Power grid cabinet perfluorohexanone release control method based on infrared and microwave measurement

Abstract

The invention discloses a power grid cabinet perfluorohexanone release control method based on infrared and microwave measurement, and belongs to the technical field of perfluorohexanone release control, and the method specifically comprises the steps: obtaining temperature data and an electric field disturbance signal in a power grid cabinet, constructing a multi-dimensional environment parameter map, dividing a time window based on the multi-dimensional environment parameter map, and obtaining a perfluorohexanone release control result; generating a time sequence expansion diagram according to track overlapping and boundary change in a time window, identifying a fault core area, performing primary judgment, when a primary judgment result exceeds a set threshold value, generating a three-dimensional release model matched with the fault core area, calculating the dosage, the diffusion rate and the release direction of the perfluorohexanone to be released, and releasing the perfluorohexanone to obtain the perfluorohexanone to be released. After releasing is completed, secondary judgment is carried out, and monitoring is carried out; the arc extinguishing efficiency is improved, meanwhile, gas waste and incidental influence on the power grid cabinet are remarkably reduced, the secondary discharge risk is effectively reduced, and the arc extinguishing device is suitable for safety protection of power grid cabinets of various voltage grades.

Description

Technical Field

[0001] This invention belongs to the field of perfluorohexanone release control technology, specifically a perfluorohexanone release control method for power grid cabinets based on infrared and microwave measurements. Background Technology

[0002] Power grid cabinets are core equipment used for control and protection circuits in power systems, and are widely used in medium and high voltage distribution networks. They typically contain high-voltage live components such as circuit breakers, disconnectors, busbars, and cable connection terminals. Under high load operation, insulation aging, loose contacts, or the influence of external environmental factors, power grid cabinets are prone to overheating, breakdown discharge, and arcing, which can lead to equipment damage, fires, or even large-scale power outages.

[0003] To suppress electric arcs and prevent fires inside electrical substations, existing technologies mostly use sulfur hexafluoride (SF6) gas as an insulating and arc-extinguishing medium, or suppress arcs after a fault occurs by triggering aerosols or gaseous extinguishing agents. However, SF6 has a strong greenhouse effect and its recycling and disposal are complex. As an environmentally friendly alternative, perfluorohexanone (PFH) is gradually being used in electrical equipment fire protection systems due to its excellent insulating properties, low greenhouse effect value, and rapid arc-extinguishing capability. However, existing PPH-based electrical substation fire extinguishing technologies generally have the following problems: traditional release mechanisms usually use full-volume spraying, making it difficult to identify potential arc evolution trends in advance, and lacking targeted control of different risk areas inside the electrical substation, resulting in low utilization and incomplete arc extinguishing. Summary of the Invention

[0004] To address the shortcomings of existing technologies, this invention proposes a method for controlling the release of perfluorohexanone from power grid cabinets based on infrared and microwave measurements.

[0005] To achieve the above objectives, the present invention provides the following technical solution: A method for controlling perfluorohexanone release from electrical substations based on infrared and microwave measurements includes: Temperature data and electric field disturbance signals inside the power grid cabinet are acquired, and a multi-dimensional environmental parameter map is constructed. The multi-dimensional environmental parameter map is used to characterize the operating status of the power grid cabinet. Based on the multidimensional environmental parameter map, time windows are divided, and a time-series expansion map is generated according to the trajectory overlap and boundary changes within the time window. The fault core region is identified, which is the central node cluster with the highest propagation chain density and continuously increasing connection strength in the time-series expansion map. A judgment is made. If the judgment result exceeds a set threshold, a three-dimensional release model matching the fault core area is generated, the expected release dose of perfluorohexanone, diffusion rate and release direction are calculated, and the release is carried out. After the release is completed, a second assessment is performed, and the release area of ​​perfluorohexanone is monitored by combining infrared temperature data and microwave disturbance signals.

[0006] Specifically, acquiring temperature data and electric field disturbance signals inside the power grid cabinet, and constructing a multi-dimensional environmental parameter map, includes: Multiple spatial measurement point coordinates are preset inside the power grid cabinet, and each spatial measurement point is marked with a unique environmental label. The environmental label is used to distinguish spatial location, electrical level and heat source adjacency relationship. Infrared temperature data and microwave disturbance signals were collected from all space measurement points; The collected infrared temperature data and microwave disturbance signals are used to construct an associated grid with spatial measurement points as nodes, generating a spatially continuous three-dimensional parameter point cloud set. Based on a three-dimensional parameter point cloud set, the thermal field fluctuations, electric field disturbances and signal delay trends of each spatial measurement point under different time slices are feature-encoded and projected onto the tensor space through nonlinear mapping to obtain multi-channel fused data. An environmental parameter map coupled with the power grid cabinet structure model is established using the multi-channel fused data to obtain a multi-dimensional environmental parameter map.

[0007] Specifically, the process of dividing time windows based on multi-dimensional environmental parameter maps, generating time-series expansion maps based on trajectory overlaps and boundary changes within the time windows, and identifying fault core regions includes: The multidimensional environmental parameter map is divided into multiple partitions, each partition corresponding to a dynamic evolution window. The dynamic evolution window describes the spatial state of the power grid cabinet that evolves continuously over time. In each dynamic evolution window, the transfer trend of the temperature change rate and the electric field disturbance waveform change rate within a continuous time period is extracted, and the relevant displacement interval within the dynamic evolution window is calculated. A temperature field disturbance mutual attraction model is established based on the relevant displacement interval, and the abnormal behavior type is determined by logical association rules, and the abnormal behavior type is marked as different fault source candidate modes. For each candidate fault source pattern, an anomaly factor trajectory cluster is constructed. The anomaly factor trajectory cluster takes the initial excitation point of the disturbance source as the origin and traces the signal propagation chain outward along the time series, marking the disturbance directionality and coverage. Based on the overlapping relationships between the trajectory clusters of anomaly factors and the changes in propagation boundaries, a time-series expansion map is generated and the core fault region is identified.

[0008] Specifically, the step of generating a time-series expansion map and identifying the core fault region based on the overlapping relationships and propagation boundary changes among the anomaly factor trajectory clusters includes: The trajectory clusters of each anomaly factor are discretized on the time axis, and the displacement and disturbance direction identifiers between trajectory nodes in adjacent time slices are vectorized and encoded. Based on the vectorized encoding, the trajectory clusters of anomalies are hierarchically clustered according to spatial overlap and directional consistency to obtain multiple trajectory aggregation units. Each trajectory aggregation unit is time-series segmented, a boundary transformation linked list is constructed, and the dynamic behavior of the trajectory is recorded in the boundary transformation linked list; The boundary transformation linked list is mapped to a multi-dimensional time series network, and a time series expansion map is generated based on the connection strength between trajectory aggregation units. In the time-series expansion diagram, the central node cluster with the highest propagation chain density and continuously increasing connection strength is identified as the fault core region.

