Temperature rise monitoring method and system for high-voltage totally-enclosed vacuum environment-friendly GIS equipment

Abstract

The invention relates to the technical field of temperature monitoring, and discloses a temperature rise monitoring method and system for high-voltage totally-closed vacuum environment-friendly GIS equipment, and the method comprises the steps: collecting temperature data through a temperature sensor, and constructing a node feature matrix and an adjacent matrix according to the spatial position relation of an SF6 gas chamber layer, a high-voltage conductor layer and a housing sealing layer; carrying out intra-layer graph convolution operation and inter-layer graph convolution operation to obtain a temperature space distribution feature vector, and carrying out time sequence fusion to obtain a temperature dynamic change time sequence feature vector; partitioning the monitoring area according to the temperature spatial distribution feature vector and the temperature dynamic change time sequence feature vector to obtain adaptive control parameters of each partition; and the adaptive control parameters are input into a multi-level compensation controller for temperature compensation coefficient calculation, and temperature correction parameters of each level are obtained, so that the integrity and accuracy of temperature field distribution feature extraction are improved, and accurate evaluation and graded early warning of temperature rise risks of different levels are realized.

Description

Technical Field

[0001] The present application relates to the field of temperature monitoring technology, and in particular to a temperature rise monitoring method and system for high-voltage, fully enclosed, vacuum, environmentally friendly GIS equipment. Background Art

[0002] High-voltage, fully enclosed, vacuum, and environmentally friendly GIS equipment is a key device in the power system, and its safe and stable operation has an important impact on the reliability of the power system. In actual operation, the temperature distribution characteristics inside the GIS equipment are complex, involving the temperature coupling of multiple layers such as the SF6 gas chamber, high-voltage conductor, and shell. The traditional single-point temperature measurement method is difficult to fully reflect the temperature field distribution characteristics of the equipment.

[0003] The existing temperature rise monitoring methods for GIS equipment mainly rely on empirical knowledge to judge the temperature threshold, and lack in-depth analysis of the dynamic correlation between the temperature at different levels of the equipment, resulting in low warning accuracy. At the same time, due to the dynamic changes in the operating conditions of the equipment, the temperature field distribution shows obvious spatiotemporal correlation characteristics, and it is difficult for traditional static monitoring methods to accurately grasp the temperature change trend. In addition, multiple factors such as SF6 gas pressure, load current and ambient humidity in GIS equipment will affect the distribution of the temperature field, and there is a complex coupling relationship between these factors. Existing monitoring methods often consider these factors separately, lack of systematic multi-parameter fusion analysis, and it is difficult to achieve accurate warning of temperature rise risks, affecting the operational reliability and safety of the equipment. Summary of the invention

[0004] The present application provides a temperature rise monitoring method and system for high-voltage, fully enclosed, vacuum, environmentally friendly GIS equipment, thereby improving the integrity and accuracy of temperature field distribution feature extraction, and realizing accurate assessment and graded warning of temperature rise risks at different levels.

[0005] The first aspect of the present application provides a temperature rise monitoring method for a high-voltage fully enclosed vacuum environmentally friendly GIS device, the temperature rise monitoring method for a high-voltage fully enclosed vacuum environmentally friendly GIS device comprising: The temperature data is collected through the temperature sensor, and the node feature matrix and adjacency matrix are constructed according to the spatial position relationship between the SF6 gas chamber layer, the high-voltage conductor layer and the shell sealing layer; Performing intra-layer graph convolution operations and inter-layer graph convolution operations on the node feature matrix and the adjacency matrix to obtain a temperature spatial distribution feature vector, and performing time series fusion to obtain a temperature dynamic change time series feature vector; Partitioning the monitoring area according to the temperature spatial distribution feature vector and the temperature dynamic change time series feature vector to obtain adaptive control parameters for each partition; The adaptive control parameters are input into the multi-level compensation controller to calculate the temperature compensation coefficient, so as to obtain the temperature correction parameters of each level.

[0006] The second aspect of the present application provides a temperature rise monitoring system for a high-voltage fully enclosed vacuum environmentally friendly GIS device, the temperature rise monitoring system for a high-voltage fully enclosed vacuum environmentally friendly GIS device comprising: The acquisition module is used to collect temperature data through a temperature sensor and construct a node feature matrix and an adjacency matrix according to the spatial position relationship between the SF6 gas chamber layer, the high-voltage conductor layer and the shell sealing layer; An operation module, used for performing intra-layer graph convolution operation and inter-layer graph convolution operation on the node feature matrix and the adjacency matrix to obtain a temperature spatial distribution feature vector, and performing time series fusion to obtain a temperature dynamic change time series feature vector; A partitioning module, used to partition the monitoring area according to the temperature spatial distribution characteristic vector and the temperature dynamic change time series characteristic vector, and obtain adaptive control parameters of each partition; The compensation module is used to input the adaptive control parameters into the multi-level compensation controller to calculate the temperature compensation coefficient and obtain the temperature correction parameters of each level.

[0007] Compared with the prior art, the present application has the following beneficial effects: by constructing a three-layer network topology structure, the systematic management of the temperature data of the SF6 gas chamber layer, the high-voltage conductor layer and the outer shell sealing layer is realized, and the integrity and accuracy of the temperature field distribution feature extraction are improved; the combination of intra-layer graph convolution and inter-layer graph convolution is adopted to effectively capture the temperature correlation between different layers and enhance the ability to extract the spatial distribution characteristics of the temperature field; combined with the timing fusion mechanism of the state update gate and the parameter association gate, the dynamic correlation analysis of temperature data with multiple parameters such as SF6 gas pressure and ambient humidity is realized; through the partition reinforcement learning control strategy, differentiated control parameters are adopted for different areas, which improves the accuracy of temperature control; a multi-level compensation control mechanism is adopted, which takes into account the coupling effects of gas pressure, load current and ambient humidity, and improves the accuracy of temperature correction; based on the risk warning mechanism of the multi-layer fully connected neural network, accurate assessment and graded warning of temperature rise risks at different levels are realized. BRIEF DESCRIPTION OF THE DRAWINGS

[0008] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative labor.

[0009] The structures, proportions, sizes, etc. illustrated in the drawings of this specification are only used to match the contents disclosed in the specification so as to facilitate understanding and reading by persons familiar with this technology. They are not used to limit the conditions under which the present invention can be implemented, and therefore have no substantive technical significance. Any structural modification, change in proportion or adjustment of size, without affecting the effects and purposes that can be achieved by the present invention, should still fall within the scope of the technical contents disclosed by the present invention.

[0010] Figure 1 It is a flow chart of a temperature rise monitoring method of a high-voltage fully enclosed vacuum environmentally friendly GIS device provided by an embodiment of the present invention; Figure 2 It is a schematic block diagram of the structure of a temperature rise monitoring system for high-voltage, fully enclosed, vacuum, environmentally friendly GIS equipment provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0011] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0012] The flowcharts shown in the accompanying drawings are only examples and do not necessarily include all the contents and operations / steps, nor must they be executed in the order described. For example, some operations / steps may also be decomposed, combined or partially merged, so the actual execution order may change according to actual conditions.

[0013] It should also be understood that the terms used in this application specification are only for the purpose of describing specific embodiments and are not intended to limit the application. As used in this application specification and the appended claims, unless the context clearly indicates otherwise, the singular forms "a", "an" and "the" are intended to include plural forms.

[0014] It should be further understood that the term "and / or" used in the specification and appended claims of this application refers to any combination and all possible combinations of one or more of the associated listed items, and includes these combinations. Figure 1 An embodiment of the temperature rise monitoring method of high-voltage fully enclosed vacuum environmentally friendly GIS equipment in the embodiment of the present application includes: Step 100: collect temperature data through a temperature sensor, and construct a node feature matrix and an adjacency matrix according to the spatial position relationship between the SF6 gas chamber layer, the high-voltage conductor layer and the shell sealing layer; It is understandable that the execution subject of the present application can be a temperature rise monitoring system of a high-voltage fully enclosed vacuum environmentally friendly GIS device, or a terminal or a server, which is not specifically limited here. The present application embodiment is described by taking a server as the execution subject as an example.

