An environmentally friendly gas switchgear cooperative control method for a microgrid

Abstract

The present application relates to the technical field of micro-grid equipment control, and discloses an environmental protection gas switch equipment cooperative control method for a micro-grid, comprising: real-time monitoring of gas state parameters and local discharge characteristic quantities of each switch equipment, and analysis, obtaining a health degree index and a risk probability of the corresponding switch equipment; determining whether to issue a warning message according to the health degree and the risk probability, if yes, determining a first correlation level table and a second correlation level table of each switch equipment issuing the warning message; generating a warning credibility of the corresponding switch equipment according to the first correlation level table, if the warning credibility is greater than a preset credibility threshold, combining the corresponding second correlation level table to cooperatively analyze and judge, and forming a cooperative analysis and judgment result; and generating a cooperative control instruction based on the cooperative analysis and judgment result and a preset interlocking control logic, realizing state mutual feedback and cooperative decision between the switch equipments, and effectively improving power supply continuity and operation reliability of the micro-grid.

Description

Technical Field

[0001] This application relates to the field of microgrid equipment control technology, and in particular to a collaborative control method for environmentally friendly gas switchgear used in microgrids. Background Technology

[0002] As an autonomous system integrating distributed power sources, loads, energy storage, and monitoring and protection devices, microgrids exhibit significant volatility and intermittency in their internal sources and loads. This multi-source fluctuation scenario places higher demands on the operational reliability of critical electrical equipment in microgrids, especially switching equipment.

[0003] In existing technologies, environmentally friendly gas-insulated switchgear widely used in microgrids relies primarily on traditional electrical quantity monitoring (such as current and voltage) and independent equipment status alarms for protection and control. However, under the complex operating conditions of microgrids, the lack of real-time sensing of the environmentally friendly gas status and partial discharge activity of the switchgear makes comprehensive judgment impossible. Furthermore, the isolated and delayed control decisions of each switchgear lead to unnecessary tripping or failure to isolate real faults in a timely manner. This hinders the achievement of status feedback and collaborative decision-making between equipment, affecting the power supply continuity and operational reliability of the microgrid. Summary of the Invention

[0004] To address the aforementioned technical issues, this application provides a collaborative control method for environmentally friendly gas switchgear in microgrids. By constructing a health assessment and risk prediction model, it provides accurate early warnings of switchgear faults; it constructs a first and second association hierarchy table to generate a collaborative assessment result including a priority-ranked list of faulty equipment and a comprehensive association impact value; and, combined with a pre-set interlocking control logic library, it intelligently generates and issues collaborative control commands to achieve state feedback and collaborative decision-making among switchgear, effectively improving the power supply continuity and operational reliability of the microgrid.

[0005] In some embodiments of this application, a method for coordinated control of environmentally friendly gas switchgear for microgrids is provided, including: Real-time monitoring and analysis of gas state parameters and partial discharge characteristics of each switchgear are performed to obtain the health index and risk probability of the corresponding switchgear. Determine whether to issue a warning message based on the health status and risk probability. If so, determine the first association level table and the second association level table for each switchgear that issued the warning message. The warning confidence level of the corresponding switching device is generated based on the first association hierarchy table. If the warning confidence level is greater than the preset confidence level threshold, the corresponding second association hierarchy table is used for collaborative analysis, and a collaborative analysis result is formed. Based on the collaborative analysis results and the preset interlocking control logic, collaborative control commands are generated.

[0006] In some embodiments of this application, the health status and risk probability of the corresponding switching device are obtained, including: The gas state parameters and partial discharge characteristics are standardized, and features are extracted from the standardized gas state parameters and partial discharge characteristics to obtain several first features and several second features. The gas state degradation index is obtained by normalizing all the first characteristics and then weighting them. The partial discharge activity index is obtained by normalizing all the second features and then weighting them. A health index is generated based on the gas state deterioration index and the partial discharge activity index. Construct a feature vector, which includes all features within a preset time period; Based on a pre-trained risk prediction model, risk prediction is performed on the feature vector to obtain the failure risk probability.

[0007] In some embodiments of this application, determining whether to issue an early warning message based on health status and risk probability includes: Pre-set preset health index thresholds and preset risk probability thresholds; When the health index of the switching equipment is lower than the preset health index threshold, a level 1 early warning message is issued; When the probability of failure risk of the switching equipment exceeds the preset risk probability threshold, a level 2 early warning message is issued. If the health index is not less than the preset health index threshold and the failure risk probability is not greater than the preset risk probability threshold, no early warning message will be issued.

[0008] In some embodiments of this application, before determining the first association table and the second association table of each switching device that issues an early warning message, the method further includes: When a Level 1 early warning message is issued, the feature vector of the corresponding switching equipment within a preset time period is matched with a preset equipment fault fingerprint database to obtain a number of matching degrees. The preset equipment fault fingerprint database includes several preset fault types, and each preset fault type is mapped to a corresponding fault feature vector. The preset fault types with a matching degree greater than the preset matching degree threshold are set as several first fault types of the corresponding switch at the current time node, and several corresponding first confidence levels are generated. When a Level II early warning message is issued, time-series predictions are performed on all first features and all second features to generate a predicted change curve for each feature in the future period. The curve is then compared with a preset fault change template library to obtain several comprehensive similarity scores. The preset fault change template library includes several preset fault types, and each preset fault type is mapped to a standard fault change curve with several preset features. Preset fault types with a comprehensive similarity greater than a preset similarity threshold are designated as the second fault type of the corresponding switching equipment in the future time period, and several corresponding second credibility scores are generated.

