A ring main unit dehumidification method and device, electronic equipment and storage medium

Abstract

The present application relates to the technical field of distribution network equipment state monitoring and intelligent moisture-proof control, and particularly relates to a ring main unit dehumidification method and device, electronic equipment and a storage medium, the method comprising: clustering analysis of all monitoring points according to historical temperature and humidity time series data and moisture characteristic vectors at each monitoring point, obtaining a plurality of dehumidification partitions; constructing a moisture space-time correlation matrix according to historical humidity time series data in the historical temperature and humidity time series data of all monitoring points in each dehumidification partition; the moisture space-time correlation matrix contains target moisture correlation coefficients and target water vapor propagation lag time between upstream and downstream dehumidification partitions; using the moisture space-time correlation matrix as a control constraint basis, the target ring main unit is dehumidified in partitions. Through the present application, the target ring main unit can be dynamically divided into dehumidification partitions, relying on the moisture space-time correlation matrix to adapt to the water vapor dynamic migration characteristics, and the dehumidification control precision is improved.

Description

Technical Field

[0001] This invention relates to the field of power distribution equipment status monitoring and intelligent moisture control technology, and more specifically, to a dehumidification method, device, electronic equipment and storage medium for ring main units. Background Technology

[0002] Urban distribution network ring main units are mostly deployed in high-humidity environments such as underground corridors, low-lying outdoor areas, and coastal areas with salt spray. The cabinet is physically divided into sealed cavities such as high-voltage component cavities, secondary wiring cavities, and cable cavities. Affected by diurnal temperature differences, rainy weather, aging of seals, and equipment heat dissipation, the temperature and humidity in each cavity of the cabinet exhibit a non-linear spatiotemporal distribution. Moisture can diffuse across the cavities through gaps and holes, easily forming condensation on the surface of insulating components, causing faults such as insulation degradation, partial discharge, and short-circuit tripping, seriously affecting the safe operation of distribution network equipment.

[0003] Traditional dehumidifiers rely on physical partitions in the cabinet to manually fix the zones. This rigid zoning method cannot adapt to the dynamic migration characteristics of water vapor and makes it difficult to accurately locate the damp areas. Summary of the Invention

[0004] In view of this, the purpose of this application is to provide a dehumidification method, device, electronic equipment and storage medium for ring main units, which can dynamically divide the target ring main unit into dehumidification zones, and improve the dehumidification control accuracy by adapting the dynamic migration characteristics of water vapor based on the spatiotemporal correlation matrix of humidity.

[0005] In a first aspect, embodiments of this application provide a dehumidification method for a ring main unit, the method comprising: Historical temperature and humidity time-series data and humidity feature vectors are obtained at each monitoring point in each sealed cavity within the target ring network cabinet; the humidity feature vectors are used to characterize the humidity change pattern at the corresponding monitoring point. Based on the historical temperature and humidity time series data and the humidity feature vector at each monitoring point, cluster analysis is performed on all monitoring points to obtain multiple dehumidification zones; the temperature and humidity time series change trends among all monitoring points contained in any dehumidification zone are the same; all monitoring points within any dehumidification zone belong to the same sealed cavity; Based on the historical humidity time series data from all monitoring points in each dehumidification zone, a humidity spatiotemporal correlation matrix is ​​constructed; the humidity spatiotemporal correlation matrix includes the target humidity correlation coefficient and the target water vapor propagation lag time between the upstream and downstream dehumidification zones. Using the aforementioned humidity spatiotemporal correlation matrix as the basis for control constraints, the target ring main unit is dehumidified by zone.

[0006] In one possible implementation, the step of performing cluster analysis on all monitoring points based on the historical temperature and humidity time-series data and the humidity feature vector at each monitoring point yields multiple dehumidification zones, including: Based on the historical temperature and humidity time series data at each monitoring point, determine the similarity of the humidity time series change trend between any two monitoring points; and based on the humidity feature vector at each monitoring point, determine the humidity feature distance between any two monitoring points. Based on the similarity of the humidity time series change trend and the humidity characteristic distance between each pair of monitoring points, all monitoring points are clustered to obtain multiple first monitoring point clusters, or multiple monitoring point clusters and at least one isolated monitoring point. The first monitoring point cluster, which contains monitoring points in at least two sealed cavities, is split to obtain at least two second monitoring point clusters; all monitoring points in any second monitoring point cluster belong to the same sealed cavity. Multiple dehumidification zones are generated based on the undivided first monitoring point clusters, second monitoring point clusters, and isolated monitoring points.

[0007] In one possible implementation, the target humidity correlation coefficient and target water vapor propagation lag time between any upstream dehumidification zone and any downstream dehumidification zone are determined according to the following steps: For any one of the upstream dehumidification zone and the downstream dehumidification zone, the comprehensive humidity time series data corresponding to the dehumidification zone is determined based on the historical humidity time series data at all monitoring points within the dehumidification zone. Based on the comprehensive humidity time series data corresponding to the upstream dehumidification zone and the comprehensive humidity time series data corresponding to the downstream dehumidification zone, the initial humidity correlation coefficient between the upstream dehumidification zone and the downstream dehumidification zone under each preset water vapor propagation lag time is determined. The maximum initial humidity correlation coefficient between the upstream dehumidification zone and the downstream dehumidification zone, and its corresponding preset water vapor propagation lag time, are determined as the target humidity correlation coefficient and target water vapor propagation lag time between the upstream dehumidification zone and the downstream dehumidification zone.

[0008] In one possible implementation, the step of using the humidity spatiotemporal correlation matrix as a control constraint to perform zoned dehumidification control on the target ring main unit includes: Acquire real-time temperature and humidity data for each dehumidification zone within the current time period; the real-time temperature and humidity data includes real-time temperature and humidity time-series data. For each dehumidification zone, the humidity spatiotemporal correlation matrix, the real-time temperature and humidity related data corresponding to the dehumidification zone, and the real-time temperature and humidity time series data of the associated zones corresponding to the dehumidification zone are input into the humidity prediction model to obtain the humidity prediction result of the dehumidification zone in the future within a preset time period; the associated zones refer to other dehumidification zones in the humidity spatiotemporal correlation matrix whose target humidity correlation coefficient with the corresponding dehumidification zone is greater than a preset humidity correlation threshold. The baseline humidity judgment threshold is corrected based on the humidity prediction results corresponding to each dehumidification zone to obtain the target humidity time sequence judgment threshold for each dehumidification zone within a preset time period in the future. Based on the humidity prediction results corresponding to each dehumidification zone and the target humidity time sequence determination threshold, the target ring main unit is dehumidified by zone.