[0015] Specifically, the temperature data of the corresponding areas are collected by temperature sensors arranged in the SF6 gas chamber layer, high-voltage conductor layer and shell sealing layer of the GIS equipment. In the SF6 gas chamber layer, multiple sensors are arranged to collect the temperature data of each monitoring point inside the gas chamber, and the specific spatial position coordinates of these sensors are recorded at the same time. The temperature values ​​and coordinate information are paired and sorted into a sensor node set of the gas chamber layer. In the high-voltage conductor layer, temperature sensors are installed at key locations such as circuit breaker contacts, disconnectors and bushing joints. The temperature data of these locations are collected, and the data is paired in combination with their spatial position coordinates to generate a sensor node set of the conductor layer. In the shell sealing layer, sensors are arranged to monitor the temperature changes of the equipment shell and supporting structure. After pairing these temperature data with the corresponding spatial positions, a sensor node set of the shell layer is formed. The spatial coordinates of the nodes of each layer are standardized, and the spatial positions of the nodes at different levels are mapped to a unified coordinate system to ensure that the data of all nodes are consistent under the same coordinate system. At the same time, the collected temperature data are normalized, and the temperature data of different nodes are adjusted to a unified numerical range to eliminate the interference of different measurement ranges on subsequent calculations, and a standardized sensor node matrix is ​​generated. On the basis of the standardized matrix, the node feature matrix is ​​constructed. The node feature matrix is ​​a key data structure that comprehensively describes the attributes of each node. It contains the normalized temperature value, spatial position and hierarchical information of each node. The hierarchical attribute of each node is clearly marked as SF6 gas chamber layer, high-voltage conductor layer or shell enclosure layer. According to the spatial position information in the standardized sensor node matrix, the distance between any two sensor nodes is calculated to establish the spatial association between the nodes. By comparing these distance values ​​with the preset threshold, it is determined whether the two nodes have a direct connection relationship. If the distance between the two nodes is within the intra-layer connection threshold range and they belong to the same layer, they are defined as intra-layer connection; if the distance between the two nodes is within the inter-layer connection threshold range and they belong to different layers, they are defined as inter-layer connection. Through this judgment process, the relationship between the nodes is converted into a binary adjacency matrix, in which each element indicates whether there is a connection between the corresponding two nodes. The binary adjacency matrix is ​​associated with the node hierarchical information. In the binary adjacency matrix, the intra-layer connection edges and inter-layer connection edges are clearly marked, for example, different identifiers or weights are used to reflect the nature of the two connections, comprehensively represent the connection relationship between nodes, and show the correlation characteristics within each layer and between layers. Through the above steps, a node feature matrix containing node temperature, spatial position and hierarchical information and an adjacency matrix reflecting the connection relationship and hierarchical structure between nodes are constructed.

[0016] Step 200: perform intra-layer graph convolution operations and inter-layer graph convolution operations on the node feature matrix and the adjacency matrix to obtain a temperature spatial distribution feature vector, and perform time series fusion to obtain a temperature dynamic change time series feature vector; Specifically, the temperature data of each sensor node in the node feature matrix is ​​normalized, and the hierarchical information of each node is combined to mark it, and all sensor nodes are divided into three different hierarchical node groups, namely, the air chamber layer node group, the conductor layer node group and the shell layer node group, to form a hierarchical grouping matrix. At the same time, based on the adjacency matrix, the connection relationship between nodes is distinguished, and all connection edges are divided into an intra-layer connection edge set and an inter-layer connection edge set according to the hierarchical label of the sensor node, and a connection edge classification matrix is ​​obtained. The initial feature transformation result is obtained by performing matrix multiplication operation on the hierarchical grouping matrix and the first convolution kernel matrix, and the ReLU activation function is applied to this operation result to introduce nonlinearity, and the initial feature matrix within the layer is generated. Each node feature vector in the initial feature matrix will contain an important representation of its local features. The attention coefficient of each node feature vector in the initial feature matrix within the layer is calculated. By comprehensively considering the temperature gradient and topological distance between nodes, the attention weight coefficient of each node is generated, and the attention weight matrix is ​​constructed. The attention weight matrix is ​​weightedly combined with the initial feature matrix within the layer, and the graph convolution operation is performed in combination with the set of intra-layer connection edges to extract the local feature interaction between nodes, and the intra-layer spatial feature matrix is ​​obtained to reflect the temperature distribution characteristics within each layer. In order to capture the feature association between nodes at different levels, the node feature vector in the intra-layer spatial feature matrix is ​​convolved with the second convolution kernel matrix, and the cross-layer feature transfer is completed in combination with the inter-layer connection edge set to generate an inter-layer correlation feature matrix, which reflects the heat conduction relationship between different levels and reveals the synergy of each layer in the overall temperature distribution. The intra-layer spatial feature matrix and the inter-layer correlation feature matrix are feature spliced ​​to obtain a spatial feature expression structure. The spliced ​​feature vector is input into the fully connected layer for feature compression to obtain the temperature spatial distribution feature vector. The temperature spatial distribution feature vector is time-series fused with the SF6 gas pressure data and the ambient humidity data. The dynamic association between temperature features and external environmental parameters is captured through the time series model, and a time series feature vector of temperature dynamic change is generated to comprehensively reflect the time change law of temperature rise inside the GIS equipment and its response to environmental conditions.

[0017] The temperature spatial distribution feature vector and SF6 gas pressure data are normalized to eliminate the inconsistency caused by different data types and numerical ranges, ensure that different parameters remain comparable within the same numerical range, and generate a temperature and pressure time series matrix. The temperature and pressure time series matrix is ​​aligned with the ambient humidity data in the time dimension to ensure the consistency of the sampling points of all parameter data on the time axis, and form a multi-parameter time series data matrix containing temperature, pressure and humidity information. On the basis of the multi-parameter time series data matrix, the proportion of historical information in the matrix is ​​calculated through the state update gate to determine the comprehensive weight relationship between the data at each moment and its historical information, and generate a state update matrix. The state update gate analyzes the dynamic change law of the data at each time step and judges the importance of historical information at the current moment, thereby effectively capturing the cumulative effect of historical information during the temperature rise process. At the same time, the parameter association gate is used to calculate the influence of each parameter in the multi-parameter time series data matrix. Based on the thermodynamic coupling relationship between SF6 gas pressure and temperature, and the correlation between ambient humidity and temperature changes, a parameter association matrix is ​​generated to quantify the influence between different parameters and dynamically adjust the weight of each parameter in the overall feature extraction process. The state update matrix and the parameter association matrix are matrix multiplied, and the key features of the data are extracted through cross-dimensional fusion calculation. The fused result is subjected to feature selection through the output gate to screen out the most representative parameter combination for the dynamic change of temperature rise and obtain the fused feature matrix. Dynamic time attention calculation is performed on the time series data in the fused feature matrix. Combined with the operating status of GIS equipment (such as the start and stop status of the equipment) and the impact of load changes on temperature characteristics, a time series weighted matrix is ​​generated. The dynamic time attention mechanism can give different weights to feature data at different times, so that the data in the key operating stage of the equipment contributes more significantly to the overall characteristics. The fused feature matrix and the time series weighted matrix are weighted superimposed to obtain the initial dynamic feature vector. The initial dynamic feature vector is extracted through the gated recurrent unit (GRU). Through its gating mechanism, GRU can effectively capture the long-term dependency of feature vectors in time series and avoid the gradient vanishing problem in traditional recurrent networks. Under the processing of GRU, the initial dynamic feature vector is converted into a time series feature vector of temperature dynamic change, which reflects the temperature rise law of GIS equipment under different environmental conditions and operating conditions.

