on the sensor group, including displacement data, vibration data, temperature data and environmental parameters;

[0011] The multi-data fusion processing module is used to build a data fusion model, perform fusion analysis on the collected multi-type monitoring parameters, and establish an equipment operation status evaluation index system based on the fusion analysis results.

[0012] The intelligent monitoring and early warning module is used to build a remote data transmission network adapted to the electromagnetic environment of the substation. At the same time, based on the preset safe operation threshold of the GIS busbar expansion joint and the fusion analysis results, it monitors the deformation of the expansion joint and the changes in the operating status of the equipment in real time. When the monitoring data exceeds the safety threshold, the early warning mechanism is triggered, and the early warning information is sent to the operation and maintenance personnel through the preset interaction channel based on the early warning mechanism.

[0013] The visualization module is used to acquire real-time monitoring data, historical monitoring data, and operational status information based on the equipment operation status evaluation index system, and to visualize and display them based on the access requirements of multiple terminals.

[0014] Furthermore, the sensor group performs the following process for acquiring multiple types of monitoring parameters:

[0015] Based on the collection requirements of multiple types of monitoring parameters, the adaptation requirements of GIS equipment to sensors are obtained. Non-contact laser sensing devices are used as the core devices for displacement data acquisition, and vibration sensing devices and temperature sensing devices adapted to the electromagnetic environment of substations are selected as auxiliary devices to construct a sensing system that corresponds one-to-one with the monitoring parameters.

[0016] Obtain the structural characteristics of the GIS busbar compartment and expansion joints, configure modular installation components to adapt to the sensor group, and determine the deployment priority of each sensor based on the inducing factors of deformation of the GIS busbar expansion joints:

[0017] Based on the requirement of no interference between sensors, the deployment angle of each sensor is optimized. At the same time, the operating environment requirements of the GIS equipment are obtained, and an interference avoidance scheme for the sensors is designed based on the operating environment requirements.

[0018] Furthermore, the monitoring parameter acquisition module also includes:

[0019] Based on the deployment status of the sensor group, standard displacement calibration components, standard vibration sources and standard temperature sources are obtained, and each sensor is calibrated according to its type in combination with the acquisition requirements of multiple types of monitoring parameters.

[0020] Based on the data synchronization requirements, a unified time reference signal is generated and sent to each sensor to ensure that the timestamps of the collected data are consistent.

[0021] Furthermore, the categorized calibration of each sensor also includes:

[0022] Obtain the actual monitoring accuracy values ​​of each sensor before and after calibration. The actual monitoring accuracy values ​​are calculated based on the deviation between the standard source parameters and the sensor output parameters during the calibration process.

[0023] Obtain the preset accuracy thresholds corresponding to multiple types of monitoring parameters, compare the actual monitoring accuracy values ​​of each sensor with the corresponding preset accuracy thresholds, and count the frequency of actual monitoring accuracy values ​​exceeding the preset accuracy thresholds during the calibration process of a preset number of consecutive calibrations.

[0024] Obtain the accuracy compliance frequency benchmark range for each sensor within its historical calibration cycle, and compare the frequency of actual monitoring accuracy values ​​exceeding the preset accuracy threshold with the accuracy compliance frequency benchmark range.

[0025] If the frequency of occurrences exceeding the preset accuracy threshold is lower than the lower limit of the accuracy compliance frequency benchmark range, the sensor accuracy decay trend is determined to be abnormal, triggering an alarm for abnormal sensor operation. The alarm information is then transmitted to the preset interaction channel and simultaneously pushed to maintenance personnel.

[0026] Furthermore, the multi-data fusion processing module constructs a data fusion model, specifically including:

[0027] Based on multiple types of monitoring parameters and structural mechanics theory, finite element analysis models of local and overall structures of GIS equipment are established. Based on the monitoring data after type-calibration, the parameters of the finite element analysis models are initialized to match the actual operating state of the GIS equipment.

[0028] The stress distribution and deformation characteristics of the expansion joint of the GIS busbar compartment were analyzed based on the finite element analysis model after parameter initialization, and the stress and deformation calculation values ​​of the key structural points of the expansion joint were extracted.

[0029] The calculated stress and deformation values ​​are correlated and mapped with the real-time collected calibrated monitoring data to construct a data fusion model.

[0030] Furthermore, the intelligent monitoring and early warning module specifically includes:

[0031] The data receiving unit is used to receive calibrated multi-type monitoring data and the fusion analysis results of the data fusion model based on a remote data transmission network adapted to the electromagnetic environment of the substation, and classifies the multi-type monitoring data and fusion analysis results for distributed storage.

[0032] The anomaly determination unit is used to determine the deformation status of the GIS busbar expansion joint. Combined with the preset safe operation threshold of the GIS busbar expansion joint, it performs secondary verification on the real-time monitoring data and fusion analysis results to determine whether the equipment is at risk of abnormal deformation.

[0033] The deformation prediction unit is used to extract real-time temperature and vibration data, combine them with the finite element analysis model and the initial operating state parameters of unstable key structural points, predict the subsequent deformation trend of unstable key structural points, and generate deformation prediction curves and risk levels for a preset time period in the future.

[0034] Furthermore, the deformation prediction unit includes:

[0035] The deformation prediction curves and material property parameters and structural constraints of the unstable key structural points are obtained, and the deformation causes of the unstable key structural points are determined based on the material property parameters and structural constraints of the unstable key structural points.

[0036] The causes of deformation are matched with a pre-defined strategy knowledge base, and the preferred intervention type corresponding to the causes of deformation is determined based on the matching results. The intervention types include thermal compensation and structural constraint point adjustment.

[0037] When the preferred intervention type is thermal compensation, the adjacent region with the largest temperature gradient within a preset radius centered on the unstable critical structural point is determined based on the finite element analysis model. The location for applying thermal compensation is determined based on the adjacent region. At the same time, the thermal compensation amount is determined based on the deviation between the real-time temperature data and the preset ideal working temperature range of the unstable critical structural point, as well as the influence coefficient of unit temperature change on the deformation of the unstable critical structural point.

