Substation operation state inference and decision method and system based on cloud analysis

Abstract

The application provides a substation operation state inference and decision method and system based on cloud analysis, relates to the technical field of power system automation, and obtains the decomposition product concentration of target gas for insulation, gas purity, gas moisture content, environmental temperature and humidity data, and leakage current data and dielectric loss factor data inside the gas insulated switchgear of the substation; through an edge node, the obtained data is processed, the fused data is obtained, and the gas decomposition characteristics, insulation electrical parameters and environmental influence factors are extracted, so as to identify the insulation deterioration degree of the gas insulated switchgear through a cloud intelligent analysis platform, accordingly adjust the overvoltage protection threshold of the gas insulated switchgear, trigger the protection decision corresponding to the insulation deterioration degree, realize dynamic identification of the insulation deterioration degree and self-adaptive adjustment of the overvoltage protection threshold, and improve the accuracy of state evaluation and the intelligent level of operation and maintenance decision.

Description

Technical Field

[0001] This application relates to the field of power system automation technology, and in particular to a method and system for reasoning and decision-making about substation operation status based on cloud analytics. Background Technology

[0002] With the continuous advancement of smart grid construction, substations, as key nodes in the power system, face higher requirements for operational safety and reliability. Especially in high-voltage, high-capacity operating environments, the insulation condition of gas-insulated switchgear directly affects the stable operation of the entire substation. Because gas-insulated switchgear is subjected to high electric field strength and complex environmental conditions for extended periods, the internal insulation materials may slowly deteriorate due to factors such as partial discharge, thermal aging, or mechanical stress, eventually leading to serious faults.

[0003] Current research and applications employ distributed sensor networks combined with local expert systems to collect and preliminarily diagnose key operating parameters of GIS equipment. These solutions deploy multiple types of sensors at critical equipment locations to acquire data such as gas composition, partial discharge signals, and temperature and humidity. They then utilize pre-defined rule bases and threshold logic to perform status identification at the station, triggering alarms or suggestive operational commands when parameters exceed limits or show abnormal trends. However, existing solutions exhibit significant limitations when dealing with complex multivariate coupling relationships. They struggle to effectively integrate the nonlinear correlations between environmental disturbances, electrical stress, and chemical changes, leading to frequent misjudgments or omissions in status assessments. Furthermore, their decision-making mechanisms often rely on static threshold triggers, lacking continuous tracking and forward-looking projection of equipment degradation trends. This results in delayed or overly conservative protection responses, failing to provide strong support for differentiated operation and maintenance strategies. Summary of the Invention

[0004] The purpose of this application is to provide a cloud-based analysis-based substation operation status reasoning and decision-making method and system to address the significant limitations of existing technologies when facing complex multivariate coupling relationships, such as the tendency for misjudgments or omissions in status assessment; the lack of continuous tracking and forward-looking projection of equipment degradation trends; and the difficulty in achieving differentiated operation and maintenance.

[0005] To address the aforementioned technical problems, firstly, this application provides a cloud-based analysis-based method for substation operation status reasoning and decision-making, including:

[0006] To obtain the concentration of decomposition products, gas purity, gas moisture content, ambient temperature and humidity data of the gas-insulated switchgear inside the gas-insulated switchgear in the substation, as well as leakage current data and dielectric loss factor data of the gas-insulated switchgear.

[0007] By configuring edge nodes around the gas-insulated switchgear, outlier removal and feature fusion processing are performed on the concentration of decomposition products, the purity of the gas, the moisture content of the gas, the ambient temperature and humidity data, the leakage current data, and the dielectric loss factor data to obtain fused data, and the fused data is uploaded to the cloud intelligent analysis platform.

[0008] Gas decomposition characteristics, insulation electrical parameters, and environmental impact factors are extracted from the fused data. The fused data is then collaboratively analyzed through the cloud-based intelligent analysis platform to identify the degree of insulation degradation of gas-insulated switchgear.

[0009] Based on the degree of insulation degradation, the overvoltage protection threshold of the gas-insulated switchgear is adjusted to obtain the adjusted overvoltage protection threshold, thereby triggering a protection decision corresponding to the degree of insulation degradation. The protection decision includes protection decisions corresponding to early warning prompts, load transfer strategies, and emergency maintenance strategies, respectively.

[0010] Optionally, gas decomposition characteristics, insulation electrical parameters, and environmental impact factors are extracted from the fused data to perform collaborative analysis on the fused data through the cloud-based intelligent analysis platform, identifying the degree of insulation degradation of the gas-insulated switchgear, including:

[0011] Gas decomposition characteristics, insulation electrical parameters, and environmental impact factors are extracted from the fused data. The gas decomposition characteristics include the concentration variation trend of the decomposition products of the target gas for insulation, the stable range of gas purity, and the fluctuation characteristics of moisture content. The insulation electrical parameters include the variation law of leakage current amplitude and the distribution characteristics of dielectric loss factor. The environmental impact factors include the synergistic variation relationship of ambient temperature and humidity.

[0012] Using the cloud-based intelligent analysis platform, the gas decomposition characteristics and the insulation electrical parameters are cross-compared to determine the degree of correlation between the gas decomposition characteristics and the insulation electrical parameters. The degree of correlation is then matched with the synergistic change relationship in the environmental impact factors to obtain the synergistic relationship between the gas decomposition characteristics, the insulation electrical parameters, and the environmental impact factors.

[0013] The collaborative relationship is compared with the historical feature data of gas-insulated switchgear under different insulation states pre-stored in the cloud-based intelligent analysis platform to determine the insulation state corresponding to the collaborative relationship;

[0014] Based on the insulation condition, the degree of insulation degradation of the gas-insulated switchgear is determined.

[0015] Optionally, using the cloud-based intelligent analysis platform, the gas decomposition characteristics and the insulation electrical parameters are cross-compared to determine the degree of correlation between the gas decomposition characteristics and the insulation electrical parameters. This degree of correlation is then matched with the synergistic changes in the environmental impact factors to obtain the synergistic relationship between the gas decomposition characteristics, insulation electrical parameters, and environmental impact factors, including:

[0016] The statistics include the percentage of times when the trend of the concentration change of the decomposition products of the target gas for insulation is consistent with the trend of the change of the leakage current amplitude during the same data acquisition period; the percentage of times when the distribution characteristics of the dielectric loss factor are within the normal factor range corresponding to the stable range of gas purity; and the percentage of times when the fluctuation characteristics of the moisture content exceed the preset normal fluctuation range and the distribution characteristics of the dielectric loss factor show abnormal distribution.

[0017] The degree of correlation between gas decomposition characteristics and insulation electrical parameters is determined based on the percentage of occurrences, the percentage of values, and the percentage of abnormal occurrences.

[0018] The aforementioned cooperative change relationship is divided into high temperature and high humidity mode, high temperature and low humidity mode, low temperature and high humidity mode, and low temperature and low humidity mode.

[0019] The correlation degree is matched with the corresponding normal correlation degree ranges under the high temperature and high humidity mode, high temperature and low humidity mode, low temperature and high humidity mode, and low temperature and low humidity mode, respectively, to determine the reasonable correlation level corresponding to the correlation degree.

[0020] Based on the synergistic change relationship, the reasonable correlation level, and the correlation degree, a synergistic relationship is generated between gas decomposition characteristics, insulation electrical parameters, and environmental impact factors.

[0021] Optionally, the collaborative relationship is compared with historical characteristic data of gas-insulated switchgear under different insulation states pre-stored in the cloud-based intelligent analysis platform to determine the insulation state corresponding to the collaborative relationship, including:

[0022] The historical characteristic data of gas-insulated switchgear under different insulation states pre-stored in the cloud-based intelligent analysis platform are classified to obtain multiple historical collaborative relationship groups;

[0023] Select a target historical collaborative relationship group from the plurality of historical collaborative relationship groups that is consistent with the collaborative change relationship of the collaborative relationship;

[0024] The reasonable association level of the aforementioned collaborative relationship is compared with the reasonable association level in each target historical collaborative relationship group to select candidate historical collaborative relationship groups.

[0025] The percentage of times, percentage of values, and percentage of anomalies in the degree of association of the collaborative relationship are compared with the normal value range of each percentage in each candidate historical collaborative relationship group, and the number of matching items in each candidate historical collaborative relationship group where each percentage is within the corresponding normal value range is counted.

[0026] The insulation state of the gas-insulated switchgear corresponding to the candidate historical cooperative relationship group with the most matching items is selected as the insulation state corresponding to the cooperative relationship.

[0027] Optionally, the degree of insulation degradation of the gas-insulated switchgear is determined based on the insulation condition, including:

[0028] Key data were extracted from the insulation state.

[0029] The key data is compared with each insulation condition judgment standard to count the number of data items that meet the first insulation condition judgment standard, the second insulation condition judgment standard, and the third insulation condition judgment standard. The insulation condition corresponding to the largest number of data items is selected as the insulation degradation degree of the gas-insulated switchgear.

[0030] Optionally, by configuring edge nodes around the gas-insulated switchgear, outlier removal and feature fusion processing are performed on the concentration of decomposition products, the purity of the gas, the moisture content of the gas, the ambient temperature and humidity data, the leakage current data, and the dielectric loss factor data to obtain fused data, including:

[0031] The concentration of the decomposition products, the purity of the gas, and the moisture content of the gas are divided into multiple data groups, and the average fluctuation value within each data group is calculated.

[0032] By configuring edge nodes around the gas-insulated switchgear, the first abnormal data in each data group whose deviation from the average fluctuation value exceeds a preset deviation range is removed to form a first type of target data.

[0033] Calculate the difference in leakage current and the difference in dielectric loss factor between adjacent acquisition times, and set a first abnormal threshold and a second abnormal threshold based on all the differences in leakage current and dielectric loss factor.

[0034] Remove the last abnormal leakage current in the leakage current data where the leakage current difference value exceeds the first abnormal threshold, and remove the last dielectric loss factor in the dielectric loss factor data where the dielectric loss factor difference value exceeds the second abnormal threshold to obtain the second type of target data.

[0035] The second abnormal data that does not conform to the preset reasonable deviation range in the environmental temperature and humidity data is removed to obtain the third type of target data;

[0036] Temporal alignment and feature concatenation are performed on the first type of target data, the second type of target data, and the third type of target data to generate fused data.

