Converter station energy efficiency prediction system based on deep learning

Abstract

The invention discloses a converter station energy efficiency prediction system based on deep learning, and the system comprises the steps: S1, collecting the data of all equipment in a converter station, and generating an original data set; s2, processing the original data set to generate a standardized data set; s3, constructing an equipment network diagram based on the topological relation of the converter station equipment; s4, the graph neural network module constructs a graph neural network model based on the equipment network graph, extracts feature information of nodes and edges, and generates a multi-dimensional embedded vector; s5, the meta learning module updates model parameters by adopting a hierarchical optimization strategy based on the multi-dimensional embedded vector generated by the graph neural network module, and generates a model parameter set capable of adapting to different input feature distributions; and S6, based on the model parameter set generated by the meta learning module, predicting the energy efficiency related indexes of the converter station. The method has the advantages of strong modeling capability, high prediction precision, strong adaptability and good real-time performance.

Description

Technical Field

[0001] The present invention relates to the technical field of converter station energy efficiency prediction, and in particular to a converter station energy efficiency prediction system based on deep learning. Background Art

[0002] With the rapid development of energy technology and power transmission technology, converter stations have become the core equipment of high-voltage direct current transmission systems. The efficient operation of converter stations is crucial for the efficient transmission of energy and the stable operation of the system. However, due to the complex operating environment of equipment in converter stations, the accurate prediction of energy efficiency-related parameters still faces many challenges.

[0003] The purpose of converter station energy efficiency prediction is to provide a scientific basis for converter station operation optimization and fault prevention by monitoring and analyzing the equipment operating status. In the prior art, traditional energy efficiency prediction methods mainly rely on rule-based mathematical models and empirical formulas. However, with the complexity of the equipment structure in the converter station and the rapid growth of its operating data, these traditional methods have exposed significant limitations: 1. Lack of complex association modeling capabilities: Traditional methods usually only model a single device or independent parameters, making it difficult to capture the complex topological relationships and dynamic interaction characteristics between devices in the converter station.

[0004] 2. Difficult to adapt to diverse operating environments: The operating status of the converter station is significantly affected by environmental conditions and operating modes, and existing methods lack the ability to quickly adapt to changes in data feature distribution under different conditions.

[0005] 3. Inefficient prediction accuracy: Traditional prediction methods based on empirical formulas have limited prediction accuracy when faced with high-dimensional, nonlinear and dynamically changing operating data, and are difficult to meet the needs of actual operation optimization.

[0006] 4. Lack of real-time and multi-dimensional analysis capabilities: Existing methods mostly rely on offline analysis and cannot achieve real-time prediction of operating status and comprehensive analysis of multi-dimensional energy efficiency parameters.

[0007] Therefore, how to provide a converter station energy efficiency prediction system based on deep learning is an issue that technical personnel in this field need to solve urgently. Summary of the invention

[0008] One purpose of the present invention is to propose a converter station energy efficiency prediction system based on deep learning. The present invention fully combines graph neural networks and meta-learning technology, and describes in detail an algorithm for accurately predicting energy efficiency-related indicators by modeling the topological relationship of converter station equipment and quickly adapting to diversified operating environments. It has the advantages of strong modeling ability, high prediction accuracy, strong adaptability and good real-time performance.

[0009] A converter station energy efficiency prediction system based on deep learning according to an embodiment of the present invention includes: S1, data acquisition module, used to collect data of various devices in the converter station and generate an original data set containing the complete equipment operating status; S2, data preprocessing module, used to process the original data set, including data cleaning, missing value filling, normalization and time series feature extraction, to generate a cleaned standardized data set; S3, a graph construction module, which constructs a device network graph based on the topological relationship of the converter station devices, wherein the device network graph represents each device as a node and the electrical connection relationship between the devices as an edge; S4, graph neural network module, builds a graph neural network model based on the device network graph, extracts feature information of nodes and edges, and generates a multi-dimensional embedding vector; S5, the meta-learning module, based on the multi-dimensional embedding vector generated by the graph neural network module, uses a hierarchical optimization strategy to update the model parameters of the graph neural network model and generate a set of model parameters that can adapt to different input feature distributions; S6, energy efficiency prediction module, predicts the energy efficiency related indicators of the converter station based on the model parameter set generated by the meta-learning module, combined with the standardized data set and multi-dimensional embedding vector.