[0018] Step 300, partition the monitoring area according to the temperature spatial distribution feature vector and the temperature dynamic change time series feature vector, and obtain the adaptive control parameters of each partition; It should be noted that the temperature spatial distribution feature vector and the temperature dynamic change time series feature vector are deeply integrated in the spatial and temporal dimensions to generate a spatiotemporal feature matrix. The spatiotemporal feature matrix contains the temperature distribution characteristics of each monitoring area inside the GIS equipment and its dynamic laws of change over time. By performing hierarchical dimensionality reduction on the matrix, redundant information is removed while retaining key features to obtain the regional classification feature vector. The regional classification feature vector is input into the classification network to perform regional division according to the thermal characteristics and temperature dynamic laws of different parts in the GIS equipment. The core heating parts such as circuit breaker contacts and disconnectors are divided into key areas; the conductor connection points, sleeve joints and other locations are divided into transition areas; and the housing and support structure areas are divided into ordinary areas. Through this step, the regional category label matrix is ​​generated. The reinforcement learning training environment is constructed for the regional category label matrix. The reinforcement learning environment clarifies the control objectives and operable range of the system by setting the state space and action space. The state space defines the temperature state of each partition, including temperature distribution characteristics, dynamic change trends, etc., while the action space covers the temperature control actions that may be taken by each partition, such as adjusting the operation mode of the cooling system and optimizing the gas flow path. By setting the temperature control requirements for different areas, the reinforcement learning environment parameters are generated. The reinforcement learning environment parameters are input into the reinforcement learning agent, and the intelligent agent explores and samples the temperature control actions of each partition to generate an action sample set. In this process, the reinforcement learning agent tries different control actions in a trial-and-error manner and evaluates each action through the constructed reward function. The design of the reward function comprehensively considers factors such as the accuracy, response speed, temperature uniformity, and long-term stability of temperature control to ensure that the control strategy of each partition can not only achieve accurate temperature rise control, but also respond quickly to environmental changes and maintain stability during long-term operation. Through the combination of action sample sets and partition reward values, the reinforcement learning agent can continuously optimize the control strategy. In order to improve the quality of the control strategy, the policy network is trained on the action sample sets and partition reward values. The training process uses the policy gradient algorithm to gradually generate policy network models suitable for each partition by optimizing the parameters of the policy network. These models can dynamically generate highly adaptable temperature control strategies based on the characteristics of different partitions. To ensure that the performance of the policy network model reaches the optimal state, performance evaluation and parameter fine-tuning are performed on it. The trained policy network model is input into the temperature control accuracy evaluator, response speed evaluator and stability evaluator respectively to evaluate its control performance in the actual environment. The evaluation results will be used as feedback to fine-tune the policy network parameters to obtain the optimized policy network model. The temperature control action sequence is generated based on the optimized policy network model. These action sequences represent the specific control measures that should be taken by each partition under different conditions. By converting the temperature control action sequence into actual control parameters, the adaptive control parameters of each partition are generated.Adaptive control parameters dynamically adjust the temperature rise control strategy according to the partition characteristics to ensure uniform temperature distribution and good thermal stability of GIS equipment during operation, thereby improving the safety and reliability of the equipment.

[0019] Step 400: input the adaptive control parameters into the multi-level compensation controller to calculate the temperature compensation coefficient, and obtain the temperature correction parameters of each level.

[0020] Specifically, the adaptive control parameters are input into the multi-level compensation controller, and the compensation controller performs hierarchical processing on the control parameters according to the structural characteristics of the GIS equipment and the thermodynamic characteristics of different levels to generate a hierarchical control parameter set. The control parameters of the gas chamber layer mainly focus on the pressure and temperature relationship of the SF6 gas, the parameters of the high-voltage conductor layer focus on analyzing the thermal effect of the load current, and the outer shell layer performs condensation analysis based on the relationship between ambient humidity and surface temperature. The hierarchical processing ensures that the control parameters of each layer can accurately match its unique thermodynamic and operating characteristics. After the hierarchical processing is completed, the compensation controller uses the corresponding physical model and calculation method to calculate the temperature compensation coefficient for different levels. For the gas chamber layer, the SF6 gas pressure-temperature coupling model is used for analysis, the gas pressure and temperature data are input into the model, and the correlation calculation is performed based on the coupling relationship between gas pressure and temperature to obtain the temperature compensation coefficient of the gas chamber layer. This coefficient can effectively reflect the thermodynamic change characteristics inside the SF6 gas and ensure the accuracy of the temperature rise monitoring and control of the gas chamber layer. At the same time, in the high-voltage conductor layer, the load current data is input into the load current heat accumulation model, and the heat accumulation relationship between the current intensity and the conductor temperature rise is calculated to obtain the temperature compensation coefficient of the conductor layer. Through calculation, the influence of load current on conductor temperature rise is quantified, so as to accurately compensate and control the thermal effect of high-voltage conductor layer. For the outer shell layer, the focus is on analyzing the influence of ambient humidity on temperature rise and condensation risk. By inputting the data collected by the ambient humidity sensor into the humidity distribution calculation module, the humidity distribution calculation result is obtained, and it is input into the ambient humidity condensation model for condensation risk analysis. Based on the relationship between humidity distribution and surface temperature, the model can accurately evaluate the condensation risk of the outer shell layer and generate the temperature compensation coefficient of the outer shell layer to ensure that the outer shell layer can still maintain temperature uniformity in an environment with large humidity changes and avoid the impact of condensation on equipment operation. After completing the calculation of the temperature compensation coefficient of each layer, the compensation control matrix is ​​generated according to the correspondence between the hierarchical control parameter set and the compensation coefficient of each layer. The matrix organically combines the control parameters of each layer with its corresponding compensation coefficient to form a unified framework that fully reflects the temperature rise compensation requirements of each layer. In order to ensure the rationality and stability of the compensation parameters, the parameters in the compensation control matrix are subjected to stability constraint processing. By analyzing the temperature change limit of each layer, the correction range of each compensation parameter is determined, and a constraint correction matrix is ​​generated to avoid monitoring deviation or system instability caused by over-compensation or under-compensation. The constraint correction matrix is ​​input into the temperature compensation calculation module, and the temperature correction amount of each layer is calculated according to the temperature compensation coefficient of each layer to generate the final temperature correction parameters.

[0021] The temperature correction parameters, temperature spatial distribution feature vectors and temperature dynamic change time series feature vectors are deeply fused, and these data are integrated by level and characteristic dimension through feature combination to ensure the temporal and spatial consistency of all parameters. The fused data is standardized to unify the numerical range of the data, eliminate the influence of dimensional differences on the calculation results, and generate a standardized feature matrix. Based on the standardized feature matrix, the temperature rise risk analysis and risk level classification are performed for the SF6 gas chamber layer, the high-voltage conductor layer and the shell sealing layer. For the SF6 gas chamber layer, the temperature deviation value of the gas chamber layer is calculated by the temperature anomaly detection module using the gas temperature characteristics and the corrected parameter values ​​in the standardized feature matrix, and these deviation values ​​are compared and analyzed with the preset insulation breakdown threshold. If the deviation value exceeds the threshold range, it is determined that there is an insulation breakdown risk. The risk level matrix of the gas chamber layer is generated in combination with the detection results. The value of each node in the matrix reflects the risk level distribution of different areas of the gas chamber. For the high-voltage conductor layer, the temperature gradient characteristics of the circuit breaker contacts, disconnectors and bushing joints are extracted from the standardized feature matrix and input into the temperature rise detection module. By calculating the temperature change rate of these key parts and combining their heat conduction characteristics, the temperature rise risk of each area in the conductor layer is evaluated, and the risk level matrix of the conductor layer is generated, which reflects the thermal safety of each key part in the high-voltage conductor layer. At the same time, the temperature rise risk analysis of the enclosure sealing layer is mainly based on temperature uniformity detection, and the surface temperature distribution characteristics in the standardized feature matrix are correlated with the ambient temperature difference. By analyzing the deviation between the surface temperature change of different areas of the equipment shell and the ambient temperature, the temperature rise uniformity and condensation risk of the shell layer are evaluated, and the risk level matrix of the shell layer is generated. Through independent analysis of the air chamber layer, conductor layer and shell layer, risk level matrices corresponding to the three layers are formed, and each matrix records the temperature rise risk information of each node in the layer. The air chamber layer risk level matrix, the conductor layer risk level matrix and the shell layer risk level matrix are input into a multi-layer fully connected neural network. In the neural network, the potential correlation in the three-layer risk matrix is ​​extracted through layer-by-layer processing of linear transformation and nonlinear activation function, and a comprehensive hierarchical risk feature vector containing risk information of different levels is generated. The hierarchical risk feature vector is input into the Softmax classification module. By calculating the probability distribution of various risks and combining the preset temperature rise risk thresholds of the air chamber layer, conductor layer and shell layer, a multi-level risk assessment is completed and an early warning probability matrix is ​​generated. Each element of the early warning probability matrix represents the early warning probability of the corresponding level and risk level. The early warning probability matrix is ​​correlated with the real-time operating status data of the GIS equipment, and the actual risk level of each area is comprehensively judged in combination with the temperature rise rate and temperature rise duration of the equipment, and the corresponding graded early warning signal is output. In this way, the early warning signal can dynamically reflect the operating status and potential risks of the equipment.