[0038] Meanwhile, when the preferred intervention type is structural constraint point adjustment, the deviation between the current displacement vector and the target displacement vector of the unstable key structural point is determined based on the finite element analysis model, and the adjustment vector for the structural constraint point is determined based on the deviation.

[0039] A virtual intervention scheme is obtained based on the location and value of thermal compensation application, structural constraint points, and corresponding adjustment vectors.

[0040] The virtual intervention scheme corresponding to the preferred intervention type and the material property parameters and structural constraints of the unstable key structural points are input into the finite element analysis model, and the preset time period is discretized into N consecutive time steps based on the input results.

[0041] Based on the finite element analysis model, the structural response of the unstable key structural point at each time step is determined sequentially according to the continuous time steps. Based on the structural response at each time step, the corresponding key deformation variables are determined. The key deformation variables at each time step are summarized to generate the deformation development trend curve of the unstable key structural point under the virtual intervention scheme.

[0042] The obtained deformation development trend curve is compared with the deformation prediction curve, and the intervention correction value of the virtual intervention plan is determined based on the comparison results.

[0043] The intervention correction value is evaluated based on the preset risk convergence assessment rules, and the virtual intervention plan is iteratively corrected when the preset requirements are not met until the preset requirements are met.

[0044] The iteratively revised virtual intervention scheme is associated and encapsulated with unstable critical structural points to obtain the execution operation and maintenance strategy.

[0045] Furthermore, the process of determining the deformation state of the expansion joints in the GIS busbar compartment includes:

[0046] Based on the initial monitoring data collected by the sensor group after classification and calibration, the initial operating status parameters of each key structural point of the GIS busbar expansion joint are obtained, including the initial displacement value and the initial strain value.

[0047] The real-time temperature and vibration data collected by the sensor group are input into the finite element analysis model to generate the stress prediction values ​​and corresponding stress setting ranges for each key structural point.

[0048] Based on the data fusion model, the initial operating state parameters of each key structural point are compared with the corresponding stress setting range. When the initial operating state parameters of the key structural point are within the setting range of the corresponding stress prediction value, the deformation state of the current key structural point is determined to be stable.

[0049] When the initial operating state parameters of a critical structural point exceed the corresponding stress setting range, the current deformation state of the critical structural point is determined to be unstable, and there is a potential deformation risk.

[0050] Based on the structural characteristics of the GIS busbar compartment and expansion joint, a three-dimensional model of the GIS busbar compartment and expansion joint is established. The key structural points that are determined to be unstable are visualized and marked in the three-dimensional model of the GIS busbar compartment and expansion joint to form risk marking areas.

[0051] Obtain the location information of the currently unstable critical structural points, and generate operation and maintenance positioning identifiers for risk-marked areas based on the location information, and associate them with the physical location codes in the substation GIS equipment ledger.

[0052] Furthermore, the deformation prediction unit also includes deformation prediction optimization of the 3D model of the GIS busbar expansion joint:

[0053] Retrieve historical deformation prediction data of the expansion joint of the GIS busbar compartment, as well as the historical actual monitoring data corresponding to the historical deformation prediction data. Calculate the deviation between each set of historical deformation prediction data and the corresponding historical actual monitoring data to obtain the historical prediction deviation value.

[0054] The frequency of historical prediction deviation values ​​exceeding the preset error threshold within a preset period is counted. When the frequency of exceeding the preset error threshold exceeds the preset frequency, the prediction accuracy of the current GIS busbar expansion joint 3D model is determined to be low.

[0055] Calculate the mean prediction deviation of all historical deformation prediction data that exceed the preset error threshold within the preset period, and obtain the error correction coefficient.

[0056] Read the real-time predicted value of the deformation prediction of the expansion joint of the GIS busbar compartment, correct the real-time predicted value according to the error correction coefficient, and store the corrected real-time predicted value with the real-time monitoring data according to the correction result.

[0057] Furthermore, the error correction coefficients are obtained, including:

[0058] Obtain all historical prediction deviation values ​​that exceed the preset error threshold within the preset period, and calculate the coefficient of variation of the historical prediction deviation based on the historical prediction deviation values;

[0059] Meanwhile, the error correction coefficient is obtained by acquiring the average prediction deviation of all historical deformation prediction data exceeding the preset error threshold within the preset period, and the obtained error correction coefficient is optimized based on the obtained discrete coefficient to obtain the standard error correction coefficient.

[0060] The real-time predicted value of the deformation prediction of the expansion joint of the GIS busbar compartment is corrected based on the standard error correction coefficient to obtain the optimized real-time predicted value.

[0061] Compared with the prior art, the beneficial effects of the present invention are:

[0062] This invention utilizes multi-source sensor calibration based on standard sources and unified time reference synchronization, coupled with accuracy trend monitoring and anomaly warning structures, to avoid monitoring data failure caused by sensor accuracy decay. It addresses the problems of traditional calibration, which only provides single-correction and lacks trend prediction, ensuring the accuracy and effectiveness of multi-source monitoring data. By employing finite element analysis models and multi-source data fusion design, it eliminates the limitations of single-sensor monitoring and data redundancy, deeply exploring the intrinsic correlation between parameters and expansion joint deformation. Combined with a dual judgment mechanism and 3D model annotation, it accurately identifies abnormal deformation risks, breaking through the bottleneck of traditional monitoring that relies solely on single data. Through an electromagnetic interference-resistant remote transmission network, dynamic deformation prediction, and a multi-terminal visualization structure, it achieves classified storage of monitoring data, early prediction of potential risks, and precise location of hidden dangers, assisting maintenance personnel in quickly developing response plans. This overcomes the limitations of traditional passive early warning and ambiguous maintenance positioning, ensuring the safe and stable operation of GIS equipment. Attached Figure Description

[0063] Figure 1 This is a module diagram of the GIS busbar expansion joint deformation online monitoring system of the present invention;

[0064] Figure 2 This is a flowchart of the online monitoring process for deformation of the expansion joints in the GIS busbar compartment according to the present invention. Detailed Implementation

[0065] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0066] To address the technical challenges of existing GIS monitoring systems not covering online monitoring of busbar expansion joint deformation, relying on single data points for inaccurate assessments, and the low efficiency and susceptibility to misjudgments associated with offline detection, as well as the lack of relevant monitoring and early warning standards and quantitative standards for deformation causes, thus hindering early warning capabilities, please refer to [link to relevant documentation]. Figures 1-2 This embodiment provides the following technical solution:

[0067] The GIS busbar expansion joint deformation online monitoring system based on multi-source data fusion includes:

[0068] The monitoring parameter acquisition module is used to collect various types of monitoring parameters that characterize the operating status of the GIS busbar compartment and expansion joint in real time based on the sensor group. These include displacement data, vibration data, temperature data, and environmental parameters, specifically the busbar compartment's displacement to the ground, the three-dimensional displacement of the expansion joint, the strain of key stress-bearing parts of the busbar compartment, the vibration signal of the GIS equipment, the temperature of key areas of the busbar compartment, and the on-site environmental parameters.