[0010] Optionally, the S3 specifically includes: S31, topology data extraction unit, extracts the topology data of converter station equipment from the standardized data set, including node set and edge set , where the node set Indicates the various devices in the converter station, and the edge collection Indicates the electrical connection relationship between devices; S32, feature mapping unit, node set The device operation status data of each node in is mapped into the node feature vector ,in is the feature vector dimension, is the set of real numbers; S33, edge weight calculation unit, based on edge set The attribute information of each edge in the calculation corresponds to the weight value ,in Representation Node With Node The connection strength between: ; in, and Respectively represent nodes and nodes The characteristic vector of represents the attribute vector of the edge, is the weight matrix of node features, is the weight vector of edge attributes, where represents the dimension of the edge attribute vector, Represents the rectified linear unit activation function, which is used to map the output value to a non-negative range; S34, graph structure generation unit, based on the node set , edge set And the corresponding weight value Build a device network diagram ,in The weight matrix representing the node features, the matrix elements is the weight value.

[0011] Optionally, the S4 specifically includes: S41, input node feature unit, receive the device network graph generated by the graph construction module , where the node set Contains the node feature vector for each node , edge set A vector containing the attributes of each edge , the weight matrix Contains the weight of each edge ; S42, feature aggregation unit, aggregates the local neighborhood information of the node based on the graph convolution operation and updates the node feature vector : ; in, Representation Node The set of neighbor nodes of Representation Node and nodes The weight value between For the The weight matrix of the layer, For Node No. Layer feature vector, is the activation function; S43, feature update unit, updates the node feature vector layer by layer based on the multi-layer graph convolutional network to generate the node embedding vector ,in represents the dimension of the final embedding vector; S44, edge feature fusion unit, generates edge embedding vector based on node embedding vector and edge attribute vector ; S45. Merge the node embedding vector and the edge embedding vector to generate a multi-dimensional embedding vector.

[0012] Optionally, the S44 specifically includes: S441. Extract the embedding vectors of the two end nodes associated with the edge from the node embedding vector and , extract the attribute vector corresponding to each edge from the attribute vector of the edge ; S442, embedding vector for nodes , The attribute vector of the edge Perform concatenation operation to generate joint feature vector : ; in, Represents vector concatenation operation, joint feature vector , is the dimension of the node embedding vector, is the dimension of the edge attribute vector; S443, joint feature vector Perform nonlinear transformation to generate edge embedding vector : ; in, is the weight matrix, is the bias vector, is the activation function, is the generated edge embedding vector, is the dimension of the edge embedding vector.

[0013] Optionally, the S45 specifically includes: S451, merge node embedding vector and edge embedding vector, according to each node The node embedding vector The edge embedding vector between the node and its neighbor , calculate the comprehensive embedding vector of the node : ; in, Representation Node The set of neighbor nodes of is the transformation matrix of the node embedding vector, is the transformation matrix of the edge embedding vector, Representation Node The number of neighbor nodes, is a nonlinear activation function; S452, fusion of global and local embedding features, comprehensive embedding vector for all nodes , construct the global embedding feature vector , and concatenated with the comprehensive embedding vector of each node to generate the final multidimensional embedding vector : ; in, , is the total number of node sets, represents the vector concatenation operation, is the final generated multi-dimensional embedding vector.

[0014] Optionally, the S5 specifically includes: S51, meta-task construction unit, constructs a meta-task set consisting of multiple meta-tasks based on the multi-dimensional embedding vector generated by the graph neural network module ,in, Represents the total number of meta-tasks, each meta-task It includes input feature distribution and corresponding objective function, which is used to simulate the input data feature distribution and its target mapping relationship in different operating environments; S52, initial parameter optimization unit, based on meta-task set , by minimizing the total loss function of all meta-tasks on the initial model parameters To optimize: ; in, Representation model in meta-task The loss function on are the model parameters to be optimized; S53, task adaptation parameter updating unit, for the meta-task set Each meta-task in , using the gradient descent method from the initial model parameters Updated to task specific parameters , to adapt the meta-task The characteristic distribution of S54, global parameter update unit, updates task specific parameters Aggregation, combining the weights of each meta-task For initial parameters Optimize and generate global parameters that reflect cross-task ; S55, global parameters With task-specific parameters Combining to form a set of model parameters that can adapt to different input feature distributions .

[0015] Optionally, the S53 specifically includes: S531. According to the meta-task set Each meta-task The loss function , calculate the initial model parameters Gradient : ; in, Meta-task The loss function is Represents the loss function with respect to the initial model parameters The gradient of S532. According to the meta-task Dynamically adjust the learning rate based on the feature distribution Optimize update step size: ; in, is the initial learning rate, is the learning rate decay factor, The number of iterations optimized for the meta-task; S533, based on learning rate and gradient , for the initial model parameters Update to generate task specific parameters : ; in, Representation and Meta-Tasks Task-specific parameters adapted to feature distribution.