[0009] In one possible implementation, the step of inputting the humidity spatiotemporal correlation matrix, the real-time temperature and humidity correlation data corresponding to the dehumidification zone, and the real-time temperature and humidity time-series data of the associated zone corresponding to the dehumidification zone into the humidity prediction model to obtain the humidity prediction result of the dehumidification zone within a preset future time period includes: The real-time temperature and humidity related data corresponding to the dehumidification zone and the real-time temperature and humidity time series data of the associated zone corresponding to the dehumidification zone are input into the causal dilated convolution module in the humidity prediction model to obtain the temperature and humidity change feature vector corresponding to the dehumidification zone. The temperature and humidity abrupt change feature vector and the humidity spatiotemporal correlation matrix are input into the spatiotemporal attention module of the humidity prediction model to obtain the spatiotemporal fused humidity feature vector corresponding to the dehumidification zone. The spatiotemporal fusion humidity feature vector is input into the prediction module of the humidity prediction model to obtain the humidity prediction result.

[0010] In one possible implementation, the target humidity determination threshold for any future time within the target humidity time-series determination threshold for any dehumidification zone is determined according to the following steps: Based on the temperature prediction value in the temperature prediction time series data of the humidity prediction result corresponding to the dehumidification zone at the future time, the benchmark humidity judgment threshold is corrected to obtain the initial humidity judgment threshold of the dehumidification zone at the future time. The initial humidity threshold is scaled based on the probability of condensation occurrence in the humidity prediction results to obtain the target humidity threshold for the dehumidification zone in the future.

[0011] In one possible implementation, the step of performing zoned dehumidification on the target ring main unit based on the humidity prediction results corresponding to each dehumidification zone and the target humidity time-series determination threshold includes: Based on the humidity prediction results and the target humidity time sequence determination threshold corresponding to each dehumidification zone, the target humidity conditions and dehumidification control commands corresponding to each dehumidification zone are determined. Based on the target humidity conditions of each dehumidification zone and the dehumidification control commands corresponding to each dehumidification zone, the target ring main unit can achieve zoned dehumidification.

[0012] Secondly, embodiments of this application also provide a Ring main unit dehumidification The apparatus includes: The acquisition module is used to acquire historical temperature and humidity time series data and humidity feature vectors at each monitoring point in each sealed cavity within the target ring network cabinet; the humidity feature vector is used to characterize the humidity change pattern at the corresponding monitoring point. The clustering module is used to perform cluster analysis on all monitoring points based on the historical temperature and humidity time series data and the humidity feature vector at each monitoring point to obtain multiple dehumidification zones; the temperature and humidity time series change trends among all monitoring points contained in any dehumidification zone are the same; and all monitoring points in any dehumidification zone belong to the same sealed cavity. The construction module is used to construct a humidity spatiotemporal correlation matrix based on the historical humidity time series data in the historical temperature and humidity time series data of all monitoring points in each dehumidification zone; the humidity spatiotemporal correlation matrix includes the target humidity correlation coefficient and the target water vapor propagation lag time between the upstream dehumidification zone and the downstream dehumidification zone. The dehumidification module is used to perform zoned dehumidification of the target ring main unit, based on the spatiotemporal correlation matrix of humidity as the control constraint.

[0013] In one possible implementation, the clustering module is specifically used to determine the similarity of the humidity time series change trend between any two monitoring points based on the historical temperature and humidity time series data at each monitoring point; and to determine the humidity feature distance between any two monitoring points based on the humidity feature vector at each monitoring point. Based on the similarity of the humidity time series change trend and the humidity characteristic distance between each pair of monitoring points, all monitoring points are clustered to obtain multiple first monitoring point clusters, or multiple monitoring point clusters and at least one isolated monitoring point. The first monitoring point cluster, which contains monitoring points in at least two sealed cavities, is split to obtain at least two second monitoring point clusters; all monitoring points in any second monitoring point cluster belong to the same sealed cavity. Multiple dehumidification zones are generated based on the undivided first monitoring point clusters, second monitoring point clusters, and isolated monitoring points.

[0014] In one possible implementation, the construction module is specifically configured to determine the target humidity correlation coefficient and the target water vapor propagation lag time between any upstream dehumidification zone and any downstream dehumidification zone according to the following steps: For any one of the upstream dehumidification zone and the downstream dehumidification zone, the comprehensive humidity time series data corresponding to the dehumidification zone is determined based on the historical humidity time series data at all monitoring points within the dehumidification zone. Based on the comprehensive humidity time series data corresponding to the upstream dehumidification zone and the comprehensive humidity time series data corresponding to the downstream dehumidification zone, the initial humidity correlation coefficient between the upstream dehumidification zone and the downstream dehumidification zone under each preset water vapor propagation lag time is determined. The maximum initial humidity correlation coefficient between the upstream dehumidification zone and the downstream dehumidification zone, and its corresponding preset water vapor propagation lag time, are determined as the target humidity correlation coefficient and target water vapor propagation lag time between the upstream dehumidification zone and the downstream dehumidification zone.

[0015] In one possible implementation, the dehumidification module is specifically used to acquire real-time temperature and humidity related data of each dehumidification zone within the current time period; the real-time temperature and humidity related data includes real-time temperature and humidity time-series data; for each dehumidification zone, the humidity spatiotemporal correlation matrix, the real-time temperature and humidity related data corresponding to the dehumidification zone, and the real-time temperature and humidity time-series data of the associated zone corresponding to the dehumidification zone are input into the humidity prediction model to obtain the humidity prediction result of the dehumidification zone within a preset future time period; the associated zone refers to other dehumidification zones in the humidity spatiotemporal correlation matrix whose target humidity correlation coefficient with the corresponding dehumidification zone is greater than a preset humidity correlation threshold; the baseline humidity judgment threshold is corrected according to the humidity prediction result corresponding to each dehumidification zone to obtain the target humidity time-series judgment threshold for each dehumidification zone within a preset future time period; the target ring main unit is dehumidified by zone according to the humidity prediction result corresponding to each dehumidification zone and the target humidity time-series judgment threshold.

[0016] In one possible implementation, the dehumidification module is specifically used to input the real-time temperature and humidity related data corresponding to the dehumidification zone and the real-time temperature and humidity time-series data of the associated zone corresponding to the dehumidification zone into the causal dilated convolution module in the humidity prediction model to obtain the temperature and humidity abrupt change feature vector corresponding to the dehumidification zone; input the temperature and humidity abrupt change feature vector and the humidity spatiotemporal correlation matrix into the spatiotemporal attention module in the humidity prediction model to obtain the spatiotemporal fused humidity feature vector corresponding to the dehumidification zone; and input the spatiotemporal fused humidity feature vector into the prediction module in the humidity prediction model to obtain the humidity prediction result.

[0017] In one possible implementation, the dehumidification module is specifically configured to determine, according to the following steps, a target humidity determination threshold for any future time within the target humidity time-series determination threshold for any dehumidification zone over the future preset duration: Based on the temperature prediction value in the temperature prediction time series data of the humidity prediction result corresponding to the dehumidification zone at the future time, the benchmark humidity judgment threshold is corrected to obtain the initial humidity judgment threshold of the dehumidification zone at the future time. The initial humidity threshold is scaled based on the probability of condensation occurrence in the humidity prediction results to obtain the target humidity threshold for the dehumidification zone in the future.