[0069] The multi-data fusion processing module is used to apply multi-source data fusion technology, combine structural mechanics and GIS equipment fault diagnosis theory to construct a data fusion model, perform fusion analysis on the collected multi-type monitoring parameters, eliminate the limitations of single sensor monitoring and the redundancy and contradiction between data, and explore the intrinsic correlation between each parameter and the deformation of the GIS busbar expansion joint; and establish an equipment operation status evaluation index system based on the fusion analysis results to determine the deformation status of the GIS busbar expansion joint, such as normal operation and potential deformation risk.

[0070] In this embodiment, the construction of a data fusion model specifically includes:

[0071] Based on multiple types of monitoring parameters and structural mechanics theory, finite element analysis models of local and overall structures of GIS equipment are established. Based on the monitoring data after type-calibration, the parameters of the finite element analysis models are initialized to match the actual operating state of the GIS equipment.

[0072] The stress distribution and deformation characteristics of the expansion joint of the GIS busbar compartment were analyzed based on the finite element analysis model after parameter initialization, and the stress and deformation calculation values ​​of the key structural points of the expansion joint were extracted.

[0073] The calculated stress and deformation values ​​are correlated and mapped with the real-time collected calibrated monitoring data to construct a data fusion model, providing model support for subsequent deformation state determination;

[0074] The intelligent monitoring and early warning module is used to build a remote data transmission network adapted to the electromagnetic environment of the substation. It classifies and stores the fusion analysis results and pre-processed multi-type monitoring parameters. At the same time, based on the preset safe operation threshold of the GIS busbar expansion joint and the fusion analysis results, it monitors the deformation of the expansion joint and changes in the operating status of the equipment in real time. When the monitoring data exceeds the safety threshold, the early warning mechanism is triggered, and the early warning information is sent to the operation and maintenance personnel through the preset interaction channel based on the early warning mechanism.

[0075] The visualization module is used to acquire real-time monitoring data, historical monitoring data, and operational status information based on the equipment operational status evaluation index system. The real-time monitoring data and historical monitoring data come from a distributed data storage unit, and the operational status information is derived from the equipment operational status evaluation index system. It is also visualized based on the access requirements of multiple terminals, allowing maintenance personnel to view monitoring data in real time, trace historical data, and visualize the equipment operational status, providing decision support for the formulation of maintenance plans and fault handling for the GIS busbar expansion joint.

[0076] In this embodiment, multiple monitoring parameters characterizing the operating status of the GIS busbar compartment and expansion joint are collected in real time by a sensor array. A finite element analysis model of the local and overall structure of the GIS equipment is constructed by combining structural mechanics and GIS equipment fault diagnosis theory. This model explores the intrinsic correlation between each parameter and the deformation of the GIS busbar compartment expansion joint, effectively overcoming the limitations of traditional single-sensor monitoring. The accuracy and reliability of GIS busbar compartment expansion joint deformation monitoring are improved through multi-source data fusion and finite element analysis models. Furthermore, an equipment operating status evaluation index system is established based on the fusion analysis results to assess the deformation status of the GIS busbar compartment expansion joint, including normal operation and potential deformation risks. The system makes judgments to provide decision support for the formulation of operation and maintenance plans and fault handling of GIS busbar expansion joints. Simultaneously, based on preset safe operation thresholds for GIS busbar expansion joints and fusion analysis results, it monitors expansion joint deformation and equipment operating status changes in real time. When monitoring data exceeds the safety threshold, an early warning mechanism is triggered, enabling timely identification of potential deformation risks and triggering of warnings. Warning information is sent to operation and maintenance personnel through preset interactive channels, allowing them to view monitoring data in real time, trace historical data, and visualize equipment operating status. This visualization provides intuitive and efficient decision-making support for operation and maintenance work, ensuring the safe and stable operation of GIS equipment.

[0077] In this embodiment, the sensor group acquires multiple types of monitoring parameters as follows:

[0078] Based on the collection requirements of various monitoring parameters, the adaptation requirements of GIS equipment to sensors are obtained. Non-contact laser sensing devices are used as the core devices for displacement data acquisition. Based on the principle of laser pulse reflection ranging, the high-precision monitoring requirements of the three-dimensional displacement of the expansion joint and the ground displacement of the busbar compartment are matched. Vibration sensing devices and temperature sensing devices adapted to the electromagnetic environment of the substation are selected as auxiliary devices to match the collection requirements of vibration signals of GIS equipment and temperature of key areas of the busbar compartment, respectively, and a sensing system corresponding to the monitoring parameters is constructed.

[0079] Obtain the structural characteristics of the GIS busbar compartment and expansion joint, configure modular installation components to adapt to the sensor group, and design adaptable modular installation components based on the structural characteristics. For non-contact laser sensing devices, design a split clamp containing a transmitting unit and a reflecting unit to adapt to the installation space of the flange plate. For vibration sensing devices and temperature sensing devices, design an adhesive or clamp-type clamp to adapt to the curved / planar structure of the busbar compartment shell and the key stress parts of the expansion joint.

[0080] All installation components are made of high-strength insulating materials. Based on the insulation performance requirements of GIS equipment, it is ensured that there is no electrical coupling between the installation components and the GIS metal shell and insulating parts, while not damaging the sealing structure of the GIS equipment.