[0037] Optionally, the overvoltage protection threshold of the gas-insulated switchgear is adjusted according to the degree of insulation degradation to obtain an adjusted overvoltage protection threshold, thereby triggering a protection decision corresponding to the degree of insulation degradation. The protection decision includes protection decisions corresponding to early warning alerts, load transfer strategies, and emergency maintenance strategies, respectively.

[0038] Based on the insulation withstand characteristics of gas-insulated switchgear, the degree of abnormality of key parameters corresponding to the degree of insulation degradation, and the operation and maintenance safety redundancy requirements of substations, voltage adjustment rules corresponding to the degree of insulation degradation are set.

[0039] Obtain the overvoltage protection threshold of the gas-insulated switchgear, and adjust the overvoltage protection threshold according to the voltage adjustment rules to obtain the adjusted overvoltage protection threshold;

[0040] When the insulation degradation level is slightly degraded, a protective decision is triggered to provide an early warning. The early warning is used to instruct the maintenance terminal to send the insulation degradation level, the adjusted overvoltage protection threshold, and the inspection cycle information.

[0041] When the insulation degradation level is moderate, the load transfer strategy is triggered to make a protection decision. The load transfer strategy is used to indicate that the target load within the risk range defined according to the adjusted overvoltage protection threshold is transferred to the preset backup line and to send an early warning to the operation and maintenance terminal.

[0042] When the insulation degradation reaches a severe degradation state, the protection decision of the emergency maintenance strategy is triggered. The emergency maintenance strategy is used to instruct the start of the equipment shutdown procedure, lock the adjusted overvoltage protection threshold, send the emergency maintenance command to the operation and maintenance terminal, and transfer the entire load to the preset backup line.

[0043] Secondly, this application provides a cloud-based analysis-based substation operation status reasoning and decision-making system, including:

[0044] The acquisition module is used to acquire the concentration of decomposition products of the target gas used for insulation inside the gas-insulated switchgear of the substation, gas purity, gas moisture content, ambient temperature and humidity data of the gas-insulated switchgear, as well as leakage current data and dielectric loss factor data of the gas-insulated switchgear.

[0045] The processing module is used to perform outlier removal and feature fusion processing on the concentration of decomposition products, the purity of gas, the moisture content of gas, the ambient temperature and humidity data, the leakage current data, and the dielectric loss factor data through edge nodes configured around the gas-insulated switchgear, to obtain fused data, and upload the fused data to the cloud intelligent analysis platform.

[0046] The identification module is used to extract gas decomposition characteristics, insulation electrical parameters and environmental impact factors from the fused data, so as to identify the degree of insulation degradation of the gas-insulated switchgear through collaborative analysis of the fused data by the cloud-based intelligent analysis platform.

[0047] The adjustment module is used to adjust the overvoltage protection threshold of the gas-insulated switchgear according to the degree of insulation degradation, so as to obtain the adjusted overvoltage protection threshold and trigger the protection decision corresponding to the degree of insulation degradation. The protection decision includes protection decisions corresponding to early warning prompts, load transfer strategies and emergency maintenance strategies, respectively.

[0048] Thirdly, this application provides an electronic device, comprising:

[0049] Memory, used to store computer programs;

[0050] A processor is used to execute the computer program to implement the steps of the cloud-based analysis-based substation operation status reasoning and decision-making method as described in the first aspect above.

[0051] Fourthly, this application provides a computer-readable storage medium storing a computer program that, when executed by a processor, can implement the steps of the cloud-based analysis-based substation operation status reasoning and decision-making method described in the first aspect above.

[0052] The cloud-based analysis-based substation operation status reasoning and decision-making method provided in this application synchronously collects multi-dimensional operational data such as the concentration of decomposition products of target gases inside gas-insulated switchgear, gas purity, moisture content, ambient temperature and humidity, leakage current, and dielectric loss factor. Outlier removal and feature fusion are performed at edge nodes around the equipment, improving the reliability and consistency of on-site data and reducing the impact of noise interference on the analysis results. The fused high-quality data is then uploaded to a cloud-based intelligent analysis platform. Leveraging the platform's powerful computing capabilities, in-depth collaborative analysis of gas decomposition characteristics, insulation electrical parameters, and environmental impact factors is conducted. This overcomes the limitations of local systems in handling multi-variable nonlinear coupling, enabling dynamic identification and trend tracking of equipment insulation degradation. Based on the degradation state, the overvoltage protection response threshold is adaptively adjusted to trigger graded protection measures, making early warning and maintenance decisions more aligned with the actual equipment condition and improving the accuracy of fault prediction and the flexibility of response strategies. For the insulation degradation identification process, gas decomposition characteristics, insulation electrical parameters, and environmental influencing factors are extracted from fused data. Further cross-comparison of gas decomposition and electrical parameters is conducted, and a collaborative relationship model is built using temperature and humidity change patterns. This distinguishes the characteristic differences between environmental disturbances and actual insulation degradation, avoiding misjudging environmental fluctuations as equipment failures. The collaborative relationship is then matched with historical feature data to accurately locate the current insulation state, thereby accurately determining the degradation stage of the equipment. This overcomes the shortcomings of traditional local expert systems, which rely on fixed rules, struggle to integrate multi-source information, and cannot adapt to equipment state evolution. It improves the ability to identify insulation degradation processes under complex operating conditions, enhances the robustness of state assessment and the foresight of decision-making, and realizes the transformation from static threshold alarms to dynamic, continuous, and intelligent state extrapolation and protection response. This provides technical support for the refined management and proactive operation and maintenance of key equipment in substations. Attached Figure Description

[0053] To more clearly illustrate the technical solutions of the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0054] Figure 1 A flowchart illustrating the cloud-based analysis-based substation operation status reasoning and decision-making method provided in this application embodiment;

[0055] Figure 2 A schematic diagram illustrating a specific implementation of the cloud-based analysis-based substation operation status reasoning and decision-making method provided in this application embodiment;

[0056] Figure 3This is a schematic diagram of the structure of a cloud-based substation operation status reasoning and decision-making system provided in an embodiment of this application. Detailed Implementation

[0057] To address the degradation of gas-insulated switchgear under long-term operation due to the combined effects of electric field stress, thermal effects, and environmental factors, which can lead to partial discharge and thermal decomposition of the internal insulation materials, resulting in strong coupling and nonlinear characteristics between trace gas composition changes and electrical parameter fluctuations, traditional methods struggle to accurately capture the evolution trend of early defects. This leads to lag and uncertainty in condition assessment, affecting the timeliness and accuracy of fault warnings. This application focuses on constructing a collaborative mechanism integrating edge preprocessing and cloud-based deep analysis to address the shortcomings of traditional local diagnostic methods in multi-source data fusion and dynamic decision-making. By deploying edge nodes around the equipment, preliminary cleaning and feature integration of multi-dimensional parameters such as gas decomposition products, gas purity, moisture content, ambient temperature and humidity, leakage current, and dielectric loss factor are achieved, overcoming data distortion caused by on-site interference. The processed high-quality data is then uploaded to a cloud platform. In the cloud, intelligent analysis models are used to deeply mine the fused data, extracting comprehensive features reflecting chemical changes, electrical performance degradation, and environmental impacts. A dynamic assessment model for insulation degradation is established, eliminating reliance on fixed thresholds and static rules. Furthermore, by combining the evaluation results, the response boundary of overvoltage protection is adaptively adjusted to trigger early warnings, load scheduling, or maintenance arrangements that match the actual state of the equipment, thereby realizing the transformation from passive alarm to proactive prevention and control, and improving the accuracy of state judgment and the flexibility of operation and maintenance strategies.

[0058] To enable those skilled in the art to better understand the present application, the present application will be further described in detail below with reference to the accompanying drawings and specific embodiments. Obviously, the described embodiments are merely some embodiments of the present application, and not all embodiments. Based on the embodiments in this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0059] The core of this application is to provide a cloud-based analysis-based method for substation operation status reasoning and decision-making. A flowchart illustrating one specific implementation method is shown below. Figure 1 As shown, the method includes:

[0060] Step 101: Obtain the concentration of decomposition products, gas purity, gas moisture content, ambient temperature and humidity data of the gas-insulated switchgear inside the substation, as well as the leakage current data and dielectric loss factor data of the gas-insulated switchgear.

[0061] In this step, the target gas for insulation refers to the specific gas filled inside the gas-insulated switchgear to achieve the insulation performance and arc-extinguishing function of the equipment. It is usually sulfur hexafluoride or a mixture thereof. Its insulation performance directly affects the operational safety of the gas-insulated switchgear. The internal insulation status of the equipment can be determined based on the monitoring of the parameters of this gas.

[0062] The concentration of decomposition products refers to the content of various decomposition substances produced when the target gas for insulation decomposes inside the gas-insulated switchgear due to factors such as partial discharge and thermal aging. It reflects the degree of deterioration of the insulation material inside the gas-insulated switchgear and potential fault conditions. Based on this concentration data, the insulation status of the equipment can be assessed.

[0063] Gas purity refers to the purity of the target gas used for insulation within gas-insulated switchgear, i.e., the proportion of the target gas among all gas components inside the equipment. Decreased gas purity will lead to a decrease in the insulation performance of the equipment. This parameter is used to measure whether the target gas used for insulation meets the insulation requirements for normal operation of the equipment.

[0064] Ambient temperature and humidity data refers to the temperature and humidity data of the operating environment of gas-insulated switchgear. Excessively high or low ambient temperatures and excessive humidity can affect the insulation performance and electrical parameter stability of the equipment. This data is used to analyze the impact of environmental factors on the operating status of the equipment.

[0065] Leakage current data refers to the magnitude and changes in the current leaked through the insulation of gas-insulated switchgear during operation. An increase in leakage current usually indicates a decrease in the insulation performance of the equipment, and this data is used to directly reflect the insulation status of the equipment.

[0066] The dielectric loss factor refers to the ratio of active power to reactive power generated by the dielectric loss of the insulating medium in gas-insulated switchgear under the action of an AC electric field. The larger the value, the greater the dielectric loss and the worse the insulation performance. This data is used to evaluate the degree of dielectric loss and insulation performance of the equipment.

[0067] In this embodiment, a gas sensor is deployed inside the gas-insulated switchgear to collect the concentration of decomposition products, gas purity, and gas moisture content of the target gas for insulation in real time; a temperature and humidity sensor is deployed in the environment where the gas-insulated switchgear is located to collect ambient temperature and humidity data in real time; and a current sensor and a dielectric loss tester are deployed at the electrical connection points of the gas-insulated switchgear to collect leakage current data and dielectric loss factor data in real time, respectively. All of the above data are transmitted in real time to the edge node for subsequent processing, and all data collected by the sensors and testing equipment are accompanied by a precise timestamp.