[0018] In one possible implementation, the dehumidification module is specifically used to determine the target humidity conditions and dehumidification control commands corresponding to each dehumidification zone based on the humidity prediction results and the target humidity time-series determination threshold for each dehumidification zone; and to realize the zoned dehumidification of the target ring main unit based on the target humidity conditions of each dehumidification zone and the dehumidification control commands corresponding to each dehumidification zone.

[0019] Thirdly, embodiments of this application also provide an electronic device, including: a processor, a storage medium, and a bus, wherein the storage medium stores machine-readable instructions executable by the processor, and when the electronic device is running, the processor communicates with the storage medium via the bus, and the processor executes the machine-readable instructions to perform the steps of the ring main unit dehumidification method as described in any of the first aspects.

[0020] Fourthly, embodiments of this application also provide a computer-readable storage medium storing a computer program, which, when executed by a processor, performs the steps of the ring main unit dehumidification method as described in any of the first aspects.

[0021] This application provides a method, device, electronic device, and storage medium for dehumidifying a ring main unit. The method includes: performing cluster analysis on all monitoring points based on historical temperature and humidity time-series data and humidity feature vectors to obtain multiple dehumidification zones; constructing a humidity spatiotemporal correlation matrix based on historical humidity time-series data from all monitoring points within each dehumidification zone; the humidity spatiotemporal correlation matrix includes the target humidity correlation coefficient and target water vapor propagation lag time between the upstream and downstream dehumidification zones; and using the humidity spatiotemporal correlation matrix as a control constraint basis to perform zoned dehumidification on the target ring main unit. This application enables dynamic division of the target ring main unit into dehumidification zones, adapting to the dynamic migration characteristics of water vapor based on the humidity spatiotemporal correlation matrix, thereby improving dehumidification control accuracy. Attached Figure Description

[0022] To more clearly illustrate the technical solutions of the embodiments of this application, the accompanying drawings used in the embodiments will be briefly introduced below. It should be understood that the following drawings only show some embodiments of this application and should not be regarded as a limitation of the scope. For those skilled in the art, other related drawings can be obtained based on these drawings without creative effort.

[0023] Figure 1 A flowchart of a ring main unit dehumidification method provided in an embodiment of this application is shown; Figure 2 This document illustrates a flowchart of the clustering process for monitoring points provided in an embodiment of this application. Figure 3 This paper shows a schematic diagram of the structure of a ring main unit dehumidification device provided in an embodiment of this application; Figure 4 A schematic diagram of the structure of an electronic device provided in an embodiment of this application is shown. Detailed Implementation

[0024] To make the objectives, technical solutions, and advantages of the embodiments of this application clearer, the technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. It should be understood that the accompanying drawings in this application are for illustrative and descriptive purposes only and are not intended to limit the scope of protection of this application. Furthermore, it should be understood that the schematic drawings are not drawn to scale. The flowcharts used in this application illustrate operations implemented according to some embodiments of this application. It should be understood that the operations in the flowcharts may not be implemented in sequence, and steps without logical contextual relationships may be reversed or implemented simultaneously. In addition, those skilled in the art, guided by the content of this application, may add one or more other operations to the flowcharts, or remove one or more operations from the flowcharts.

[0025] Furthermore, the described embodiments are merely some, not all, of the embodiments of this application. The components of the embodiments of this application described and illustrated herein can typically be arranged and designed in various different configurations. Therefore, the following detailed description of the embodiments of this application provided in the accompanying drawings is not intended to limit the scope of the claimed application, but merely to illustrate selected embodiments of the application. All other embodiments obtained by those skilled in the art based on the embodiments of this application without inventive effort are within the scope of protection of this application.

[0026] To enable those skilled in the art to utilize the content of this application, and in conjunction with the specific application scenario of "distribution network equipment status monitoring and intelligent moisture-proof control technology," the following embodiments are provided. For those skilled in the art, the general principles defined herein can be applied to other embodiments and application scenarios without departing from the spirit and scope of this application. Although this application primarily describes the field of "distribution network equipment status monitoring and intelligent moisture-proof control technology," it should be understood that this is merely an exemplary embodiment.

[0027] It should be noted that the term "comprising" will be used in the embodiments of this application to indicate the presence of the features declared thereafter, but does not exclude the addition of other features.

[0028] The following is a detailed description of a ring main unit dehumidification method provided in the embodiments of this application.

[0029] Reference Figure 1 The diagram shown is a flowchart illustrating a dehumidification method for a ring main unit according to an embodiment of this application. The exemplary steps of this embodiment are described below: S101. Obtain historical temperature and humidity time series data and humidity feature vectors at each monitoring point in each sealed cavity of the target ring network cabinet.

[0030] In this embodiment, the target ring main unit refers to a ring main unit that requires dehumidification. Multiple monitoring points are pre-positioned in each sealed cavity within the target ring main unit. During operation, temperature and humidity data are continuously collected at each monitoring point. For example, the latest temperature and humidity data collected at each monitoring point can be obtained within a preset update cycle and a preset time period (one month, six months, or one year, etc.) to obtain historical temperature and humidity time-series data at each monitoring point. The historical temperature and humidity time-series data includes temperature and humidity data arranged in chronological order of collection time.

[0031] In addition, the humidity feature vector at each monitoring point is used to characterize the humidity change pattern of the corresponding monitoring point. This includes the temperature amplitude (used to characterize the regional environmental stability), humidity amplitude (used to characterize the regional environmental stability), and average humidity growth rate per unit time (used to characterize the speed of moisture intrusion and accumulation, whether it is fast or slow), which are obtained by statistically analyzing the corresponding historical temperature and humidity data. It also includes the number of condensation failures that occurred at the corresponding monitoring point within a preset time period (used to characterize the inherent humidity risk level), and the airflow resistance in the closed cavity where the corresponding monitoring point is located (used to characterize the ease of water vapor diffusion across regions).

[0032] Temperature amplitude refers to the difference between the maximum and minimum temperature values ​​in the corresponding historical temperature and humidity data. Humidity amplitude refers to the difference between the maximum and minimum humidity values ​​in the corresponding historical temperature and humidity data.

[0033] S102. Based on the historical temperature and humidity time series data and the humidity feature vector at each monitoring point, perform cluster analysis on all monitoring points to obtain multiple dehumidification zones.

[0034] In this application embodiment, a hard constraint on the physical control of the ring main unit and a soft constraint on the temporal humidity trend are designed. Cluster analysis is performed on all monitoring points to dehumidify the target ring main unit. Among them, (1) the hard constraint on the physical control of the ring main unit: means that the cluster is not allowed to cross the original sealed partition of the cabinet. If the candidate clusters identified by the clustering algorithm are distributed in two independent sealed cavities, the clusters are automatically split according to the physical partition to fit the actual spatial isolation structure of the ring main unit and avoid incorrect partitioning across cavities; (2) the soft constraint on the temporal humidity trend: only the monitoring points with the correlation coefficient of the humidity temporal change trend higher than the set threshold are divided into the same cluster to ensure that the water vapor diffusion in the same dehumidification zone has a strong temporal correlation and to filter the invalid clustering results caused by the instantaneous single-point humidity interference. Therefore, the temperature and humidity temporal change trends of all monitoring points contained in any dehumidification zone are the same; all monitoring points in any dehumidification zone belong to the same sealed cavity.