[0016] Optionally, the S54 specifically includes: S541, based on meta-task set Each meta-task The feature distribution and task importance of , calculate the task weight , the task weights satisfy the following constraints: ; in, Represents the total number of meta-tasks, weight Assign tasks based on their specific importance indicators; S542. Gradients of all meta-tasks Perform weighted summation to generate global gradient : ; in, Indicates the task Task-specific parameters The corresponding gradient, is the task weight; S543, based on initial model parameters and the global gradient , update the global parameters and generate optimized global parameters : ; in, is the global optimization step size.

[0017] Optionally, the S6 specifically includes: S61, feature fusion unit, a set of model parameters generated based on the meta-learning module , combining the equipment operation features and multi-dimensional embedding vectors in the standardized dataset to construct the input feature matrix : ; in, For Node The equipment operating characteristics, is the corresponding multi-dimensional embedding vector, Represents feature concatenation operation; S62, energy efficiency prediction model selection unit, from the model parameter set Select the corresponding global parameter and task-specific parameters Building task-specific prediction models , select the optimal task model according to the equipment operation characteristics; S63, prediction calculation unit, based on the selected task-specific prediction model And the input feature matrix , calculate the energy efficiency related indicators of the converter station: ; in, Indicates the energy efficiency related indicators of the converter station, represents a task-specific prediction model, are task-specific parameters.

[0018] The beneficial effects of the present invention are: (1) The present invention combines graph neural networks and meta-learning technology to comprehensively model the topological relationships and dynamic interaction characteristics of equipment in the converter station, enabling the system to accurately capture complex equipment associations and global characteristics, thereby effectively improving the prediction accuracy of energy efficiency-related indicators, especially for the processing requirements of multi-dimensional, high-dimensional and dynamically changing data.

[0019] (2) By introducing meta-learning technology, the present invention enables the model to quickly adapt to different operating environments and changes in data feature distribution, so that when the operating mode or environmental conditions of the converter station change, it can still efficiently predict the energy efficiency indicators, avoiding the performance degradation problem caused by environmental changes in traditional methods.

[0020] (3) The present invention realizes real-time monitoring and efficient calculation of converter station energy efficiency related indicators by combining standardized data, multi-dimensional embedded vectors and dynamically updated model parameter sets in real time. This not only significantly shortens the prediction delay time, but also supports real-time optimization of the operating status of the converter station, effectively improving the operating efficiency and safety of the system.

[0021] (4) The present invention provides a comprehensive analysis capability for energy efficiency-related issues through task-specific prediction models and dynamic optimization of global parameters, enabling the system to understand the operating status of the converter station in multiple dimensions, automatically complete the prediction and analysis of energy efficiency indicators, reduce dependence on manual intervention, and improve the intelligence and applicability of energy efficiency prediction. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] The accompanying drawings are used to provide a further understanding of the present invention and constitute a part of the specification. Together with the embodiments of the present invention, they are used to explain the present invention and do not constitute a limitation of the present invention. In the accompanying drawings: Figure 1 This is an overall framework diagram of a converter station energy efficiency prediction system based on deep learning proposed in the present invention. DETAILED DESCRIPTION

[0023] The present invention will now be described in further detail with reference to the accompanying drawings. These drawings are simplified schematic diagrams, which only illustrate the basic structure of the present invention in a schematic manner, and therefore only show the components related to the present invention.

[0024] refer to Figure 1 , a converter station energy efficiency prediction system based on deep learning, comprising: S1, data acquisition module, used to collect data of various devices in the converter station and generate an original data set containing the complete equipment operating status; In this implementation, S1 specifically includes: The data acquisition module is used to collect operating status data, environmental variables and topological relationship data from various devices in the converter station, and generate an original data set containing the complete equipment operating status.

[0025] S2, data preprocessing module, used to process the original data set, including data cleaning, missing value filling, normalization and time series feature extraction, to generate a cleaned standardized data set; In this implementation, S2 specifically includes: The data preprocessing module is used to comprehensively process the original data set generated by the data acquisition module. The data cleaning process includes detecting and removing outliers and noise data, filling in the gaps in the data through interpolation, and linearly transforming the data through normalization processing, mapping all data values ​​​​to a unified range, and eliminating the impact of dimensional differences on subsequent analysis. Time series feature extraction extracts trend features, periodic features, and time series correlation from the original data set based on the time dependence of the data to generate a standardized data set.

