Method, device and equipment for transient voltage stability evaluation of multi-dc hub power grid

Abstract

This invention discloses a method, apparatus, and device for assessing transient voltage stability in a multi-DC hub power grid. The method includes inputting the topological characteristics of the multi-DC hub power grid, as well as the operating mode characteristics and disturbance characteristics of each node in the power grid, into a pre-constructed multi-dimensional structure-aware graph neural network to generate multi-dimensional structural features characterizing the grid's structural characteristics and operating state. The multi-dimensional structural features from different perspectives are then sequentially modeled for importance and adaptively weighted and fused to obtain enhanced structure-aware features reflecting the relative importance of the multi-dimensional structural features. These enhanced structure-aware features are then input into a trained transient voltage stability assessment model, outputting the transient voltage stability assessment result. This invention addresses the technical problems of low computational efficiency and insufficient adaptability to changes in various features in the transient voltage stability assessment of multi-DC hub power grids.

Description

Technical Field

[0001] This invention relates to the field of power grid technology, and in particular to a method, apparatus and equipment for assessing transient voltage stability in a multi-DC hub power grid. Background Technology

[0002] During the construction of the new power system, renewable energy from outside the region is continuously transmitted to the near-regional load centers of the receiving end through multiple conventional DC transmission channels. Renewable energy power plants in the receiving-end grid are connected to the grid on a large scale in both centralized and distributed modes, and the proportion of power electronic equipment in the grid is constantly increasing. Under the combined effect of multiple DC feeds and a high proportion of renewable energy access, the operating capacity of synchronous generators in the receiving-end grid is continuously decreasing, and the dynamic reactive power support capacity is relatively insufficient. The grid as a whole exhibits typical "weak grid" operating characteristics. When severe disturbances occur in the grid, the risk of transient voltage instability increases significantly. In extreme cases, it may even trigger continuous commutation failures on multiple DC lines, leading to a chain reaction of accidents in the sending and receiving end grids, posing a major threat to the safe and stable operation of the new power system.

[0003] The transient voltage stability characteristics of a multi-DC hub power grid are affected by a variety of factors. On the one hand, the DC feed-in power level and the output distribution and operating status of renewable energy power plants and conventional generating units directly affect the reactive power balance characteristics and voltage support capability of the system. On the other hand, spatial factors such as the local topology of the AC grid and the location of disturbances further affect the propagation path and scope of transient voltage instability, making the transient voltage stability mechanism exhibit obvious nonlinear and spatially correlated characteristics, and significantly increasing the difficulty of transient voltage stability assessment.

[0004] Traditional transient voltage stability assessment methods are mainly based on mechanistic models and time-domain simulation calculations. They determine whether transient voltage instability has occurred by performing detailed simulation analysis on specific operating modes and disturbance scenarios. While these methods have high accuracy in single-condition analysis, they require extensive offline simulation calculations when dealing with highly diverse operating modes and a large number of disturbance scenario combinations in multi-DC hub power grids. This results in a huge computational workload, making it difficult to meet the needs of online transient voltage stability assessment.

[0005] In recent years, with the development of artificial intelligence, data-driven transient voltage stability assessment methods have gradually attracted attention. These methods typically extract electrical quantity characteristics that reflect the operating state of the power grid and utilize deep learning models to uncover the nonlinear mapping relationship between operating state characteristics and transient voltage stability characteristics, thereby achieving transient voltage stability discrimination or stability margin prediction. However, existing data-driven methods mostly focus on learning the numerical characteristics of the operating state, failing to fully consider the influence of structural information such as topology, branch power flow distribution, and disturbance location in multi-DC hub power grids. This results in insufficient generalization ability of the model under conditions of topology changes or fault location changes, limiting the reliability and engineering applicability of the assessment results. Summary of the Invention

[0006] This invention provides a transient voltage stability assessment method for multi-DC hub power grids, which addresses the technical problems of low computational efficiency and insufficient adaptability to changes in operating mode, topology, and fault location in transient voltage stability assessment of multi-DC hub power grids under conditions of high proportion of renewable energy access.

[0007] According to one aspect of the present invention, a method for assessing transient voltage stability in a multi-DC hub power grid is provided, comprising: The topological characteristics of the multi-DC hub power grid, as well as the operating mode characteristics and disturbance characteristics of each node in the multi-DC hub power grid, are input into the constructed multi-dimensional structure perception graph neural network to generate multi-dimensional structural features that characterize the power grid structure characteristics and operating status. The importance of the multidimensional structural features from different perspectives is modeled and adaptively weighted and fused in sequence to obtain structure-aware enhanced features that reflect the relative importance of the multidimensional structural features. The structure-aware enhanced features are input into the trained transient voltage stability evaluation model, and the transient voltage stability evaluation results are output.