[0035] Furthermore, referring to Figure 2 The diagram shown is a flowchart of the clustering process for monitoring points provided in this application embodiment. The detailed implementation process is as follows: S201. Based on the historical temperature and humidity time series data of each monitoring point, determine the similarity of the humidity time series change trend between any two monitoring points; and based on the humidity feature vector of each monitoring point, determine the humidity feature distance between any two monitoring points.

[0036] In this embodiment, the similarity of the temporal variation trend of humidity between any two monitoring points is determined as follows: The Pearson correlation coefficient is calculated between the historical temperature time series data (temperature data arranged in chronological order of collection time) at the first monitoring point and the historical temperature time series data at the second monitoring point to obtain the similarity of the temperature temporal variation trend; the Pearson correlation coefficient is calculated between the historical humidity time series data (humidity data arranged in chronological order of collection time) at the first monitoring point and the historical humidity time series data at the second monitoring point to obtain the similarity of the humidity temporal variation trend; the similarity of the temperature temporal variation trend and the similarity of the humidity temporal variation trend are weighted and summed to obtain the similarity of the humidity temporal variation trend between the first and second monitoring points.

[0037] Among them, considering the condensation formation mechanism, humidity has a stronger dominant role in the risk of dampness. Therefore, the fusion weight corresponding to the similarity of temperature time series change trends (e.g., 0.4) is smaller than the fusion weight corresponding to the similarity of humidity time series change trends (e.g., 0.6).

[0038] In addition, determine the humidity characteristic distance between any two monitoring points: calculate the Euclidean distance between two monitoring points based on the humidity characteristic vectors at each monitoring point; and determine the Euclidean distance between two monitoring points as the humidity characteristic distance between the two monitoring points.

[0039] S202. Based on the similarity of the humidity time-series change trends and the humidity characteristic distance between each pair of monitoring points, all monitoring points are clustered to obtain multiple first monitoring point clusters, or multiple monitoring point clusters and at least one isolated monitoring point.

[0040] In this embodiment, relying solely on humidity characteristic distance is insufficient to distinguish between humid conditions that have similar statistical characteristics but different temporal evolution patterns. Therefore, a soft constraint based on the similarity of humidity temporal change trends is introduced as a clustering threshold to ensure that the humidity rise and fall trends within the same area originate from the same source. Thus, this embodiment employs a dual mandatory constraint of "humidity temporal change trend similarity + humidity characteristic distance," ensuring that monitoring points within a cluster not only have similar long-term humidity steady-state attributes but also exhibit completely consistent humidity dynamic evolution patterns under diurnal and rainy conditions. Specifically, the similarity of humidity temporal change trends among all monitoring points within any first monitoring point cluster is greater than or equal to a preset trend similarity threshold, and the humidity characteristic distance among all monitoring points is less than or equal to a preset distance threshold.

[0041] Optionally, this application embodiment divides the working scenarios of the ring main unit into three types: coastal high humidity, underground enclosed, and outdoor open-air. Corresponding clustering hyperparameters (neighborhood radius and minimum number of sample points) are set for each working scenario. Therefore, clustering is performed on all monitoring points, including: based on the similarity of the humidity time-series change trends and the humidity characteristic distance between pairs of monitoring points, using the preset clustering hyperparameters corresponding to the working scenario of the target ring main unit, all monitoring points are clustered to obtain multiple first monitoring point clusters, or multiple monitoring point clusters and at least one isolated monitoring point.

[0042] Here, during cluster analysis, there may be monitoring points that cannot simultaneously meet the dual thresholds of similarity and feature distance with all other monitoring points. Specifically, the similarity of the humidity time-series change trend must be greater than or equal to a preset similarity threshold, while the humidity feature distance must be less than or equal to a preset distance threshold. In such cases, the monitoring point cannot be included in any cluster. These points often correspond to special locations such as wall-penetrating terminals in cabinets or independent small compartments in corners. Their moisture entry and exit paths are unique, and their temperature and humidity change patterns are not common to other points. The clustering algorithm classifies them as isolated monitoring points, which are subsequently managed separately as dehumidification zones.

[0043] S203. The first monitoring point cluster, which includes monitoring points in at least two sealed cavities, is split to obtain at least two second monitoring point clusters.

[0044] In this embodiment of the application, all first monitoring point clusters are traversed. If a first monitoring point cluster contains monitoring points in at least two sealed cavities, the monitoring points in the first monitoring point cluster are split according to the sealed cavities to obtain at least two second monitoring point clusters. All monitoring points in any second monitoring point cluster belong to the same sealed cavity. The monitoring points in any two second monitoring point clusters belong to different sealed cavities.

[0045] S204. Generate multiple dehumidification zones based on the undivided first monitoring point clusters, second monitoring point clusters, and isolated monitoring points.

[0046] In this application embodiment, each monitoring point in each of the undivided first monitoring point clusters, the second monitoring point clusters containing a number of monitoring points greater than or equal to the minimum number of sample points, the second monitoring point clusters containing a number of monitoring points less than the minimum number of sample points, and each isolated monitoring point are each treated as a separate dehumidification zone.

[0047] S103. Construct a humidity spatiotemporal correlation matrix based on the historical humidity time series data from the historical temperature and humidity time series data of all monitoring points in each dehumidification zone.

[0048] In this embodiment, the humid spatiotemporal correlation matrix is ​​an N*N matrix, where N is the number of dehumidification zones. The humid spatiotemporal correlation matrix includes the target humidity correlation coefficient and the target water vapor propagation lag time between the upstream and downstream dehumidification zones. The upstream dehumidification zone can be any dehumidification zone, and the downstream dehumidification zone can be any dehumidification zone.

[0049] Among them, the target humidity correlation coefficient is used to quantify the impact of humidity changes in the upstream dehumidification zone on the fluctuations in the downstream dehumidification zone; the target water vapor propagation lag time is used to characterize the time interval required for moisture to diffuse from the upstream dehumidification zone to the downstream dehumidification zone, reflecting the temporal sequence characteristics of water vapor propagation across the cavity.

[0050] In addition, if the upstream dehumidification zone and the downstream dehumidification zone are the same dehumidification zone, the target humidity correlation coefficient between the upstream dehumidification zone and the downstream dehumidification zone is set to 1, and the target water vapor propagation lag time is set to 0.

[0051] Specifically, if the upstream dehumidification zone and the downstream dehumidification zone are different dehumidification zones, the target humidity correlation coefficient and the target water vapor propagation lag time between any upstream dehumidification zone and any downstream dehumidification zone are determined according to the following steps: Step 1: For any one of the upstream dehumidification zone and the downstream dehumidification zone, determine the comprehensive humidity time series data corresponding to the dehumidification zone based on the historical humidity time series data of all monitoring points within the dehumidification zone.