[0026] S3, a graph construction module, which constructs a device network graph based on the topological relationship of the converter station devices, wherein the device network graph represents each device as a node and the electrical connection relationship between the devices as an edge; In this implementation, S3 specifically includes: S31, topology data extraction unit, extracts the topology data of converter station equipment from the standardized data set, including node set and edge set , where the node set Indicates the various devices in the converter station, and the edge collection Indicates the electrical connection relationship between devices; S32, feature mapping unit, node set The device operation status data of each node in is mapped into the node feature vector ,in is the feature vector dimension, is the set of real numbers; S33, edge weight calculation unit, based on edge set The attribute information of each edge in the calculation corresponds to the weight value ,in Representation Node With Node The connection strength between: ; in, and Respectively represent nodes and nodes The characteristic vector of represents the attribute vector of the edge, is the weight matrix of node features, is the weight vector of edge attributes, where represents the dimension of the edge attribute vector, Represents the rectified linear unit activation function, which is used to map the output value to a non-negative range; S34, graph structure generation unit, based on the node set , edge set And the corresponding weight value Build a device network diagram ,in The weight matrix representing the node features, the matrix elements is the weight value.

[0027] This implementation method extracts the topological structure data of the converter station equipment from the standardized data set, constructs the equipment network diagram, and uses the node feature vector and edge weight calculation to generate complete equipment relationship modeling information, which can accurately describe the electrical connection relationship and dynamic characteristics between the equipment in the converter station. This method integrates the node operation status and edge attribute characteristics, which can not only reflect the local equipment characteristics, but also capture the global interaction characteristics through topological relationships, thereby providing high-quality input data for subsequent deep learning modeling, improving the system's modeling ability and prediction accuracy for the complex energy efficiency characteristics of the converter station, and providing a reliable data foundation and structural support for the energy efficiency optimization of the converter station.

[0028] S4, graph neural network module, builds a graph neural network model based on the device network graph, extracts feature information of nodes and edges, and generates a multi-dimensional embedding vector; In this implementation, S4 specifically includes: S41, input node feature unit, receive the device network graph generated by the graph construction module , where the node set Contains the node feature vector for each node , edge set A vector containing the attributes of each edge , the weight matrix Contains the weight of each edge ; S42, feature aggregation unit, aggregates the local neighborhood information of the node based on the graph convolution operation and updates the node feature vector : ; in, Representation Node The set of neighbor nodes of Representation Node and nodes The weight value between For the The weight matrix of the layer, For Node No. Layer feature vector, is the activation function; S43, feature update unit, updates the node feature vector layer by layer based on the multi-layer graph convolutional network to generate the node embedding vector ,in represents the dimension of the final embedding vector; S44, edge feature fusion unit, generates edge embedding vector based on node embedding vector and edge attribute vector ; S45. Merge the node embedding vector and the edge embedding vector to generate a multi-dimensional embedding vector.

[0029] S441. Extract the embedding vectors of the two end nodes associated with the edge from the node embedding vector and , extract the attribute vector corresponding to each edge from the attribute vector of the edge ; S442, embedding vector for nodes , The attribute vector of the edge Perform concatenation operation to generate joint feature vector : ; in, Represents vector concatenation operation, joint feature vector , is the dimension of the node embedding vector, is the dimension of the edge attribute vector; S443, joint feature vector Perform nonlinear transformation to generate edge embedding vector : ; in, is the weight matrix, is the bias vector, is the activation function, is the generated edge embedding vector, is the dimension of the edge embedding vector.

[0030] S451, merge node embedding vector and edge embedding vector, according to each node The node embedding vector The edge embedding vector between the node and its neighbor , calculate the comprehensive embedding vector of the node : ; in, Representation Node The set of neighbor nodes of is the transformation matrix of the node embedding vector, is the transformation matrix of the edge embedding vector, Representation Node The number of neighbor nodes, is a nonlinear activation function; S452, fusion of global and local embedding features, comprehensive embedding vector for all nodes , construct the global embedding feature vector , and concatenated with the comprehensive embedding vector of each node to generate the final multidimensional embedding vector : ; in, , is the total number of node sets, represents the vector concatenation operation, is the final generated multi-dimensional embedding vector.