[0068] Step 102: By configuring edge nodes around the gas-insulated switchgear, outlier removal and feature fusion processing are performed on the concentration of decomposition products, the purity of the gas, the moisture content of the gas, the ambient temperature and humidity data, the leakage current data, and the dielectric loss factor data to obtain fused data, and the fused data is uploaded to the cloud intelligent analysis platform.

[0069] In this step, edge nodes configured around the gas-insulated switchgear refer to hardware devices deployed near the switchgear, capable of data reception, preliminary processing, and data uploading. These typically include data acquisition, processing, and communication modules, used to receive various monitoring data from the equipment nearby, perform preliminary processing such as outlier removal and feature fusion, reducing data transmission volume and improving data processing efficiency. The fused data refers to a dataset containing multi-dimensional monitoring information, formed by the edge nodes after removing outliers from the collected decomposition product concentration, gas purity, gas moisture content, ambient temperature and humidity, leakage current, and dielectric loss factor data, and then integrating them according to timestamps. The cloud-based intelligent analysis platform refers to a software system deployed on a cloud server, possessing data storage, feature extraction, collaborative analysis, and decision generation functions. It receives the fused data uploaded from the edge nodes, extracts data features through intelligent algorithms, performs collaborative analysis, identifies the degree of insulation degradation of the equipment, and generates threshold adjustment and protection decision commands, enabling remote intelligent monitoring and management of the gas-insulated switchgear's operating status.

[0070] Step 103: Extract gas decomposition characteristics, insulation electrical parameters, and environmental impact factors from the fused data, and conduct collaborative analysis on the fused data through the cloud-based intelligent analysis platform to identify the degree of insulation degradation of the gas-insulated switchgear.

[0071] In this step, gas decomposition characteristics refer to the set of characteristic parameters related to the decomposition of the target gas for insulation, extracted from the fused data. These include the trend of decomposition product concentration changes, the stable range of gas purity, and the fluctuation characteristics of moisture content. This set of characteristics reflects the state changes of the target gas for insulation, and thus indirectly reflects the deterioration of the internal insulation materials of the gas-insulated switchgear. Insulation electrical parameters refer to the set of parameters related to the electrical insulation performance of the gas-insulated switchgear, extracted from the fused data. These include the variation law of leakage current amplitude and the distribution characteristics of dielectric loss factor. This set of parameters is used to directly evaluate the electrical insulation performance of the equipment and is the core basis for judging the insulation status of the equipment. Environmental impact factors refer to the characteristic parameters reflecting the impact of environmental factors on the operating status of the gas-insulated switchgear, extracted from the fused data. These are usually synergistically changing. This factor is used to analyze the impact of changes in ambient temperature and humidity on the insulation performance and electrical parameters of the equipment, providing an environmental dimension reference for equipment status assessment. The degree of insulation degradation refers to the extent to which the internal insulation performance of gas-insulated switchgear deteriorates, as determined by a cloud-based intelligent analysis platform through collaborative analysis of gas decomposition characteristics, insulation electrical parameters, and environmental impact factors. It is typically classified into three levels: slight degradation, moderate degradation, and severe degradation. This degree is the core basis for subsequent adjustments to overvoltage protection thresholds and trigger protection decisions.

[0072] Step 104: Adjust the overvoltage protection threshold of the gas-insulated switchgear according to the degree of insulation degradation to obtain the adjusted overvoltage protection threshold, so as to trigger the protection decision corresponding to the degree of insulation degradation. The protection decision includes protection decisions corresponding to early warning prompts, load transfer strategies and emergency maintenance strategies, respectively.

[0073] In this step, the overvoltage protection threshold refers to a pre-set voltage protection critical value to prevent insulation breakdown faults in gas-insulated switchgear due to excessive voltage. When the operating voltage of the equipment exceeds this threshold, the protection device will activate its protection action. This threshold is usually set based on the normal insulation performance of the equipment and is the core parameter for overvoltage protection. The adjusted overvoltage protection threshold refers to the new protection threshold obtained by adjusting the initial overvoltage protection threshold according to the degree of insulation degradation of the gas-insulated switchgear. Since insulation degradation leads to a decrease in the equipment's voltage withstand capability, lowering the protection threshold can improve the sensitivity of the equipment's overvoltage protection and prevent equipment failure at the original threshold due to insulation degradation. The protection decision refers to the set of response strategies generated based on the degree of insulation degradation of the gas-insulated switchgear and the adjusted overvoltage protection threshold to ensure the safe operation of the equipment. This includes early warning prompts, load transfer strategies, and emergency maintenance strategies. This decision can provide differentiated operation and maintenance solutions based on the differences in equipment status to ensure the safe operation of the equipment. Early warning alerts are triggered when the insulation degradation of gas-insulated switchgear is minor, prompting maintenance personnel to monitor the equipment status and conduct subsequent inspections. This typically includes information such as the degree of insulation degradation, adjusted overvoltage protection thresholds, and recommended inspection cycles, transmitted to maintenance personnel via maintenance terminals to provide early warning of equipment failure. Load transfer strategies are triggered when the insulation degradation of gas-insulated switchgear is moderate, reducing the operating load and preventing further insulation degradation. This involves transferring target loads within the equipment's risk zone to preset backup lines, reducing the equipment's load capacity, and simultaneously sending early warning alerts to buy time for subsequent maintenance. Emergency maintenance strategies are triggered when the insulation degradation of gas-insulated switchgear is severe, quickly stopping the equipment and preventing serious failures. These include initiating equipment shutdown procedures, locking adjusted overvoltage protection thresholds, sending emergency maintenance commands, and transferring all loads to ensure timely removal of the equipment for maintenance and prevent further escalation of the fault.

[0074] This application's embodiments, by collecting multi-source data, provide a complete data foundation for subsequent multi-dimensional analysis, avoiding the analytical bias caused by single data collection in existing solutions; they achieve preliminary fusion of multi-dimensional data on environmental disturbances, electrical stress, and chemical changes, breaking through the limitations of multi-variable nonlinear correlation fracture in existing solutions; they dynamically adapt to the state evolution characteristics during equipment aging, reducing the probability of misjudgment and omission in state assessment; they enable continuous tracking and forward-looking extrapolation of equipment degradation trends, avoiding the problems of delayed or overly conservative protection responses, providing precise support for differentiated operation and maintenance strategies, improving the safety and reliability of gas-insulated switchgear operation, and ensuring the stable operation of substations and even the entire power system.

[0075] This application provides a specific embodiment. Step 102 involves using edge nodes configured around the gas-insulated switchgear to perform outlier removal and feature fusion processing on the concentration of decomposition products, the purity of the gas, the moisture content of the gas, the ambient temperature and humidity data, the leakage current data, and the dielectric loss factor data to obtain fused data. The specific steps include:

[0076] Step 201: Divide the concentration of the decomposition products, the purity of the gas, and the moisture content of the gas into multiple data groups, and calculate the average fluctuation value within each data group.

[0077] In this step, the fixed acquisition duration refers to the uniform time interval set for collecting data on the concentration of decomposition products, gas purity, and gas moisture content of the target gas used for insulation. This interval is used to divide the continuously collected data of these three gas parameters into independently processable data groups, and is preset based on substation operation and maintenance experience or data processing efficiency requirements. A data group refers to a set containing continuously collected values ​​of a specific gas parameter, divided according to the fixed acquisition duration. This set is used to achieve segmented anomaly detection of gas parameters and is based on the fixed acquisition duration and continuously collected data. The average fluctuation value refers to the arithmetic mean of the deviations of all collected values ​​of a specific gas parameter within a single data group from the group's average value. It reflects the overall fluctuation level of the gas parameter within that data group and is used as a benchmark for judging whether a single data point is abnormal.

[0078] In this embodiment, based on a fixed acquisition duration pre-set for the acquisition of target gas parameters for insulation, the three types of continuously acquired data—concentration of decomposition products, gas purity, and gas moisture content—received by the edge node are divided into multiple data groups according to the fixed acquisition duration. For example, if the fixed acquisition duration is 1 hour, then the three types of data within each 1 hour constitute a corresponding data group. For each data group, the average value of all acquired values ​​of the parameter within the group is calculated first, then the difference between each acquired value within the group and the average value is calculated, and finally, the sum of all individual data fluctuation values ​​within the group is divided by the number of data points to obtain the average fluctuation value of the parameter within each data group. Ultimately, each data group corresponds to one average fluctuation value.

[0079] Step 202: By configuring edge nodes around the gas-insulated switchgear, remove the first abnormal data in each data group whose deviation from the average fluctuation value exceeds a preset deviation range, so as to form the first type of target data.

[0080] In this step, the preset deviation range refers to a pre-defined threshold range for judging whether the fluctuation value of a single gas parameter data point is abnormal. It is used to filter the first abnormal data within the data group. The first abnormal data refers to the collected data within the gas parameter data group where the deviation of a single data fluctuation value from its corresponding average fluctuation value exceeds the preset deviation range, reflecting the existence of collection errors or transient interference. This data is selected based on the preset deviation range. The data group after removing the abnormal data refers to the set of clean gas parameter collected values ​​remaining after removing the first abnormal data. This set is used for summarizing the subsequent first-category target data. The first-category target data refers to the clean gas parameter dataset formed by integrating all gas parameter data groups after removing the abnormal data according to the collection time period. This dataset is used for subsequent fusion with electrical and environmental parameter data.

[0081] In this embodiment, by configuring edge nodes around the gas-insulated switchgear, the individual data fluctuation values ​​of decomposition product concentration, gas purity, and gas moisture content in each data group are compared with the average fluctuation value of the corresponding data group. If the deviation between the individual data fluctuation value and the average fluctuation value exceeds a preset deviation range, the collected data corresponding to the individual data fluctuation value is the first abnormal data. After removing the first abnormal data in each data group, the abnormal data group is obtained. Then, the abnormal data groups of all parameters are summarized and integrated according to the collection time period to form the first type of target data containing clean gas parameter data.

[0082] Step 203: Calculate the difference in leakage current and the difference in dielectric loss factor between adjacent acquisition times, and set a first abnormal threshold and a second abnormal threshold based on all the differences in leakage current and dielectric loss factor.