[0052] In this embodiment of the application, the comprehensive humidity time series data includes humidity data arranged in chronological order of collection time. Specifically, the comprehensive humidity data at any collection time in the comprehensive humidity time series data is calculated according to the following steps: the average value of all historical humidity data at that collection time in the historical humidity time series data of all monitoring points is calculated to obtain the comprehensive humidity data at that collection time in the comprehensive humidity time series data.

[0053] Step 2: Based on the comprehensive humidity time series data corresponding to the upstream dehumidification zone and the comprehensive humidity time series data corresponding to the downstream dehumidification zone, determine the initial humidity correlation coefficient between the upstream and downstream dehumidification zones under each preset water vapor propagation lag time.

[0054] In this embodiment of the application, the upstream dehumidification zone is... Corresponding comprehensive humidity time series data and downstream dehumidification zones Corresponding comprehensive humidity time series data Substituting into the following formula, we obtain the water vapor propagation lag time between the upstream and downstream dehumidification zones at any preset time. Initial humidity correlation coefficient : ; in, upstream dehumidification zone The corresponding comprehensive humidity time series data at the sampling time Comprehensive humidity data; For downstream dehumidification zones Corresponding comprehensive humidity time series data During the sampling time Comprehensive humidity data; upstream dehumidification zone The average humidity data at all sampling times in the corresponding comprehensive humidity time series data; Downstream dehumidification zone Corresponding comprehensive humidity time series data The average value of humidity data at all sampling times.

[0055] here, Characterizing the upstream dehumidification zone Humidity changes earlier than in the downstream dehumidification zone. The humidity changes, and the direction of water vapor propagation is upstream dehumidification zone. Downstream dehumidification zone . Characterizing downstream dehumidification zones The humidity change occurred earlier than in the upstream dehumidification zone. The humidity changes, and the direction of water vapor propagation is downstream dehumidification zone. Upstream dehumidification zone ; It characterizes the synchronous fluctuation of humidity between the upstream and downstream dehumidification zones, with no time difference in water vapor propagation.

[0056] Step 3: Determine the maximum initial humidity correlation coefficient between the upstream dehumidification zone and the downstream dehumidification zone, along with its corresponding preset water vapor propagation lag time, as the target humidity correlation coefficient and target water vapor propagation lag time between the upstream and downstream dehumidification zones.

[0057] S104. Using the spatiotemporal correlation matrix of humidity as the basis for control constraints, the target ring main unit is dehumidified by zone.

[0058] In this embodiment, the water vapor migration characteristics between different dehumidification zones are introduced using a humidity spatiotemporal correlation matrix to perform zoned dehumidification on the target ring main unit. The specific implementation process is as follows: Step 1: Obtain real-time temperature and humidity data for each dehumidification zone within the current time period; the real-time temperature and humidity data includes real-time temperature and humidity time-series data.

[0059] In this embodiment, the humidity-related data for any dehumidification zone within the current time period (e.g., 24 hours) includes real-time temperature and humidity time-series data within the dehumidification zone, temperature difference time-series data between the sealed cavity where the dehumidification zone is located and the outside of the target ring main unit, rainfall time-series data for the area where the target ring main unit is located, and real-time temperature and humidity time-series data for the associated zones corresponding to the dehumidification zone. Associated zones refer to other dehumidification zones in the humidity spatiotemporal correlation matrix whose target humidity correlation coefficient with the corresponding dehumidification zone is greater than a preset humidity correlation threshold.

[0060] Step 2: For each dehumidification zone, input the humidity spatiotemporal correlation matrix, the real-time temperature and humidity related data corresponding to the dehumidification zone, and the temperature and humidity time series data of the associated zone corresponding to the dehumidification zone into the humidity prediction model to obtain the humidity prediction result of the dehumidification zone in the future within a preset time period.

[0061] In this embodiment, the humidity prediction model is a pre-trained deep learning model used to predict the humidity of each dehumidification zone within a preset future time period. The implementation process of this humidity prediction model is detailed below: i. Input the real-time temperature and humidity related data corresponding to the dehumidification zone and the real-time temperature and humidity time series data of the associated zone corresponding to the dehumidification zone into the causal dilation convolution module in the humidity prediction model to obtain the temperature and humidity change feature vector corresponding to the dehumidification zone.

[0062] In this embodiment, the causal dilation convolution module consists of multiple causal dilation convolution layers, which are used to extract the short-term abrupt changes in temperature and humidity within a single dehumidification zone, capture the rapid accumulation pattern of water vapor caused by rainfall and equipment start-up and shutdown, and obtain a feature vector of temperature and humidity abrupt changes.

[0063] ii. Input the temperature and humidity abrupt change feature vector and the humidity spatiotemporal correlation matrix into the spatiotemporal attention module of the humidity prediction model to obtain the spatiotemporal fused humidity feature vector corresponding to the dehumidification zone.

[0064] In this embodiment, the spatiotemporal attention module relies on the humidity spatiotemporal correlation matrix to constrain the cross-zone water vapor weight allocation, adaptively strengthening the water vapor diffusion correlation features between zones, and obtaining the spatiotemporal fused humidity feature vector corresponding to the dehumidification zone. The formula is expressed as follows: ; ; in, This is the spatiotemporal attention weight matrix after incorporating the humid spatiotemporal correlation matrix; For activation functions; The query matrix is ​​obtained by linear mapping of the feature vectors of sudden temperature and humidity changes; The key matrix is ​​obtained by linear mapping of the eigenvectors of abrupt changes in temperature and humidity. This is the value matrix obtained by linear mapping of the eigenvectors of abrupt temperature and humidity changes; For transpose, The spatiotemporal correlation matrix of humidity; The dimension of the key vector.

[0065] iii. Input the spatiotemporal fusion humidity feature vector into the prediction module of the humidity prediction model to obtain the humidity prediction result.

[0066] In this embodiment, the prediction module is composed of an LSTM network, which is responsible for mining the long-term temporal dependencies of temperature, humidity, and meteorological parameters to achieve multi-step humidity level prediction for the next 6 hours. The humidity prediction results include temperature prediction time series data (temperature prediction values ​​sorted by time), humidity prediction time series data (humidity prediction values ​​sorted by time), and condensation occurrence time series probability (condensation occurrence probability sorted by time).

[0067] In addition, the embodiment of this application also provides a training process for the humidity prediction model: obtaining a humidity spatiotemporal correlation sample matrix, temperature and humidity related sample data corresponding to each dehumidification zone, and humidity labels corresponding to the temperature and humidity related sample data; using the temperature and humidity related sample data corresponding to each dehumidification zone as sample data, and the humidity labels corresponding to the temperature and humidity related sample data as labels, updating and iterating the humidity prediction model with the minimum loss function value as the optimization objective, to obtain the trained humidity prediction model.