[0009] Specifically, the process involves making a judgment. When the judgment result exceeds a set threshold, a three-dimensional release model matching the fault core area is generated. The expected release dose, diffusion rate, and release direction of perfluorohexanone are calculated, and the release is then performed, including: Extract the spatial boundary information, disturbance source type, and evolution rate in the temporal expansion diagram of the fault core region to generate the geometric framework of the release trigger domain; The degree of correlation between parameters within the release trigger domain is classified into different levels to obtain the release priority factor matrix; In the release priority factor matrix, release path nodes and progressive direction instructions are generated based on the cumulative intensity of disturbance and spatial directional trend. Each path node is assigned a dose increase / decrease indicator and a propagation rate adjustment index to form a dose mapping array; By utilizing the linkage between the dose mapping array and the time-series expansion map, the entire release trigger domain is decomposed and assembled to generate a three-dimensional release model, and then the release is performed.

[0010] Specifically, assigning a release dose increase / decrease identifier and a propagation rate adjustment index to each path node to form a dose mapping array includes: Each path node is divided into multiple dynamic node groups according to its spatial distribution and channel branching relationship, and a perfluorohexanone dose allocation table is established for each dynamic node group. Collect the environmental parameter sequence related to the dynamic node group, and segment the environmental parameter sequence to obtain multiple environmental status windows; The displacement calculation is performed on the environmental parameter change trend within each environmental state window. The dose increment and dose decrement indicators are respectively bound to the path nodes in the dynamic node group, and a dose flow direction mapping rule is established by the displacement direction and the disturbance trend. Based on the channel branching relationship between path nodes, the propagation rate adjustment index of each path node is calculated according to the dose flow mapping rule, and the dose adjustment link is generated according to the weight relationship between path nodes. The link result is written into the perfluorohexanone dose allocation table. The perfluorohexanone dose allocation tables of all dynamic node groups are merged, and a dose mapping array is generated based on the spatial distribution of path nodes.

[0011] Specifically, the step of utilizing the linkage between the dose mapping array and the time-series expansion map to decompose and assemble the entire release trigger domain, generate a three-dimensional release model, and then perform the release includes: Based on the dose mapping array, the release triggering domain is divided into multiple initial spatial units, and a dose flow identifier corresponding to the dose mapping array is embedded in each spatial unit to form a spatial unit dose matrix. The temporal expansion map is processed by time slicing, and the expansion boundaries of each node cluster in the trajectory network in different time periods are mapped to the spatial unit dose matrix to generate a temporal overlap index. Based on the temporal overlap index, the directional sequence of dose flow between adjacent spatial units is calculated, and the progressive path of dose across spatial units is marked according to the propagation chain density of the node cluster. The release trigger domain is iteratively decomposed along the main dose flow direction to generate subdomains with irregular boundaries, and a local release rhythm sequence is marked in each subdomain. The local release rhythm sequences of all decomposed subdomains are topologically combined in three dimensions to obtain a three-dimensional release model.

[0012] Specifically, after the release is completed, a secondary judgment is performed, and the perfluorohexanone release area is monitored by combining infrared temperature data and microwave perturbation signals, including: After release, the degree of deviation of the actual monitoring results is determined based on the multidimensional environmental parameter spectrum before release and the dose mapping array during release. Data is reacquired from the release area, and a residual signal image stream is constructed at the same spatial measurement point. The residual signal image stream is used to track incompletely suppressed disturbance behavior or signal rebound signs. The residual signal image stream is time-folded, and the current data is projected and compared with the historical disturbance trajectory of the same area before release to extract the current offset factor and historical recurrence fragments. Based on the actual offset factor and historical recurrence fragments, and combined with the disturbance triggering probability weight within the gas distribution boundary, a residual risk index field is constructed. When the local weights in the residual risk index field exceed a set threshold, or when no convergence trend of the residual risk index field is observed within a continuous detection period, a release command is triggered.

[0013] A perfluorohexanone release control system for a power grid cabinet based on infrared and microwave measurements is used to implement the perfluorohexanone release control method for a power grid cabinet based on infrared and microwave measurements. The system includes: a spectrum construction module, a region identification module, a primary judgment control module, and a secondary judgment control module. The map construction module is used to acquire temperature data and electric field disturbance signals inside the power grid cabinet, and to construct a multi-dimensional environmental parameter map. The region identification module is used to divide time windows based on a multi-dimensional environmental parameter map, generate a time-series expansion map based on the trajectory overlap and boundary changes within the time window, and identify the core fault region. The single-judgment control module is used to make a single judgment. When the result of the single judgment exceeds a set threshold, it generates a three-dimensional release model that matches the core area of ​​the fault, calculates the expected release dose, diffusion rate and release direction of perfluorohexanone, and then releases it. The secondary judgment control module is used to perform a secondary judgment after the release is completed, and to monitor the perfluorohexanone release area by combining infrared temperature data and microwave disturbance signals.

[0014] Specifically, the primary judgment control module includes: a preliminary region generation unit, a dose mapping array construction unit, and a release model establishment unit; The preliminary region generation unit is used to extract the spatial boundary information, disturbance source type and evolution rate in the temporal expansion diagram of the fault core region, generate the geometric framework of the release trigger domain, and generate release path nodes and progressive direction instructions. The dose mapping array construction unit is used to assign a release dose increase / decrease identifier and a propagation rate adjustment index to each path node to form a dose mapping array. The release model establishment unit is used to decompose and combine the entire release trigger domain by utilizing the linkage relationship between the dose mapping array and the time-series expansion diagram, generate a three-dimensional release model, and then release the data.

[0015] Compared with the prior art, the beneficial effects of the present invention are: This invention proposes a perfluorohexanone (PFH) release control method for power grid cabinets based on infrared and microwave measurements. It generates a time-series expansion map through trajectory cluster construction and boundary aggregation calculation, accurately identifying the core fault region of arcing or discharge. When the predicted fault risk exceeds a set threshold, a three-dimensional release model matching the fault core region is generated. The required PPH dose, diffusion rate, and release direction are automatically calculated, and a multi-path, zoned quantitative release control strategy is executed. After release, closed-loop control is formed through environmental parameter recovery, residual signal image stream construction, and residual risk index field calculation. This method significantly reduces gas waste and collateral impacts on the power grid cabinet while improving arc extinguishing efficiency. It enables intelligent and dynamic multi-stage release strategies in complex power grid cabinet environments, effectively reducing the risk of secondary discharge and is suitable for power grid cabinet safety protection at various voltage levels. Attached Figure Description

[0016] Figure 1 Flowchart of the perfluorohexanone release control method for power grid cabinets based on infrared and microwave measurements provided by the present invention; Figure 2 This invention provides a fault core area identification map; Figure 3 The three-dimensional release model structure diagram provided by the present invention; Figure 4 The diagram shows the architecture of the perfluorohexanone release control system for the power grid cabinet based on infrared and microwave measurements provided by this invention. Detailed Implementation

[0017] The present application will now be described in detail with reference to specific embodiments. These embodiments will help those skilled in the art to further understand the present application, but do not limit the present application in any way. It should be noted that those skilled in the art can make several modifications and improvements without departing from the concept of the present application. These all fall within the protection scope of the present application.