[0009] In some embodiments of this application, determining a first association hierarchy table and a second association hierarchy table for each switching device that issues an early warning message includes: Randomly select one of the switching devices that issued the early warning message as the target device; Based on the fault type of the target device, several relevant historical fault logs are filtered out, and the first and second associated time periods are set in each historical fault log. Extract the historical health index of other switchgear that has an electrical connection, data communication relationship or topological proximity to the target device during the first and second correlation periods from the selected historical fault logs; Calculate the frequency of historical health index being less than a preset health index threshold, identify other switching devices with a frequency greater than a preset frequency threshold in all first associated time periods as the first associated devices of the target device, and identify other switching devices with a frequency greater than a preset frequency threshold in all second associated time periods as the second associated devices of the target device; Generate a first association hierarchy table of the target device based on all first associated devices of the same fault type, and generate a second association hierarchy table of the target device based on all second associated devices of the same fault type. Several first association level tables and several second association level tables are generated sequentially for each switching device that issues an early warning message.

[0010] In some embodiments of this application, the target device, all first associated devices under each fault type, and all second associated devices are marked in a preset microgrid electrical topology diagram to obtain several first associated device topology annotation diagrams and several second associated device topology annotation diagrams. Based on the topology annotation map of the first associated device, the first topological distance between each first associated device and the target device is extracted, and the association weight of each first associated device is generated by combining the frequency of collaborative failures with the target device and the probability of occurrence of the corresponding first failure type. According to the mapping relationship between the association weight of the first associated device and the corresponding preset first association weight range, a first association hierarchy table is formed for each fault type of the target device. The first association hierarchy table includes several first association levels, and each first association level corresponds to at least one first associated device and its association weight. Based on the topology annotation map of the second associated device, the second topology distance between each second associated device and the target device is extracted, and the association weight of each second associated device is generated by combining the frequency of collaborative failures with the target device and the probability of occurrence of the second failure type. Based on the mapping relationship between the association weight of the second associated device and the corresponding preset second association weight range, a second association hierarchy table is formed for each fault type of the target device. The second association hierarchy table includes several second association levels, and each second association level corresponds to at least one second associated device and its association weight.

[0011] In some embodiments of this application, generating the warning confidence level of the corresponding switching device based on the first association hierarchy table includes: Set the actual first associated time period for the current time node; Obtain the health index of each first associated device of the switching equipment that issued the early warning message during the actual first associated period; The number of normal states and the number of abnormal states of all first-association devices are counted, and combined with the corresponding association weights, the early warning reliability compensation coefficient for each first association level in the corresponding first association hierarchy table for the corresponding fault type is generated. Based on the early warning reliability compensation coefficient of each first association level and the weight coefficient of the corresponding level, a comprehensive early warning reliability compensation coefficient for the corresponding fault type is generated. The warning credibility of the switchgear warning message is generated based on the comprehensive warning credibility compensation coefficient for each fault type and the corresponding credibility.

[0012] In some embodiments of this application, collaborative judgment is performed in conjunction with the corresponding second association hierarchy table to form a collaborative judgment result, including: Remove switching devices whose warning confidence level is not greater than the preset confidence level threshold. Based on the edge computing node cluster, use time priority, signal strength gradient and causality test to identify the faulty switching devices and output a list of faulty devices. The list of faulty devices includes the initial impact level for each remaining switchgear; Based on the number of second-related devices in the second-related hierarchy table for each remaining switching device, the weight coefficient of each second-related device, and the related weight, calculate the predicted related influence value for each second-related hierarchy table. Based on the predicted association impact value of each second association level and the weight coefficient of the corresponding level, the comprehensive association impact value of the corresponding switchgear is generated. An adjustment instruction for the initial impact level is generated based on the comprehensive correlation impact value, and the remaining switching equipment is prioritized according to the adjusted impact level to obtain a collaborative assessment result. The collaborative assessment result includes the priority order of the faulty equipment, the comprehensive correlation impact value of each switching equipment, and the adjusted impact level.

[0013] In some embodiments of this application, collaborative control instructions are generated based on collaborative judgment results and a preset interlocking control logic library, including: The collaborative analysis results are used as input. The condition matching engine retrieves all interlocking rules that meet the triggering conditions from the preset interlocking control logic library, instantiates the parameters, and generates the initial collaborative control command. The initial collaborative control command is conflict detected. If a conflict exists, the command is optimized according to the preset conflict resolution strategy, and the final collaborative control command is generated and sent to the corresponding environmental gas switchgear for execution through the edge computing node. The execution status of the instructions is fed back to the microgrid monitoring center in real time, and it is determined whether the coordinated control instructions need to be corrected.

[0014] In some embodiments of this application, an electronic device is also included, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, when the processor executes the program, it implements the collaborative control method for environmentally friendly gas switching devices for microgrids as described in any one of claims 1 to 9.

[0015] The method for coordinated control of environmentally friendly gas switchgear in microgrids according to an embodiment of this application has the following advantages compared with the prior art: By constructing a health assessment and risk prediction model, accurate early warnings are provided for switchgear failures; a first association hierarchy table and a second association hierarchy table are constructed to generate a collaborative judgment result that includes a priority-ranked list of faulty equipment and a comprehensive association impact value; and combined with a preset interlocking control logic library, collaborative control commands are intelligently generated and issued to realize status feedback and collaborative decision-making among switchgear, effectively improving the power supply continuity and operational reliability of the microgrid. Attached Figure Description

[0016] Figure 1 This is a flowchart illustrating a collaborative control method for environmentally friendly gas switchgear in a microgrid, as described in an embodiment of this application. Figure 2 This is a schematic diagram of the structure of the electronic device in the embodiments of this application. Detailed Implementation

[0017] The specific embodiments of this application will be described in further detail below with reference to the accompanying drawings and examples. The following examples are used to illustrate this application, but are not intended to limit the scope of this application.