[0081] Based on the inducing factors of GIS busbar expansion joint deformation, such as temperature field changes, stress concentration, and foundation settlement, the deployment priority of each sensor is determined:

[0082] The transmitting and reflecting units of the non-contact laser sensing device are arranged on the flanges on both sides of the expansion joint to ensure that the laser optical path covers the three-dimensional displacement monitoring range of the expansion joint.

[0083] Vibration sensors are deployed at stress concentration points such as busbar ends and bends to directly collect vibration signals under stress.

[0084] Temperature sensors are placed close to the weld seams, sealing surfaces, and bellows surfaces of the busbar compartment to collect temperature change data.

[0085] Based on the requirement of no interference between sensors, the placement angle of each sensor is optimized to avoid laser light path obstruction and vibration signal crosstalk; at the same time, the operating environment requirements of GIS equipment are obtained, including insulation performance, electromagnetic compatibility, and sealing performance, and interference avoidance schemes for sensors are designed based on the operating environment requirements.

[0086] In this embodiment, the sensor power supply module uses an isolated power supply to avoid electrical interference with the GIS equipment power supply system. The signal transmission line uses shielded cable and is laid along the non-high voltage area of ​​the GIS equipment to avoid interference from alternating electric fields. The non-contact laser sensing device selects a laser wavelength with no ionizing radiation to ensure that it does not affect the insulation performance of SF6 gas. The installation process uses a detachable connection and does not involve drilling or welding the GIS metal shell to avoid damaging the sealing structure and causing SF6 gas leakage.

[0087] In this embodiment, the sensor group adopts a sensor system with a non-contact laser sensing device as the core and vibration sensing device and temperature sensing device as auxiliary. The sensors are deployed by the installation components adapted to the structure of GIS equipment, so as to realize the real-time and high-precision acquisition of monitoring parameters while avoiding interference with the insulation performance and mechanical stability of GIS equipment.

[0088] In this embodiment, the monitoring parameter acquisition module further includes:

[0089] Based on the deployment status of the sensor group, standard displacement calibration components, standard vibration sources and standard temperature sources are obtained, and each sensor is calibrated according to its type in combination with the acquisition requirements of multiple types of monitoring parameters.

[0090] For the non-contact laser sensing device, the known displacement value of the standard displacement calibrator is used as a reference to correct the ranging deviation; for the vibration sensing device and the temperature sensing device, the output data is verified by the standard vibration source and the standard temperature source, respectively.

[0091] Based on the data synchronization requirements, a unified time reference signal is generated and sent to each sensor to ensure that the timestamps of the collected data are consistent and to avoid synchronization errors when fusing multi-source data.

[0092] In this embodiment, the categorized calibration of each sensor also includes:

[0093] Obtain the actual monitoring accuracy values ​​of each sensor before and after calibration. The actual monitoring accuracy values ​​are calculated based on the deviation between the standard source parameters and the sensor output parameters during the calibration process.

[0094] Preset accuracy thresholds for various monitoring parameters are obtained and set based on the deformation judgment requirements of the expansion joint of the GIS busbar compartment. The preset accuracy threshold for displacement data matches the minimum effective judgment unit for three-dimensional displacement monitoring of the expansion joint. The preset accuracy thresholds for vibration data and temperature data match the minimum monitoring accuracy for the sensitivity of vibration signal anomaly identification of GIS equipment and the correlation between temperature change and deformation in key areas of the busbar compartment, respectively. The actual monitoring accuracy values ​​of each sensor are compared with the corresponding preset accuracy thresholds. During the calibration process of a series of preset times, the frequency of actual monitoring accuracy values ​​exceeding the preset accuracy thresholds is counted.

[0095] Obtain the accuracy compliance frequency benchmark range for each sensor within its historical calibration cycle, based on the historical calibration records of similar sensors on GIS equipment; compare the frequency of actual monitoring accuracy values ​​exceeding the preset accuracy threshold with the accuracy compliance frequency benchmark range.

[0096] If the frequency of occurrences exceeding the preset accuracy threshold is lower than the lower limit of the accuracy compliance frequency benchmark range, the sensor accuracy decay trend is determined to be abnormal, triggering an alarm for abnormal sensor operation. The alarm information is then transmitted to a preset interaction channel, including the type of abnormal sensor, the frequency of exceeding the threshold, and details of the deviation between the actual monitoring accuracy value and the preset accuracy threshold. This information is also simultaneously pushed to maintenance personnel, prompting them to promptly repair or replace the abnormal sensor to avoid monitoring data failure due to sensor accuracy issues.

[0097] In this embodiment, the multi-source sensor categorized calibration and accuracy anomaly judgment structure, through sensor standard source calibration adapted to GIS equipment and unified time reference synchronization, combined with accuracy trend monitoring and anomaly alarm, avoids monitoring data failure caused by sensor accuracy decay, ensures the reliability of multi-source data acquisition, and solves the problem of traditional calibration only providing single correction and lacking trend prediction. The finite element model and multi-source data fusion structure design eliminates the limitations of single sensor monitoring and data redundancy, deeply explores the intrinsic relationship between parameters and deformation, improves the accuracy of expansion joint deformation state judgment, and breaks through the monitoring bottleneck of traditional single data reliance. The anti-electromagnetic interference remote transmission and multi-terminal visualization structure realizes classified storage of monitoring data, real-time early warning and visualization presentation, helping operation and maintenance personnel to quickly locate risks, formulate disposal plans, ensure the safe and stable operation of GIS equipment, and improve the efficiency and pertinence of operation and maintenance decisions.

[0098] In this embodiment, the intelligent monitoring and early warning module specifically includes:

[0099] The data receiving unit is used to receive calibrated multi-type monitoring data and the fusion analysis results of the data fusion model based on a remote data transmission network adapted to the electromagnetic environment of the substation; and to classify and distribute the data and results to ensure that the data storage format is consistent with that of historical monitoring data.

[0100] The anomaly determination unit is used to determine the deformation status of the GIS busbar expansion joint, including information on stable / unstable key structural points, marked areas and area indexes. Combined with preset safe operation thresholds for the GIS busbar expansion joint, such as stress thresholds and displacement thresholds, the unit performs secondary verification on real-time monitoring data and fusion analysis results. When the number of unstable key structural points exceeds the preset number threshold, or the stress / displacement value of a certain key structural point exceeds the safe operation threshold, the unit determines that the equipment has an abnormal deformation risk.