[0022] In the embodiment of the present application, by constructing a three-layer network topology structure, systematic management of temperature data of the SF6 gas chamber layer, high-voltage conductor layer and outer shell sealing layer is achieved, and the completeness and accuracy of the temperature field distribution feature extraction are improved; the combination of intra-layer graph convolution and inter-layer graph convolution is adopted to effectively capture the temperature correlation between different layers, and enhance the ability to extract the spatial distribution characteristics of the temperature field; combined with the timing fusion mechanism of the state update gate and the parameter association gate, the dynamic correlation analysis of temperature data with multiple parameters such as SF6 gas pressure and ambient humidity is realized; through the partitioned reinforcement learning control strategy, differentiated control parameters are adopted for different areas, which improves the accuracy of temperature control; a multi-level compensation control mechanism is adopted, which takes into account the coupling effects of gas pressure, load current and ambient humidity, and improves the accuracy of temperature correction; based on the risk warning mechanism of a multi-layer fully connected neural network, accurate assessment and graded warning of temperature rise risks at different levels are realized.

[0023] In a specific embodiment, the process of executing step 100 may specifically include the following steps: Pair the data collected by the temperature sensor at the internal monitoring point of the SF6 gas chamber layer with the spatial position coordinates of the sensor to obtain a set of sensor nodes in the gas chamber layer; Pair the temperature sensor data collected by the circuit breaker contacts, disconnectors and bushing joints in the high-voltage conductor layer with the spatial position coordinates of the sensors to obtain a set of conductor layer sensor nodes; Pair the data collected by the temperature sensors of the equipment shell and the supporting structure in the shell closed layer with the spatial position coordinates of the sensors to obtain a set of sensor nodes in the shell layer; The space coordinate system of the air chamber layer sensor node set, the conductor layer sensor node set and the shell layer sensor node set are unified and the temperature data is normalized to obtain a standardized sensor node matrix; Based on the temperature data in the standardized sensor node matrix, the node feature matrix is ​​constructed and the hierarchical attribute is marked to obtain the node feature matrix containing hierarchical information; The three-dimensional Euclidean distance between any two sensor nodes is calculated according to the spatial coordinate information in the standardized sensor node matrix to obtain the node distance matrix, and the distance value in the node distance matrix is ​​compared with the preset intra-layer connection threshold and inter-layer connection threshold to generate a binary adjacency matrix. The binary adjacency matrix is ​​associated with the node level information, and the intra-layer connection edges and inter-layer connection edges are marked in the binary adjacency matrix to obtain the adjacency matrix.

[0024] Specifically, the monitoring points in the SF6 gas chamber layer are composed of multiple temperature sensors arranged inside the gas chamber. These sensors collect temperature data at different positions of the gas chamber. ,in Indicates At the same time, the spatial position of each sensor is expressed by three-dimensional coordinates It is defined as the actual position of the sensor in three-dimensional space. By pairing the temperature data collected by each sensor with its corresponding spatial position coordinates, the following set is formed: ; in, is the set of sensor nodes at the air chamber layer, Represents the total number of sensors in the air chamber layer. Similarly, in the high-voltage conductor layer, sensors are arranged at key locations such as circuit breaker contacts, disconnectors, and bushing joints to collect temperature data at these locations. and the corresponding spatial position coordinates The sensor node set of the conductor layer is defined as: ; in, represents the set of sensor nodes of the conductor layer, is the number of sensors in the conductor layer. For the enclosure layer, the sensors are arranged to collect temperature data of the equipment enclosure and supporting structure. and spatial position coordinates , whose node set is defined as: ; in, represents the set of sensor nodes in the conductor layer, is the number of sensors in the outer shell layer. In order to unify the spatial coordinate system of each layer and ensure that the sensor nodes in different layers have a consistent reference system, , and Perform a unified coordinate transformation, assuming the reference coordinate system The transformed coordinates are expressed as: ; in, is the coordinate in the unified coordinate system, is the rotation matrix, is the translation vector. By unifying the coordinate system, a standardized sensor node matrix is ​​generated : ; Each row represents a sensor node, whose features include temperature data, spatial coordinates, and hierarchical attributes. , calculate the three-dimensional Euclidean distance between any two sensor nodes : ; in, and Represents the index of two nodes respectively. By performing the above calculations on all node pairs, a node distance matrix is ​​generated. , whose elements Indicates and The distance between nodes. According to the preset intra-layer connection threshold and inter-layer connection threshold ,right The distance value in is used to judge and generate a binary adjacency matrix ; in, Elements in Representation Node and Whether to connect. Perform an association operation on the binary adjacency matrix A and the hierarchical information in the node feature matrix, mark the intra-layer connection edges and inter-layer connection edges, and generate an adjacency matrix containing hierarchical attributes.

[0025] In a specific embodiment, the process of executing step 200 may specifically include the following steps: The temperature data of each sensor node in the node feature matrix is ​​normalized and hierarchically labeled, and the sensor nodes are divided into an air chamber layer node group, a conductor layer node group and an outer shell layer node group according to the hierarchical information of the sensor nodes, and a hierarchical grouping matrix is ​​obtained; The connection relationship in the adjacency matrix is ​​distinguished, and the connection edges are divided into the intra-layer connection edge set and the inter-layer connection edge set according to the hierarchical labels between the sensor nodes, and the connection edge classification matrix is ​​obtained; Perform matrix multiplication on the hierarchical grouping matrix and the first convolution kernel matrix to obtain the operation result, and perform ReLU activation function on the operation result to obtain the initial feature matrix within the layer; The attention coefficient is calculated for each node feature vector in the initial feature matrix within the layer, and the node weight coefficient is generated according to the node temperature gradient and topological distance to obtain the attention weight matrix; The attention weight matrix is ​​weightedly combined with the initial feature matrix within the layer, and a graph convolution operation is performed with the set of connected edges within the layer to obtain the spatial feature matrix within the layer; The feature vectors between nodes at different levels in the intra-layer spatial feature matrix are convolved with the second convolution kernel matrix, and the features are transferred with the inter-layer connection edge set to obtain the inter-layer correlation feature matrix; The intra-layer spatial feature matrix and the inter-layer correlation feature matrix are concatenated and feature compressed through the fully connected layer to obtain the temperature spatial distribution feature vector; The temperature spatial distribution feature vector is fused with the SF6 gas pressure data and the ambient humidity data in time series to obtain the temperature dynamic change time series feature vector.