[0018] In the description of this application, it should be understood that the terms "center", "upper", "lower", "front", "rear", "left", "right", "vertical", "horizontal", "top", "bottom", "inner", "outer", etc., indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings. They are only for the convenience of describing this application and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, they should not be construed as limitations on this application.

[0019] The terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of technical features indicated. Therefore, a feature defined as "first" or "second" may explicitly or implicitly include one or more of that feature. In the description of this application, unless otherwise stated, "a plurality of" means two or more.

[0020] In the description of this application, it should be noted that, unless otherwise expressly specified and limited, the terms "installation," "connection," and "linking" should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral connection; they can refer to a mechanical connection or an electrical connection; they can refer to a direct connection or an indirect connection through an intermediate medium; and they can refer to the internal connection between two components. Those skilled in the art can understand the specific meaning of the above terms in this application based on the specific circumstances.

[0021] like Figure 1 As shown in the figure, an embodiment of this application provides a collaborative control method for environmentally friendly gas switchgear in a microgrid, comprising: S101: Real-time monitoring of gas state parameters and partial discharge characteristics of each switchgear, and analysis to obtain the health index and risk probability of the corresponding switchgear. S102: Determine whether to issue a pre-alarm message based on the health status and risk probability. If so, determine the first association level table and the second association level table for each switchgear that issued the pre-alarm message. S103: Generate the warning confidence level of the corresponding switching device based on the first association hierarchy table. If the warning confidence level is greater than the preset confidence level threshold, perform collaborative analysis in conjunction with the corresponding second association hierarchy table and form a collaborative analysis result. S104: Generate collaborative control instructions based on collaborative analysis results and preset interlocking control logic.

[0022] In this embodiment, the gas state parameters include pressure P, temperature T, and humidity H, and the partial discharge characteristic quantities include pulse sequence (amplitude q, phase). (Time t).

[0023] In some embodiments of this application, the health status and risk probability of the corresponding switching device are obtained, including: The gas state parameters and partial discharge characteristics are standardized, and features are extracted from the standardized gas state parameters and partial discharge characteristics to obtain several first features and several second features. The gas state degradation index is obtained by normalizing all the first characteristics and then weighting them. The partial discharge activity index is obtained by normalizing all the second features and then weighting them. A health index is generated based on the gas state deterioration index and the partial discharge activity index. Construct a feature vector, which includes all features within a preset time period; Based on a pre-trained risk prediction model, risk prediction is performed on the feature vector to obtain the failure risk probability.

[0024] In this embodiment, the first feature includes the relative change rate of gas density Δρ, the humidity dew point offset Ddp (calculated by comparing the dew point temperature at the current pressure with the maximum allowable dew point of the device), and the pressure-temperature change correlation, where ρ = P / R T, Δρ(t) = (ρ(t)) ρ_rated) / ρ_rated × 100%.

[0025] In this embodiment, the second feature includes the average discharge amplitude q. - The parameters include: maximum discharge amplitude, discharge repetition rate N (number of discharge pulses per unit time), discharge asymmetry Asymmetry (difference in discharge quantity or number of discharges during positive and negative half-cycles at power frequency), spectral statistical characteristics (describing the concentration and dispersion of discharge pulses in phase distribution), and trend characteristics (the slope of the above characteristics within the sliding time window), where Asym = |N + -N ∣ / ∣N + +N |

[0026] In this embodiment, the partial discharge activity index = α1 f(Δρ)+α2 g(Ddp)+α3 h (correlation outliers), where f, g, and h are functions that map physical quantities to scores of 0-100 (e.g., f is 0 when density decreases by 5%, 100 when it decreases by 20%, with linear interpolation in between), α1 is 0.4, α2 is 0.3, and α3 is 0.3.

[0027] In this embodiment, the partial discharge activity index = β1 u(q - )+β2 v(N)+β3 w(Asym)+β4 x (spectral feature changes), where u, v, w, x are mapping functions, β1 is 0.3, β2 is 0.25, β3 is 0.2, and β4 is 0.25.

[0028] In this embodiment, the health index is obtained by weighted summation of the gas state deterioration index and the partial discharge activity index, wherein the weight of the gas state deterioration index is 0.45 and the weight of the partial discharge activity index is 0.55.

[0029] In this embodiment, the feature vector includes all first features, all second features, as well as original values, statistical values ​​(mean, variance), and trend values.

[0030] In this embodiment, the risk prediction model adopts a deep learning model based on LSTM. A historical dataset is constructed by using normal operating condition data and data on the development of anomalies into failures. The historical dataset is labeled with risk levels and trained to obtain the risk prediction model. Its input is a feature vector containing all first and second features within a preset time period (such as the past 24 hours), and the output is the probability of failure occurring in the next 24 hours.

[0031] In this embodiment, the health index and fault risk probability are calculated to comprehensively determine whether a pre-alarm message for the corresponding switching equipment needs to be issued, laying the foundation for the subsequent generation of coordinated control commands. This can effectively improve the safety and reliability of the switching equipment operation, reduce the fault rate, and ensure the stable operation of the microgrid system.

[0032] In some embodiments of this application, determining whether to issue an early warning message based on health status and risk probability includes: Pre-set preset health index thresholds and preset risk probability thresholds; When the health index of the switching equipment is lower than the preset health index threshold, a level 1 early warning message is issued; When the probability of failure risk of the switching equipment exceeds the preset risk probability threshold, a level 2 early warning message is issued. If the health index is not less than the preset health index threshold and the failure risk probability is not greater than the preset risk probability threshold, no early warning message will be issued.