[0031] This implementation method effectively combines local topological characteristics with global features by fusing node embedding vectors and edge embedding vectors in the device network graph to generate a multi-dimensional embedding vector. This method can not only accurately capture the complex interactive relationships between converter station equipment, but also enhance the understanding of the overall network dynamics through global features, making the expression of converter station energy efficiency related information more comprehensive and accurate. Through the comprehensive modeling of node and edge features, the representation capability of the energy efficiency prediction model is further improved, providing efficient and reliable data foundation support for energy efficiency optimization in complex converter station environments.

[0032] S5, the meta-learning module, based on the multi-dimensional embedding vector generated by the graph neural network module, uses a hierarchical optimization strategy to update the model parameters of the graph neural network model and generate a set of model parameters that can adapt to different input feature distributions; In this implementation, S5 specifically includes: S51, meta-task construction unit, constructs a meta-task set consisting of multiple meta-tasks based on the multi-dimensional embedding vector generated by the graph neural network module ,in, Represents the total number of meta-tasks, each meta-task It includes input feature distribution and corresponding objective function, which is used to simulate the input data feature distribution and its target mapping relationship in different operating environments; S52, initial parameter optimization unit, based on meta-task set , by minimizing the total loss function of all meta-tasks on the initial model parameters To optimize: ; in, Representation model in meta-task The loss function on are the model parameters to be optimized; S53, task adaptation parameter updating unit, for the meta-task set Each meta-task in , using the gradient descent method from the initial model parameters Updated to task specific parameters , to adapt the meta-task The characteristic distribution of S54, global parameter update unit, updates task specific parameters Aggregation, combining the weights of each meta-task For initial parameters Optimize and generate global parameters that reflect cross-task ; S55, global parameters With task-specific parameters Combining to form a set of model parameters that can adapt to different input feature distributions .

[0033] S531. According to the meta-task set Each meta-task The loss function , calculate the initial model parameters Gradient : ; in, Meta-task The loss function is Represents the loss function with respect to the initial model parameters The gradient of S532. According to the meta-task Dynamically adjust the learning rate based on the feature distribution Optimize update step size: ; in, is the initial learning rate, is the learning rate decay factor, The number of iterations optimized for the meta-task; S533, based on learning rate and gradient , for the initial model parameters Update to generate task specific parameters : ; in, Representation and Meta-Tasks Task-specific parameters adapted to feature distribution.

[0034] S541, based on meta-task set Each meta-task The feature distribution and task importance of , calculate the task weight , the task weights satisfy the following constraints: ; in, Represents the total number of meta-tasks, weight Assign tasks based on their specific importance indicators; S542. Gradients of all meta-tasks Perform weighted summation to generate global gradient : ; in, Representation Task Task-specific parameters The corresponding gradient, The corresponding gradient, and the global gradient , update the global parameters and generate optimized global parameters : ; in, is the global optimization step size.

[0035] This implementation introduces a hierarchical optimization strategy through a meta-learning module, and dynamically adjusts the learning rate and model parameter update path in combination with the task characteristics in the meta-task set, so that the model has the ability to quickly adapt to different input feature distributions, which can not only capture the global change characteristics of the converter station operating environment, but also refine the local feature distribution of specific tasks to generate a more accurate task-specific parameter set. In addition, the effective combination of global parameters and task-specific parameters combines the universality and personalized adaptation capabilities across tasks, significantly improving the generalization performance and prediction accuracy of the converter station energy efficiency prediction system, and providing efficient and reliable support for energy efficiency optimization decisions in complex operating scenarios.

[0036] S6, energy efficiency prediction module, predicts the energy efficiency related indicators of the converter station based on the model parameter set generated by the meta-learning module, combined with the standardized data set and multi-dimensional embedding vector.

[0037] In this implementation, S6 specifically includes: S61, feature fusion unit, a set of model parameters generated based on the meta-learning module , combining the equipment operation features and multi-dimensional embedding vectors in the standardized dataset to construct the input feature matrix : ; in, For Node The equipment operating characteristics, is the corresponding multi-dimensional embedding vector, Represents feature concatenation operation; S62, energy efficiency prediction model selection unit, from the model parameter set Select the corresponding global parameter and task-specific parameters Building task-specific prediction models , select the optimal task model according to the equipment operation characteristics; S63, prediction calculation unit, based on the selected task-specific prediction model And the input feature matrix , calculate the energy efficiency related indicators of the converter station: ; in, Indicates the energy efficiency related indicators of the converter station, represents a task-specific prediction model, are task-specific parameters.