[0008] Optionally, the topological characteristics of the multi-DC hub power grid, as well as the operating mode characteristics and disturbance characteristics of each node in the multi-DC hub power grid, include: Representing a multi-DC hub power grid as a structured network:

[0009] In the formula, V represents the set of power grid nodes, E represents the set of power grid edges, and A represents the adjacency matrix obtained from the power grid topology connection relationship; For each node Construct the initial feature vector of the node:

[0010] In the formula, V i P represents the node voltage amplitude.i Q i These are the active and reactive power at the nodes, respectively. The available reactive power regulation capability of a node is used to characterize the reactive power regulation margin that a node can still provide or absorb under the current operating mode. Type i For node type identification, including AC bus nodes, conventional power supply nodes, new energy power supply access nodes, load nodes, etc.; The node-level perturbation embedding is obtained by perturbation feature mapping, where d represents the perturbation feature;

[0011] In the formula, E T Disturbance types include single-phase ground faults, two-phase short-circuit faults, and two-phase short-circuit to ground faults; E L This indicates the location where the disturbance occurred.

[0012] Optionally, the step of inputting the topological characteristics of the multi-DC hub power grid, as well as the operating mode characteristics and disturbance characteristics of each node in the multi-DC hub power grid, into the constructed multi-dimensional structure perception graph neural network to generate multi-dimensional structural features characterizing the power grid's structural characteristics and operating state includes: The node features are updated layer by layer using a multi-layer stacked neural network; where the feature of node i in the l-th layer is represented as... l represents the network layer number; the initial node representation is obtained from the node feature vectors:

[0013] In the formula, For a multilayer perceptron, it is a function used to map node features to initial feature representations; Different structural perspectives Construct the corresponding structure weight matrices respectively. ;in, Indicates the first s Nodes from a structural perspective j For nodes i The structural association weights are used to characterize the connection strength or influence between nodes from the perspective of that structure; in the... l In the layer, the first s Nodes from a structural perspective i The feature aggregation is represented as:

[0014] In the formula, Indicates the first s From a structural perspective, nodes i A set of nodes that are related; Indicates the first l In the layer for the firsts Learnable feature transformation matrices are set for each structural perspective to characterize the mapping relationship of node features under different structural perspectives; For activation functions; Fusion of aggregation results from different structural perspectives: By fusing feature information from different perspectives through the output of a multi-layered perceptual graph neural network, a multi-dimensional structural feature representation of each node is finally obtained:

[0015] In the formula, Indicates the first l The first in the layer s Feature fusion weights corresponding to each structural perspective; Represents a node i In the s Updated feature representation from a structural perspective; The multidimensional structural feature representation of the node is obtained:

[0016] Through the modeling process of the multidimensional structural perceptual graph neural network, the obtained multidimensional structural feature representation H can characterize the operating state and spatial characteristics of the power grid from multiple structural perspectives.

[0017] Optionally, the importance modeling and adaptive weighted fusion of the multidimensional structural features from different perspectives are performed sequentially to obtain structure-aware enhanced features that reflect the relative importance of the multidimensional structural features, including: Importance modeling of the multidimensional structural features from different perspectives includes: For the s-th structural perspective, a structural importance scoring function is introduced:

[0018] In the formula, H (s) This represents the feature subspace corresponding to the s-th structural viewpoint. This is a global pooling operation; Then, adaptive weighted fusion is performed to obtain structure-aware enhanced features that reflect the relative importance of multidimensional structural features, including: The importance scores of all structural perspectives are normalized to obtain the structural perception weights:

[0019] In the formula, Indicates the first s The importance-normalized weights of each structural perspective are used to characterize the relative contribution of different structural perspectives to the transient voltage stability assessment task; these weights are combined with the aforementioned multi-perspective feature fusion weights. They have different meanings.

[0020] Based on structure-aware weights, multi-dimensional structural features are reconstructed using weighted averages.

[0021] In the formula, This is a structure-aware enhancement feature that integrates structural importance information.

[0022] Optionally, the transient voltage stability evaluation model is a Transformer model with a structure-aware module; The transient voltage stability assessment model is used to model the relationship between different features and reflect the differences in importance of multidimensional structural features.

[0023] Optionally, the step of inputting the structure-aware enhanced features into the trained transient voltage stability evaluation model and outputting the transient voltage stability evaluation result includes: The structure-aware enhancement features Mapped to the input sequence of the Transformer model:

[0024] Structure-aware self-attention mechanism: In the Transformer self-attention calculation process, structure-aware weights are introduced to modulate the attention calculation, and their calculation form is as follows:

[0025] In the formula, Q is the query matrix; K is the key matrix; V is the value matrix; d is the feature dimension; and B is the structure-aware weight. The constructed structural bias matrix is ​​used to reflect the differences in importance of multidimensional structural features; High-level feature representations of the input sequence are extracted using a multi-layer perceptual Transformer encoder. The transient voltage stability evaluation results are obtained by evaluating the output layer:

[0026] In the formula, This indicates the result of transient voltage stability judgment or transient voltage stability margin assessment.