[0033] In this embodiment, the preset health index threshold and the preset risk probability threshold can be dynamically adjusted and set according to the operating requirements of the microgrid system, the importance of the switching equipment, and historical fault data to adapt to the early warning needs in different scenarios. In this application, the preset health index threshold is 60 and the preset risk probability threshold is 30%.

[0034] In this embodiment, the warning level of the Level 1 warning message is higher than that of the Level 2 warning message. By setting up a multi-level warning mechanism, different levels of warnings can be issued according to the actual health status and fault risk of the switching equipment, so that maintenance personnel can take targeted measures and improve the accuracy and effectiveness of the warnings.

[0035] In some embodiments of this application, before determining the first association table and the second association table of each switching device that issues an early warning message, the method further includes: When a Level 1 early warning message is issued, the feature vector of the corresponding switching equipment within a preset time period is matched with a preset equipment fault fingerprint database to obtain a number of matching degrees. The preset equipment fault fingerprint database includes several preset fault types, and each preset fault type is mapped to a corresponding fault feature vector. The preset fault types with a matching degree greater than the preset matching degree threshold are set as several first fault types of the corresponding switch at the current time node, and several corresponding first confidence levels are generated. When a Level II early warning message is issued, time-series predictions are performed on all first features and all second features to generate a predicted change curve for each feature in the future period. The curve is then compared with a preset fault change template library to obtain several comprehensive similarity scores. The preset fault change template library includes several preset fault types, and each preset fault type is mapped to a standard fault change curve with several preset features. Preset fault types with a comprehensive similarity greater than a preset similarity threshold are designated as the second fault type of the corresponding switching equipment in the future time period, and several corresponding second credibility scores are generated.

[0036] In this embodiment, the fault feature vectors stored in the preset equipment fault fingerprint database are historical feature vectors of different historical fault types of the corresponding switching equipment extracted from a large number of historical fault logs. The preset fault change template database is the historical change curve of each feature of different historical fault types of the corresponding switching equipment extracted from a large number of historical fault logs. The historical change curve refers to the feature change trend curve from the historical fault start node to the historical stable node. Several preset fault types include Type I gas leak, Type II gas leak, moisture, suspension discharge, gas gap discharge, surface discharge, etc.

[0037] In this embodiment, the matching degree refers to the Euclidean distance or cosine similarity between the feature vector and the fault feature vector of each preset fault type. By calculating the distance between the current feature vector and the feature vectors of each fault type in the fault fingerprint database, the matching degree is greater when the distance is shorter and smaller when the distance is greater. The comprehensive similarity is obtained by performing dynamic time warping (DTW) on the predicted change curve of each feature and the standard fault change curve, obtaining the DTW distance, and normalizing it to obtain the similarity between each feature and the standard fault change curve of the same preset fault type. The comprehensive similarity is obtained by using a weighted average method (such as assigning coefficients according to the influence of the feature on the fault).

[0038] In this embodiment, the preset matching degree threshold is 85%, the preset similarity threshold is 80%, and the first confidence degree and the second confidence degree are obtained by normalization of the matching degree or the comprehensive similarity degree, with a value range of 0-100%.

[0039] In this embodiment, by determining the current fault type and the predicted fault type, the foundation is laid for the subsequent determination of the first association table and the second association table for each switching device. The construction of these two association tables makes the early warning information no longer isolated state data, but structured information associated with specific fault modes, providing a clear analytical direction and data support for subsequent collaborative judgment.

[0040] In some embodiments of this application, determining a first association hierarchy table and a second association hierarchy table for each switching device that issues an early warning message includes: Randomly select one of the switching devices that issued the early warning message as the target device; Based on the fault type of the target device, several relevant historical fault logs are filtered out, and the first and second associated time periods are set in each historical fault log. Extract the historical health index of other switchgear that has an electrical connection, data communication relationship or topological proximity to the target device during the first and second correlation periods from the selected historical fault logs; Calculate the frequency of historical health index being less than a preset health index threshold, identify other switching devices with a frequency greater than a preset frequency threshold in all first associated time periods as the first associated devices of the target device, and identify other switching devices with a frequency greater than a preset frequency threshold in all second associated time periods as the second associated devices of the target device; Generate a first association hierarchy table of the target device based on all first associated devices of the same fault type, and generate a second association hierarchy table of the target device based on all second associated devices of the same fault type. Several first association level tables and several second association level tables are generated sequentially for each switching device that issues an early warning message.

[0041] In this embodiment, the first associated time period refers to the preset time period before the occurrence of the first fault type in the historical fault log. The preset time period refers to the 24 hours before the historical start time node of the fault type and the historical start time node. The second associated time period refers to the 48 hours after the historical start time node of the fault type in the historical fault log.

[0042] In this embodiment, when the target device corresponds to one first fault type, there is only one first association level table and one second association level table. If it corresponds to multiple first fault types, a corresponding first association level table and a second association level table are generated for each first fault type. The same applies when the target device corresponds to a second fault type, which will not be elaborated here.

[0043] In this embodiment, by constructing a first association hierarchy table, it is possible to identify the associated devices and their association weights that have exhibited abnormal health status in the first association period in historical data when the target device experiences a specific first fault type or a predicted second fault type. By constructing a second association hierarchy table, it is possible to identify the associated devices and their association weights that have exhibited abnormal health status in the second association period in historical data when the target device experiences a specific second fault type. This links the fault warning of the target device with the status of other devices that have spatiotemporal correlations, expanding the warning information of a single device into a status correlation analysis of a device cluster. This provides a structured basis for subsequent collaborative judgment based on the status of associated devices, avoiding the limitations of single device warnings, realizing a systematic assessment of the scope of fault impact and potential associated risks, avoiding misjudgments or omissions that may result from isolated judgments, and improving the comprehensiveness and accuracy of the warning information.