[0026] Specifically, the node feature matrix represents the main features of each sensor node, including temperature data, spatial coordinates, and the level information to which it belongs, in the form of: ; in, is the node feature matrix, Indicates The temperature data of each node, is the three-dimensional space coordinate of the node, is the node level label (such as air chamber layer, conductor layer, shell layer), Indicates the total number of nodes. For temperature data Normalization is performed to eliminate the differences in the numerical range. The normalization formula is: ; in, is the normalized temperature value, and are the minimum and maximum values ​​of all node temperatures respectively. Map the temperature values ​​to the interval [0,1]. Label the nodes according to their level , the nodes are divided into three groups, namely the air chamber layer node group, the conductor layer node group and the shell layer node group, forming a hierarchical grouping matrix , which is of the form: ; in, Each contains the node features belonging to the corresponding level. At the same time, the adjacency matrix For any two nodes and , if they belong to the same layer and meet the connection conditions (such as the Euclidean distance is less than the intra-layer threshold), their connection belongs to the intra-layer connection; otherwise, if they belong to different layers and meet the cross-layer connection conditions, their connection belongs to the inter-layer connection. The intra-layer connection set and the inter-layer connection set are defined as: ; ; Grouping the levels matrix With the first convolution kernel matrix Perform matrix multiplication to extract the initial feature transformation results of each layer: ; in, is the convolution kernel matrix, used for feature transformation, is the initial feature matrix within the layer. For each node feature vector in , calculate its attention coefficient to highlight the importance of key nodes. Calculated based on node temperature gradient and topological distance: ; in, is the learned attention vector, and Is a node and nodes The feature vector of , || represents the vector concatenation operation. Constructing the attention weight matrix The attention weight matrix is ​​weightedly combined with the initial feature matrix within the layer, and the graph convolution operation is performed on the set of connected edges within the layer to generate the spatial feature matrix within the layer: ; in, is the convolution kernel weight, is the activation function. The node features in the spatial feature matrix within the layer are combined with the second convolution kernel matrix Convolution is performed, and features are transferred with the inter-layer connection edge set to generate the inter-layer correlation feature matrix: ; Concatenate the intra-layer spatial feature matrix and the inter-layer correlation feature matrix: ; The concatenated features are compressed through the fully connected layer to generate the final temperature spatial distribution feature vector: ; in, and is the weight and bias of the fully connected layer. The temperature spatial distribution feature vector is combined with the SF6 gas pressure data and ambient humidity data Perform temporal fusion and use the temporal attention mechanism to extract dynamic features: ; Through this process, a temperature dynamics time series feature vector is generated which fully describes the temperature distribution and dynamics.

[0027] In a specific embodiment, the execution step performs time series fusion of the temperature spatial distribution feature vector with the SF6 gas pressure data and the ambient humidity data to obtain the temperature dynamic change time series feature vector may specifically include the following steps: The temperature spatial distribution characteristic vector and the SF6 gas pressure data are numerically normalized to obtain a temperature and pressure time series matrix, and the temperature and pressure time series matrix is ​​aligned with the ambient humidity data to obtain a multi-parameter time series data matrix; The historical information ratio of the multi-parameter time series data matrix is ​​calculated through the state update gate to obtain the state update matrix; The parameter influence degree of the multi-parameter time series data matrix is ​​calculated through the parameter association gate, and the parameter association matrix is ​​generated according to the coupling relationship between SF6 gas pressure and temperature and the correlation between ambient humidity and temperature; Perform matrix multiplication on the state update matrix and the parameter association matrix, and perform feature selection through the output gate to obtain a fused feature matrix; Perform dynamic time attention calculation on the time series data in the fusion feature matrix, and generate a time series weighted matrix according to the equipment start / stop status and load changes; The fusion feature matrix and the timing weighted matrix are weighted superimposed to obtain the initial dynamic feature vector, and the timing feature of the initial dynamic feature vector is extracted through a gated cyclic unit to obtain the temperature dynamic change timing feature vector.

[0028] Specifically, the temperature spatial distribution eigenvector and SF6 gas pressure data They represent the spatial distribution characteristics of the internal temperature of the equipment and the changing trend of the SF6 gas pressure. Assuming the temperature characteristics and gas pressure data is the vector corresponding to the time, respectively The parameter changes are recorded at time steps. In order to eliminate the influence of different dimensions on data analysis, the two vectors are numerically normalized. The normalization process uses the following formula: ; in, and They are Normalized temperature and pressure values ​​for time steps, and Represent the minimum and maximum values ​​of temperature and pressure respectively. After normalization, the two normalized vectors are combined to form the temperature and pressure time series matrix: ; The ambient humidity data Data alignment is performed with the temperature and pressure time series matrix to form a multi-parameter time series data matrix. Data alignment ensures that The synchronization of humidity data in the time dimension. The aligned multi-parameter time series data matrix is ​​expressed as: ; Multi-parameter time series data matrix through state update gate Calculate the historical information ratio to generate the state update matrix The state update gate controls the impact of the historical state on the current time step through a nonlinear transformation, and its calculation formula is: ; in, is the weight matrix of the state update gate, is the bias vector, is the activation function (such as Sigmoid). Each element of represents the influence ratio of historical information on the current moment parameter. At the same time, the multi-parameter time series data matrix Calculate the parameter influence. Generate a parameter correlation matrix based on the thermodynamic coupling relationship between SF6 gas pressure and temperature, and the correlation between ambient humidity and temperature condensation risk. The calculation formula is: ; in, and are the weights and biases of the parameter association gates. The softmax function is used to map the results into probability distributions and quantify the contribution of different parameters to the temperature rise. Matrix associated with parameters Perform matrix multiplication and select features on the results through the output gate to generate a fused feature matrix : ; Where · represents element-by-element multiplication. Dynamic time attention calculation is performed on the time series data in the fusion feature matrix to capture the dynamic impact of the equipment start-stop state and load change on temperature rise. Dynamic time attention weight matrix The calculation formula is: ; in, and is the weight and bias of the temporal attention module. The initial dynamic feature vector is obtained by weighted superposition of the dynamic temporal attention weight matrix and the fusion feature matrix: ; The initial dynamic feature vector is input into the gated recurrent unit (GRU) to extract the timing features and obtain the final temperature dynamic change timing feature vector. The update formula of GRU is: ; in, It is the update gate. is a candidate hidden state, is the hidden state at the previous moment, Represents element-wise multiplication.

[0029] In a specific embodiment, the process of executing step 300 may specifically include the following steps: According to the temperature spatial distribution feature vector and the temperature dynamic change time series feature vector, feature fusion is performed in the spatial and temporal dimensions to obtain the spatiotemporal feature matrix, and the spatiotemporal feature matrix is ​​subjected to hierarchical dimensionality reduction processing to obtain the regional classification feature vector; The regional classification feature vector is input into the classification network, the core heating parts of the circuit breaker contacts and disconnectors are divided into key areas, the conductor connection and bushing joint positions are divided into transition areas, and the housing and support structure parts are divided into ordinary areas, and the regional category label matrix is ​​obtained; The reinforcement learning training environment is constructed for the regional category label matrix, and the state space and action space are set according to the temperature control requirements of each partition to obtain the reinforcement learning environment parameters; The reinforcement learning environment parameters are input into the reinforcement learning agent, and the temperature control actions of each partition are explored and sampled to obtain the action sample set. The reward function is constructed according to the temperature control accuracy, response speed, temperature uniformity and long-term stability to obtain the partition reward value. The policy network is trained on the action sample set and partition reward value, and the policy gradient algorithm is used to update the network parameters to obtain the policy network model of each partition; The strategy network model is input into the temperature control accuracy evaluator, the response speed evaluator and the stability evaluator respectively to obtain the control performance evaluation results, and the strategy network parameters are fine-tuned according to the control performance evaluation results to obtain the optimized strategy network model; The temperature control action sequence is generated according to the optimized strategy network model, and the temperature control action sequence is converted into adaptive control parameters of each partition.