[0068] Here, this embodiment of the application abandons the general mean squared error (MSE) loss and designs an asymmetric sparse loss function for condensation risk. The cost of high condensation risk far outweighs the cost of misprediction. Higher loss weights are assigned to high-humidity samples, and sparse penalties are applied to stable low-humidity conditions to suppress frequent false alarms from the model and adapt to the business constraints of prioritizing the safety of distribution network equipment. Specifically, the loss function expression is as follows: ; ; ; in, The value of the loss function; This represents the piecewise asymmetric error loss value. The loss value is the sparsity penalty term; It represents the average humidity value of all future times in the humidity label corresponding to the i-th temperature and humidity related sample data; This represents the average humidity prediction value predicted by the model corresponding to the i-th temperature and humidity related sample data; This is the threshold for determining high humidity risk, used to distinguish between hazardous condensation conditions and normal dry conditions. The weighting coefficient for missed predictions satisfies This is used to amplify the penalty for missed reporting errors in high humidity. For mispredicted weighting coefficients, satisfying This is used to mitigate the penalty for false alarms caused by high humidity. The sparsity penalty coefficient controls the sparsity constraint strength under low humidity conditions. The low humidity stability error tolerance threshold is used to filter out minor monitoring fluctuations under normal operating conditions. This represents the total number of temperature and humidity-related sample data.

[0069] Here, for the above-mentioned segmented asymmetric error loss term, combined with the condensation mechanism of the ring main unit and the equipment safety operation standards, the operating data is divided into two types of intervals, and differentiated loss constraints are implemented: (1) High humidity risk interval: The equipment in the interval has safety hazards such as condensation, insulation dampness, and discharge short circuit, which are the key optimization scenarios for the model; (2) Stable low humidity interval: The equipment is in a normal, dry operating condition with no risk of condensation. The optimization goal is to suppress false alarms and stabilize model predictions.

[0070] This application's embodiments abandon the traditional MSE symmetric equal penalty logic and set differentiated weight penalties for two types of errors, namely, missed prediction and misprediction, in the high-humidity risk range, which fits the core constraint of "safety first" for distribution network equipment; the basic squared error constraint is retained in the low-humidity range to ensure the basic prediction accuracy. The error in leakage prediction under high humidity conditions is amplified by weighting, which greatly increases the penalty for missed detections and forces the model to prioritize the identification of condensation risk, thus avoiding equipment safety accidents caused by underestimating humidity. The system reduces the weight of false alarms in high humidity conditions, tolerates a small number of false alarms, sacrifices some prediction accuracy, and minimizes the risk of missed condensation detection to meet the high reliability requirements of power equipment. In low humidity conditions, there is no additional weight correction, and the system uses the basic squared error to ensure prediction stability under normal operating conditions.

[0071] Furthermore, regarding the aforementioned sparse penalty term loss, the ring main unit operates in a stable low-humidity state for most of its time, with minor normal fluctuations in temperature and humidity. This can easily lead to frequent small false alarms from the model and frequent start-ups and shutdowns of the dehumidification equipment. To address this issue, a new sparse penalty term is added that only applies to the low-humidity range. This term filters out invalid small errors through L1 sparse constraints, suppressing redundant warnings.

[0072] Step 3: Correct the baseline humidity judgment threshold based on the humidity prediction results corresponding to each dehumidification zone to obtain the target humidity time sequence judgment threshold for each dehumidification zone within a preset time period in the future.

[0073] In this embodiment, the reference humidity determination threshold is specifically corrected according to the following steps: i. Based on the temperature prediction value in the temperature prediction time series data of the humidity prediction result corresponding to the dehumidification zone at the future time, the baseline humidity judgment threshold is corrected to obtain the initial humidity judgment threshold of the dehumidification zone at the future time.

[0074] In this embodiment, the critical humidity for condensation changes non-linearly with ambient temperature, and condensation is more likely to occur at lower temperatures. Therefore, the predicted temperature of the dehumidification zone at the future time is used as the basis for this calculation. and reference humidity threshold Substituting (the condensation humidity threshold under standard operating conditions at 25℃) into the following formula, we obtain the initial humidity determination threshold for the dehumidification zone at the future time. : ; in, This is the temperature and humidity coupling coefficient. Physical meaning: when the predicted temperature value... When it decreases, Automatic reduction means that the system automatically tightens the humidity safety limit under low temperature conditions, which is in line with the thermodynamic law of condensation.

[0075] ii. Based on the probability of condensation occurrence in the humidity prediction results, the initial humidity determination threshold is scaled to obtain the target humidity determination threshold of the dehumidification zone in the future.

[0076] In this embodiment, three condensation risk levels are pre-defined: low condensation risk, medium condensation risk, and high condensation risk. Based on the maximum probability of condensation occurrence in the humidity prediction time series, the condensation risk level of the dehumidification zone is determined. The product of the preset scaling factor corresponding to the condensation risk level of the dehumidification zone and the initial humidity judgment threshold is determined as the target humidity judgment threshold for the dehumidification zone in the future.

[0077] Here, if the maximum probability of condensation occurrence in the time-series probability of condensation occurrence in the humidity prediction result is less than the first condensation probability threshold, then the dehumidification zone is at a low condensation risk level; if the maximum probability of condensation occurrence in the time-series probability of condensation occurrence in the humidity prediction result is greater than or equal to the first condensation probability threshold and less than the second condensation probability threshold, then the dehumidification zone is at a medium condensation risk level; if the maximum probability of condensation occurrence is greater than or equal to the second condensation probability threshold, then the dehumidification zone is at a high condensation risk level.

[0078] In this strategy, the preset scaling factor for low condensation risk level is 1, keeping the initial humidity threshold unchanged and employing a conventional dehumidification strategy. For medium condensation risk level, the preset scaling factor is used to slightly lower the initial humidity threshold, initiating pre-dehumidification earlier. For high condensation risk level, the preset scaling factor is used to significantly lower the initial humidity threshold, maximizing early intervention to prevent moisture accumulation. This strategy only dynamically adjusts the humidity threshold, keeping the temperature threshold unchanged. Pre-dehumidification is achieved by tightening the humidity critical conditions, avoiding unnecessary temperature control disturbances.

[0079] Step 4: Based on the humidity prediction results and the target humidity time sequence determination threshold corresponding to each dehumidification zone, perform zoned dehumidification on the target ring network cabinet.

[0080] In this embodiment of the application, the target ring main unit is dehumidified by utilizing the humidity prediction results corresponding to each dehumidification zone and the target humidity time-series determination threshold through the following steps: i. Based on the humidity prediction results and the target humidity time sequence determination threshold corresponding to each dehumidification zone, determine the target humidity conditions and dehumidification control commands corresponding to each dehumidification zone.

[0081] In this embodiment, the current humidity value of each dehumidification zone at the current moment, the humidity prediction result corresponding to each dehumidification zone, and the target humidity time-series determination threshold corresponding to each initial zone are obtained and input into the dehumidification control model to obtain the target humidity conditions and dehumidification control instructions corresponding to each dehumidification zone. The dehumidification control model is pre-trained to plan the corresponding target humidity conditions and dehumidification control instructions for each dehumidification zone using the current humidity value, the humidity prediction result corresponding to each dehumidification zone, and the target humidity time-series determination threshold corresponding to each initial zone.