[0101] The deformation prediction unit is used to predict the subsequent deformation trend of unstable key structural points based on the finite element analysis model and the initial stress value of unstable key structural points, and input real-time collected temperature data and vibration data to reflect the temperature field changes and stress concentration trends. It generates deformation prediction curves and risk levels (low / medium / high) for a preset time period in the future, providing a predictive basis for triggering the early warning mechanism.

[0102] In this embodiment, the process of determining the deformation state of the expansion joint of the GIS busbar compartment includes:

[0103] Based on the initial monitoring data collected by the sensor group after classification and calibration, the initial operating status parameters of each key structural point of the GIS busbar expansion joint are obtained, including the initial displacement value and the initial strain value.

[0104] The real-time temperature and vibration data collected by the sensor group are input into the finite element analysis model to generate the stress prediction value and corresponding stress setting range of each key structural point, based on the GIS equipment structural mechanics design standard and the expansion joint material tolerance limit setting.

[0105] Based on the data fusion model, the initial operating state parameters of each key structural point are compared with the corresponding stress setting range. When the initial operating state parameters of the key structural point are within the setting range of the corresponding stress prediction value, the deformation state of the current key structural point is determined to be stable.

[0106] When the initial operating state parameters of a critical structural point exceed the corresponding stress setting range, the deformation state of the current critical structural point is determined to be unstable, and there is a potential deformation risk, such as weld cracking or bellows fatigue damage risk.

[0107] Based on the structural characteristics of the GIS busbar compartment and expansion joint, a three-dimensional model of the GIS busbar compartment and expansion joint is established. The key structural points that are determined to be unstable are visualized and marked in the three-dimensional model of the GIS busbar compartment and expansion joint to form risk marking areas.

[0108] The system acquires the location information of currently unstable critical structural points and generates operation and maintenance positioning identifiers for risk-marked areas based on the location information. These identifiers are then linked to the physical location codes in the substation's GIS equipment ledger to facilitate subsequent operation and maintenance positioning. This allows operation and maintenance personnel to quickly locate potential fault locations.

[0109] In this embodiment, a remote transmission and distributed storage structure adapted to the electromagnetic environment of the substation is used to achieve stable reception and classified storage of calibrated monitoring data and fusion analysis results. This ensures that the data storage format is consistent with historical data, guaranteeing the anti-interference capability and historical compatibility of data transmission. The dual judgment structure of critical structural point status determination and secondary verification of safety thresholds, combined with the comparison results of initial operating state parameters and stress setting range, accurately identifies abnormal deformation risks, avoiding false alarms or missed alarms caused by a single judgment dimension, and improving the reliability of risk judgment. A three-dimensional model is constructed based on the structural features of GIS equipment and unstable critical structural points are marked. Location identifiers are generated by associating with the physical codes of the equipment ledger, helping maintenance personnel to quickly locate potential hazards. The dynamic capture of temperature field changes and stress concentration trends generates deformation prediction curves and risk levels, providing a forward-looking basis for the early warning mechanism. This breaks through the limitations of traditional passive early warning that relies solely on real-time monitoring, and enables early prediction of potential risks.

[0110] In this embodiment, the deformation prediction unit further includes deformation prediction optimization of the 3D model of the GIS busbar expansion joint:

[0111] Retrieve historical deformation prediction data of the expansion joint of the GIS busbar compartment, as well as the historical actual monitoring data corresponding to the historical deformation prediction data. Calculate the deviation between each set of historical deformation prediction data and the corresponding historical actual monitoring data to obtain the historical prediction deviation value.

[0112] The frequency of historical prediction deviation values ​​exceeding the preset error threshold within a preset period is counted. When the frequency of exceeding the preset error threshold exceeds the preset frequency, the prediction accuracy of the current GIS busbar expansion joint 3D model is determined to be low.

[0113] Calculate the mean prediction deviation of all historical deformation prediction data that exceed the preset error threshold within the preset period, and obtain the error correction coefficient.

[0114] Read the real-time predicted value of the deformation prediction of the expansion joint of the GIS busbar compartment, correct the real-time predicted value according to the error correction coefficient, and store the corrected real-time predicted value with the real-time monitoring data according to the correction result.

[0115] In this embodiment, by retrieving historical deformation prediction data and corresponding historical actual monitoring data and calculating the deviation, the accuracy of the prediction model is retrospectively evaluated. The accuracy of the model prediction is judged based on the frequency of deviation exceeding the threshold within a preset period. The downward trend of the 3D model prediction accuracy can be identified in a timely manner, avoiding prediction failure due to insufficient model accuracy. This breaks through the limitations of traditional methods that rely solely on single prediction results and cannot actively determine the applicability of the model. Correction coefficients are calculated for deviation data exceeding the threshold, and the current real-time prediction value is specifically corrected to improve the fit between the prediction value and the actual situation. This solves the problem that traditional predictions lack a dynamic correction mechanism and the accuracy is difficult to guarantee continuously. Moreover, the correction results are correlated with the actual monitoring data, avoiding the lack of optimization basis caused by isolated data.

[0116] In this embodiment, the deformation prediction unit includes:

[0117] The deformation prediction curves and material property parameters and structural constraints of the unstable key structural points are obtained, and the deformation causes of the unstable key structural points are determined based on the material property parameters and structural constraints of the unstable key structural points.

[0118] The causes of deformation are matched with a pre-defined strategy knowledge base, and the preferred intervention type corresponding to the causes of deformation is determined based on the matching results. The intervention types include thermal compensation and structural constraint point adjustment.

[0119] When the preferred intervention type is thermal compensation, the adjacent region with the largest temperature gradient within a preset radius centered on the unstable critical structural point is determined based on the finite element analysis model. The location for applying thermal compensation is determined based on the adjacent region. At the same time, the thermal compensation amount is determined based on the deviation between the real-time temperature data and the preset ideal working temperature range of the unstable critical structural point, as well as the influence coefficient of unit temperature change on the deformation of the unstable critical structural point.