[0030] Specifically, the temperature spatial distribution eigenvector And the temperature dynamic change time series characteristic vector They represent the static distribution characteristics and time variation of the temperature in the device respectively. Assume is a -dimensional vector The matrix of is the number of time steps, is the dimension of spatial features. In order to fuse these two features in spatial and temporal dimensions, and Perform feature alignment and splicing to generate a spatiotemporal feature matrix : ; Among them, concat represents the feature concatenation operation, and the result is a The first row is the spatial feature, and the remaining The rows are time series features. The spatiotemporal feature matrix is ​​subjected to hierarchical dimensionality reduction to reduce redundant information and extract key features. Dimensionality reduction is achieved through linear transformation, and its formula is: ; in, is the dimension reduction weight matrix, is the bias vector, is the activation function (such as ReLU). The output is a low-dimensional matrix that contains the key features after spatiotemporal fusion. The reduced feature matrix is ​​input into the classification network to classify the regions according to the structural characteristics of the equipment and the temperature distribution law. Assume that the output of the classification network is the regional classification feature vector , the formula is: ; in, and are the weights and biases of the classification network, and the softmax function maps the output to a probability distribution. Based on the classification results, the core heating parts such as circuit breaker contacts and disconnectors in the equipment are divided into key areas, the conductor connections, bushing joints and other locations are divided into transition areas, and the housing and supporting structure are divided into ordinary areas to form a regional category label matrix. After the regional classification is completed, the regional category label matrix Construct a reinforcement learning training environment. The core of reinforcement learning is to set a reasonable state space. and action space The state space contains parameters such as temperature, pressure, humidity, etc. of each partition, and the action space is defined as a set of control actions such as adjusting the radiator switch and gas flow. It is expressed as: ; in, is a reward function used to evaluate the quality of an action. The reinforcement learning agent interacts with the environment to explore and sample the temperature control actions of each partition and generate an action sample set D. At the same time, a reward function is constructed based on temperature control accuracy, response speed, temperature uniformity, and long-term stability: ; in, It is The reward value of the step, They represent the control accuracy, response speed, temperature uniformity and stability scores respectively. is the weight. Based on the action sample set and reward value , use the policy gradient algorithm to train the policy network and update the network parameters : ; in, is the learning rate, is the policy network in state Next select action After the training is completed, the policy network model is input into the temperature control accuracy evaluator, response speed evaluator and stability evaluator to evaluate the control performance. Assume that the evaluation result is , by further fine-tuning the policy network parameters, an optimized policy network model is generated. The temperature control action sequence is generated based on the optimized policy network model: ; in, It is an optimized strategy network model. The action sequence is converted into adaptive control parameters of each partition to accurately control the temperature rise of GIS equipment.

[0031] In a specific embodiment, the process of executing step 400 may specifically include the following steps: The adaptive control parameters are input into the multi-level compensation controller, and the control parameters are hierarchically processed according to the hierarchical attributes of the controller to obtain a hierarchical control parameter set; The gas pressure and temperature data are input into the SF6 gas pressure-temperature coupling model, and a correlation analysis is performed based on the coupling relationship between gas pressure and temperature to obtain the gas chamber layer temperature compensation coefficient; The load current data of the high-voltage conductor layer is input into the load current heat accumulation model, and the heat accumulation calculation is performed according to the relationship between the current intensity and the conductor temperature rise to obtain the conductor layer temperature compensation coefficient; The humidity distribution of the data collected by the environmental humidity sensor is calculated to obtain the humidity distribution calculation result, and the humidity distribution calculation result is input into the environmental humidity condensation model, and the condensation risk analysis is performed according to the relationship between humidity and surface temperature to obtain the shell temperature compensation coefficient; Parameter matching is performed according to the correspondence between the hierarchical control parameter set and the temperature compensation coefficient of each layer to obtain a compensation control matrix; The parameters in the compensation control matrix are subjected to stability constraint processing, and the parameter correction range is determined according to the temperature change limit of each layer to obtain the constraint correction matrix; The constraint correction matrix is ​​input into the temperature compensation calculation module, and the temperature correction amount is calculated according to the compensation coefficient of each level to obtain the temperature correction parameters of each level.

[0032] Specifically, the adaptive control parameters It is a comprehensive set of adjustment parameters obtained through previous analysis, including the regulation requirements of the air chamber layer, high-voltage conductor layer and shell sealing layer, which is expressed in the form of: ; in, The control parameter sets corresponding to the air chamber layer, conductor layer and shell layer respectively. In the multi-level compensation controller, according to the hierarchical attributes of the controller, the input control parameters are separated and grouped according to the hierarchical attributes to generate a hierarchical control parameter set. The control parameter vector of each level is defined as: ; Each element represents a control requirement at a specific level. For example, the parameters of the gas chamber layer include the gas flow adjustment coefficient and the valve adjustment coefficient, the parameters of the high-voltage conductor layer involve the current distribution ratio, and the shell layer includes the radiator opening parameters and the shell thermal conductivity adjustment coefficient. For the gas chamber layer, the SF6 gas pressure-temperature coupling model is used to analyze the correlation between gas pressure and temperature. Gas pressure data and temperature data Input into the coupled model, the basic formula of the model is: ; in, represents the temperature change of the air chamber layer, is the sensitivity coefficient of gas pressure to temperature change, is the actual gas pressure, is the reference gas pressure value. Converted to air chamber layer temperature compensation coefficient , used to correct the temperature rise behavior of the air chamber layer. For the high voltage conductor layer, the load current data Input the load current heat accumulation model to calculate the effect of current intensity on the temperature rise of the conductor layer. The heat accumulation formula of the model is: ; in, is the accumulated heat, is the resistance of the conductor, is the load current, It's time. Based on The calculation result of the conductor layer temperature change is expressed as: ; in, is the heat capacity of the conductor. The result is to generate the conductor layer temperature compensation coefficient For the outer shell, the ambient humidity data Input the humidity distribution calculation module to calculate the distribution characteristics of humidity on the shell surface. The results of the humidity distribution calculation are combined with the ambient humidity condensation model. The basic formula for condensation risk analysis is: ; in, is the condensation risk factor, is the ambient humidity, is the dew point temperature, is the shell surface temperature. The shell temperature compensation coefficient is generated by the calculation result of the condensation risk factor The temperature compensation coefficients of the above three levels are combined, and the compensation control matrix is ​​generated according to the corresponding relationship between the hierarchical control parameter set and the temperature compensation coefficient: ; Among them, the compensation control matrix The diagonal elements of represent the temperature compensation coefficients of each layer. In order to ensure the stability of the compensation parameters, The parameters in the are processed with stability constraints. Based on the temperature change limits of each level, the parameter correction range is determined: ; in, and Respectively represent the minimum and maximum allowable values ​​of the compensation coefficient, is the constraint correction matrix. Enter the temperature compensation calculation module and calculate the temperature correction amount according to the temperature compensation coefficient of each level: ; in, Represents the final temperature correction parameter of each layer.

[0033] In a specific embodiment, the temperature rise monitoring method for high-pressure fully enclosed vacuum environmentally friendly GIS equipment further includes the following steps: The temperature correction parameters, the temperature spatial distribution feature vector and the temperature dynamic change time series feature vector are subjected to feature combination and data standardization to obtain a standardized feature matrix; The standardized characteristic matrix is ​​used to detect the temperature anomaly of the SF6 gas chamber layer, and the gas chamber layer risk level matrix is ​​generated based on the comparative analysis of the gas temperature deviation value and the insulation breakdown threshold; Conduct temperature rise detection of high-voltage conductor layer on the standardized characteristic matrix, and generate conductor layer risk level matrix based on temperature gradient calculation of circuit breaker contacts, disconnectors and bushing joints; Conduct temperature uniformity test of the enclosure layer on the standardized characteristic matrix, and generate the enclosure layer risk level matrix based on the calculation of surface temperature distribution and ambient temperature difference; The air chamber layer risk level matrix, the conductor layer risk level matrix and the shell layer risk level matrix are input into a multi-layer fully connected neural network, and processed by linear transformation and nonlinear activation function to obtain a hierarchical risk feature vector; Perform Softmax classification on the hierarchical risk feature vectors, perform multi-level warning calculations based on the preset temperature rise risk thresholds of the air chamber layer, conductor layer, and shell layer, and obtain a warning probability matrix; The warning data in the warning probability matrix are correlated with the equipment operation status data, and the corresponding graded warning signals are output according to the temperature rise rate and duration.