[0018] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application.

[0019] It should be noted that, unless there is a conflict, the various features in the embodiments of this application can be combined with each other, all of which are within the protection scope of this application. Furthermore, although functional modules are divided in the device schematic diagram and a logical order is shown in the flowchart, in some cases, the steps shown or described can be executed in a different order than the module division in the device or the order in the flowchart. In addition, the terms "first," "second," and "third" used in this application do not limit the data or execution order, but only distinguish identical or similar items with essentially the same function and effect.

[0020] Unless otherwise defined, all technical and scientific terms used in this specification have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs. The terminology used in this specification is for the purpose of describing particular embodiments only and is not intended to limit the scope of this application. The term "and / or" as used in this specification includes any and all combinations of one or more of the associated listed items.

[0021] Example 1: Please see Figure 1 The present invention provides an embodiment of a method for controlling the release of perfluorohexanone from a power grid cabinet based on infrared and microwave measurements, comprising the following specific steps: Step S1: Obtain temperature data and electric field disturbance signals inside the power grid cabinet, and construct a multi-dimensional environmental parameter map, which is used to characterize the operating status of the power grid cabinet.

[0022] The specific steps of step S1 are as follows: Step S101: Preset the coordinates of multiple spatial measurement points inside the power grid cabinet, and mark each spatial measurement point with a unique environmental label. The environmental label is used to distinguish the spatial location, electrical level and heat source adjacency relationship.

[0023] In this embodiment, multiple spatial measurement point coordinates are set at key locations inside the power grid cabinet, with each spatial measurement point corresponding to a fixed physical location. By dividing the interior of the power grid cabinet into zones in a three-dimensional coordinate system, the busbar, circuit breaker contacts, adjacent areas of disconnecting switches, and locations with concentrated heat sources are respectively set as priority measurement points. Each measurement point is assigned a unique environmental label, which includes spatial coordinate information, electrical hierarchy attributes, and the number of adjacent heat-generating components, enabling independent identification and rapid recognition of the environmental conditions at different locations. Through this arrangement, precise positioning of the complex space inside the power grid cabinet can be achieved.

[0024] Step S102: Collect infrared temperature data and microwave disturbance signals from all spatial measurement points.

[0025] Step S103: Construct an associated grid with the collected infrared temperature data and microwave disturbance signal using spatial measurement points as nodes to generate a spatially continuous three-dimensional parameter point cloud set.

[0026] In this embodiment, the infrared temperature data and microwave disturbance signals obtained from various spatial measurement points inside the power grid cabinet are jointly processed. Specifically, each spatial measurement point is used as a node, and a connection relationship is established between it and its adjacent spatial measurement points to form a multi-node associated grid structure. Through this associated grid, the temperature gradient, microwave reflection intensity, and signal delay characteristics of each spatial measurement point are uniformly encoded and mapped to a three-dimensional coordinate system. The spatial association strength between nodes is weighted, and interpolation is performed through the weighted grid to obtain a continuous spatial temperature distribution and electric field disturbance distribution, thereby realizing the construction of a three-dimensional parameter point cloud set.

[0027] Step S104: Based on the three-dimensional parameter point cloud set, feature encoding is performed on the thermal field fluctuation, electric field disturbance morphology and signal delay trend of each spatial measurement point under different time slices, and the data is projected into the tensor space through nonlinear mapping to obtain multi-channel fused data.

[0028] In this embodiment, the three-dimensional parameter point cloud set obtained in the previous step is used to perform time-series processing on the multi-dimensional parameters such as temperature change, electric field disturbance waveform, and microwave signal propagation delay of each spatial measurement point in multiple consecutive time slices. For the data sequence of each spatial measurement point, key attributes such as thermal field fluctuation intensity, electric field disturbance waveform feature points, and delay change rate are extracted, and these attributes are feature-encoded so that the multi-source information of each spatial measurement point in different time slices can be represented by a unified multi-dimensional feature vector. Subsequently, a nonlinear mapping method is used to project these feature vectors into the tensor space, while preserving the inherent correlation between time, space, and signal dimensions, and finally generating a fusion dataset containing multiple independent channels.

[0029] Step S105: Use the multi-channel fused data to establish an environmental parameter map coupled with the power grid cabinet structure model to obtain a multi-dimensional environmental parameter map.

[0030] In this embodiment, by mapping the encoded feature vector of each spatial measurement point to the geometric grid nodes of the internal structure of the power grid cabinet, a matching relationship between spatial coordinates and feature attributes is formed. Through this matching relationship, features such as temperature, electric field disturbance and signal delay are projected into a coordinate system consistent with the internal structure of the equipment. By using multidimensional interpolation and node fusion algorithms to fill in structural units that are not directly measured, a multidimensional environmental parameter map with dynamic features of temperature, electric field and signal is finally generated in three-dimensional space.

[0031] Step S2: Based on the multidimensional environmental parameter map, divide the time window, generate a time series expansion map according to the trajectory overlap and boundary changes within the time window, and identify the fault core area. The fault core area is the central node cluster with the highest propagation chain density and continuous increase in connection strength in the time series expansion map.

[0032] It should be noted that propagation chain density refers to the number of nodes that a propagation chain from a disturbance source node along the time series in a time-series expansion diagram. It is a comprehensive measure of the link connectivity strength and reflects the degree of aggregation of the fault signal in the spatial and temporal dimensions, that is, whether the signal propagation is dispersed or concentrated in a specific area or path.

[0033] like Figure 2 As shown, the specific steps of step S2 are as follows: Step S201: Divide the multidimensional environmental parameter map into several stability partitions, each partition corresponding to a dynamic evolution window, the dynamic evolution window describing the spatial state of the power grid cabinet that evolves continuously on the time axis.

[0034] In this embodiment, based on the established multidimensional environmental parameter map, the internal operating space of the power grid cabinet is divided into multiple stability zones. During the division process, according to the gradient change characteristics of environmental parameters in space, multidimensional indicators such as temperature, electric field disturbance and signal delay are hierarchically clustered. Combined with the physical structural boundary of the power grid cabinet, several zones that can maintain relatively stable parameter fluctuations in local areas are obtained. Subsequently, an independent dynamic evolution window is set for each zone, and the continuous change of parameters in the zone is recorded using a time-series method, so that each dynamic evolution window can independently characterize the evolution state of a specific spatial region inside the power grid cabinet in the time dimension.

[0035] Step S202: In each dynamic evolution window, extract the transfer trend of the temperature change rate and the electric field disturbance waveform change rate within the continuous time period, and calculate the relevant displacement interval within the dynamic evolution window.