[0083] In this step, the leakage current difference value refers to the difference between the leakage current values ​​at two adjacent acquisition times in the leakage current data, reflecting the instantaneous change amplitude of the leakage current and used to determine whether there are abnormal abrupt changes in the leakage current data. The dielectric loss factor difference value refers to the difference between the dielectric loss factor values ​​at two adjacent acquisition times in the dielectric loss factor data, reflecting the instantaneous change amplitude of the dielectric loss factor and used to determine whether there are abnormal abrupt changes in the dielectric loss factor data. The first abnormal threshold is a critical value used to determine whether the leakage current difference value is abnormal, used to filter abnormal leakage currents. The second abnormal threshold is a critical value used to determine whether the dielectric loss factor difference value is abnormal, used to filter abnormal dielectric loss factor values.

[0084] In this embodiment, for the leakage current data received by the edge node, the difference between the leakage current values ​​at two adjacent acquisition times is calculated to obtain multiple leakage current difference values; similarly, for the dielectric loss factor data, the difference between the dielectric loss factor values ​​at adjacent acquisition times is calculated to obtain multiple dielectric loss factor difference values; through statistical analysis, the sum of all leakage current difference values ​​is divided by the number of difference values ​​to obtain a first average difference value, and the maximum and minimum values ​​among all leakage current difference values ​​are found, the difference between the two being the first fluctuation range; the above operation is repeated for the dielectric loss factor difference values ​​to obtain a second average difference value and a second fluctuation range; the first average difference value is added to the maximum value in the first fluctuation range to obtain a first anomaly threshold; the second average difference value is added to the maximum value in the second fluctuation range to obtain a second anomaly threshold.

[0085] Step 204: Remove the next abnormal leakage current in the leakage current data where the leakage current difference value exceeds the first abnormal threshold, and remove the next dielectric loss factor in the dielectric loss factor data where the dielectric loss factor difference value exceeds the second abnormal threshold, to obtain the second type of target data.

[0086] In this step, abnormal leakage current refers to the leakage current value at the next acquisition time after the leakage current difference value exceeds the first abnormal threshold, reflecting an abnormal sudden change in leakage current at that time. The dielectric loss factor is a parameter reflecting the degree of dielectric loss in gas-insulated switchgear; its data is used to evaluate the equipment's insulation performance. Here, it specifically refers to the original acquisition data before outlier removal or the clean data after outlier removal. The second type of target data refers to the clean electrical parameter dataset formed by integrating the leakage current data after removing abnormal leakage current and the dielectric loss factor data after removing abnormal dielectric loss factor values.

[0087] In this embodiment, each leakage current difference value corresponding to the leakage current data is checked one by one. If a certain leakage current difference value exceeds the first abnormal threshold, the leakage current value at the next acquisition time corresponding to the difference value is an abnormal leakage current and is removed from the leakage current data. Similarly, the dielectric loss factor difference value is checked. If a certain dielectric loss factor difference value exceeds the second abnormal threshold, the dielectric loss factor value at the next acquisition time corresponding to the difference value is removed from the dielectric loss factor data. The leakage current data after removing abnormal data is integrated with the dielectric loss factor data to form a second type of target data containing clean electrical parameter data.

[0088] Step 205: Remove the second abnormal data in the environmental temperature and humidity data that does not conform to the preset reasonable deviation range to obtain the third type of target data.

[0089] In this step, the reasonable deviation range refers to the range within which the current environmental temperature and humidity data is allowed to fluctuate, based on the conventional value range. For example, the upper limit of the conventional value range is set at +5%, and the lower limit at -5%. The conventional value range refers to the range in which the vast majority of normal data in historical environmental temperature and humidity data falls, such as the 95% confidence interval, reflecting the normal fluctuation range of environmental temperature and humidity. The second abnormal data refers to the collected data in the current environmental temperature and humidity data that exceeds the reasonable deviation range, reflecting environmental interference or collection errors. It is obtained by filtering based on the reasonable deviation range. The third type of target data refers to the environmental temperature and humidity data after removing the second abnormal data, i.e., the clean environment parameter dataset, used for subsequent multi-dimensional data fusion. It is obtained by removing the second abnormal data from the environmental temperature and humidity data.

[0090] In this embodiment, the historical environmental temperature and humidity data of the substation over the past year are retrieved through the database interface of the edge node. The 95% confidence interval of the historical data is calculated and used as the normal value interval. A reasonable deviation range corresponding to the normal value interval is set according to the operation and maintenance requirements. The currently collected environmental temperature and humidity data are compared with the reasonable deviation range one by one. If a data point exceeds the range, it is determined to be the second abnormal data. After removing it, the third type of target data is obtained.

[0091] Step 206: Perform temporal alignment and feature splicing on the first type of target data, the second type of target data, and the third type of target data to generate fused data.

[0092] In this embodiment of the application, the data processing module of the edge node first performs time-series alignment on the first type of target data, the second type of target data, and the third type of target data: extracts the timestamp of each collected value in the three types of data, and associates the three types of data at the same time or adjacent times with the timestamp as the basis; then performs feature splicing: integrates the feature fields of the three types of data after time-series alignment into the same data structure, and finally generates fused data.

[0093] This application's embodiments solve the problems of poor data quality caused by coarse data processing and broken multi-dimensional data correlation by accurately removing outliers by type and fusing time-series alignment features. It can avoid the accidental deletion of valid data or the retention of abnormal data caused by a single anomaly removal method. It breaks through the limitations of existing solutions where multi-dimensional data are independent and unrelated, and provides high-quality, strongly correlated basic data for subsequent collaborative analysis on cloud-based intelligent analysis platforms. It reduces the risk of misjudgment and omission in status assessment caused by poor data quality, and lays a data foundation for accurate identification of the degree of equipment insulation degradation.

[0094] This application provides a specific embodiment, such as Figure 2As shown, step 103 involves extracting gas decomposition characteristics, insulation electrical parameters, and environmental impact factors from the fused data. This data is then used in a collaborative analysis via the cloud-based intelligent analysis platform to identify the degree of insulation degradation in the gas-insulated switchgear. Specifically, this includes the following steps:

[0095] Step 301: Extract gas decomposition characteristics, insulation electrical parameters, and environmental impact factors from the fused data. The gas decomposition characteristics include the concentration variation trend of the decomposition products of the target gas for insulation, the stable range of gas purity, and the fluctuation characteristics of moisture content. The insulation electrical parameters include the variation law of leakage current amplitude and the distribution characteristics of dielectric loss factor. The environmental impact factors include the synergistic variation relationship of ambient temperature and humidity.

[0096] In this step, the trend of the concentration change of the decomposition products of the target gas for insulation refers to the direction and rate of change of the concentration of the decomposition products of the target gas for insulation with the acquisition time, extracted from the fused data, reflecting the dynamic process of the deterioration of the internal insulation material of the gas-insulated switchgear.

[0097] The stable range of gas purity refers to the range of values ​​in which the purity of the target gas for insulation remains stable over a period of time, selected from the fused data, reflecting whether the gas purity meets the equipment insulation requirements.

[0098] Moisture content fluctuation characteristics refer to the frequency and amplitude of fluctuations in the moisture content of the target gas for insulation, extracted from the fused data, reflecting the degree of influence of moisture on the gas insulation performance. Leakage current amplitude variation refers to the range and trend of leakage current amplitude of gas-insulated switchgear over time, extracted from the fused data, reflecting the electrical characteristics of the equipment's insulation performance.

[0099] The dielectric loss factor distribution characteristics refer to the numerical distribution range and concentration trend of the dielectric loss factor of gas-insulated switchgear extracted from the fused data over a period of time, reflecting the degree of loss of the insulating dielectric. The coordinated variation relationship refers to the coordinated variation relationship between the ambient temperature and humidity of the gas-insulated switchgear extracted from the fused data, reflecting the comprehensive impact of environmental factors on the equipment's condition.

[0100] In this embodiment, firstly, data related to the target gas for insulation in the fused data are screened, and the direction of increase and decrease of the concentration of decomposition products and the amount of change per unit time during the continuous acquisition period are statistically analyzed to obtain the trend of decomposition product concentration change; then, the numerical range of gas purity over a period of time is statistically analyzed, and the interval with fluctuation amplitude less than the preset value is screened as the stable interval of gas purity; the number of fluctuations and the maximum and minimum differences of moisture content per unit time are statistically analyzed to obtain the moisture content fluctuation characteristics, and the above three constitute the gas decomposition characteristics; next, data related to electrical performance in the fused data are screened, and the fluctuation range and increase and decrease trend of leakage current amplitude in different time periods are statistically analyzed to obtain the change law of leakage current amplitude; the numerical concentration interval and dispersion of dielectric loss factor are statistically analyzed to obtain the dielectric loss factor distribution characteristics, and the above two constitute the insulation electrical parameters; finally, the temperature and humidity data collected synchronously in the fused data are screened, and the correlation between temperature and humidity changes is analyzed to obtain the synergistic change relationship, and this model constitutes the environmental impact factor.

[0101] Step 302: Using the cloud-based intelligent analysis platform, cross-compare the gas decomposition characteristics with the insulation electrical parameters to determine the degree of correlation between the gas decomposition characteristics and the insulation electrical parameters. Match the degree of correlation with the synergistic change relationship in the environmental impact factors to obtain the synergistic relationship between the gas decomposition characteristics, insulation electrical parameters, and environmental impact factors.

[0102] In this step, the degree of correlation refers to the closeness of the correlation between two types of features determined by statistical results of multiple sets of correspondences between gas decomposition characteristics and insulation electrical parameters. It reflects the coupling relationship between changes in gas state and changes in electrical performance and is used for subsequent matching of environmental impact factors. The synergistic relationship refers to the correlation between gas decomposition characteristics, insulation electrical parameters, and environmental impact factors established by combining synergistic change relationships, reasonable correlation levels, and degree of correlation. It reflects the comprehensive law of equipment status being affected by multiple factors and is used to compare historical data to determine insulation status. It is obtained by integrating the results of temperature and humidity pattern classification, correlation degree matching, and level determination.

[0103] In this embodiment of the application, step 302 specifically includes the following steps:

[0104] Step 311: Statistically calculate the percentage of times when the trend of the concentration change of the decomposition products of the target gas for insulation is consistent with the trend of the change of the leakage current amplitude during the same data acquisition period, the percentage of times when the distribution characteristics of the dielectric loss factor are within the normal factor range corresponding to the stable range of gas purity, and the percentage of times when the fluctuation characteristics of the moisture content exceed the preset normal fluctuation range, the distribution characteristics of the dielectric loss factor show abnormal distribution.