[0034] Specifically, the temperature correction parameter, the temperature spatial distribution feature vector, and the temperature dynamic change time series feature vector are combined and data normalized to generate a standardized feature matrix. Assume that the temperature correction parameter is , the temperature spatial distribution characteristic vector is , the time series characteristic vector of temperature dynamic change is , is a The matrix of is the number of time steps, is the feature dimension. and and Perform feature concatenation to form a joint feature matrix: ; Among them, concat represents the feature concatenation operation. To ensure the comparability of different data sources, Normalization is performed to map each eigenvalue to the [0,1] interval. The normalization formula is: ; in, is the normalized eigenvalue, is the original value, and They are The minimum and maximum values ​​of the column features are input into the temperature anomaly detection module to analyze the temperature of the SF6 gas chamber layer. Assume that the SF6 gas chamber layer temperature data is , by calculating the temperature deviation value of each monitoring point: ; in, It is The temperature deviation value of each node, is the reference temperature threshold. For comparison: ; Generate a risk level matrix for air chamber layers , each element reflects the risk level of the node. For the high-voltage conductor layer, the temperature rise of the circuit breaker contacts, disconnectors and bushing joints is analyzed. Assume that the conductor layer temperature data is , calculate the temperature gradient: ; in, It is The temperature gradient of each node. The risk level is divided according to the gradient size: ; Generate Conductor Layer Risk Level Matrix The temperature uniformity of the enclosure is calculated by analyzing the surface temperature distribution and the ambient temperature difference. Assuming the enclosure surface temperature is , the ambient temperature is , define the temperature difference as: ; Calculate the standard deviation of the temperature difference Indicates uniformity and according to Risk level classification: ; Generate Shell Risk Level Matrix . The three risk level matrices , , Input into a multi-layer fully connected neural network, and use linear transformation and nonlinear activation function to extract hierarchical risk features: ; in, and are the weights and biases of the neural network, is the activation function. Softmax classification is performed on the hierarchical risk feature vector to calculate the probability of each risk level: ; Generate warning probability matrix Will Conduct correlation analysis with equipment operating status data (such as temperature rise rate and duration) and output graded warning signals.

[0035] The above describes the temperature rise monitoring method of the high-voltage fully enclosed vacuum environmentally friendly GIS equipment in the embodiment of the present application. The following describes the temperature rise monitoring system 10 of the high-voltage fully enclosed vacuum environmentally friendly GIS equipment in the embodiment of the present application. Figure 2 In the embodiment of the present application, a temperature rise monitoring system 10 for high-pressure fully enclosed vacuum environmentally friendly GIS equipment includes: The acquisition module 11 is used to collect temperature data through a temperature sensor, and construct a node feature matrix and an adjacency matrix according to the spatial position relationship between the SF6 gas chamber layer, the high-voltage conductor layer and the shell sealing layer; The operation module 12 is used to perform intra-layer graph convolution operations and inter-layer graph convolution operations on the node feature matrix and the adjacency matrix to obtain the temperature spatial distribution feature vector, and perform time series fusion to obtain the temperature dynamic change time series feature vector; A partitioning module 13 is used to partition the monitoring area according to the temperature spatial distribution feature vector and the temperature dynamic change time series feature vector to obtain the adaptive control parameters of each partition; The compensation module 14 is used to input the adaptive control parameters into the multi-level compensation controller to calculate the temperature compensation coefficient and obtain the temperature correction parameters of each level.

[0036] Through the coordinated cooperation of the above-mentioned components and by constructing a three-layer network topology structure, the systematic management of the temperature data of the SF6 gas chamber layer, the high-voltage conductor layer and the shell sealing layer is realized, and the completeness and accuracy of the temperature field distribution feature extraction are improved; the combination of intra-layer graph convolution and inter-layer graph convolution is adopted to effectively capture the temperature correlation between different layers and enhance the ability to extract the spatial distribution characteristics of the temperature field; combined with the timing fusion mechanism of the state update gate and the parameter association gate, the dynamic correlation analysis of temperature data with multiple parameters such as SF6 gas pressure and ambient humidity is realized; through the partition reinforcement learning control strategy, differentiated control parameters are adopted for different areas to improve the accuracy of temperature control; the multi-level compensation control mechanism is adopted to consider the coupling effect of gas pressure, load current and ambient humidity, and improve the accuracy of temperature correction; based on the risk warning mechanism of the multi-layer fully connected neural network, the accurate assessment and graded warning of temperature rise risks at different levels are realized.

[0037] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the above-described systems, systems and units can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0038] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application is essentially or the part that contributes to the prior art or all or part of the technical solution can be embodied in the form of a software product. The computer software product is stored in a storage medium, including several instructions to enable an electronic device (which can be a personal computer, server, or network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present application. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM), random access memory (RAM), disk or optical disk, etc., various media that can store program codes.

[0039] As described above, the above embodiments are only used to illustrate the technical solutions of the present application, rather than to limit them. Although the present application has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present application.

Claims

1. A temperature rise monitoring method for high-voltage fully enclosed vacuum environmentally friendly GIS equipment, characterized in that: The method comprises: The temperature data is collected through the temperature sensor, and the node feature matrix and adjacency matrix are constructed according to the spatial position relationship between the SF6 gas chamber layer, the high-voltage conductor layer and the shell sealing layer; Performing intra-layer graph convolution operations and inter-layer graph convolution operations on the node feature matrix and the adjacency matrix to obtain a temperature spatial distribution feature vector, and performing time series fusion to obtain a temperature dynamic change time series feature vector; Partitioning the monitoring area according to the temperature spatial distribution feature vector and the temperature dynamic change time series feature vector to obtain adaptive control parameters for each partition; The adaptive control parameters are input into the multi-level compensation controller to calculate the temperature compensation coefficient, so as to obtain the temperature correction parameters of each level.

2. The temperature rise monitoring method of high-voltage fully enclosed vacuum environmentally friendly GIS equipment according to claim 1 is characterized in that: The temperature data is collected by the temperature sensor, and a node feature matrix and an adjacency matrix are constructed according to the spatial position relationship between the SF6 gas chamber layer, the high-voltage conductor layer and the shell sealing layer, including: Pair the data collected by the temperature sensor at the internal monitoring point of the SF6 gas chamber layer with the spatial position coordinates of the sensor to obtain a set of sensor nodes in the gas chamber layer; Pair the temperature sensor data collected by the circuit breaker contacts, disconnectors and bushing joints in the high-voltage conductor layer with the spatial position coordinates of the sensors to obtain a set of conductor layer sensor nodes; Pair the data collected by the temperature sensors of the equipment shell and the supporting structure in the shell closed layer with the spatial position coordinates of the sensors to obtain a set of sensor nodes in the shell layer; Performing spatial coordinate system unification and temperature data normalization processing on the air chamber layer sensor node set, the conductor layer sensor node set and the shell layer sensor node set to obtain a standardized sensor node matrix; Based on the temperature data in the standardized sensor node matrix, a node feature matrix is ​​constructed and hierarchical attribute marking is performed to obtain a node feature matrix containing hierarchical information; Calculating the three-dimensional Euclidean distance between any two sensor nodes according to the spatial coordinate information in the standardized sensor node matrix to obtain a node distance matrix, and comparing the distance values ​​in the node distance matrix with the preset intra-layer connection threshold and inter-layer connection threshold to generate a binary adjacency matrix; The binary adjacency matrix is ​​associated with the node level information, and the intra-layer connection edges and inter-layer connection edges are marked in the binary adjacency matrix to obtain an adjacency matrix.

3. The temperature rise monitoring method of high-voltage fully enclosed vacuum environmentally friendly GIS equipment according to claim 1 is characterized in that: The intra-layer graph convolution operation and the inter-layer graph convolution operation are performed on the node feature matrix and the adjacency matrix to obtain the temperature spatial distribution feature vector, and the time series fusion is performed to obtain the temperature dynamic change time series feature vector, including: Normalizing and hierarchically marking the temperature data of each sensor node in the node feature matrix, dividing the sensor nodes into an air chamber layer node group, a conductor layer node group and an outer shell layer node group according to the hierarchical information to which the sensor nodes belong, and obtaining a hierarchical grouping matrix; The connection relationship in the adjacency matrix is ​​distinguished, and the connection edges are divided into an intra-layer connection edge set and an inter-layer connection edge set according to the layer labels between the sensor nodes, so as to obtain a connection edge classification matrix; Performing a matrix multiplication operation on the hierarchical grouping matrix and the first convolution kernel matrix to obtain an operation result, and performing ReLU activation function processing on the operation result to obtain an initial feature matrix within the layer; Calculating the attention coefficient for each node feature vector in the initial feature matrix in the layer, generating a node weight coefficient according to the node temperature gradient and the topological distance, and obtaining an attention weight matrix; The attention weight matrix is ​​weightedly combined with the initial feature matrix in the layer, and a graph convolution operation is performed with the set of connection edges in the layer to obtain a spatial feature matrix in the layer; Performing convolution operation on feature vectors between nodes at different levels in the intra-layer spatial feature matrix and a second convolution kernel matrix, and performing feature transfer with the inter-layer connection edge set to obtain an inter-layer correlation feature matrix; Performing feature concatenation on the intra-layer spatial feature matrix and the inter-layer correlation feature matrix, and performing feature compression through a fully connected layer to obtain a temperature spatial distribution feature vector; The temperature spatial distribution characteristic vector is time-series fused with the SF6 gas pressure data and the ambient humidity data to obtain a temperature dynamic change time-series characteristic vector.