[0036] In this embodiment, within each dynamic evolution window formed by step S201, the infrared temperature data and microwave perturbation signal collected over a continuous time period are differentially processed to calculate the temperature change rate and electric field perturbation waveform change rate between each time slice. By establishing a time series mapping relationship, the temperature change rate curve and the electric field perturbation waveform change rate curve are synchronously aligned, and the transfer trend information between the two is extracted. Based on the transfer trend, the relative displacement of the temperature and electric field perturbation signals is statistically calculated to form a correlation displacement result describing the relative movement interval of the two signals within the time window.

[0037] Step S203: Establish a temperature field disturbance mutual attraction model based on the relevant displacement interval, determine the abnormal behavior type through logical association rules, and mark the abnormal behavior type as different fault source candidate modes.

[0038] In this embodiment, based on the relevant displacement interval obtained in step S202, a mutual attraction model between the temperature field and the electric field disturbance is constructed. Specifically, by establishing a two-way coupling relationship within each dynamic evolution window, the temperature change rate is used as the thermal driving factor, and the electric field disturbance waveform change rate is used as the electric driving factor. The degree of mutual influence between the two is analyzed within the coupling framework. Through a logical association rule base, the combination patterns of the thermal driving factor and the electric driving factor are classified, and each type of abnormal behavior corresponds to a potential fault source. Finally, a mapping relationship is established between the abnormal behavior type and its corresponding potential fault source, and the results are output as different candidate fault mode types.

[0039] Step S204: Construct an anomaly factor trajectory cluster for each fault source candidate mode. The anomaly factor trajectory cluster takes the initial excitation point of the disturbance source as the origin and traces the signal propagation chain outward along the time series, marking the disturbance directionality and coverage area.

[0040] In this embodiment, for each candidate fault mode marked in step S203, an initial disturbance source is selected as the origin of the trajectory cluster in the dynamic evolution window. Specifically, by analyzing the maximum point of the infrared temperature gradient or the peak point of the microwave electric field disturbance in the time slice, the excitation position of the fault source is determined. Starting from this origin, the propagation path of the signal in space is traced outward along the continuous time slice, the diffusion direction and intensity change of the disturbance signal are recorded, and a disturbance factor label is generated at each propagation node. Multiple continuous nodes and their directional features are integrated into a trajectory cluster to describe the spatial propagation chain of the fault mode. Finally, the direction vector of the disturbance and the coverage boundary are marked on each trajectory cluster.

[0041] Step S205: Based on the overlapping relationship between the trajectory clusters of abnormal factors and the changes in the propagation boundary, generate a time-series expansion map and identify the core fault region.

[0042] The specific steps of step S205 are as follows: Step S2051: Discretize and sample the trajectory clusters of each anomaly factor on the time axis, and vectorize and encode the displacement and disturbance direction identifiers between trajectory nodes in adjacent time slices.

[0043] In this embodiment, each anomaly factor trajectory cluster obtained in step S204 is discretized along the time axis. Specifically, the continuous time series is divided into multiple fixed time segments, and the spatial coordinates and disturbance direction attributes of all nodes in the trajectory cluster are extracted in each time segment. For each pair of nodes in adjacent time segments, the displacement difference between the nodes is calculated, and the displacement difference is combined with the disturbance direction information to form directional vector data. Finally, the displacement vectors of all nodes in the time segment are arranged and encoded in chronological order to obtain a vectorized trajectory dataset that can be used for subsequent trajectory pattern analysis.

[0044] Step S2052: Based on the vectorized encoding, the trajectory clusters of the anomaly factors are hierarchically clustered according to spatial overlap and directional consistency to obtain multiple trajectory aggregation units.

[0045] In this embodiment, the vectorized encoded trajectory cluster data obtained in step S2051 is input into the hierarchical clustering module. Specifically, the overlap degree of each anomaly factor trajectory cluster in space is first calculated, including the spatial coverage and proximity index between trajectory nodes. Then, a directional consistency metric is established based on the angle between trajectory propagation direction vectors and the rate of directional change. By combining the two parameters of spatial overlap degree and directional consistency, a similarity matrix between trajectories is constructed. Using a top-down hierarchical clustering method, trajectory clusters with high similarity are merged step by step to finally obtain multiple trajectory aggregation units, each unit representing a set of trajectories with similar spatial paths and consistent directional evolution.

[0046] Step S2053: Perform temporal segmentation on each trajectory aggregation unit, construct a boundary transformation linked list, and record the dynamic behavior of the trajectory in the boundary transformation linked list.

[0047] In this embodiment, each trajectory aggregation unit generated in step S2052 is subjected to temporal segmentation processing. First, the continuous time series of the trajectory aggregation unit is divided into several dynamic stages, each stage corresponding to a specific evolutionary state of trajectory propagation. Then, within each time period, dynamic features such as spatial expansion, contraction, and directional rotation of the trajectory boundary are extracted, and these features are recorded as node information. By connecting the nodes of each time period in chronological order, a boundary transformation linked list is formed, which can store the dynamic change behavior of the trajectory in the time dimension in a serialized form.

[0048] Step S2054: Map the boundary transformation linked list to a multidimensional time series network, and generate a time series expansion map based on the connection strength between trajectory aggregation units.

[0049] In this embodiment, the boundary transformation linked list constructed in step S2053 is used as input and mapped to a multidimensional time series network. Specifically, firstly, a time dimension index and a spatial dimension index in the network are assigned to each node in the boundary transformation linked list, and the dynamic stages of the trajectory aggregation unit are defined as different state nodes in the multidimensional network. Subsequently, the connection strength between nodes is calculated based on the spatial proximity, boundary overlap rate, and evolution synchronization between trajectory aggregation units. By using these strength parameters as weights of network edges, directed connections between nodes are established in the multidimensional time series network. Finally, based on the high-weight connection paths between different nodes in the network, a time-series expansion graph reflecting the evolution relationship of trajectory aggregation units is generated to describe the multidimensional dynamic links of fault propagation.

[0050] Step S2055: In the time-series expansion diagram, identify the central node cluster with the highest propagation chain density and continuously increasing connection strength as the fault core region.

[0051] In this embodiment, a joint analysis of node density and connection strength is performed on the time-series expansion graph constructed in step S2054. First, the propagation chain density of each node in the time-series expansion graph is calculated. This density is determined by the number of directed connections around the node and the weighted path length. Then, the changes in the connection strength of each node in the time series are dynamically monitored, and the node set that maintains an increasing trend over multiple consecutive time periods is selected. By combining the high density index with the continuously increasing connection strength, the central node cluster located on the main trunk of the propagation chain and possessing a highly concentrated propagation capability is identified. This central node cluster represents the core risk source area of ​​potential electric arc or discharge inside the power grid cabinet.