[0038] This implementation method combines the equipment operation data and multi-dimensional embedding vectors through the feature fusion unit to construct an input feature matrix, comprehensively integrate the converter station operation status and global topological features, and can capture complex system dynamic relationships; by selecting the optimal task-specific prediction model from the model parameter set, dynamically adapting to different operation data features, the accuracy of the prediction model is ensured; using the prediction calculation unit to calculate energy efficiency related indicators, such as the top oil temperature of the converter transformer and the system energy loss, high-precision modeling of energy efficiency characteristics is achieved. This implementation method combines the global characteristic modeling capabilities of deep learning with the rapid adaptation capabilities of meta-learning, providing an accurate and efficient solution for converter station energy efficiency prediction, while providing reliable data support for energy efficiency optimization and operation decisions. Embodiment 1:

[0039] In order to verify the feasibility of the present invention, the present invention is applied to the high-voltage DC converter station system of a large power grid operating company. The company uses multiple converter stations in power transmission. As core nodes, the energy efficiency management of the converter stations directly affects the operating efficiency and stability of the entire power grid. However, the company has found the following problems in energy efficiency management in recent years: the prediction deviation of energy efficiency-related indicators of some converter stations is large, and it is impossible to effectively support operation optimization and early fault warning. In order to solve this problem, the company decided to introduce the converter station energy efficiency prediction system based on deep learning of the present invention.

[0040] The deployment process of this system includes data collection, graph construction and modeling, training of meta-learning modules, and real-time application of energy efficiency prediction modules. The operating data in the converter station is first collected by acquisition equipment, including converter station equipment topology, electrical connection data, operating status parameters, and environmental variables. Through the graph construction module, these data are mapped into a device network graph, where nodes represent converter station equipment and edges represent electrical relationships between devices. On this basis, node features are extracted through the graph neural network module, and the meta-learning module is used to generate adaptation parameter sets for different operating environments. Finally, the energy efficiency prediction module performs real-time predictions based on these parameters.

[0041] A converter station is located in a coastal area. Due to long-term exposure to high humidity and high temperature, the top oil temperature of its converter transformer frequently increases abnormally. However, the traditional empirical model cannot accurately predict the oil temperature, resulting in several equipment failures caused by overheating. After the energy efficiency prediction system of the present invention is deployed in the converter station, it uses the real-time collected equipment operation data and environmental data to construct an equipment network diagram, and quickly adapts to the characteristic distribution of the coastal environment through the meta-learning module. In the energy efficiency prediction module, the top oil temperature and energy loss of the converter transformer are predicted in real time by combining standardized data and multi-dimensional embedded vectors, providing data support for the operation optimization of the converter station.

[0042] In actual applications, the system first extracts the complex interaction characteristics between devices through graph neural networks to generate embedded vectors for the converter station. These embedded vectors are fused with standardized operation data and input into the energy efficiency prediction module, which selects task-specific models for real-time prediction. For example, in a scenario where the load current rises sharply, the system accurately predicts that the top oil temperature will reach the alarm threshold in 45 minutes, providing operators with sufficient time to take load reduction measures in advance. In order to evaluate the system performance, the company compared the energy efficiency prediction results and operation optimization effects before and after the system deployment. The specific data records are shown in Table 1.

[0043] Table 1 Performance verification report of converter station energy efficiency prediction system ; As can be seen from Table 1, after deploying the system of the present invention, the error of the energy efficiency prediction of the converter station is significantly reduced, especially the performance improvement under complex operating environment and high dynamic load is the most significant. Specifically, the average prediction error of the top oil temperature is reduced from 7.3°C of the traditional method to 2.1°C, which greatly improves the prediction accuracy and avoids the equipment operation risk caused by inaccurate prediction. For the energy loss index, the prediction error is reduced from the original 15.4% to 3.8%, which provides more reliable data support for the operation department to optimize energy management and formulate energy-saving strategies. In addition, the system prediction delay is shortened from 45 seconds to 12 seconds, achieving a significant improvement in real-time performance.

[0044] The system's early warning capability provides an important guarantee for the safe operation of the converter station. In a scenario of sudden load changes, the system successfully predicted that the top oil temperature would reach the alarm threshold within 45 minutes, providing the operation and maintenance team with sufficient time to take countermeasures. Compared with the limitation of traditional methods that cannot provide effective early warnings, the present invention can provide a warning time window of at least 30 minutes, greatly reducing the possibility of equipment failures due to overheating. In actual applications, the number of equipment failures caused by abnormal oil temperature has dropped from 3 times per month before deployment to 0 times after deployment, which significantly improves the reliability of the equipment and the stability of grid operation.