4. The temperature rise monitoring method of high-voltage fully enclosed vacuum environmentally friendly GIS equipment according to claim 3 is characterized in that: The time series fusion of the temperature spatial distribution feature vector with the SF6 gas pressure data and the ambient humidity data to obtain the temperature dynamic change time series feature vector includes: The temperature spatial distribution characteristic vector and the SF6 gas pressure data are numerically normalized to obtain a temperature and pressure time series matrix, and the temperature and pressure time series matrix is ​​data aligned with the ambient humidity data to obtain a multi-parameter time series data matrix; Perform historical information ratio calculation on the multi-parameter time series data matrix through a state update gate to obtain a state update matrix; Calculating the parameter influence of the multi-parameter time series data matrix through a parameter association gate, and generating a parameter association matrix according to the coupling relationship between SF6 gas pressure and temperature and the degree of association between ambient humidity and temperature; Performing matrix multiplication operation on the state update matrix and the parameter association matrix, and performing feature selection through an output gate to obtain a fused feature matrix; Performing dynamic time attention calculation on the time series data in the fusion feature matrix, and generating a time series weighted matrix according to the start / stop state of the device and the load change; The fusion feature matrix and the timing weighted matrix are weighted superimposed to obtain an initial dynamic feature vector, and the timing feature of the initial dynamic feature vector is extracted through a gated cycle unit to obtain a temperature dynamic change timing feature vector.

5. The temperature rise monitoring method of high-voltage fully enclosed vacuum environmentally friendly GIS equipment according to claim 1 is characterized in that: The partitioning of the monitoring area according to the temperature spatial distribution feature vector and the temperature dynamic change time series feature vector to obtain the adaptive control parameters of each partition includes: Performing feature fusion in spatial and temporal dimensions according to the temperature spatial distribution feature vector and the temperature dynamic change time series feature vector to obtain a spatiotemporal feature matrix, and performing hierarchical dimensionality reduction processing on the spatiotemporal feature matrix to obtain a regional classification feature vector; The regional classification feature vector is input into the classification network, the core heating parts of the circuit breaker contacts and disconnectors are divided into key areas, the conductor connection and the bushing joint are divided into transition areas, and the housing and the supporting structure are divided into ordinary areas, so as to obtain a regional category label matrix; Constructing a reinforcement learning training environment for the regional category label matrix, setting a state space and an action space according to the temperature control requirements of each partition, and obtaining reinforcement learning environment parameters; Input the reinforcement learning environment parameters into the reinforcement learning agent, perform exploration sampling on the temperature control action of each partition to obtain an action sample set, and construct a reward function according to the temperature control accuracy, response speed, temperature uniformity and long-term stability to obtain a partition reward value; Performing policy network training on the action sample set and the partition reward value, and updating network parameters using a policy gradient algorithm to obtain a policy network model for each partition; The strategy network model is respectively input into a temperature control accuracy evaluator, a response speed evaluator and a stability evaluator to obtain a control performance evaluation result, and the strategy network parameters are fine-tuned according to the control performance evaluation result to obtain an optimized strategy network model; A temperature control action sequence is generated according to the optimized policy network model, and the temperature control action sequence is converted into adaptive control parameters of each partition.

6. The temperature rise monitoring method of high-voltage fully enclosed vacuum environmentally friendly GIS equipment according to claim 1 is characterized in that: The step of inputting the adaptive control parameters into the multi-level compensation controller to calculate the temperature compensation coefficient to obtain the temperature correction parameters of each level includes: Inputting the adaptive control parameters into a multi-level compensation controller, performing hierarchical processing on the control parameters according to hierarchical attributes of the controller, and obtaining a hierarchical control parameter set; The gas pressure and temperature data are input into the SF6 gas pressure-temperature coupling model, and a correlation analysis is performed based on the coupling relationship between gas pressure and temperature to obtain the gas chamber layer temperature compensation coefficient; The load current data of the high-voltage conductor layer is input into the load current heat accumulation model, and the heat accumulation calculation is performed according to the relationship between the current intensity and the conductor temperature rise to obtain the conductor layer temperature compensation coefficient; Performing humidity distribution calculation on the data collected by the environmental humidity sensor to obtain a humidity distribution calculation result, and inputting the humidity distribution calculation result into the environmental humidity condensation model, performing condensation risk analysis based on the relationship between humidity and surface temperature, and obtaining an outer shell temperature compensation coefficient; Performing parameter matching according to the correspondence between the hierarchical control parameter set and the temperature compensation coefficients of each layer to obtain a compensation control matrix; Performing stability constraint processing on the parameters in the compensation control matrix, determining the parameter correction range according to the temperature change limit of each level, and obtaining a constraint correction matrix; The constraint correction matrix is ​​input into the temperature compensation calculation module, and the temperature correction amount is calculated according to the compensation coefficient of each level to obtain the temperature correction parameters of each level.

7. The temperature rise monitoring method of high-voltage fully enclosed vacuum environmentally friendly GIS equipment according to claim 1 is characterized in that: The method of calculating the temperature rise risk probability at different levels according to the temperature correction parameter, the temperature spatial distribution feature vector and the temperature dynamic change time series feature vector, and outputting a graded warning signal includes: Performing feature combination and data standardization processing on the temperature correction parameter, the temperature spatial distribution feature vector and the temperature dynamic change time series feature vector to obtain a standardized feature matrix; Performing temperature anomaly detection on the SF6 gas chamber layer on the standardized characteristic matrix, and generating a gas chamber layer risk level matrix based on comparative analysis of the gas temperature deviation value and the insulation breakdown threshold; Performing temperature rise detection on the high-voltage conductor layer of the standardized characteristic matrix, and generating a conductor layer risk level matrix according to temperature gradient calculation of the circuit breaker contacts, disconnectors and bushing joints; Performing temperature uniformity detection on the enclosure sealing layer of the standardized characteristic matrix, and generating an enclosure layer risk level matrix based on calculation of surface temperature distribution and ambient temperature difference; Inputting the air chamber layer risk level matrix, the conductor layer risk level matrix and the shell layer risk level matrix into a multi-layer fully connected neural network, and processing them through linear transformation and nonlinear activation function to obtain a hierarchical risk feature vector; Performing Softmax classification on the hierarchical risk feature vector, performing multi-level warning calculation according to preset temperature rise risk thresholds of the air chamber layer, the conductor layer and the outer shell layer, and obtaining a warning probability matrix; The warning data in the warning probability matrix are correlated with the equipment operation status data and analyzed, and corresponding graded warning signals are output according to the temperature rise rate and duration.

8. A temperature rise monitoring system for high-voltage, fully enclosed, vacuum, environmentally friendly GIS equipment, characterized in that: Used to implement the temperature rise monitoring method of high-voltage fully enclosed vacuum environmentally friendly GIS equipment as claimed in claim 1, the temperature rise monitoring system of the high-voltage fully enclosed vacuum environmentally friendly GIS equipment comprises: The acquisition module is used to collect temperature data through a temperature sensor and construct a node feature matrix and an adjacency matrix according to the spatial position relationship between the SF6 gas chamber layer, the high-voltage conductor layer and the shell sealing layer; An operation module, used for performing intra-layer graph convolution operation and inter-layer graph convolution operation on the node feature matrix and the adjacency matrix to obtain a temperature spatial distribution feature vector, and performing time series fusion to obtain a temperature dynamic change time series feature vector; A partitioning module, used to partition the monitoring area according to the temperature spatial distribution characteristic vector and the temperature dynamic change time series characteristic vector, and obtain adaptive control parameters of each partition; The compensation module is used to input the adaptive control parameters into the multi-level compensation controller to calculate the temperature compensation coefficient and obtain the temperature correction parameters of each level.

Images