[0052] like Figure 2 As shown, the internal environment of the power grid cabinet is collected from multiple points using infrared temperature measurement points (labeled T) and microwave disturbance measurement points (labeled D), forming a three-dimensional spatial grid. Based on the temperature changes, electric field disturbance signals, and gas distribution characteristics of these collection points, the internal space is divided into stability zones. The gray blocks on the right side of the figure represent the division of the internal environment of the power grid cabinet into multiple stability zones. Each zone maintains a relatively stable state of environmental parameters in a short period of time. The dark gray area represents the lowest rate of change of environmental parameters and the smallest disturbance fluctuation. The medium gray area represents the environmental parameters with some fluctuation but not reaching an unstable state. The light gray area represents the environmental parameters with relatively large fluctuations but still within a controllable range. It should be noted that the dark gray area is the upper right corner of the stability zone, the light gray area is the lower left middle two areas of the stability zone, and the other areas are medium gray areas. The time window correlation, the time window bar chart indicated by the arrow in the figure, illustrates the continuous evolution of different stability zones on the time axis, providing accurate input for subsequent generation of trajectory clusters, time series expansion maps, and fault core area identification.

[0053] Step S3: Make a judgment. When the judgment result exceeds the set threshold, generate a three-dimensional release model that matches the fault core area, calculate the expected release dose, diffusion rate and release direction of perfluorohexanone, and release it.

[0054] It should be noted that the threshold is set based on the actual situation or experimental simulation.

[0055] like Figure 3 As shown, the specific steps of step S3 are as follows: Step S301: Extract the spatial boundary information of the fault core region, the type of disturbance source, and the evolution rate in its temporal expansion diagram to generate the geometric framework of the release trigger domain.

[0056] In this embodiment, the spatial boundary, disturbance source type, and evolution rate information in the time-series expansion diagram are extracted from the fault core region obtained in step S2055. Specifically, the external geometric contour of the fault core region is obtained by fitting the three-dimensional boundary of the spatial distribution of the central node cluster. Combined with the trajectory cluster analysis results, the disturbance source type causing the fault is identified, such as increased contact resistance, insulation breakdown, or partial discharge. At the same time, the expansion speed and acceleration of the spatial boundary in different time slices are calculated using the time series data in the time-series expansion diagram. Based on the above information, a release trigger domain geometric framework adapted to the actual power grid cabinet structure is constructed.

[0057] Step S302: Classify the degree of correlation between parameters within the release trigger domain to obtain the release priority factor matrix.

[0058] In this embodiment, the influence of different parameters on the spread of arc risk is evaluated by calculating the cross-change rate and spatial correlation coefficient between each parameter. Then, the correlation degree is classified into three levels: high, medium and low, and priority weights are marked for each parameter node in the spatial coordinate system. Finally, the weight values ​​of all parameters are organized into a two-dimensional matrix structure, which is the release priority factor matrix.

[0059] Step S303: In the release priority factor matrix, generate release path nodes and progressive direction instructions based on the cumulative intensity of disturbance and spatial directional trend.

[0060] In this embodiment, the weighted values ​​of multi-dimensional risk factors such as cumulative temperature, electric field disturbance and gas dilution are accumulated for each spatial node, and the nodes with weighted cumulative values ​​higher than a set threshold are selected as candidate release path nodes; then, based on the spatial coordinate distribution, the direction vectors and priority order between adjacent nodes are determined to form multiple release direction links pointing to the center of the high-risk area.

[0061] It should be noted that the release effect of the arc-quenching gas depends not only on the total dose, but also on the rational planning of the release path.

[0062] Step S304: Assign each path node a release dose increase / decrease identifier and a propagation rate adjustment index to form a dose mapping array.

[0063] The specific steps of step S304 are as follows: Step S3041: Divide each path node into multiple dynamic node groups according to spatial distribution and channel branching relationship, and establish a perfluorohexanone dosage allocation table for each dynamic node group.

[0064] In this embodiment, the arc risk area has spatial heterogeneity, and uniform distribution during gas release would lead to insufficient target coverage efficiency. Based on the spatial coordinate distribution of path nodes inside the power grid cabinet and the branch topology of the channel, the release path is divided into several dynamic node groups. Each dynamic node group includes several spatially adjacent nodes with connected release paths, reflecting a sub-region where the gas progresses along a specific direction. Subsequently, a perfluorohexanone dose allocation table is constructed for each dynamic node group. This allocation table records the dose ratio of each node in the group, the expected release sequence, and the priority order of dose transfer between nodes.

[0065] Step S3042: Collect the environmental parameter sequence related to the dynamic node group, and segment the environmental parameter sequence to obtain multiple environmental status windows.

[0066] In this embodiment, for each dynamic node group obtained in step S3041, an environmental parameter sequence related to it is collected. This environmental parameter sequence includes multi-dimensional operating status information such as temperature fluctuation value, electric field disturbance intensity, gas density change, and local pressure dynamics in the spatial region where the dynamic node group is located. After continuously monitoring the collected original environmental parameter sequence according to the time axis, the entire time series is divided into multiple segments by setting dynamic thresholds and trend inflection points. Each segment is defined as an independent environmental state window to describe the stable or abrupt characteristics of the dynamic node group within a specific time range. Finally, an index relationship is established between multiple environmental state windows and dynamic node groups.

[0067] Step S3043: Perform displacement calculation on the environmental parameter change trend within each environmental state window, bind the dose increment identifier and dose reduction identifier to specific path nodes within the dynamic node group respectively, and establish a dose flow direction mapping rule through displacement direction and disturbance trend.

[0068] In this embodiment, displacement calculations are performed on the dynamic parameter change trends within each environmental state window obtained in step S3042. Specifically, time series difference analysis is performed on the temperature change rate, electric field disturbance intensity gradient, and gas density fluctuation direction to obtain the dominant displacement direction of each parameter. Based on these directional characteristics, the dose increment identifier is bound to the path node whose displacement direction is similar to that of the fault core region; the dose reduction identifier is bound to the path node whose displacement direction deviates from the disturbance trend or has a low diffusion rate. Subsequently, a dose flow mapping rule is established by synthesizing the displacement direction vectors of all path nodes. Step S3044: Based on the channel branching relationship between path nodes, calculate the propagation rate adjustment index of each path node according to the dose flow mapping rule, generate the dose adjustment link according to the weight relationship between path nodes, and write the link result into the perfluorohexanone dose allocation table.

[0069] In this embodiment, based on the dose flow direction mapping rule established in step S3043, the propagation rate adjustment calculation is performed on each path node in the dynamic node group. First, according to the channel branching relationship between path nodes, the structural characteristics of the main channel, secondary channel and intersection point are analyzed, and the weight parameters between nodes are extracted, including spatial distance, directional consistency and risk priority. Then, according to the dose flow direction mapping rule, the propagation rate adjustment index of each path node is calculated to describe the gas transmission speed and relative release intensity between path nodes. Next, according to the weight relationship between path nodes, each path node is connected in sequence to form a dose adjustment link, and the link structure and its corresponding propagation rate adjustment index are written into the perfluorohexanone dose allocation table to complete the dynamic dose allocation planning of the entire path network.