[0045] In addition, the energy efficiency prediction system of the present invention provides a comprehensive perspective for energy efficiency analysis through the efficient fusion of multi-dimensional embedded vectors and standardized operating data. In actual operation, the system can not only accurately predict a single indicator, but also conduct a comprehensive analysis of multi-dimensional data related to energy efficiency, providing more insightful decision support for the operation team. For example, by analyzing the correlation between top oil temperature and energy loss, the system can provide optimization suggestions for load distribution and equipment scheduling under different operating conditions. This prediction method based on multi-dimensional data fusion not only reduces the limitations of operation and maintenance relying on manual experience, but also significantly improves the intelligence and refinement of converter station energy efficiency management.

[0046] The above description is only a preferred specific implementation manner of the present invention, but the protection scope of the present invention is not limited thereto. Any technician familiar with the technical field can make equivalent replacements or changes according to the technical scheme and inventive concept of the present invention within the technical scope disclosed by the present invention, which should be covered by the protection scope of the present invention.

Claims

1. A converter station energy efficiency prediction system based on deep learning, characterized in that: include: S1, data acquisition module, used to collect data of various devices in the converter station and generate an original data set containing the complete equipment operating status; S2, data preprocessing module, used to process the original data set, including data cleaning, missing value filling, normalization and time series feature extraction, to generate a cleaned standardized data set; S3, a graph construction module, which constructs a device network graph based on the topological relationship of the converter station devices, wherein the device network graph represents each device as a node and the electrical connection relationship between the devices as an edge; S4, graph neural network module, builds a graph neural network model based on the device network graph, extracts feature information of nodes and edges, and generates a multi-dimensional embedding vector; S5, the meta-learning module, based on the multi-dimensional embedding vector generated by the graph neural network module, uses a hierarchical optimization strategy to update the model parameters of the graph neural network model and generate a set of model parameters that can adapt to different input feature distributions; S6, energy efficiency prediction module, predicts the energy efficiency related indicators of the converter station based on the model parameter set generated by the meta-learning module, combined with the standardized data set and multi-dimensional embedding vector.

2. A converter station energy efficiency prediction system based on deep learning according to claim 1, characterized in that: The S3 specifically includes: S31, topology data extraction unit, extracts the topology data of converter station equipment from the standardized data set, including node set and edge set , where the node set Indicates the various devices in the converter station, and the edge collection Indicates the electrical connection relationship between devices; S32, feature mapping unit, node set The device operation status data of each node in is mapped into the node feature vector ,in is the feature vector dimension, is the set of real numbers; S33, edge weight calculation unit, based on edge set The attribute information of each edge in the calculation corresponds to the weight value ,in Representation Node With Node Strength of connection between: ； in, and Respectively represent nodes and nodes The characteristic vector of represents the attribute vector of the edge, is the weight matrix of node features, is the weight vector of edge attributes, where represents the dimension of the edge attribute vector, Represents the rectified linear unit activation function, which is used to map the output value to a non-negative range; S34, graph structure generation unit, based on the node set , edge set And the corresponding weight value Build a device network diagram ,in The weight matrix representing the node features, the matrix elements is the weight value.

3. The converter station energy efficiency prediction system based on deep learning according to claim 2, characterized in that: The S4 specifically includes: S41, input node feature unit, receive the device network graph generated by the graph construction module , where the node set Contains the node feature vector for each node , edge set A vector containing the attributes of each edge , the weight matrix Contains the weight of each edge ; S42, feature aggregation unit, aggregates the local neighborhood information of the node based on the graph convolution operation and updates the node feature vector : ； in, Representation Node The set of neighbor nodes of Representation Node and nodes The weight value between For the The weight matrix of the layer, For Node No. Layer feature vector, is the activation function; S43, feature update unit, updates the node feature vector layer by layer based on the multi-layer graph convolutional network to generate the node embedding vector ,in represents the dimension of the final embedding vector; S44, edge feature fusion unit, generates edge embedding vector based on node embedding vector and edge attribute vector ; S45. Merge the node embedding vector and the edge embedding vector to generate a multi-dimensional embedding vector.