[0044] In some embodiments of this application, the target device, all first associated devices under each fault type, and all second associated devices are marked in a preset microgrid electrical topology diagram to obtain several first associated device topology annotation diagrams and several second associated device topology annotation diagrams. Based on the topology annotation map of the first associated device, the first topological distance between each first associated device and the target device is extracted, and the association weight of each first associated device is generated by combining the frequency of collaborative failures with the target device and the probability of occurrence of the corresponding first failure type. According to the mapping relationship between the association weight of the first associated device and the corresponding preset first association weight range, a first association hierarchy table is formed for each fault type of the target device. The first association hierarchy table includes several first association levels, and each first association level corresponds to at least one first associated device and its association weight. Based on the topology annotation map of the second associated device, the second topology distance between each second associated device and the target device is extracted, and the association weight of each second associated device is generated by combining the frequency of collaborative failures with the target device and the probability of occurrence of the second failure type. Based on the mapping relationship between the association weight of the second associated device and the corresponding preset second association weight range, a second association hierarchy table is formed for each fault type of the target device. The second association hierarchy table includes several second association levels, and each second association level corresponds to at least one second associated device and its association weight.

[0045] In this embodiment, topological distance refers to the number of nodes contained in the shortest path connecting two switching devices via lines in the electrical topology diagram. For example, the topological distance between the target device and a directly connected device is 1, and the topological distance between the target device and a device connected by one device is 2. Cooperative fault frequency refers to the number of times the target device and associated devices have successively failed in the same historical fault log. A higher cooperative fault frequency indicates a stronger correlation and a greater correlation weight between the two. The probability of occurrence of the first fault type refers to the probability that a historical fault type of the first associated device can cause a fault type of the target device to occur, and the probability of occurrence of the second fault type refers to the probability that a fault type of the target device causes a corresponding historical fault type to occur in the second associated device.

[0046] In this embodiment, the association weight is calculated based on a combination of topological distance, frequency of cooperative failures, and probability of failure type occurrence. The specific calculation formula is: Association weight = ω1 (1 / topological distance) + ω2 (Frequency of collaborative failures / Total number of collaborative failures) + ω3 The probability of fault type occurrence is given by ω1, ω2, and ω3, which are the weight coefficients of each factor, and ω1 + ω2 + ω3 = 1. In this application, ω1 = 0.3, ω2 = 0.4, and ω3 = 0.3. This formula quantifies the correlation factors of different dimensions into a unified weight value, more accurately reflecting the degree of correlation between devices.

[0047] In this embodiment, the preset first association weight interval is divided into three intervals: [0.7, 1.0], [0.4, 0.7], and [0.1, 0.4], which correspond to the first association level one, the first association level two, and the first association level three, respectively. The preset second association weight interval is divided into three intervals: [0.6, 1.0], [0.3, 0.6], and [0.0, 0.3], which correspond to the second association level one, the second association level two, and the second association level three, respectively.

[0048] In this embodiment, by constructing an association hierarchy table, the degree of association between the target device and other switching devices can be clearly displayed, providing structured association information support for the subsequent early warning credibility and collaborative judgment, making the judgment process more targeted and hierarchical.

[0049] In some embodiments of this application, generating the warning confidence level of the corresponding switching device based on the first association hierarchy table includes: Set the actual first associated time period for the current time node; Obtain the health index of each first associated device of the switching equipment that issued the early warning message during the actual first associated period; The number of normal states and the number of abnormal states of all first-association devices are counted, and combined with the corresponding association weights, the early warning reliability compensation coefficient for each first association level in the corresponding first association hierarchy table for the corresponding fault type is generated. Based on the early warning reliability compensation coefficient of each first association level and the weight coefficient of the corresponding level, a comprehensive early warning reliability compensation coefficient for the corresponding fault type is generated. The warning credibility of the switchgear warning message is generated based on the comprehensive warning credibility compensation coefficient for each fault type and the corresponding credibility.

[0050] In this embodiment, when there is only one first fault type or one second fault type, the comprehensive early warning credibility compensation coefficient generated by the first association level table corresponding to the fault type and the credibility of the corresponding fault type are directly used to generate the early warning credibility. If there is not one fault type, the early warning credibility generated by each fault type is then weighted and averaged to obtain the final early warning credibility.

[0051] In this embodiment, the setting of the actual first associated time period is consistent with the definition of the first associated time period in the first associated hierarchy table. That is, when the target device corresponds to the first fault type, the actual first associated time period is the 24 hours before the current time node to the current time node.

[0052] In this embodiment, if the health index is less than a preset health index threshold, the first associated device is determined to be a device in a normal state; otherwise, it is a device in an abnormal state. The association weights corresponding to normal state devices at the same level in the same hierarchical table are summed, and the association weights corresponding to abnormal state devices are summed to obtain the sum of normal weights and the sum of abnormal weights for that level. The formula for calculating the early warning reliability compensation coefficient is: Early Warning Reliability Compensation Coefficient = γ (Normal weight sum - Abnormal weight sum) / (Abnormal weight sum + Normal weight sum), where γ is a mapping function that maps the ratio to 0.8-1.2. When the calculation result is negative, the early warning credibility compensation coefficient ranges from 0.8 to 1. When the calculation result is positive, the early warning credibility compensation coefficient ranges from 1 to 1.2. When the calculation result is 0, the early warning credibility compensation coefficient is 1.