[0105] In this step, the data acquisition period refers to the continuous time interval corresponding to the synchronous acquisition of various monitoring parameters of the gas-insulated switchgear. It is a time reference used to uniformly compare the data change relationships such as the trend of decomposition product concentration change, the law of leakage current amplitude change, and the distribution characteristics of dielectric loss factor. It is uniformly divided based on the preset acquisition cycle of the equipment.

[0106] Consistent direction of change means that within the same data collection period, the increasing or decreasing direction of the concentration of decomposition products is the same as the increasing or decreasing direction of the amplitude of leakage current.

[0107] The percentage of times refers to the proportion of the total number of collection periods in which the trend of decomposition product concentration changes is consistent with the trend of leakage current amplitude changes. It reflects the probability that the two types of characteristic changes are synchronized.

[0108] The normal factor range refers to the pre-set range of normal values ​​for the medium loss factor that matches the stable range of gas purity, reflecting the reasonable fluctuation range of the medium loss factor when the gas purity is stable.

[0109] The numerical percentage refers to the proportion of the number of values ​​in the distribution characteristics of the medium loss factor within the normal range during the same data collection period, relative to the total number of values ​​of the medium loss factor during that period. It reflects the normality of the medium loss factor when the gas purity is stable.

[0110] The preset normal fluctuation range refers to the pre-set numerical range of normal fluctuations in moisture content. It reflects the reasonable fluctuation range in which moisture content does not affect the insulation of the equipment and is used to determine whether the moisture content is abnormal. It is set based on the common fluctuation range of moisture content that has not caused insulation problems in historical data.

[0111] Abnormal distribution refers to the distribution characteristics of the medium loss factor where the value exceeds the normal range, reflecting the degree of abnormality of the medium loss factor when the moisture content is abnormal.

[0112] The percentage of abnormal occurrences refers to the proportion of times the distribution characteristics of the media loss factor show abnormal distribution among the number of times the moisture content fluctuation characteristics exceed the preset normal fluctuation range within the same collection period, reflecting the degree of impact of moisture anomalies on media loss.

[0113] In this embodiment, firstly, using the data acquisition period as a unified time reference, the increasing and decreasing directions of the decomposition product concentration change trend and the leakage current amplitude change pattern are compared for each period. The number of periods with the same direction of change is counted, and then this number is divided by the total number of acquisition periods to obtain the percentage of occurrences. Subsequently, the distribution characteristics of the medium loss factor are compared with the normal factor range corresponding to the stable range of gas purity for each period. The number of medium loss factor values ​​within this range is counted, and this number is divided by the total number of medium loss factor values ​​within this period to obtain the percentage of values. Finally, the total number of times the moisture content fluctuation characteristics exceed the preset normal fluctuation range is counted for each period, as well as the number of times the medium loss factor distribution characteristics show abnormal distribution among the total number of times. The number of abnormal distributions is divided by the total number of abnormal moisture content values ​​to obtain the percentage of abnormal occurrences.

[0114] Step 312: Determine the degree of correlation between gas decomposition characteristics and insulation electrical parameters based on the percentage of occurrences, the percentage of numerical values, and the percentage of abnormal occurrences.

[0115] In this embodiment, firstly, the weight coefficients of the frequency ratio, the numerical ratio, and the abnormal frequency ratio are preset. The weights of the three can be set according to the actual operation and maintenance needs, and the sum must be 1. Then, the three ratios are multiplied by their corresponding weight coefficients, and the product results are added together to obtain a comprehensive quantitative value. This comprehensive quantitative value is the degree of correlation between gas decomposition characteristics and insulation electrical parameters. The higher the comprehensive quantitative value, the closer the correlation between the two types of characteristics.

[0116] Step 313: Divide the cooperative change relationship into high temperature and high humidity mode, high temperature and low humidity mode, low temperature and high humidity mode, and low temperature and low humidity mode.

[0117] In this step, the high temperature and high humidity mode refers to the mode in which the ambient temperature is higher than the preset high temperature threshold and the humidity is higher than the preset high humidity threshold in the coordinated change relationship. It reflects the impact of the high temperature and high humidity environment on the equipment status and is used to match the degree of correlation to determine the reasonable level. It is divided based on the case where both temperature and humidity are in the high value range in the temperature and humidity data.

[0118] The high temperature and low humidity mode refers to a mode in which the ambient temperature is higher than the preset high temperature threshold and the humidity is lower than the preset low humidity threshold in the coordinated change relationship. It reflects the impact of the high temperature and low humidity environment on the equipment status and is used to match the degree of correlation to determine the reasonable level. It is divided based on the temperature being in the high value range and the humidity being in the low value range in the temperature and humidity data.

[0119] The low temperature and high humidity mode refers to a mode in which the ambient temperature is lower than the preset low temperature threshold and the humidity is higher than the preset high humidity threshold in the coordinated change relationship. It reflects the impact of the low temperature and high humidity environment on the equipment status and is used to match the degree of correlation to determine the reasonable level. It is divided based on the cases where the temperature is in the low value range and the humidity is in the high value range in the temperature and humidity data.

[0120] The low temperature and low humidity mode refers to the mode in which the ambient temperature is lower than the preset low temperature threshold and the humidity is lower than the preset low humidity threshold in the coordinated change relationship. It reflects the impact of the low temperature and low humidity environment on the equipment status and is used to match the degree of correlation to determine the reasonable level. It is divided based on the case where both temperature and humidity are in the low value range in the temperature and humidity data.

[0121] In this embodiment, firstly, based on preset high temperature threshold, low temperature threshold, high humidity threshold, and low humidity threshold, the ambient temperature and humidity data in the current collaborative change relationship are extracted. The temperature data is compared with the high temperature threshold and low temperature threshold, and the humidity data is compared with the high humidity threshold and low humidity threshold, respectively. Based on the comparison results, the current collaborative change relationship is divided into one of the four modes: high temperature and high humidity, high temperature and low humidity, low temperature and high humidity, and low temperature and low humidity.

[0122] Step 314: Match the correlation degree with the corresponding normal correlation degree ranges for the high temperature and high humidity mode, the high temperature and low humidity mode, the low temperature and high humidity mode, and the low temperature and low humidity mode, respectively, to determine the reasonable correlation level corresponding to the correlation degree.

[0123] In this step, the "normal correlation range" refers to the pre-defined normal range of values ​​for the correlation between gas decomposition characteristics and insulation electrical parameters under different cooperative change relationships. It reflects the reasonable level of correlation between the two types of characteristics under a specific environment and is used to determine the reasonable correlation level. This range is set based on common value intervals of the correlation level under different temperature and humidity patterns in historical data. The reasonable correlation level refers to the level of correlation determined after matching the current correlation level with the normal correlation range of the corresponding temperature and humidity pattern. It is typically divided into three levels: high, medium, and low, reflecting whether the correlation between the two types of characteristics is normal or not under a specific environment.

[0124] In this embodiment of the application, firstly, based on the temperature and humidity mode corresponding to the current cooperative change relationship, the preset normal correlation range under the mode is retrieved, and then the correlation degree is compared with the normal range. If the correlation degree is higher than the upper limit of the normal range, the reasonable correlation level is determined to be high; if the correlation degree falls within the normal range, the reasonable correlation level is determined to be medium; if the correlation degree is lower than the lower limit of the normal range, the reasonable correlation level is determined to be low.

[0125] Step 315: Based on the synergistic change relationship, the reasonable correlation level, and the correlation degree, generate the synergistic relationship between gas decomposition characteristics, insulation electrical parameters, and environmental impact factors.

[0126] In this step, the collaborative relationship refers to the comprehensive correlation result obtained by integrating the collaborative change relationship, reasonable correlation level, and correlation degree, which is used to determine the current insulation status of the equipment.

[0127] In this embodiment of the application, the cooperative change relationship, reasonable correlation level, and correlation degree are bound and integrated to form a three-dimensional cooperative relationship that includes environmental temperature and humidity pattern, feature correlation level, and feature correlation strength. This cooperative relationship is the comprehensive correlation relationship between gas decomposition characteristics, insulation electrical parameters, and environmental impact factors.

[0128] Step 303: Compare the cooperative relationship with the historical feature data of gas-insulated switchgear under different insulation states pre-stored in the cloud-based intelligent analysis platform to determine the insulation state corresponding to the cooperative relationship.

[0129] In this step, different insulation states refer to the state categories of gas-insulated switchgear according to the degree of insulation performance degradation. They typically include slightly degraded state, moderately degraded state, and severely degraded state, reflecting different health levels of the equipment's insulation performance. They are used to classify and store historical characteristic data and are classified based on the criteria for judging the degree of insulation degradation of the equipment.

[0130] Historical characteristic data refers to the gas decomposition characteristics, insulation electrical parameters, environmental impact factors, and synergistic relationship data of gas-insulated switchgear under different insulation states, which are pre-stored in the cloud-based intelligent analysis platform. It reflects the multi-dimensional characteristic patterns of the equipment under different health states in the past and is used to compare the current synergistic relationship to determine the insulation state. It is obtained by summarizing the equipment status data collected in historical operation and maintenance and the corresponding insulation degradation judgment results.

[0131] The insulation status corresponding to the cooperative relationship refers to the insulation status category of the current equipment determined by comparing the current cooperative relationship with historical feature data. It reflects the insulation health level of the current equipment and is used to subsequently determine the degree of insulation degradation. It is determined based on the matching results of the current cooperative relationship and the historical cooperative relationship group.

[0132] In this embodiment of the application, step 303 specifically includes the following steps:

[0133] Step 321: Classify the historical characteristic data of gas-insulated switchgear under different insulation states pre-stored in the cloud-based intelligent analysis platform to obtain multiple historical collaborative relationship groups.

[0134] In this step, multiple historical collaborative relationship groups include a first historical collaborative relationship group, a second historical collaborative relationship group, and a third historical collaborative relationship group. The first historical collaborative relationship group refers to the set of historical collaborative relationships pre-stored in the cloud-based intelligent analysis platform that corresponds to the slight degradation state of gas-insulated switchgear. It reflects the correlation between gas decomposition characteristics, insulation electrical parameters, and environmental impact factors under the slight degradation state. The slight degradation state refers to the insulation state of gas-insulated switchgear where the insulation performance is slightly reduced but does not affect normal operation, reflecting that the equipment insulation is in the early degradation stage.