[0070] Step S3045: Merge the perfluorohexanone dose allocation tables of all dynamic node groups and generate a dose mapping array based on the spatial distribution relationship of the path nodes.

[0071] Step S305: Utilize the linkage between the dose mapping array and the time-series expansion map to decompose and assemble the entire release trigger domain, generate a three-dimensional release model, and then release the data.

[0072] The specific steps of step S305 are as follows: Step S3051: Based on the dose mapping array, the release trigger domain is divided into multiple initial spatial units, and a dose flow identifier corresponding to the dose mapping array is embedded in each spatial unit to form a spatial unit dose matrix.

[0073] In this embodiment, the release trigger domain is spatially discretized using the dose mapping array obtained in the preceding steps. Specifically, the release trigger domain is divided into multiple initial spatial units according to structural features and risk gradients. The boundary of each spatial unit is determined based on the dynamic node group position, channel branch density, and perfluorohexanone flow direction. Subsequently, the dose flow identifiers of each path node recorded in the dose mapping array are embedded into the corresponding spatial units, giving each spatial unit attribute information characterizing dose flow direction, velocity, and priority. In this way, a spatial unit dose matrix is ​​formed.

[0074] Step S3052: Perform time slicing processing on the time-series expansion map, map the expansion boundaries of each node cluster in the trajectory network to the spatial unit dose matrix at different time periods, and generate a time-series overlap index.

[0075] In this embodiment, the time axis of the temporal expansion map is divided into multiple consecutive time periods, each time period corresponding to the evolution stage of the node cluster in the trajectory network. Subsequently, expansion boundary information, including spatial coverage, boundary propagation velocity, and direction vector, is extracted for the node cluster in each time period. The extracted expansion boundary is projected onto the spatial unit dose matrix constructed in step S3051 using a geometric mapping method, and the degree of overlap between each spatial unit and the trajectory node cluster in that time period is calculated. Finally, the overlap results of all time periods are combined in chronological order to generate a temporal overlap index.

[0076] Step S3053: Based on the temporal overlap index, calculate the directional sequence of dose flow between adjacent spatial units, and mark the progressive path of dose across spatial units according to the propagation chain density of the node cluster.

[0077] In this embodiment, based on the temporal overlap index generated in step S3052, the dose flow directionality sequence between adjacent spatial units is calculated. Specifically, by analyzing the gradient change of the overlap weight of spatial units in each time period, the main direction of dose migration in space is extracted, and the flow relationship between adjacent spatial units is recorded in vector form. Subsequently, based on the propagation chain density of the trajectory node cluster, high-frequency channels for dose transfer are identified between spatial units, and progressive path markers are set on these channels to represent the dynamic diffusion sequence of gas across spatial units during the release process.

[0078] Step S3054: Iteratively decompose the release trigger domain along the main dose flow direction to generate decomposed subdomains with irregular boundaries, and mark the local release rhythm sequence in each decomposed subdomain.

[0079] In this embodiment, based on the dose flow topology data obtained in step S3053, the release trigger domain is iteratively decomposed along the main flow direction. Specifically, firstly, the dose flow direction vector of the overall release trigger domain is determined, and the spatial units are directionally sorted according to this vector. Subsequently, using the progressive path relationship in the flow chain list, the release trigger domain is divided into multiple subdomains. Each subdomain is constructed around a local energy accumulation area in the main flow direction, maintaining an irregular boundary contour in space. Within each subdomain, based on the dose gradient change and propagation chain density, a local release rhythm sequence is marked for key nodes to guide the time-sharing and partitioned release of gas within the subdomain. Finally, multiple independently controllable release subdomain structures are formed, providing input for multi-level dose control of the three-dimensional release model.

[0080] Step S3055: Perform three-dimensional topological stitching of the local release rhythm sequences of all decomposed subdomains to obtain a three-dimensional release model.

[0081] In this embodiment, an adjacency matrix between subdomains is established based on the spatial relationships and dose flow topology information of each subdomain. Subsequently, according to the temporal order of the local release rhythm sequence and the dose delivery priority, the rhythm data of the decomposed subdomains are stitched together to form a multi-channel release trajectory on the global time axis. Through three-dimensional spatial coordinate mapping, the boundaries between the decomposed subdomains are smoothly connected to form a continuous release region distribution. Finally, a complete three-dimensional release model is generated, which includes the dose control parameters, release order, and topological connection relationships of each decomposed subdomain, and can be used to control the multidimensional targeted release of perfluorohexanone.

[0082] like Figure 3 As shown, the internal structure of the power grid cabinet is divided into multiple initial spatial units by establishing a three-dimensional spatial grid. Dose flow markers are arranged in each spatial unit to indicate the possible flow direction and diffusion trend of perfluorohexanone during release. In the cubic grid on the left side of the figure, the triangle symbol is the direction marker of the dose flow, the black solid dot represents the release path point, and the dashed line between the nodes represents the dose flow link. Specific release dose and diffusion rate parameters are assigned to each node, thus forming a spatial unit dose matrix.

[0083] The irregular gray area on the right is a subdomain of the release trigger domain, which is obtained through iterative decomposition and has irregular spatial boundaries. The black dots inside the subdomain are local release nodes, and the dashed arrows indicate the progressive release direction and path of the gas in the subdomain. By integrating the release nodes and flow links of all subdomains, a complete three-dimensional release model can be generated to achieve multi-directional and hierarchical control of perfluorohexanone release.

[0084] Step S4: After the release is completed, a second judgment is made, and the release area of ​​perfluorohexanone is monitored by combining infrared temperature data and microwave disturbance signal.

[0085] The specific steps of step S4 are as follows: Step S401: After release, based on the multidimensional environmental parameter spectrum before release and the dose mapping array during release, determine the degree of deviation of the actual monitoring results.

[0086] In this embodiment, the key indicators (including temperature, electric field disturbance, and gas concentration) in the multidimensional environmental parameter spectrum before release are first paired with the monitoring values ​​after release on spatial coordinates, and the variation differences of each indicator in different spatial units are calculated. Then, the offset vector between the monitoring results and the theoretical dose distribution is analyzed by combining the expected release trajectory and dose flow path recorded in the dose mapping array.

[0087] Step S402: Data is reacquired from the release area, and a residual signal image stream is constructed at the same spatial measurement point. The residual signal image stream is used to track incompletely suppressed disturbance behavior or signal rebound signs.

[0088] In this embodiment, after the release process is completed, environmental data is collected from each release area inside the power grid cabinet. Specifically, infrared temperature signals, microwave disturbance signals, and gas concentration data are re-acquired through the original spatial measuring points, and the signal sequence is reconstructed on the time axis of the same spatial measuring point. Subsequently, a multi-parameter fusion algorithm is used to superimpose the temperature fluctuations, residual electric field disturbance signals, and gas concentration changes of each spatial measuring point into a continuous residual signal image stream. This image stream can dynamically display the disturbance behavior that was not completely suppressed after release and the possible signs of signal rebound.