[0053] In this embodiment, the weight coefficient of each first association level is set according to the importance of the level. For example, the weight coefficient of the first association level one is 0.5, the weight coefficient of the first association level two is 0.3, and the weight coefficient of the first association level three is 0.2. The comprehensive early warning credibility compensation coefficient is obtained by weighted summation of the early warning credibility compensation coefficient of each level and its weight coefficient.

[0054] In this embodiment, the early warning reliability of the switching equipment is calculated by the formula "early warning reliability = first reliability × comprehensive early warning reliability compensation coefficient". This value can comprehensively reflect the matching degree of the target equipment's own fault characteristics and the influence of the status of related equipment on the early warning, further improving the reliability of the early warning information.

[0055] In this embodiment, by calculating the early warning credibility compensation coefficient, the early warning credibility of the target device is dynamically adjusted based on the actual state of the associated device. When the associated device exhibits an abnormal state consistent with the historical fault mode, the early warning credibility is increased; otherwise, the credibility is reduced or maintained, making the early warning result more consistent with the actual operating conditions.

[0056] In some embodiments of this application, collaborative judgment is performed in conjunction with the corresponding second association hierarchy table to form a collaborative judgment result, including: Remove switching devices whose warning confidence level is not greater than the preset confidence level threshold. Based on the edge computing node cluster, use time priority, signal strength gradient and causality test to identify the faulty switching devices and output a list of faulty devices. The list of faulty devices includes the initial impact level for each remaining switchgear; Based on the number of second-related devices in the second-related hierarchy table for each remaining switching device, the weight coefficient of each second-related device, and the related weight, calculate the predicted related influence value for each second-related hierarchy table. Based on the predicted association impact value of each second association level and the weight coefficient of the corresponding level, the comprehensive association impact value of the corresponding switchgear is generated. An adjustment instruction for the initial impact level is generated based on the comprehensive correlation impact value, and the remaining switching equipment is prioritized according to the adjusted impact level to obtain a collaborative assessment result. The collaborative assessment result includes the priority order of the faulty equipment, the comprehensive correlation impact value of each switching equipment, and the adjusted impact level.

[0057] In this embodiment, time priority refers to the initial sorting of warning messages issued by the switching equipment according to the order in which they are sent, with the equipment that issues the warning first having a higher initial priority. Signal strength gradient refers to the gradient division based on the strength of fault characteristic signals (such as the rate of gas pressure drop, the amplitude of partial discharge signals, etc.) contained in the warning messages; the higher the signal strength, the higher the priority. Causality test analyzes the causal relationship between warning signals from different switching equipment using methods such as Granger causality test to determine the possible path of fault propagation and the source equipment. By comprehensively weighting these three methods (e.g., time priority weight 0.3, signal strength gradient weight 0.4, causality test weight 0.3), the final priority of the faulty equipment is determined.

[0058] In this embodiment, the initial impact level is set based on the topological importance of the switching equipment in the microgrid and the weighting results. Topological importance is comprehensively assessed by factors such as the equipment's connection location in the microgrid topology, its load capacity share, and whether it is a critical interconnection node. Weighting results assign higher initial impact level base scores to equipment with higher weighting priorities. A 0-10 score initial impact level rating system is formed by quantifying and superimposing preset weighting coefficients (topological importance weight 0.6, weighting result weight 0.4). A higher score indicates a greater impact of equipment failure on the stable operation of the microgrid.

[0059] In this embodiment, the formula for calculating the predicted association impact value is: Predicted association impact value = Σ (association weight of the second associated device) × weight coefficient of the corresponding second association level, where Σ represents the summation of the association weights of all second associated devices within the same second association level.

[0060] In this embodiment, the comprehensive correlation impact value is the weighted sum of the predicted correlation impact values ​​of each second correlation level, with a weight coefficient of 0.5 for the first correlation level, 0.3 for the second level, and 0.2 for the third level.

[0061] In this embodiment, the degree of cascading impact that switching equipment may cause to related equipment can be quantitatively assessed by comprehensively evaluating the associated impact value. The higher the value, the greater the potential risk of associated failure.

[0062] In this embodiment, if the overall associated impact value is greater than the preset impact threshold (0.6 in this application), the initial impact level of the switchgear is increased by 2 points or more; if it is less than the preset low impact threshold (0.4 in this application), it is decreased by 1 point; if it is between the two, the initial level is maintained.

[0063] In this embodiment, the adjusted impact level serves as the core basis for priority ranking. The final collaborative assessment result can directly provide microgrid operation and maintenance personnel with clear guidance on the order of fault handling, prioritizing the handling of equipment with high impact levels and large comprehensive correlation impact values, thereby minimizing the risk of fault propagation and improving the efficiency and accuracy of microgrid fault emergency response.

[0064] In some embodiments of this application, collaborative control instructions are generated based on collaborative judgment results and a preset interlocking control logic library, including: The collaborative analysis results are used as input. The condition matching engine retrieves all interlocking rules that meet the triggering conditions from the preset interlocking control logic library, instantiates the parameters, and generates the initial collaborative control command. The initial collaborative control command is conflict detected. If a conflict exists, the command is optimized according to the preset conflict resolution strategy, and the final collaborative control command is generated and sent to the corresponding environmental gas switchgear for execution through the edge computing node. The execution status of the instructions is fed back to the microgrid monitoring center in real time, and it is determined whether the coordinated control instructions need to be corrected.

[0065] In this embodiment, the preset interlocking control logic library includes device-level autonomous rules, pair-level coordination rules, cluster-level collaborative rules, and system-level optimization rules, covering the full-scale control requirements from single device self-protection to microgrid global optimization. Each rule includes triggering conditions (such as faulty device type, impact level, associated device status, etc.), execution steps, and parameter thresholds.