[0135] The second historical collaborative relationship group refers to the set of historical collaborative relationships pre-stored in the cloud-based intelligent analysis platform that correspond to the moderate degradation state of gas-insulated switchgear. The moderate degradation state refers to an insulation state where the insulation performance of the gas-insulated switchgear has significantly decreased, requiring attention to maintenance, reflecting that the equipment insulation is in the mid-stage of degradation. The third historical collaborative relationship group refers to the set of historical collaborative relationships pre-stored in the cloud-based intelligent analysis platform that correspond to the severe degradation state of gas-insulated switchgear. The severe degradation state refers to an insulation state where the insulation performance of the gas-insulated switchgear has significantly decreased, requiring emergency maintenance, reflecting that the equipment insulation is in the late-stage of degradation.

[0136] In this embodiment of the application, all historical characteristic data of gas-insulated switchgear pre-stored in the cloud-based intelligent analysis platform are first retrieved and classified into three categories: slight deterioration state, moderate deterioration state, and severe deterioration state. The gas decomposition characteristics, insulation electrical parameters, environmental impact factors, and synergistic relationship data under the same insulation state are integrated into a historical synergistic relationship group, forming the first historical synergistic relationship group, the second historical synergistic relationship group, and the third historical synergistic relationship group, respectively.

[0137] Step 322: Select the target historical collaborative relationship group from the multiple historical collaborative relationship groups that is consistent with the collaborative change relationship of the collaborative relationship.

[0138] In this step, the target historical collaborative relationship group refers to the historical collaborative relationship group with consistent temperature and humidity patterns selected after matching the current collaborative relationship with the temperature and humidity patterns of the first, second, and third historical collaborative relationship groups. This reflects the historical equipment status patterns consistent with the current environmental conditions and is used for subsequent matching of reasonable association levels.

[0139] In this embodiment of the application, the collaborative change relationship (i.e. temperature and humidity pattern) in the current collaborative relationship is first extracted, and then compared with the collaborative change relationships contained in the first historical collaborative relationship group, the second historical collaborative relationship group, and the third historical collaborative relationship group one by one. The historical collaborative relationship group whose temperature and humidity pattern is completely consistent with the current collaborative change relationship is selected and determined as the target historical collaborative relationship group.

[0140] Step 323: Compare the reasonable association level of the collaborative relationship with the reasonable association level in each target historical collaborative relationship group to select candidate historical collaborative relationship groups.

[0141] In this step, the candidate historical collaborative relationship group refers to the target historical collaborative relationship group with the reasonable association level of the current collaborative relationship compared with the reasonable association level of each target historical collaborative relationship group, which reflects the historical equipment status pattern that is consistent with the current feature association level.

[0142] In this embodiment of the application, the reasonable association level of the current collaborative relationship is extracted, and it is compared with the reasonable association level recorded in each target historical collaborative relationship group one by one. The target historical collaborative relationship group whose reasonable association level is completely consistent with the current collaborative relationship is selected and determined as the candidate historical collaborative relationship group.

[0143] Step 324: Compare the frequency percentage, numerical percentage, and abnormal frequency percentage of the degree of association of the collaborative relationship with the normal numerical range of the corresponding percentages in each candidate historical collaborative relationship group, and count the number of matching items in each candidate historical collaborative relationship group where each percentage is within the corresponding normal numerical range.

[0144] In this step, the normal numerical range refers to the pre-set normal range of values ​​for the proportion of frequency, numerical value, and abnormal frequency in the degree of association for candidate historical collaborative relationship groups. It reflects the reasonable level of the three proportions under this type of historical state and is used to count the number of matching items. It is obtained based on the common value range of the three proportions in the candidate historical collaborative relationship groups.

[0145] In this embodiment, the normal range of frequency percentage, normal range of value percentage, and normal range of abnormal frequency percentage for each candidate historical collaborative relationship group are retrieved first. Then, the frequency percentage, value percentage, and abnormal frequency percentage of the current collaborative relationship are compared with the normal range of the corresponding candidate group. The number of items in each candidate group whose three percentages fall within the corresponding normal range is counted. This number is the number of matching items for the candidate historical collaborative relationship group.

[0146] Step 325: Select the insulation state of the gas-insulated switchgear corresponding to the candidate historical cooperative relationship group with the most matching items as the insulation state corresponding to the cooperative relationship.

[0147] In this step, the number of matching items refers to the proportion of the current collaborative relationship's correlation frequency, numerical proportion, and abnormal frequency that are within the normal numerical range of the corresponding candidate historical collaborative relationship group. It reflects the degree of matching between the current collaborative relationship and the candidate group and is used to determine the corresponding insulation state. It is obtained by comparing the three types of proportions with the normal numerical range.

[0148] In this embodiment of the application, the number of matching items of all candidate historical collaborative relationship groups is compared, and the candidate historical collaborative relationship group with the most matching items is selected. The insulation state of the candidate group is pre-marked as the insulation state corresponding to the current collaborative relationship.

[0149] Step 304: Determine the degree of insulation degradation of the gas-insulated switchgear based on the insulation condition.

[0150] In this embodiment of the application, step 304, determining the degree of insulation degradation of the gas-insulated switchgear based on the insulation state, specifically includes the following steps:

[0151] Step 331: Extract key data from the insulation state.

[0152] In this step, key data refers to the core parameter data extracted from the insulation state corresponding to the synergistic relationship, which can reflect the degree of insulation degradation of the equipment. These include the actual rate of change of the concentration of decomposition products, the actual fluctuation range of the stable range of gas purity, the actual fluctuation frequency of the moisture content fluctuation characteristics, the actual fluctuation range of the leakage current amplitude, and the actual numerical range of the dielectric loss factor distribution characteristics.

[0153] In this embodiment of the application, the actual rate of change of the concentration of decomposition products, the actual fluctuation range of the stable range of gas purity, the actual fluctuation frequency of the fluctuation characteristics of moisture content, the actual fluctuation range of the change law of leakage current amplitude, and the actual numerical range of the distribution characteristics of dielectric loss factor are extracted one by one from the characteristic data corresponding to the insulation state. Then, the above five parameters are integrated to form key data.

[0154] Step 332: Compare the key data with each insulation condition judgment standard to count the number of data items that meet the first insulation condition judgment standard, the number of data items that meet the second insulation condition judgment standard, and the number of data items that meet the third insulation condition judgment standard. Select the insulation condition corresponding to the largest number of data items as the insulation degradation degree of the gas-insulated switchgear.

[0155] In this step, each insulation condition judgment standard refers to a pre-set set of standards used to determine the degradation level of the insulation condition of gas-insulated switchgear. These standards include first, second, and third insulation condition judgment standards, reflecting the reasonable range of key data under different degradation levels. The first insulation condition judgment standard corresponds to the judgment standard for a slightly degraded state, reflecting the reasonable range of key data under this state. The second insulation condition judgment standard corresponds to the judgment standard for a moderately degraded state, reflecting the reasonable range of key data under this state. The third insulation condition judgment standard corresponds to the judgment standard for a severely degraded state, reflecting the reasonable range of key data under this state. The number of data items refers to the number of parameter items in the key data that meet a certain insulation condition judgment standard, reflecting the degree of matching between the key data and that degradation level, and is used to determine the degree of insulation degradation.

[0156] In this embodiment of the application, the five parameters in the key data are first compared with the three judgment criteria one by one. The number of data items that meet the first insulation state judgment criteria, the number of data items that meet the second insulation state judgment criteria, and the number of data items that meet the third insulation state judgment criteria are counted. The size of the three data item quantities is compared, and the insulation state corresponding to the data item with the largest value is selected and determined as the final insulation degradation degree of the gas-insulated switchgear.

[0157] This application's embodiments extract multi-dimensional features to provide a comprehensive foundation for subsequent collaborative analysis, avoiding the one-sided analysis caused by single feature extraction; it generates collaborative relationships through statistical analysis of multiple sets of feature correspondences and environmental pattern matching, integrating the nonlinear correlation of multiple variables such as gas, electrical, and environment, overcoming the limitations of existing technologies in multi-variable coupling; it gradually filters historical collaborative relationship groups according to environmental patterns, correlation levels, and proportions, avoiding the defects of local fixed rules being unable to adapt to different environments, and improving the adaptability of insulation status determination; it determines the degree of degradation based on key data and comparison with multiple standards to reduce the risk of misjudgment and omission.

[0158] This application provides a specific embodiment. Step 104 involves adjusting the overvoltage protection threshold of the gas-insulated switchgear according to the degree of insulation degradation to obtain an adjusted overvoltage protection threshold, thereby triggering a protection decision corresponding to the degree of insulation degradation. The protection decision includes protection decisions corresponding to early warning prompts, load transfer strategies, and emergency maintenance strategies, specifically including the following steps:

[0159] Step 401: Based on the insulation withstand characteristics of the gas-insulated switchgear, the degree of abnormality of key parameters corresponding to the degree of insulation degradation, and the operation and maintenance safety redundancy requirements of the substation, set the voltage adjustment rules corresponding to the degree of insulation degradation.

[0160] In this step, insulation withstand characteristic refers to the maximum voltage that gas-insulated switchgear can withstand, an inherent property that gradually decreases as the insulation deteriorates, reflecting the correspondence between the insulation state and the withstand voltage. It is obtained through statistical analysis of withstand voltage test data and historical operational fault data at different deterioration stages. The degree of anomaly in key parameters refers to the extent to which the core monitoring parameters of the gas-insulated switchgear exceed the normal range as the insulation deterioration progresses, reflecting the impact of insulation deterioration on the core parameters. It is calculated based on the ratio of the actual monitored values ​​of key parameters under different deterioration states to the normal range. Operational safety redundancy requirements refer to the safety margin requirements reserved by the substation when setting operating parameters to avoid equipment failures due to sudden overvoltage, load fluctuations, or other unexpected situations. This ensures the reliability of equipment operation and is determined based on substation historical accident analysis, equipment safety operation standards, and operational experience. Voltage adjustment rules refer to the specific rules for adjusting the overvoltage protection threshold of the gas-insulated switchgear according to the degree of insulation deterioration, ensuring that the equipment will not be overvoltage-damped due to a decrease in withstand voltage under deterioration conditions.