[0082] The dehumidification control commands include five types of parameters: duct valve on / off status, fan operating speed, single dehumidification duration, interval between adjacent dehumidification operations, and maximum allowable fan speed limit. For example: In safe standby mode, all ducts are closed, fans are stopped, dehumidification duration is zero, and dehumidification intervals are extended indefinitely; in localized mild warning mode, the corresponding dehumidification zone ducts are opened, fans operate at low speed, matching a shorter single dehumidification duration with a longer dehumidification interval; in single-zone high humidity mode, only the corresponding dehumidification zone ducts are opened, fans operate at medium to high speed, extending the single dehumidification duration and shortening the dehumidification interval; in multi-zone contiguous high humidity mode, all ducts of the corresponding dehumidification zone and its associated zones are opened, implementing full-area circulation, limiting the maximum fan speed based on temperature prediction, and simultaneously adapting the longest dehumidification duration and shortest operating interval according to the duration of humidity.

[0083] Here, this application's embodiments divide the humid conditions into four categories: safe standby, localized mild warning, single-zone high humidity, and multi-zone contiguous high humidity. For safe standby, the real-time humidity of a zone is below the target humidity time-series judgment threshold, and the humidity prediction results show no high probability of condensation and no trend of moisture accumulation. For localized mild warning, the real-time humidity of a zone does not exceed the threshold but is close to the target humidity time-series judgment threshold, and the predicted short-term humidity shows an upward trend, with a low risk of condensation. For high-risk scenarios in a single zone, a directional branch circulation scheme is planned based on the model. The predicted duration of humidity is adaptively matched with the fan speed and dehumidification duration, and the dry airflow is precisely supplied to the dense and narrow areas of components such as current transformers and terminals to eliminate airflow dead zones. For multi-zone high humidity scenarios, based on the pressure difference characteristics of the closed space of the ring network cabinet, a top-down closed-loop circulation path is designed to form a water vapor isolation barrier by using directional airflow to suppress the diffusion of moisture across the cavity. At the same time, airflow disturbance constraints are introduced: if the model predicts that the regional temperature will drop sharply in a short period of time, the maximum fan speed is forcibly limited to avoid strong airflow inducing local condensation.

[0084] ii. Based on the target humidity conditions of each dehumidification zone and the dehumidification control commands corresponding to each dehumidification zone, the target ring main unit is dehumidified in zones.

[0085] In the embodiments of this application, corresponding dehumidification control commands are issued according to the independent humid conditions of each dehumidification zone, and airflow path allocation such as directional branch circulation and whole-area closed-loop circulation is completed in a targeted manner.

[0086] Based on the same inventive concept, this application also provides a ring main unit dehumidification device corresponding to the ring main unit dehumidification method. Since the principle of the device in this application is similar to the above-mentioned ring main unit dehumidification method in this application, the implementation of the device can refer to the implementation of the method, and the repeated parts will not be described again.

[0087] Reference Figure 3 The diagram shown is a schematic of a ring main unit dehumidification device provided in an embodiment of this application. The ring main unit dehumidification device includes: The acquisition module 301 is used to acquire historical temperature and humidity time series data and humidity feature vectors at each monitoring point in each sealed cavity of the target ring network cabinet; the humidity feature vector is used to characterize the humidity change pattern at the corresponding monitoring point. Clustering module 302 is used to perform cluster analysis on all monitoring points based on the historical temperature and humidity time series data and the humidity feature vector at each monitoring point to obtain multiple dehumidification zones; the temperature and humidity time series change trends among all monitoring points contained in any dehumidification zone are the same; all monitoring points in any dehumidification zone belong to the same sealed cavity. The construction module 303 is used to construct a humidity spatiotemporal correlation matrix based on the historical humidity time series data in the historical temperature and humidity time series data of all monitoring points in each dehumidification zone; the humidity spatiotemporal correlation matrix includes the target humidity correlation coefficient and the target water vapor propagation lag time between the upstream dehumidification zone and the downstream dehumidification zone. The dehumidification module 304 is used to perform zoned dehumidification on the target ring main unit, based on the humidity spatiotemporal correlation matrix as the control constraint.

[0088] like Figure 4 As shown in the embodiment of this application, an electronic device 400 includes a processor 401, a memory 402, and a bus. The memory 402 stores machine-readable instructions that can be executed by the processor 401. When the electronic device is running, the processor 401 communicates with the memory 402 via the bus. The processor 401 executes the machine-readable instructions to perform the steps of the ring main unit dehumidification method described above.

[0089] Specifically, the memory 402 and processor 401 mentioned above can be general-purpose memory and processor, without any specific limitations. When the processor 401 runs the computer program stored in the memory 402, it can execute the dehumidification method of the ring main unit mentioned above.

[0090] Corresponding to the above-described ring main unit dehumidification method, this application embodiment also provides a computer-readable storage medium storing a computer program, which, when executed by a processor, performs the steps of the above-described ring main unit dehumidification method.

[0091] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working processes of the systems and devices described above can be referred to the corresponding processes in the method embodiments, and will not be repeated here. In the several embodiments provided in this application, it should be understood that the disclosed systems, devices, and methods can be implemented in other ways. The device embodiments described above are merely illustrative. For example, the division of modules is only a logical functional division, and in actual implementation, there may be other division methods. Furthermore, multiple modules or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the displayed or discussed mutual coupling or direct coupling or communication connection can be through some communication interfaces; the indirect coupling or communication connection of devices or modules can be electrical, mechanical, or other forms.

[0092] The modules described as separate components may or may not be physically separate. The components shown as modules may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0093] In addition, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit.

[0094] If the aforementioned functions are implemented as software functional units and sold or used as independent products, they can be stored in a processor-executable, non-volatile, computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or a portion 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 ring main unit dehumidification method described in the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, ROM, RAM, magnetic disks, or optical disks.

[0095] The above are merely specific embodiments of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

Claims

1. A method for dehumidifying a ring main unit, characterized in that, The method includes: Historical temperature and humidity time-series data and humidity feature vectors are obtained at each monitoring point in each sealed cavity within the target ring network cabinet; the humidity feature vectors are used to characterize the humidity change pattern at the corresponding monitoring point. Based on the historical temperature and humidity time series data and the humidity feature vector at each monitoring point, cluster analysis is performed on all monitoring points to obtain multiple dehumidification zones; the temperature and humidity time series change trends among all monitoring points contained in any dehumidification zone are the same; all monitoring points within any dehumidification zone belong to the same sealed cavity; Based on the historical humidity time series data from all monitoring points in each dehumidification zone, a humidity spatiotemporal correlation matrix is ​​constructed; the humidity spatiotemporal correlation matrix includes the target humidity correlation coefficient and the target water vapor propagation lag time between the upstream and downstream dehumidification zones. Using the aforementioned humidity spatiotemporal correlation matrix as the basis for control constraints, the target ring main unit is dehumidified by zone.