[0089] Step S403: Perform time folding processing on the residual signal image stream, project and compare the current data with the historical disturbance trajectory of the same area before release, and extract the current offset factor and historical recurrence fragments.

[0090] In this embodiment, time folding processing is an analysis method that aligns, overlays, and compares the residual signal data at the current moment with the historical dynamic trajectory of the same area in the past.

[0091] Step S404: Based on the actual offset factor and the historical recurrence fragment, and combined with the disturbance triggering probability weight within the gas distribution boundary, construct the residual risk index field.

[0092] In this embodiment, the gas distribution boundary is redefined after release, and the detected abnormal fluctuations in temperature, electric field disturbances, and gas concentration are mapped as offset factors. Then, disturbance segments with time-reproducible characteristics are extracted from the residual signal image stream to reflect potential secondary arc triggering behavior. By combining the offset factors and reproducible segments with the disturbance triggering probability weights of different spatial units within the gas distribution boundary, the risk coefficient of each unit is calculated. Finally, a residual risk index field is formed in three-dimensional space.

[0093] Step S405: When the local weight in the residual risk index field exceeds the set threshold, or when no convergence trend of the residual risk index field is observed within the continuous detection period, a release command is triggered.

[0094] Example 2: Please see Figure 2 Another embodiment of the present invention provides: a perfluorohexanone release control system for power grid cabinets based on infrared and microwave measurements, comprising: a spectrum construction module, a region identification module, a primary judgment control module, and a secondary judgment control module; The map construction module is used to acquire temperature data and electric field disturbance signals inside the power grid cabinet, and to construct a multi-dimensional environmental parameter map. The region identification module is used to divide time windows based on a multi-dimensional environmental parameter map, generate a time-series expansion map based on the trajectory overlap and boundary changes within the time window, and identify the core fault region. The single-judgment control module is used to make a single judgment. When the result of the single judgment exceeds a set threshold, it generates a three-dimensional release model that matches the core area of ​​the fault, calculates the expected release dose, diffusion rate and release direction of perfluorohexanone, and then releases it. The secondary judgment control module is used to perform a secondary judgment after the release is completed, and to monitor the perfluorohexanone release area by combining infrared temperature data and microwave disturbance signals.

[0095] The single-judgment control module includes: a preliminary region generation unit, a dose mapping array construction unit, and a release model establishment unit; The preliminary region generation unit is used to extract the spatial boundary information, disturbance source type and evolution rate in the temporal expansion diagram of the fault core region, generate the geometric framework of the release trigger domain, and generate the release path nodes and progressive direction instructions. The dose mapping array construction unit is used to assign a release dose increase / decrease identifier and a propagation rate adjustment index to each path node to form a dose mapping array. The release model establishment unit is used to decompose and combine the entire release trigger domain by utilizing the linkage relationship between the dose mapping array and the time-series expansion map, generate a three-dimensional release model, and then release the data.

[0096] In addition, the parts of the technical solutions provided in the embodiments of this application that are consistent with the implementation principles of the corresponding technical solutions in the prior art have not been described in detail, so as to avoid excessive elaboration.

[0097] The specific embodiments described above further illustrate the purpose, technical solution, and beneficial effects of the present invention. It should be understood that the above descriptions are merely specific embodiments of the present invention and are not intended to limit the invention. Any modifications, equivalent substitutions, or improvements made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. A method for perfluorocyclohexane release control of power grid cabinets based on infrared and microwave measurements, characterized by, include: Temperature data and electric field disturbance signals inside the power grid cabinet are acquired, and a multi-dimensional environmental parameter map is constructed. The multi-dimensional environmental parameter map is used to characterize the operating status of the power grid cabinet. Based on the multidimensional environmental parameter map, time windows are divided, and a time-series expansion map is generated according to the trajectory overlap and boundary changes within the time window. The fault core region is identified, which is the central node cluster with the highest propagation chain density and continuously increasing connection strength in the time-series expansion map. A judgment is made. If the judgment result exceeds a set threshold, a three-dimensional release model matching the fault core area is generated, the expected release dose of perfluorohexanone, diffusion rate and release direction are calculated, and the release is carried out. After the release is completed, a second assessment is performed, and the release area of ​​perfluorohexanone is monitored by combining infrared temperature data and microwave disturbance signals.

2. The method for controlling the release of perfluorohexanone from a power grid cabinet based on infrared and microwave measurements as described in claim 1, characterized in that, The acquisition of temperature data and electric field disturbance signals inside the power grid cabinet, and the construction of a multi-dimensional environmental parameter map, includes: Multiple spatial measurement point coordinates are preset inside the power grid cabinet, and each spatial measurement point is marked with a unique environmental label. The environmental label is used to distinguish spatial location, electrical level and heat source adjacency relationship. Infrared temperature data and microwave disturbance signals were collected from all space measurement points; The collected infrared temperature data and microwave disturbance signals are used to construct an associated grid with spatial measurement points as nodes, generating a spatially continuous three-dimensional parameter point cloud set. Based on a three-dimensional parameter point cloud set, the thermal field fluctuations, electric field disturbances and signal delay trends of each spatial measurement point under different time slices are feature-encoded and projected onto the tensor space through nonlinear mapping to obtain multi-channel fused data. An environmental parameter map coupled with the power grid cabinet structure model is established using the multi-channel fused data to obtain a multi-dimensional environmental parameter map.

3. The method for controlling the release of perfluorohexanone from a power grid cabinet based on infrared and microwave measurements as described in claim 2, characterized in that, The process involves dividing time windows based on multidimensional environmental parameter maps, generating time-series expansion maps based on trajectory overlaps and boundary changes within the time windows, and identifying core fault regions, including: The multidimensional environmental parameter map is divided into multiple partitions, each partition corresponding to a dynamic evolution window. The dynamic evolution window describes the spatial state of the power grid cabinet that evolves continuously over time. In each dynamic evolution window, the transfer trend of the temperature change rate and the electric field disturbance waveform change rate within a continuous time period is extracted, and the relevant displacement interval within the dynamic evolution window is calculated. A temperature field disturbance mutual attraction model is established based on the relevant displacement interval, and the abnormal behavior type is determined by logical association rules, and the abnormal behavior type is marked as different fault source candidate modes. For each candidate fault source pattern, an anomaly factor trajectory cluster is constructed. The anomaly factor trajectory cluster takes the initial excitation point of the disturbance source as the origin and traces the signal propagation chain outward along the time series, marking the disturbance directionality and coverage. Based on the overlapping relationships between the trajectory clusters of anomaly factors and the changes in propagation boundaries, a time-series expansion map is generated and the core fault region is identified.