[0161] In this embodiment, for example, by analyzing historical equipment degradation data, it is found that the insulation withstand voltage decreases by 5%-8% in slight degradation, 10%-15% in moderate degradation, and 20%-25% in severe degradation. Also, in slight degradation, the concentration of decomposition products exceeds the normal range by 10%-15%, and the leakage current exceeds the normal range by 5%-10%; in moderate degradation, the concentration exceeds 20%-30%, and the leakage current exceeds 15%-20%; and in severe degradation, the concentration exceeds 35% or more, and the leakage current exceeds 25%. Considering the need for substations to reserve a 10%-15% voltage safety margin to avoid sudden overvoltage breakdown and maintain operational safety redundancy, voltage adjustment rules are set corresponding to the degree of insulation degradation. For example, a 5% reduction in the overvoltage protection threshold corresponds to a slight degradation state, a 12% reduction to a moderate degradation state, and a 20% reduction to a severe degradation state.

[0162] Step 402: Obtain the overvoltage protection threshold of the gas-insulated switchgear, and adjust the overvoltage protection threshold according to the voltage adjustment rules to obtain the adjusted overvoltage protection threshold.

[0163] In this embodiment of the application, the current initial overvoltage protection threshold of the equipment is obtained through the protection device interface of the gas-insulated switchgear. According to the voltage adjustment rule corresponding to the determined insulation degradation degree, the adjusted overvoltage protection threshold is calculated as follows: Adjusted overvoltage protection threshold = Initial overvoltage protection threshold × (1 - Adjustment ratio), where the adjustment ratio is determined according to the voltage adjustment rule.

[0164] Step 403: When the insulation degradation level is slightly degraded, a protection decision is triggered to provide an early warning. The early warning is used to instruct the maintenance terminal to send the insulation degradation level, the adjusted overvoltage protection threshold, and the inspection cycle information.

[0165] In this step, the inspection cycle information refers to the current degree of insulation degradation of the gas-insulated switchgear, and the set time interval for maintenance personnel to conduct on-site inspections and parameter retests of the equipment, in order to track the degradation trend of the equipment in a timely manner.

[0166] In this embodiment of the application, when the insulation degradation level is slightly degraded, a protective decision is triggered to issue an early warning. The early warning is sent to the maintenance personnel's maintenance terminal via the communication module of the cloud-based intelligent analysis platform. The early warning includes the insulation degradation level, the adjusted overvoltage protection threshold, and inspection cycle information, ensuring that the maintenance personnel are aware of the equipment status in a timely manner.

[0167] Step 404: When the insulation degradation level is moderate, the load transfer strategy is triggered to make a protection decision. The load transfer strategy is used to indicate that the target load within the risk range defined according to the adjusted overvoltage protection threshold is transferred to the preset backup line and to send an early warning to the operation and maintenance terminal.

[0168] In this step, the risk zone refers to the range of operating voltage that the gas-insulated switchgear must avoid, defined by the adjusted overvoltage protection threshold. It is typically set to ±5% of the adjusted threshold, reflecting the voltage range where the equipment may fail under its current deteriorated state. This is used to determine the target load to be transferred, and is determined based on the adjusted overvoltage protection threshold and the equipment's insulation margin. The target load refers to the electrical load within the risk zone that the gas-insulated switchgear is currently supplying, such as industrial or residential loads on a specific line. Continuing to supply power to this equipment may cause a fault due to overvoltage. This is used to clarify the specific object of load transfer, and is determined based on load monitoring data within the risk zone and the equipment's power supply range. The pre-planned backup line refers to a backup power line pre-planned and constructed by the substation to take over the load of the main power supply equipment in the event of failure or deterioration. It has the capacity and power supply range matched to the main power supply line, ensuring the continuity of power supply after load transfer. This configuration is based on the substation's power supply reliability planning and load forecasting results.

[0169] In this embodiment, when the insulation degradation is moderate, the load transfer strategy is triggered to make a protection decision. First, a risk range is defined based on the adjusted overvoltage protection threshold. For example, the operating voltage of the equipment needs to be controlled within the range of 221.76kV±5%. If it exceeds this range, it is considered a risk voltage. The target load of the gas-insulated switchgear within the risk range is queried through the power dispatching system, and a load transfer instruction is generated and sent to the dispatching execution module to instruct the target load to be transferred to a preset backup line. At the same time, an early warning is sent to the operation and maintenance terminal through the cloud communication module.

[0170] Step 405: When the insulation degradation level is severe, the protection decision of the emergency maintenance strategy is triggered. The emergency maintenance strategy is used to instruct the start of the equipment shutdown procedure, lock the adjusted overvoltage protection threshold, send the emergency maintenance command to the operation and maintenance terminal, and transfer the entire load to the preset backup line.

[0171] In this step, the equipment shutdown procedure refers to a series of operational processes performed to ensure the safe removal of gas-insulated switchgear from operation when it is severely deteriorated, such as [disconnecting the load switch, disconnecting the circuit breaker, opening the isolating switch, and verifying grounding]. These procedures are used to cut off the voltage and load connection to the equipment, preventing the fault from escalating, and are based on equipment safety operating specifications and power system outage procedures. The emergency maintenance instruction refers to the maintenance instructions generated for gas-insulated switchgear in a severely deteriorated state, including the equipment number, degree of deterioration, fault risk, recommended maintenance time, and maintenance items. These instructions guide maintenance personnel to quickly carry out maintenance work and are generated based on fault risk assessment and maintenance requirement analysis when the equipment is severely deteriorated. The total load refers to the total electrical load currently carried by the gas-insulated switchgear during operation. When the equipment is severely deteriorated and requires emergency shutdown, all of this load must be transferred to avoid power outages, and is calculated based on the current load monitoring data of the equipment.

[0172] In this embodiment, when the insulation degradation is severe, an emergency maintenance strategy is triggered. First, an equipment shutdown command is generated and sent to the control module of the gas-insulated switchgear to instruct the equipment shutdown procedure to be started. Simultaneously, a locking command is sent to the protection device to lock the adjusted overvoltage protection threshold, prohibiting the protection device from modifying the threshold or the equipment from operating under load. The emergency maintenance command is sent to the operation and maintenance terminal via the cloud communication module. Finally, a full load transfer command is generated and sent to the power dispatching system to instruct the transfer of all loads carried by the equipment to a preset backup line to ensure uninterrupted power supply.

[0173] This application's embodiments address the issues of delayed response or excessive conservatism caused by fixed overvoltage protection thresholds and singular protection decisions through dynamic threshold adjustment and tiered protection decision-making. It enables dynamic adaptation of overvoltage protection thresholds to the degree of insulation degradation, avoiding the breakdown risk caused by using normal thresholds after equipment degradation or excessively lowering thresholds that impact power supply. It provides early intervention through warnings in cases of minor degradation, balances equipment safety and power supply continuity through load transfer in cases of moderate degradation, and prevents fault escalation through emergency maintenance in cases of severe degradation. This enhances the operational safety of gas-insulated switchgear and the reliability of substation power supply, providing precise support for differentiated operation and maintenance.

[0174] Figure 3 This is a schematic diagram of a specific implementation of the cloud-based analysis-based substation operation status reasoning and decision-making system provided in this application embodiment, with reference to... Figure 3 The system may include:

[0175] The acquisition module 21 is used to acquire the concentration of decomposition products of the target gas for insulation inside the gas-insulated switchgear of the substation, the gas purity, the gas moisture content, the ambient temperature and humidity data of the gas-insulated switchgear, as well as the leakage current data and dielectric loss factor data of the gas-insulated switchgear.

[0176] Processing module 22 is used to perform outlier removal and feature fusion processing on the concentration of decomposition products, the purity of gas, the moisture content of gas, the ambient temperature and humidity data, the leakage current data and the dielectric loss factor data through edge nodes configured around the gas-insulated switchgear, to obtain fused data, and to upload the fused data to the cloud intelligent analysis platform.

[0177] The identification module 23 is used to extract gas decomposition characteristics, insulation electrical parameters and environmental impact factors from the fused data, so as to conduct collaborative analysis of the fused data through the cloud intelligent analysis platform and identify the degree of insulation degradation of the gas-insulated switchgear.

[0178] The adjustment module 24 is used to adjust the overvoltage protection threshold of the gas-insulated switchgear according to the degree of insulation degradation, so as to obtain the adjusted overvoltage protection threshold and trigger the protection decision corresponding to the degree of insulation degradation. The protection decision includes protection decisions corresponding to early warning prompts, load transfer strategies and emergency maintenance strategies, respectively.

[0179] The cloud-based substation operation status reasoning and decision-making system of this application is used to implement the aforementioned cloud-based substation operation status reasoning and decision-making method. Therefore, the specific implementation of the cloud-based substation operation status reasoning and decision-making system can be found in the embodiment section of the cloud-based substation operation status reasoning and decision-making method above. The specific implementation can be referred to the description of the corresponding embodiments, which will not be repeated here.

[0180] This application also provides an electronic device, including: a memory for storing a computer program; and a processor for executing the computer program to implement the steps of the cloud-based analysis-based substation operation status reasoning and decision-making method described above.

[0181] This application also provides a computer-readable storage medium storing a computer program, which, when executed by a processor, implements the steps of any of the above-described cloud-based analysis-based substation operation status reasoning and decision-making methods.

[0182] In one exemplary embodiment, the aforementioned computer-readable storage medium may include, but is not limited to, various media capable of storing computer programs, such as USB flash drives, read-only memory, random access memory, portable hard drives, magnetic disks, or optical disks.

[0183] The embodiments of this application also provide a computer program product, which includes a computer program that, when executed by a processor, implements the steps in any of the embodiments of the cloud-based analysis-based substation operation status reasoning and decision-making method described above.

[0184] Those skilled in the art will further recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, computer software, or a combination of both. To clearly illustrate the interchangeability of hardware and software, the components and steps of the various examples have been generally described in terms of functionality in the foregoing description. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0185] The foregoing has provided a detailed description of the cloud-based analysis-based substation operation status reasoning and decision-making method and system provided in this application. Specific examples have been used to illustrate the principles and implementation methods of this application. The descriptions of the embodiments above are merely for the purpose of helping to understand the method and its core ideas. It should be noted that those skilled in the art can make various improvements and modifications to this application without departing from its principles, and these improvements and modifications also fall within the protection scope of this application.