2. The dehumidification method for a ring main unit according to claim 1, characterized in that, The process involves clustering all monitoring points based on historical temperature and humidity time-series data and humidity feature vectors to obtain multiple dehumidification zones, including: Based on the historical temperature and humidity time series data at each monitoring point, determine the similarity of the humidity time series change trend between any two monitoring points; and based on the humidity feature vector at each monitoring point, determine the humidity feature distance between any two monitoring points. Based on the similarity of the humidity time series change trend and the humidity characteristic distance between each pair of monitoring points, all monitoring points are clustered to obtain multiple first monitoring point clusters, or multiple monitoring point clusters and at least one isolated monitoring point. The first monitoring point cluster, which contains monitoring points in at least two sealed cavities, is split to obtain at least two second monitoring point clusters; all monitoring points in any second monitoring point cluster belong to the same sealed cavity. Multiple dehumidification zones are generated based on the undivided first monitoring point clusters, second monitoring point clusters, and isolated monitoring points.

3. The dehumidification method for a ring main unit according to claim 1, characterized in that, The target humidity correlation coefficient and target water vapor propagation lag time between any upstream dehumidification zone and any downstream dehumidification zone are determined according to the following steps: For any one of the upstream dehumidification zone and the downstream dehumidification zone, the comprehensive humidity time series data corresponding to the dehumidification zone is determined based on the historical humidity time series data at all monitoring points within the dehumidification zone. Based on the comprehensive humidity time series data corresponding to the upstream dehumidification zone and the comprehensive humidity time series data corresponding to the downstream dehumidification zone, the initial humidity correlation coefficient between the upstream dehumidification zone and the downstream dehumidification zone under each preset water vapor propagation lag time is determined. The maximum initial humidity correlation coefficient between the upstream dehumidification zone and the downstream dehumidification zone, and its corresponding preset water vapor propagation lag time, are determined as the target humidity correlation coefficient and target water vapor propagation lag time between the upstream dehumidification zone and the downstream dehumidification zone.

4. The dehumidification method for a ring main unit according to any one of claims 1 to 3, characterized in that, The step of using the humidity spatiotemporal correlation matrix as the control constraint basis to perform zoned dehumidification control on the target ring main unit includes: Acquire real-time temperature and humidity data for each dehumidification zone within the current time period; the real-time temperature and humidity data includes real-time temperature and humidity time-series data. For each dehumidification zone, the humidity spatiotemporal correlation matrix, the real-time temperature and humidity related data corresponding to the dehumidification zone, and the real-time temperature and humidity time series data of the associated zones corresponding to the dehumidification zone are input into the humidity prediction model to obtain the humidity prediction result of the dehumidification zone in the future within a preset time period; the associated zones refer to other dehumidification zones in the humidity spatiotemporal correlation matrix whose target humidity correlation coefficient with the corresponding dehumidification zone is greater than a preset humidity correlation threshold. The baseline humidity judgment threshold is corrected based on the humidity prediction results corresponding to each dehumidification zone to obtain the target humidity time sequence judgment threshold for each dehumidification zone within a preset time period in the future. Based on the humidity prediction results corresponding to each dehumidification zone and the target humidity time sequence determination threshold, the target ring main unit is dehumidified by zone.

5. The dehumidification method for a ring main unit according to claim 4, characterized in that, The step of inputting the humidity spatiotemporal correlation matrix, the real-time temperature and humidity related data corresponding to the dehumidification zone, and the real-time temperature and humidity time-series data of the associated zone corresponding to the dehumidification zone into the humidity prediction model to obtain the humidity prediction result of the dehumidification zone within a preset future time period includes: The real-time temperature and humidity related data corresponding to the dehumidification zone and the real-time temperature and humidity time series data of the associated zone corresponding to the dehumidification zone are input into the causal dilated convolution module in the humidity prediction model to obtain the temperature and humidity change feature vector corresponding to the dehumidification zone. The temperature and humidity abrupt change feature vector and the humidity spatiotemporal correlation matrix are input into the spatiotemporal attention module of the humidity prediction model to obtain the spatiotemporal fused humidity feature vector corresponding to the dehumidification zone. The spatiotemporal fusion humidity feature vector is input into the prediction module of the humidity prediction model to obtain the humidity prediction result.

6. The dehumidification method for a ring main unit according to claim 4, characterized in that, The target humidity threshold for any future time period within the target humidity time series determination threshold of any dehumidification zone is determined according to the following steps: Based on the temperature prediction value in the temperature prediction time series data of the humidity prediction result corresponding to the dehumidification zone at the future time, the benchmark humidity judgment threshold is corrected to obtain the initial humidity judgment threshold of the dehumidification zone at the future time. The initial humidity threshold is scaled based on the probability of condensation occurrence in the humidity prediction results to obtain the target humidity threshold for the dehumidification zone in the future.

7. The dehumidification method for a ring main unit according to claim 4, characterized in that, The step of performing zoned dehumidification on the target ring main unit based on the humidity prediction results corresponding to each dehumidification zone and the target humidity time-series determination threshold includes: Based on the humidity prediction results and the target humidity time sequence determination threshold corresponding to each dehumidification zone, the target humidity conditions and dehumidification control commands corresponding to each dehumidification zone are determined. Based on the target humidity conditions of each dehumidification zone and the dehumidification control commands corresponding to each dehumidification zone, the target ring main unit can achieve zoned dehumidification.

8. A ring main unit dehumidification device, characterized in that, The device includes: The acquisition module is used to acquire historical temperature and humidity time series data and humidity feature vectors at each monitoring point in each sealed cavity within the target ring network cabinet; the humidity feature vector is used to characterize the humidity change pattern at the corresponding monitoring point. The clustering module is used to perform cluster analysis on all monitoring points based on the historical temperature and humidity time series data and the humidity feature vector at each monitoring point to obtain multiple dehumidification zones; the temperature and humidity time series change trends among all monitoring points contained in any dehumidification zone are the same; and all monitoring points in any dehumidification zone belong to the same sealed cavity. The construction module is used to construct a humidity spatiotemporal correlation matrix based on the historical humidity time series data in the historical temperature and humidity time series data of all monitoring points in each dehumidification zone; the humidity spatiotemporal correlation matrix includes the target humidity correlation coefficient and the target water vapor propagation lag time between the upstream dehumidification zone and the downstream dehumidification zone. The dehumidification module is used to perform zoned dehumidification of the target ring main unit, based on the spatiotemporal correlation matrix of humidity as the control constraint.

9. An electronic device, characterized in that, include: The device includes a processor, a storage medium, and a bus, wherein the storage medium stores machine-readable instructions executable by the processor, and when the electronic device is running, the processor communicates with the storage medium via the bus, and the processor executes the machine-readable instructions to perform the steps of the ring main unit dehumidification method as described in any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program that, when executed by a processor, performs the steps of the ring main unit dehumidification method as described in any one of claims 1 to 7.

Images