[0120] Meanwhile, when the preferred intervention type is structural constraint point adjustment, the deviation between the current displacement vector and the target displacement vector of the unstable key structural point is determined based on the finite element analysis model, and the adjustment vector for the structural constraint point is determined based on the deviation.

[0121] A virtual intervention scheme is obtained based on the location and value of thermal compensation application, structural constraint points, and corresponding adjustment vectors.

[0122] The virtual intervention scheme corresponding to the preferred intervention type and the material property parameters and structural constraints of the unstable key structural points are input into the finite element analysis model, and the preset time period is discretized into N consecutive time steps based on the input results.

[0123] Based on the finite element analysis model, the structural response of the unstable key structural point at each time step is determined sequentially according to the continuous time steps. Based on the structural response at each time step, the corresponding key deformation variables are determined. The key deformation variables at each time step are summarized to generate the deformation development trend curve of the unstable key structural point under the virtual intervention scheme.

[0124] The obtained deformation development trend curve is compared with the deformation prediction curve, and the intervention correction value of the virtual intervention plan is determined based on the comparison results.

[0125] The intervention correction value is evaluated based on the preset risk convergence assessment rules, and the virtual intervention plan is iteratively corrected when the preset requirements are not met until the preset requirements are met.

[0126] The iteratively revised virtual intervention scheme is associated and encapsulated with unstable critical structural points to obtain the execution operation and maintenance strategy.

[0127] In this embodiment, the deformation prediction curve refers to a graphical representation of the future deformation trend of unstable key structural points predicted by an algorithm based on historical monitoring data.

[0128] In this embodiment, the unstable critical structural point refers to the specific location in the expansion joint of the GIS busbar compartment that is prone to deformation and has a significant impact on the overall structural safety.

[0129] In this embodiment, the material property parameters refer to the physical property parameters of the material constituted by the unstable critical structural points, such as elastic modulus and coefficient of thermal expansion.

[0130] In this embodiment, structural constraints refer to the external constraints that unstable critical structural points are subject to during installation or operation, such as fixed supports and connection methods.

[0131] In this embodiment, the cause of deformation refers to the main factors that cause deformation of unstable critical structural points, such as temperature changes and mechanical stress.

[0132] In this embodiment, the preset strategy knowledge base refers to a database of intervention strategies stored in advance for different causes and conditions of deformation.

[0133] In this embodiment, the preferred intervention type refers to the most suitable intervention category determined by matching the deformation cause with the strategy knowledge base, such as thermal compensation or structural constraint point adjustment.

[0134] In this embodiment, thermal compensation refers to an intervention method that uses temperature adjustment to counteract or reduce deformation caused by thermal effects.

[0135] In this embodiment, structural constraint point adjustment refers to an intervention method that adjusts deformation by changing the position or state of structural constraint points.

[0136] In this embodiment, the finite element analysis model refers to a numerical calculation model used to simulate and analyze the response of a structure under load.

[0137] In this embodiment, the adjacent region with the largest temperature gradient within the preset radius refers to the surrounding region with the most significant temperature change within a certain distance from the unstable critical structural point.

[0138] In this embodiment, the location where thermal compensation is applied refers to the specific location where thermal compensation measures are implemented.

[0139] In this embodiment, the thermal compensation value refers to the amount of temperature adjustment that needs to be applied to compensate for deformation.

[0140] In this embodiment, the deviation between the current displacement vector and the target displacement vector refers to the difference between the actual displacement and the ideal displacement of the unstable critical structural point.

[0141] In this embodiment, the adjustment vector refers to the direction and magnitude of the adjustment of the structural constraint points.

[0142] In this embodiment, a virtual intervention scheme refers to a computationally generated, yet-to-be-implemented, intervention plan.

[0143] In this embodiment, a time step refers to a single time interval after discretizing a preset time period in finite element analysis.

[0144] In this embodiment, structural response refers to the deformation or stress state of unstable critical structural points at each time step under virtual intervention.

[0145] In this embodiment, the critical deformation refers to the main deformation measurement value of the unstable critical structural point at each time step.

[0146] In this embodiment, the deformation development trend curve refers to the curve of the deformation of unstable key structural points changing over time after simulated virtual intervention.

[0147] In this embodiment, the intervention correction value refers to the adjustment amount of the virtual intervention scheme parameters based on the comparison between the deformation development trend curve and the deformation prediction curve.

[0148] In this embodiment, the preset risk convergence assessment rule refers to the set of rules used to assess whether the risk of the intervention plan meets the acceptable standard.

[0149] In this embodiment, iterative correction refers to the process of repeatedly adjusting the virtual intervention plan until the preset requirements are met.

[0150] In this embodiment, the operation and maintenance strategy refers to the finalized and executable intervention plan for the deformation of unstable critical structural points.

[0151] The working principle and beneficial effects of the above technical solution are as follows: by intelligently analyzing the causes of deformation and automatically matching the optimal intervention strategy, accurate diagnosis and efficient handling of deformation are achieved. At the same time, based on finite element simulation, virtual intervention schemes can be generated and their effects predicted. Through comparative evaluation and iterative optimization, the safety and effectiveness of intervention measures are ensured, significantly improving the predictability of deformation monitoring and the pertinence of operation and maintenance, reducing the error and cost of manual intervention, enhancing the reliability and lifespan of equipment operation, and thus ensuring the safety and stability of power facilities.

[0152] In this embodiment, obtaining the error correction coefficient includes:

[0153] Obtain all historical prediction deviation values ​​that exceed the preset error threshold within the preset period, and calculate the coefficient of variation of the historical prediction deviation based on the historical prediction deviation values;

[0154]

[0155] in, represents the mean of historical prediction deviations; m represents the total number of historical prediction deviations; j represents the index of the current historical prediction deviation, and its value ranges from [1, m]. This represents the j-th historical prediction deviation value; The standard deviation of historical forecast deviations; The coefficient of variation represents the deviation from historical predictions;

[0156] Meanwhile, the error correction coefficient is obtained by acquiring the average prediction deviation of all historical deformation prediction data exceeding the preset error threshold within the preset period, and the obtained error correction coefficient is optimized based on the obtained discrete coefficient to obtain the standard error correction coefficient. ;

[0157] in, This represents the standard error correction coefficient after optimizing the error correction coefficient based on the discrete coefficients; This represents the error correction coefficient obtained based on the average prediction deviation of all historical deformation prediction data that exceed the preset error threshold within a preset period. Indicates the weighting factor for the degree of dispersion;

[0158] The real-time predicted value of the deformation prediction of the expansion joint of the GIS busbar compartment is corrected based on the standard error correction coefficient to obtain the optimized real-time predicted value.