Claims

1. A substation operation status reasoning and decision-making method based on cloud analytics, characterized in that, include: To obtain the concentration of decomposition products, gas purity, gas moisture content, ambient temperature and humidity data of the gas-insulated switchgear inside the gas-insulated switchgear in the substation, as well as leakage current data and dielectric loss factor data of the gas-insulated switchgear. By configuring edge nodes around the gas-insulated switchgear, outlier removal and feature fusion processing are performed on the concentration of decomposition products, the purity of the gas, the moisture content of the gas, the ambient temperature and humidity data, the leakage current data, and the dielectric loss factor data to obtain fused data, and the fused data is uploaded to the cloud intelligent analysis platform. Gas decomposition characteristics, insulation electrical parameters, and environmental impact factors are extracted from the fused data. The fused data is then collaboratively analyzed through the cloud-based intelligent analysis platform to identify the degree of insulation degradation of gas-insulated switchgear. Based on the degree of insulation degradation, the overvoltage protection threshold of the gas-insulated switchgear is adjusted to obtain the adjusted overvoltage protection threshold, so as to trigger the protection decision corresponding to the degree of insulation degradation. The protection decision includes protection decisions corresponding to early warning prompts, load transfer strategies, and emergency maintenance strategies, respectively. Gas decomposition characteristics, insulation electrical parameters, and environmental impact factors are extracted from the fused data. These are then used in a collaborative analysis via the cloud-based intelligent analysis platform to identify the degree of insulation degradation in gas-insulated switchgear, including: Gas decomposition characteristics, insulation electrical parameters, and environmental impact factors are extracted from the fused data. The gas decomposition characteristics include the concentration variation trend of the decomposition products of the target gas for insulation, the stable range of gas purity, and the fluctuation characteristics of moisture content. The insulation electrical parameters include the variation law of leakage current amplitude and the distribution characteristics of dielectric loss factor. The environmental impact factors include the synergistic variation relationship of ambient temperature and humidity. Using the cloud-based intelligent analysis platform, the gas decomposition characteristics and the insulation electrical parameters are cross-compared to determine the degree of correlation between the gas decomposition characteristics and the insulation electrical parameters. The degree of correlation is then matched with the synergistic change relationship in the environmental impact factors to obtain the synergistic relationship between the gas decomposition characteristics, the insulation electrical parameters, and the environmental impact factors. The collaborative relationship is compared with the historical feature data of gas-insulated switchgear under different insulation states pre-stored in the cloud-based intelligent analysis platform to determine the insulation state corresponding to the collaborative relationship; Based on the insulation condition, the degree of insulation degradation of the gas-insulated switchgear is determined.

2. The method according to claim 1, characterized in that, Using the cloud-based intelligent analysis platform, the gas decomposition characteristics and the insulation electrical parameters are cross-compared to determine the degree of correlation between them. This correlation is then matched with the synergistic changes in the environmental impact factors to obtain the synergistic relationships among the gas decomposition characteristics, insulation electrical parameters, and environmental impact factors, including: The statistics include the percentage of times when the trend of the concentration change of the decomposition products of the target gas for insulation is consistent with the trend of the change of the leakage current amplitude during the same data acquisition period; the percentage of times when the distribution characteristics of the dielectric loss factor are within the normal factor range corresponding to the stable range of gas purity; and the percentage of times when the fluctuation characteristics of the moisture content exceed the preset normal fluctuation range and the distribution characteristics of the dielectric loss factor show abnormal distribution. The degree of correlation between gas decomposition characteristics and insulation electrical parameters is determined based on the percentage of occurrences, the percentage of values, and the percentage of abnormal occurrences. The aforementioned cooperative change relationship is divided into high temperature and high humidity mode, high temperature and low humidity mode, low temperature and high humidity mode, and low temperature and low humidity mode. The correlation degree is matched with the corresponding normal correlation degree ranges under the high temperature and high humidity mode, high temperature and low humidity mode, low temperature and high humidity mode, and low temperature and low humidity mode, respectively, to determine the reasonable correlation level corresponding to the correlation degree. Based on the synergistic change relationship, the reasonable correlation level, and the correlation degree, a synergistic relationship is generated between gas decomposition characteristics, insulation electrical parameters, and environmental impact factors.

3. The method according to claim 1, characterized in that, The collaborative relationship is compared with historical characteristic data of gas-insulated switchgear under different insulation states pre-stored in the cloud-based intelligent analysis platform to determine the insulation state corresponding to the collaborative relationship, including: The historical characteristic data of gas-insulated switchgear under different insulation states pre-stored in the cloud-based intelligent analysis platform are classified to obtain multiple historical collaborative relationship groups; Select a target historical collaborative relationship group from the plurality of historical collaborative relationship groups that is consistent with the collaborative change relationship of the collaborative relationship; The reasonable association level of the aforementioned collaborative relationship is compared with the reasonable association level in each target historical collaborative relationship group to select candidate historical collaborative relationship groups. The percentage of times, percentage of values, and percentage of anomalies in the degree of association of the collaborative relationship are compared with the normal value range of each percentage in each candidate historical collaborative relationship group, and the number of matching items in each candidate historical collaborative relationship group where each percentage is within the corresponding normal value range is counted. The insulation state of the gas-insulated switchgear corresponding to the candidate historical cooperative relationship group with the most matching items is selected as the insulation state corresponding to the cooperative relationship.

4. The method according to claim 1, characterized in that, Based on the insulation condition, the degree of insulation degradation of the gas-insulated switchgear is determined, including: Key data were extracted from the insulation state. The key data is compared with each insulation condition judgment standard to count the number of data items that meet the first insulation condition judgment standard, the second insulation condition judgment standard, and the third insulation condition judgment standard. The insulation condition corresponding to the largest number of data items is selected as the insulation degradation degree of the gas-insulated switchgear.

5. The method according to claim 1, characterized in that, By configuring edge nodes around the gas-insulated switchgear, outlier removal and feature fusion processing are performed on the concentration of decomposition products, the purity of the gas, the moisture content of the gas, the ambient temperature and humidity data, the leakage current data, and the dielectric loss factor data to obtain fused data, including: The concentration of the decomposition products, the purity of the gas, and the moisture content of the gas are divided into multiple data groups, and the average fluctuation value within each data group is calculated. By configuring edge nodes around the gas-insulated switchgear, the first abnormal data in each data group whose deviation from the average fluctuation value exceeds a preset deviation range is removed to form a first type of target data. Calculate the difference in leakage current and the difference in dielectric loss factor between adjacent acquisition times, and set a first abnormal threshold and a second abnormal threshold based on all the differences in leakage current and dielectric loss factor. Remove the last abnormal leakage current in the leakage current data where the leakage current difference value exceeds the first abnormal threshold, and remove the last dielectric loss factor in the dielectric loss factor data where the dielectric loss factor difference value exceeds the second abnormal threshold to obtain the second type of target data. The second abnormal data that does not conform to the preset reasonable deviation range in the environmental temperature and humidity data is removed to obtain the third type of target data; Temporal alignment and feature concatenation are performed on the first type of target data, the second type of target data, and the third type of target data to generate fused data.

6. The method according to claim 1, characterized in that, Based on the degree of insulation degradation, the overvoltage protection threshold of the gas-insulated switchgear is adjusted to obtain the adjusted overvoltage protection threshold, thereby triggering a protection decision corresponding to the degree of insulation degradation. The protection decision includes protection decisions corresponding to early warning alerts, load transfer strategies, and emergency maintenance strategies, respectively. Based on the insulation withstand characteristics of gas-insulated switchgear, the degree of abnormality of key parameters corresponding to the degree of insulation degradation, and the operation and maintenance safety redundancy requirements of substations, voltage adjustment rules corresponding to the degree of insulation degradation are set. Obtain the overvoltage protection threshold of the gas-insulated switchgear, and adjust the overvoltage protection threshold according to the voltage adjustment rules to obtain the adjusted overvoltage protection threshold; When the insulation degradation level is slightly degraded, a protective decision is triggered to provide an early warning. The early warning is used to instruct the maintenance terminal to send the insulation degradation level, the adjusted overvoltage protection threshold, and the inspection cycle information. When the insulation degradation level is moderate, the load transfer strategy is triggered to make a protection decision. The load transfer strategy is used to indicate that the target load within the risk range defined according to the adjusted overvoltage protection threshold is transferred to the preset backup line and to send an early warning to the operation and maintenance terminal. When the insulation degradation reaches a severe degradation state, the protection decision of the emergency maintenance strategy is triggered. The emergency maintenance strategy is used to instruct the start of the equipment shutdown procedure, lock the adjusted overvoltage protection threshold, send the emergency maintenance command to the operation and maintenance terminal, and transfer the entire load to the preset backup line.

7. A cloud-based substation operation status reasoning and decision-making system, used to execute the cloud-based substation operation status reasoning and decision-making method according to any one of claims 1 to 6, characterized in that, include: The acquisition module is used to acquire the concentration of decomposition products of the target gas used for insulation inside the gas-insulated switchgear of the substation, gas purity, gas moisture content, ambient temperature and humidity data of the gas-insulated switchgear, as well as leakage current data and dielectric loss factor data of the gas-insulated switchgear. The processing module is used to perform outlier removal and feature fusion processing on the concentration of decomposition products, the purity of gas, the moisture content of gas, the ambient temperature and humidity data, the leakage current data, and the dielectric loss factor data through edge nodes configured around the gas-insulated switchgear, to obtain fused data, and upload the fused data to the cloud intelligent analysis platform. The identification module is used to extract gas decomposition characteristics, insulation electrical parameters and environmental impact factors from the fused data, so as to identify the degree of insulation degradation of the gas-insulated switchgear through collaborative analysis of the fused data by the cloud-based intelligent analysis platform. The adjustment module is used to adjust the overvoltage protection threshold of the gas-insulated switchgear according to the degree of insulation degradation, so as to obtain the adjusted overvoltage protection threshold and trigger the protection decision corresponding to the degree of insulation degradation. The protection decision includes protection decisions corresponding to early warning prompts, load transfer strategies and emergency maintenance strategies, respectively.

8. An electronic device, characterized in that, include: Memory, used to store computer programs; A processor, configured to execute the computer program to implement the steps of the cloud-based analysis-based substation operation status reasoning and decision-making method as described in any one of claims 1 to 6.

9. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program that, when executed by a processor, enables the substation operation status reasoning and decision-making method based on cloud analysis as described in any one of claims 1 to 6.

Images