[0066] In this embodiment, the condition matching engine adopts a combination of fuzzy matching and precise matching. First, it performs precise matching based on key parameters such as the faulty equipment type and comprehensive correlation impact value in the collaborative judgment results. Then, it instantiates the parameters of rules with a matching degree of more than 80%, and replaces the variables in the rules with specific values ​​such as actual equipment ID and action time.

[0067] In this embodiment, the initial collaborative control command is generated based on the collaborative judgment results and the interlocking rules matched in the preset interlocking control logic library. For rules that meet the triggering conditions, the specific parameters in the collaborative judgment results (such as the faulty device ID, the range of associated devices to be disconnected, and the time delay of the action execution) are substituted into the execution steps of the rule to complete parameter instantiation, thereby forming the initial collaborative control command for the faulty device. In the case of multiple faulty devices, the initial collaborative control command will contain the control logic corresponding to each device in order of priority, ensuring that the control commands for high-priority faulty devices are generated and executed first.

[0068] In this embodiment, conflict detection includes resource conflict, target conflict, timing conflict and permission conflict detection, which is achieved by constructing a directed graph of instruction execution. If the execution paths of two instructions have device resource competition or logical contradictions, they are determined to be in conflict.

[0069] In this embodiment, the conflict resolution strategy includes ensuring the stability of the microgrid main grid, prioritizing control targets, filtering by the size of the fault impact range, and following the order of interlocking rules taking effect.

[0070] In this embodiment, the control commands are simulated and rehearsed in accordance with the guidelines for safe and stable operation of power systems. If risks such as overvoltage, overcurrent or equipment overload occur during the rehearsal, the command optimization process is returned to readjust the action parameters.

[0071] In this embodiment, the execution result is uploaded in real time through status feedback messages, forming a closed-loop control process of "analysis-command-execution-feedback", which realizes rapid and precise coordinated control of the microgrid environmental gas switchgear.

[0072] In some embodiments of this application, such as Figure 2 As shown, the system also includes an electronic device, which may include a processor 201, a communications interface 202, a memory 203, and a communication bus 204. The processor 201, communications interface 202, and memory 203 communicate with each other via the communication bus 204. The processor 201 can call logic instructions stored in the memory 203 to execute a collaborative control method for environmentally friendly gas switching equipment in a microgrid.

[0073] Furthermore, the logical instructions in the aforementioned memory 203 can be implemented as software functional units and, when sold or used as independent products, can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, essentially, or the part that contributes to the prior art, or a part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of the present invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0074] On the other hand, the present invention also provides a computer program product, which includes a computer program that can be stored on a non-transitory computer-readable storage medium. When the computer program is executed by a processor, the computer is able to execute the environmentally friendly gas switching device collaborative control method for microgrids provided by the above methods.

[0075] In another aspect, the present invention also provides a non-transitory computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, is implemented to perform the environmentally friendly gas switching device collaborative control method for microgrids provided by the methods described above.

[0076] The above description is only a preferred embodiment of this application. It should be noted that for those skilled in the art, several improvements and substitutions can be made without departing from the technical principles of this application, and these improvements and substitutions should also be considered within the scope of protection of this application.

Claims

1. A method for coordinated control of environmentally friendly gas switchgear for microgrids, characterized in that, include: Real-time monitoring and analysis of gas state parameters and partial discharge characteristics of each switchgear are performed to obtain the health index and risk probability of the corresponding switchgear. Determine whether to issue a warning message based on the health status and risk probability. If so, determine the first association level table and the second association level table for each switchgear that issued the warning message. The warning confidence level of the corresponding switching device is generated based on the first association hierarchy table. If the warning confidence level is greater than the preset confidence level threshold, the corresponding second association hierarchy table is used for collaborative analysis, and a collaborative analysis result is formed. Based on the collaborative analysis results and the preset interlocking control logic, collaborative control commands are generated.

2. The method for coordinated control of environmentally friendly gas switchgear for microgrids as described in claim 1, characterized in that, Obtain the health status and risk probability of the corresponding switching equipment, including: The gas state parameters and partial discharge characteristics are standardized, and features are extracted from the standardized gas state parameters and partial discharge characteristics to obtain several first features and several second features. The gas state degradation index is obtained by normalizing all the first characteristics and then weighting them. The partial discharge activity index is obtained by normalizing all the second features and then weighting them. A health index is generated based on the gas state deterioration index and the partial discharge activity index. Construct a feature vector, which includes all features within a preset time period; Based on a pre-trained risk prediction model, risk prediction is performed on the feature vector to obtain the failure risk probability.

3. The method for coordinated control of environmentally friendly gas switchgear for microgrids as described in claim 2, characterized in that, Whether to issue an early warning message is determined based on health status and risk probability, including: Pre-set preset health index thresholds and preset risk probability thresholds; When the health index of the switching equipment is lower than the preset health index threshold, a level 1 early warning message is issued; When the probability of failure risk of the switching equipment exceeds the preset risk probability threshold, a level 2 early warning message is issued. If the health index is not less than the preset health index threshold and the failure risk probability is not greater than the preset risk probability threshold, no early warning message will be issued.

4. The method for coordinated control of environmentally friendly gas switchgear for microgrids as described in claim 3, characterized in that, Before determining the first and second association tables for each switching device that issues an early warning message, the process also includes: When a Level 1 early warning message is issued, the feature vector of the corresponding switching equipment within a preset time period is matched with a preset equipment fault fingerprint database to obtain a number of matching degrees. The preset equipment fault fingerprint database includes several preset fault types, and each preset fault type is mapped to a corresponding fault feature vector. The preset fault types with a matching degree greater than the preset matching degree threshold are set as several first fault types of the corresponding switch at the current time node, and several corresponding first confidence levels are generated. When a Level II early warning message is issued, time-series predictions are performed on all first features and all second features to generate a predicted change curve for each feature in the future period. The curve is then compared with a preset fault change template library to obtain several comprehensive similarity scores. The preset fault change template library includes several preset fault types, and each preset fault type is mapped to a standard fault change curve with several preset features. Preset fault types with a comprehensive similarity greater than a preset similarity threshold are designated as the second fault type of the corresponding switching equipment in the future time period, and several corresponding second credibility scores are generated.