[0159] In this embodiment, the coefficient of variation of the historical prediction deviation is used to reflect the degree of fluctuation of the deviation value.

[0160] In this embodiment, It is determined based on historical data, and its value range is [0.3, 0.5], which is used to adjust the degree of influence of the dispersion coefficient on the correction coefficient.

[0161] The working principle and beneficial effects of the above technical solution are as follows: By considering the dispersion of historical deviation values, the influence of drastic fluctuations in historical deviations on the final determined error correction coefficient is avoided. By calculating the dispersion coefficient of historical prediction deviations, and correcting the error coefficient obtained by considering only the mean value of prediction deviations based on the dispersion coefficient, the over-correction or under-correction of real-time prediction values ​​is avoided, making the error correction coefficient more consistent with the actual distribution law of historical deviations, further ensuring the accuracy of real-time prediction values, and improving the online monitoring effect of deformation of expansion joints in GIS busbar compartments.

[0162] The above description is only a preferred embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any equivalent substitutions or modifications made by those skilled in the art within the scope of the technology disclosed in the present invention, based on the technical solution and inventive concept of the present invention, should be covered within the scope of protection of the present invention.

Claims

1. An online monitoring system for deformation of expansion joints in GIS busbar compartments based on multi-source data fusion, characterized in that, include: The monitoring parameter acquisition module is used to collect multiple types of monitoring parameters in real time based on the sensor group, including displacement data, vibration data, temperature data and environmental parameters; The multi-data fusion processing module is used to build a data fusion model, perform fusion analysis on the collected multi-type monitoring parameters, and establish an equipment operation status evaluation index system based on the fusion analysis results. The intelligent monitoring and early warning module is used to monitor the deformation of the expansion joint and changes in the operating status of the equipment in real time based on the preset safe operation threshold of the GIS busbar compartment expansion joint and the fusion analysis results. When the monitoring data exceeds the safety threshold, an early warning mechanism is triggered. The visualization module is used to acquire real-time monitoring data, historical monitoring data, and operational status information based on the equipment operational status evaluation index system, and to display them visually. The intelligent monitoring and early warning module specifically includes: The data receiving unit is used to receive calibrated multi-type monitoring data and the fusion analysis results of the data fusion model based on a remote data transmission network adapted to the electromagnetic environment of the substation. The anomaly determination unit is used to determine the deformation status of the GIS busbar expansion joint. Combined with the preset safe operation threshold of the GIS busbar expansion joint, it performs secondary verification on the real-time monitoring data and fusion analysis results to determine whether there is a risk of abnormal deformation. The deformation prediction unit is used to extract real-time temperature and vibration data, and combine them with a finite element analysis model to predict the subsequent deformation trend of unstable key structural points, generating deformation prediction curves and risk levels for a preset time period in the future; Deformation prediction unit, including: The deformation prediction curves and material property parameters and structural constraints of the unstable key structural points are obtained, and the deformation causes of the unstable key structural points are determined based on the material property parameters and structural constraints of the unstable key structural points. The causes of deformation are matched with a pre-defined strategy knowledge base, and the preferred intervention type corresponding to the causes of deformation is determined based on the matching results. The intervention types include thermal compensation and structural constraint point adjustment. When the preferred intervention type is thermal compensation, the adjacent region with the largest temperature gradient within a preset radius centered on the unstable critical structural point is determined based on the finite element analysis model. The location for applying thermal compensation is determined based on the adjacent region. At the same time, the thermal compensation amount is determined based on the deviation between the real-time temperature data and the preset ideal working temperature range of the unstable critical structural point, as well as the influence coefficient of unit temperature change on the deformation of the unstable critical structural point. Meanwhile, when the preferred intervention type is structural constraint point adjustment, the deviation between the current displacement vector and the target displacement vector of the unstable key structural point is determined based on the finite element analysis model, and the adjustment vector for the structural constraint point is determined based on the deviation. A virtual intervention scheme is obtained based on the location and value of thermal compensation application, structural constraint points, and corresponding adjustment vectors. The virtual intervention scheme corresponding to the preferred intervention type and the material property parameters and structural constraints of the unstable key structural points are input into the finite element analysis model, and the preset time period is discretized into N consecutive time steps based on the input results. Based on the finite element analysis model, the structural response of the unstable key structural point at each time step is determined sequentially according to the continuous time steps. Based on the structural response at each time step, the corresponding key deformation variables are determined. The key deformation variables at each time step are summarized to generate the deformation development trend curve of the unstable key structural point under the virtual intervention scheme. The obtained deformation development trend curve is compared with the deformation prediction curve, and the intervention correction value of the virtual intervention plan is determined based on the comparison results. The intervention correction value is evaluated based on the preset risk convergence assessment rules, and the virtual intervention plan is iteratively corrected when the preset requirements are not met until the preset requirements are met. The iteratively revised virtual intervention scheme is associated and encapsulated with unstable critical structural points to obtain the execution operation and maintenance strategy.

2. The GIS busbar expansion joint deformation online monitoring system based on multi-source data fusion as described in claim 1, characterized in that, The sensor group acquires multiple types of monitoring parameters as follows: Based on the collection requirements of multiple types of monitoring parameters, the adaptation requirements of GIS equipment to sensors are obtained. Non-contact laser sensing devices are used as the core devices for displacement data acquisition, and vibration sensing devices and temperature sensing devices adapted to the electromagnetic environment of substations are selected as auxiliary devices to construct a sensing system that corresponds one-to-one with the monitoring parameters. Obtain the structural characteristics of the GIS busbar compartment and expansion joint, configure modular installation components to adapt to the sensor group, and determine the deployment priority of each sensor based on the inducing factors of deformation of the GIS busbar expansion joint. Based on the requirement of no interference between sensors, the deployment angle of each sensor is optimized. At the same time, the operating environment requirements of the GIS equipment are obtained, and an interference avoidance scheme for the sensors is designed based on the operating environment requirements.