4. The method for controlling the release of perfluorohexanone from a power grid cabinet based on infrared and microwave measurements as described in claim 3, characterized in that, The process of generating a time-series expansion map and identifying the core fault region based on the overlapping relationships and propagation boundary changes among the trajectory clusters of anomaly factors includes: The trajectory clusters of each anomaly factor are discretized on the time axis, and the displacement and disturbance direction identifiers between trajectory nodes in adjacent time slices are vectorized and encoded. Based on the vectorized encoding, the trajectory clusters of anomalies are hierarchically clustered according to spatial overlap and directional consistency to obtain multiple trajectory aggregation units. Each trajectory aggregation unit is time-series segmented, a boundary transformation linked list is constructed, and the dynamic behavior of the trajectory is recorded in the boundary transformation linked list; The boundary transformation linked list is mapped to a multi-dimensional time series network, and a time series expansion map is generated based on the connection strength between trajectory aggregation units. In the time-series expansion diagram, the central node cluster with the highest propagation chain density and continuously increasing connection strength is identified as the fault core region.

5. The method for controlling the release of perfluorohexanone from a power grid cabinet based on infrared and microwave measurements as described in claim 4, characterized in that, The process involves making a judgment, and when the judgment result exceeds a set threshold, generating a three-dimensional release model matching the core fault region, calculating the expected release dose, diffusion rate, and release direction of perfluorohexanone, and then releasing it, including: Extract the spatial boundary information, disturbance source type, and evolution rate in the temporal expansion diagram of the fault core region to generate the geometric framework of the release trigger domain; The degree of correlation between parameters within the release trigger domain is classified into different levels to obtain the release priority factor matrix; In the release priority factor matrix, release path nodes and progressive direction instructions are generated based on the cumulative intensity of disturbance and spatial directional trend. Each path node is assigned a dose increase / decrease indicator and a propagation rate adjustment index to form a dose mapping array; By utilizing the linkage between the dose mapping array and the time-series expansion map, the entire release trigger domain is decomposed and assembled to generate a three-dimensional release model, and then the release is performed.

6. The method for controlling the release of perfluorohexanone from a power grid cabinet based on infrared and microwave measurements as described in claim 5, characterized in that, Assigning release dose increase / decrease identifiers and propagation rate adjustment indexes to each path node constitutes a dose mapping array, including: Each path node is divided into multiple dynamic node groups according to its spatial distribution and channel branching relationship, and a perfluorohexanone dose allocation table is established for each dynamic node group. Collect the environmental parameter sequence related to the dynamic node group, and segment the environmental parameter sequence to obtain multiple environmental status windows; The displacement calculation is performed on the environmental parameter change trend within each environmental state window. The dose increment and dose decrement indicators are respectively bound to the path nodes in the dynamic node group, and a dose flow mapping rule is established by the displacement direction and the disturbance trend. Based on the channel branching relationship between path nodes, the propagation rate adjustment index of each path node is calculated according to the dose flow mapping rule, and the dose adjustment link is generated according to the weight relationship between path nodes. The link result is written into the perfluorohexanone dose allocation table. The perfluorohexanone dose allocation tables of all dynamic node groups are merged, and a dose mapping array is generated based on the spatial distribution of path nodes.

7. The method for controlling the release of perfluorohexanone from a power grid cabinet based on infrared and microwave measurements as described in claim 6, characterized in that, The process of decomposing and assembling the entire release trigger domain using the linkage between the dose mapping array and the time-series expansion map to generate a three-dimensional release model, and then releasing the data, includes: Based on the dose mapping array, the release triggering domain is divided into multiple initial spatial units, and a dose flow identifier corresponding to the dose mapping array is embedded in each spatial unit to form a spatial unit dose matrix. The temporal expansion map is processed by time slicing, and the expansion boundaries of each node cluster in the trajectory network in different time periods are mapped to the spatial unit dose matrix to generate a temporal overlap index. Based on the temporal overlap index, the directional sequence of dose flow between adjacent spatial units is calculated, and the progressive path of dose across spatial units is marked according to the propagation chain density of the node cluster. The release trigger domain is iteratively decomposed along the main dose flow direction to generate subdomains with irregular boundaries, and a local release rhythm sequence is marked in each subdomain. The local release rhythm sequences of all decomposed subdomains are topologically combined in three dimensions to obtain a three-dimensional release model.

8. The method for controlling the release of perfluorohexanone from a power grid cabinet based on infrared and microwave measurements as described in claim 7, characterized in that, After release, a secondary assessment is performed, and the perfluorohexanone release area is monitored by combining infrared temperature data and microwave perturbation signals, including: After release, the degree of deviation of the actual monitoring results is determined based on the multidimensional environmental parameter spectrum before release and the dose mapping array during release. Data is reacquired from the release area, and a residual signal image stream is constructed at the same spatial measurement point. The residual signal image stream is used to track incompletely suppressed disturbance behavior or signal rebound signs. The residual signal image stream is time-folded, and the current data is projected and compared with the historical disturbance trajectory of the same area before release to extract the current offset factor and historical recurrence fragments. Based on the actual offset factor and historical recurrence fragments, and combined with the disturbance triggering probability weight within the gas distribution boundary, a residual risk index field is constructed. When the local weights in the residual risk index field exceed a set threshold, or when no convergence trend of the residual risk index field is observed within a continuous detection period, a release command is triggered.

9. A perfluorohexanone release control system for a power grid cabinet based on infrared and microwave measurements, used to implement the perfluorohexanone release control method for a power grid cabinet based on infrared and microwave measurements as described in any one of claims 1-8, characterized in that, include: The system includes a map construction module, a region identification module, a primary judgment control module, and a secondary judgment control module. The map construction module is used to acquire temperature data and electric field disturbance signals inside the power grid cabinet, and to construct a multi-dimensional environmental parameter map. The region identification module is used to divide time windows based on a multi-dimensional environmental parameter map, generate a time-series expansion map based on the trajectory overlap and boundary changes within the time window, and identify the core fault region. The single-judgment control module is used to make a single judgment. When the result of the single judgment exceeds a set threshold, it generates a three-dimensional release model that matches the fault core area, calculates the expected release dose, diffusion rate and release direction of perfluorohexanone, and then releases it. The secondary judgment control module is used to perform a secondary judgment after the release is completed, and to monitor the perfluorohexanone release area by combining infrared temperature data and microwave disturbance signals.

10. The perfluorohexanone release control system for power grid cabinets based on infrared and microwave measurements as described in claim 9, characterized in that, The primary judgment control module includes: a preliminary region generation unit, a dose mapping array construction unit, and a release model establishment unit; The preliminary region generation unit is used to extract the spatial boundary information, disturbance source type and evolution rate in the temporal expansion diagram of the fault core region, generate the geometric framework of the release trigger domain, and generate the release path nodes and progressive direction instructions. The dose mapping array construction unit is used to assign a release dose increase / decrease identifier and a propagation rate adjustment index to each path node to form a dose mapping array. The release model building unit is used to decompose and combine the entire release trigger domain by utilizing the linkage relationship between the dose mapping array and the time-series expansion map, generate a three-dimensional release model, and then release the data.

Images