4. A converter station energy efficiency prediction system based on deep learning according to claim 3, characterized in that: The S44 specifically includes: S441. Extract the embedding vectors of the two end nodes associated with the edge from the node embedding vector and , extract the attribute vector corresponding to each edge from the attribute vector of the edge ; S442, embedding vector for nodes , The attribute vector of the edge Perform concatenation operation to generate joint feature vector : ； in, Represents vector concatenation operation, joint feature vector , is the dimension of the node embedding vector, is the dimension of the edge attribute vector; S443, joint feature vector Perform nonlinear transformation to generate edge embedding vector : ； in, is the weight matrix, is the bias vector, is the activation function, is the generated edge embedding vector, is the dimension of the edge embedding vector.

5. The converter station energy efficiency prediction system based on deep learning according to claim 3, characterized in that: The S45 specifically includes: S451, merge node embedding vector and edge embedding vector, according to each node The node embedding vector The edge embedding vector between the node and its neighbor , calculate the comprehensive embedding vector of the node : ; in, Representation Node The set of neighbor nodes of is the transformation matrix of the node embedding vector, is the transformation matrix of the edge embedding vector, Representation Node The number of neighbor nodes, is a nonlinear activation function; S452, fusion of global and local embedding features, comprehensive embedding vector for all nodes , construct the global embedding feature vector , and concatenated with the comprehensive embedding vector of each node to generate the final multidimensional embedding vector : ； in, , is the total number of node sets, represents the vector concatenation operation, is the final generated multi-dimensional embedding vector.

6. A converter station energy efficiency prediction system based on deep learning according to claim 1, characterized in that: The S5 specifically includes: S51, meta-task construction unit, constructs a meta-task set consisting of multiple meta-tasks based on the multi-dimensional embedding vector generated by the graph neural network module ,in, Represents the total number of meta-tasks, each meta-task It includes input feature distribution and corresponding objective function, which is used to simulate the input data feature distribution and its target mapping relationship in different operating environments; S52, initial parameter optimization unit, based on meta-task set , by minimizing the total loss function of all meta-tasks on the initial model parameters To optimize: ； in, Representation model in meta-task The loss function on are the model parameters to be optimized; S53, task adaptation parameter updating unit, for the meta-task set Each meta-task in , using the gradient descent method from the initial model parameters Updated to task specific parameters , to adapt the meta-task The characteristic distribution of S54, global parameter update unit, updates task specific parameters Aggregation, combining the weights of each meta-task For initial parameters Optimize and generate global parameters that reflect cross-task ; S55, global parameters With task-specific parameters Combining to form a set of model parameters that can adapt to different input feature distributions .

7. A converter station energy efficiency prediction system based on deep learning according to claim 6, characterized in that: The S53 specifically includes: S531. According to the meta-task set Each meta-task The loss function , calculate the initial model parameters Gradient : ； in, Meta-task The loss function is Represents the loss function with respect to the initial model parameters The gradient of S532. According to the meta-task Dynamically adjust the learning rate based on the feature distribution Optimize update step size: ； in, is the initial learning rate, is the learning rate decay factor, The number of iterations optimized for the meta-task; S533, based on learning rate and gradient , for the initial model parameters Update to generate task specific parameters : ； in, Representation and Meta-Tasks Task-specific parameters adapted to feature distribution.

8. A converter station energy efficiency prediction system based on deep learning according to claim 6, characterized in that: The S54 specifically includes: S541, based on meta-task set Each meta-task The feature distribution and task importance of , calculate the task weight , the task weights satisfy the following constraints: ； in, Represents the total number of meta-tasks, weight Assign tasks based on their specific importance indicators; S542. Gradients of all meta-tasks Perform weighted summation to generate global gradient : ； in, Indicates the task Task-specific parameters The corresponding gradient, is the task weight; S543, based on initial model parameters and the global gradient , update the global parameters and generate optimized global parameters : ； in, is the global optimization step size.

9. A converter station energy efficiency prediction system based on deep learning according to claim 6, characterized in that: The S6 specifically includes: S61, feature fusion unit, a set of model parameters generated based on the meta-learning module , combining the equipment operation features and multi-dimensional embedding vectors in the standardized dataset to construct the input feature matrix : ； in, For Node The equipment operating characteristics, is the corresponding multi-dimensional embedding vector, Represents feature concatenation operation; S62, energy efficiency prediction model selection unit, from the model parameter set Select the corresponding global parameter and task-specific parameters Building task-specific prediction models , select the optimal task model according to the equipment operation characteristics; S63, prediction calculation unit, based on the selected task-specific prediction model And the input feature matrix , calculate the energy efficiency related indicators of the converter station: ； in, Indicates the energy efficiency related indicators of the converter station, represents a task-specific prediction model, are task-specific parameters.

Images