3. The GIS busbar expansion joint deformation online monitoring system based on multi-source data fusion as described in claim 2, characterized in that, The monitoring parameter acquisition module also includes: Based on the deployment status of the sensor group, standard displacement calibration components, standard vibration sources and standard temperature sources are obtained, and each sensor is calibrated according to its type in combination with the acquisition requirements of multiple types of monitoring parameters. Based on the data synchronization requirements, a unified time reference signal is generated and sent to each sensor to ensure that the timestamps of the collected data are consistent.

4. The GIS busbar expansion joint deformation online monitoring system based on multi-source data fusion as described in claim 3, characterized in that, The categorized calibration of each sensor also includes: Obtain the actual monitoring accuracy values ​​of each sensor before and after calibration. The actual monitoring accuracy values ​​are calculated based on the deviation between the standard source parameters and the sensor output parameters during the calibration process. Obtain the preset accuracy thresholds corresponding to multiple types of monitoring parameters, compare the actual monitoring accuracy values ​​of each sensor with the corresponding preset accuracy thresholds, and count the frequency of actual monitoring accuracy values ​​exceeding the preset accuracy thresholds during the calibration process of a preset number of consecutive calibrations. Obtain the accuracy compliance frequency benchmark range for each sensor within its historical calibration cycle, and compare the frequency of actual monitoring accuracy values ​​exceeding the preset accuracy threshold with the accuracy compliance frequency benchmark range. If the frequency of occurrences exceeding the preset accuracy threshold is lower than the lower limit of the accuracy compliance frequency benchmark range, the sensor accuracy decay trend is determined to be abnormal, triggering an alarm for abnormal sensor operation. The alarm information is then transmitted to the preset interaction channel and simultaneously pushed to maintenance personnel.

5. The GIS busbar expansion joint deformation online monitoring system based on multi-source data fusion as described in claim 1, characterized in that, The multi-data fusion processing module constructs a data fusion model, specifically including: Based on multiple types of monitoring parameters and structural mechanics theory, finite element analysis models of local and overall structures of GIS equipment are established. Based on the monitoring data after type-calibration, the parameters of the finite element analysis models are initialized to match the actual operating state of the GIS equipment. The stress distribution and deformation characteristics of the expansion joint of the GIS busbar compartment were analyzed based on the finite element analysis model after parameter initialization, and the stress and deformation calculation values ​​of the key structural points of the expansion joint were extracted. The calculated stress and deformation values ​​are correlated and mapped with the real-time collected calibrated monitoring data to construct a data fusion model.

6. The GIS busbar expansion joint deformation online monitoring system based on multi-source data fusion as described in claim 1, characterized in that, The process of determining the deformation state of the expansion joints in the GIS busbar compartment includes: Based on the initial monitoring data collected by the sensor group after classification and calibration, the initial operating status parameters of each key structural point of the GIS busbar expansion joint are obtained, including the initial displacement value and the initial strain value. The real-time temperature and vibration data collected by the sensor group are input into the finite element analysis model to generate the stress prediction values ​​and corresponding stress setting ranges for each key structural point. Based on the data fusion model, the initial operating state parameters of each key structural point are compared with the corresponding stress setting range. When the initial operating state parameters of the key structural point are within the setting range of the corresponding stress prediction value, the deformation state of the current key structural point is determined to be stable. When the initial operating state parameters of a critical structural point exceed the corresponding stress setting range, the current deformation state of the critical structural point is determined to be unstable, and there is a potential deformation risk. Based on the structural characteristics of the GIS busbar compartment and expansion joint, a three-dimensional model of the GIS busbar compartment and expansion joint is established. The key structural points that are determined to be unstable are visualized and marked in the three-dimensional model of the GIS busbar compartment and expansion joint to form risk marking areas. Obtain the location information of the currently unstable critical structural points, and generate operation and maintenance positioning identifiers for risk-marked areas based on the location information, and associate them with the physical location codes in the substation GIS equipment ledger.

7. The GIS busbar expansion joint deformation online monitoring system based on multi-source data fusion as described in claim 1, characterized in that, The deformation prediction unit also includes deformation prediction optimization of the 3D model of the expansion joint of the GIS busbar compartment: Retrieve historical deformation prediction data of the expansion joint of the GIS busbar compartment, as well as the historical actual monitoring data corresponding to the historical deformation prediction data. Calculate the deviation between each set of historical deformation prediction data and the corresponding historical actual monitoring data to obtain the historical prediction deviation value. The frequency of historical prediction deviation values ​​exceeding the preset error threshold within a preset period is counted. When the frequency of exceeding the preset error threshold exceeds the preset frequency, the prediction accuracy of the current GIS busbar expansion joint 3D model is determined to be low. Calculate the mean prediction deviation of all historical deformation prediction data that exceed the preset error threshold within the preset period, and obtain the error correction coefficient. Read the real-time predicted value of the deformation prediction of the expansion joint of the GIS busbar compartment, correct the real-time predicted value according to the error correction coefficient, and store the corrected real-time predicted value with the real-time monitoring data according to the correction result.

8. The GIS busbar expansion joint deformation online monitoring system based on multi-source data fusion as described in claim 7, characterized in that, To obtain the error correction coefficients, including: Obtain all historical prediction deviation values ​​that exceed the preset error threshold within the preset period, and calculate the coefficient of variation of the historical prediction deviation based on the historical prediction deviation values; Meanwhile, the error correction coefficient is obtained by acquiring the average prediction deviation of all historical deformation prediction data exceeding the preset error threshold within the preset period, and the obtained error correction coefficient is optimized based on the obtained discrete coefficient to obtain the standard error correction coefficient. The real-time predicted value of the deformation prediction of the expansion joint of the GIS busbar compartment is corrected based on the standard error correction coefficient to obtain the optimized real-time predicted value.

Images