5. The method for coordinated control of environmentally friendly gas switchgear for microgrids as described in claim 1, characterized in that, Determine the first association level table and the second association level table for each switching device that issues the early warning message, including: Randomly select one of the switching devices that issued the early warning message as the target device; Based on the fault type of the target device, several relevant historical fault logs are filtered out, and the first and second associated time periods are set in each historical fault log. Extract the historical health index of other switchgear that has an electrical connection, data communication relationship or topological proximity to the target device during the first and second correlation periods from the selected historical fault logs; Calculate the frequency of historical health index being less than a preset health index threshold, identify other switching devices with a frequency greater than a preset frequency threshold in all first associated time periods as the first associated devices of the target device, and identify other switching devices with a frequency greater than a preset frequency threshold in all second associated time periods as the second associated devices of the target device; Generate a first association hierarchy table of the target device based on all first associated devices of the same fault type, and generate a second association hierarchy table of the target device based on all second associated devices of the same fault type. Several first association level tables and several second association level tables are generated sequentially for each switching device that issues an early warning message.

6. The method for coordinated control of environmentally friendly gas switchgear for microgrids as described in claim 5, characterized in that, The target device, all first associated devices and all second associated devices under each fault type are marked on the preset microgrid electrical topology diagram, resulting in several first associated device topology annotation diagrams and several second associated device topology annotation diagrams. Based on the topology annotation map of the first associated device, the first topological distance between each first associated device and the target device is extracted, and the association weight of each first associated device is generated by combining the frequency of collaborative failures with the target device and the probability of occurrence of the corresponding first failure type. Based on the mapping relationship between the association weight of the first associated device and the corresponding preset first association weight range, a first association hierarchy table is formed for each fault type of the corresponding target device; The first association hierarchy table includes several first association levels, and each first association level corresponds to at least one first association device and its association weight; Based on the topology annotation map of the second associated device, the second topology distance between each second associated device and the target device is extracted, and the association weight of each second associated device is generated by combining the frequency of collaborative failures with the target device and the probability of occurrence of the second failure type. Based on the mapping relationship between the association weight of the second associated device and the corresponding preset second association weight range, a second association hierarchy table is formed for each fault type of the target device. The second association hierarchy table includes several second association levels, and each second association level corresponds to at least one second associated device and its association weight.

7. The method for coordinated control of environmentally friendly gas switchgear for microgrids as described in claim 6, characterized in that, The warning confidence level of the corresponding switching device is generated based on the first association hierarchy table, including: Set the actual first associated time period for the current time node; Obtain the health index of each first associated device of the switching equipment that issued the early warning message during the actual first associated period; The number of normal states and the number of abnormal states of all first-association devices are counted, and combined with the corresponding association weights, the early warning reliability compensation coefficient for each first association level in the corresponding first association hierarchy table for the corresponding fault type is generated. Based on the early warning reliability compensation coefficient of each first association level and the weight coefficient of the corresponding level, a comprehensive early warning reliability compensation coefficient for the corresponding fault type is generated. The warning credibility of the switchgear warning message is generated based on the comprehensive warning credibility compensation coefficient for each fault type and the corresponding credibility.

8. The method for coordinated control of environmentally friendly gas switchgear for microgrids as described in claim 7, characterized in that, The corresponding second-level related tables are used for collaborative analysis, and the collaborative analysis results are generated, including: Remove switching devices whose warning confidence level is not greater than the preset confidence level threshold. Based on the edge computing node cluster, use time priority, signal strength gradient and causality test to identify the faulty switching devices and output a list of faulty devices. The list of faulty devices includes the initial impact level for each remaining switchgear; Based on the number of second-related devices in the second-related hierarchy table for each remaining switching device, the weight coefficient of each second-related device, and the related weight, calculate the predicted related influence value for each second-related hierarchy table. Based on the predicted association impact value of each second association level and the weight coefficient of the corresponding level, the comprehensive association impact value of the corresponding switchgear is generated. Based on the comprehensive correlation impact value, an adjustment instruction for the initial impact level is generated, and the remaining switching equipment is prioritized according to the adjusted impact level to obtain the collaborative assessment result; The collaborative assessment results include the priority order of faulty equipment, the comprehensive correlation impact value of each switching device, and the adjusted impact level.

9. The method for coordinated control of environmentally friendly gas switchgear for microgrids as described in claim 8, characterized in that, Based on the collaborative analysis results and the pre-set interlocking control logic library, collaborative control instructions are generated, including: The collaborative analysis results are used as input. The condition matching engine retrieves all interlocking rules that meet the triggering conditions from the preset interlocking control logic library, instantiates the parameters, and generates the initial collaborative control command. The initial collaborative control command is conflict detected. If a conflict exists, the command is optimized according to the preset conflict resolution strategy, and the final collaborative control command is generated and sent to the corresponding environmental gas switchgear for execution through the edge computing node. The execution status of the instructions is fed back to the microgrid monitoring center in real time, and it is determined whether the coordinated control instructions need to be corrected.

10. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the program, it implements the collaborative control method for environmentally friendly gas switchgear for microgrids as described in any one of claims 1 to 9.

Images