[0027] Optionally, it also includes: Acquire a large amount of historical topology data, operation mode data and disturbance data of each node of the multi-DC hub power grid, and construct a dataset; A transient voltage stability assessment model for a multi-DC hub power grid is trained based on a dataset, resulting in a well-trained transient voltage stability assessment model for a multi-DC hub power grid. The transient voltage stability assessment model for a multi-DC hub power grid includes a multi-dimensional structure-aware graph neural network and a Transformer transient voltage stability assessment model with a structure-aware module.

[0028] According to another aspect of the present invention, a transient voltage stability assessment device for a multi-DC hub power grid is provided, comprising: The multidimensional feature extraction unit is used to input the topological features of the multi-DC hub power grid, as well as the operating mode features and disturbance features of each node in the multi-DC hub power grid, into the constructed multidimensional structure perception graph neural network to generate multidimensional structural features that characterize the power grid structure characteristics and operating status. The structure-aware enhancement unit is used to sequentially perform importance modeling and adaptive weighted fusion on the multidimensional structural features from different perspectives to obtain structure-aware enhancement features that reflect the relative importance of the multidimensional structural features. The evaluation unit is used to input the structure-aware enhanced features into the trained transient voltage stability evaluation model and output the transient voltage stability evaluation result.

[0029] According to another aspect of the present invention, an electronic device is provided, the electronic device comprising: At least one processor; and A memory communicatively connected to the at least one processor; wherein, The memory stores a computer program that can be executed by the at least one processor, which enables the at least one processor to perform the multi-DC hub power grid transient voltage stability assessment method according to any embodiment of the present invention.

[0030] According to another aspect of the present invention, a computer-readable storage medium is provided, the computer-readable storage medium storing computer instructions for causing a processor to execute and implement the multi-DC hub power grid transient voltage stability assessment method according to any embodiment of the present invention.

[0031] The technical solution of this invention involves inputting the topological features of a multi-DC hub power grid, as well as the operating mode features and disturbance features of each node in the multi-DC hub power grid, into a pre-constructed multi-dimensional structure-aware graph neural network to generate multi-dimensional structural features characterizing the power grid's structural characteristics and operating state. Importance modeling and adaptive weighted fusion are then performed on the multi-dimensional structural features from different perspectives to obtain enhanced structure-aware features reflecting the relative importance of the multi-dimensional structural features. These enhanced structure-aware features are then input into a trained transient voltage stability assessment model, embedding the adaptive perception information of the structural features into the feature association calculation process. This allows the assessment model to explicitly perceive the topological characteristics of the power grid and spatial information such as disturbance locations while learning electrical quantity features, thereby achieving efficient assessment of the transient voltage stability characteristics of the multi-DC hub power grid.

[0032] It should be understood that the description in this section is not intended to identify key or essential features of the embodiments of the present invention, nor is it intended to limit the scope of the invention. Other features of the invention will become readily apparent from the following description. Attached Figure Description

[0033] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0034] Figure 1 This is a flowchart of a method for evaluating transient voltage stability in a multi-DC hub power grid according to Embodiment 1 of the present invention; Figure 2 This is an architecture diagram of a multi-DC hub power grid transient voltage stability model provided in one embodiment of the present invention; Figure 3 This is a structural diagram of a multi-DC hub power grid transient voltage stability assessment device according to Embodiment 2 of the present invention; Figure 4 This is a schematic diagram of the structure of an electronic device for implementing the transient voltage stability assessment method for multi-DC hub power grids according to embodiments of the present invention. Detailed Implementation

[0035] To enable those skilled in the art to better understand the present invention, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of the present invention.

[0036] It should be noted that the terms "first," "second," etc., in the specification, claims, and accompanying drawings of this invention are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of the invention described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover a non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0037] Example 1 Figure 1 This document provides a flowchart of a method for evaluating transient voltage stability in a multi-DC hub power grid, as described in Embodiment 1 of the present invention. Figure 1 As shown, the method includes: S101. Input the topological characteristics of the multi-DC hub power grid, as well as the operating mode characteristics and disturbance characteristics of each node in the multi-DC hub power grid, into the constructed multi-dimensional structural perception graph neural network to generate multi-dimensional structural features that characterize the power grid structural characteristics and operating status.

[0038] Among them, the operating mode characteristics can include features such as DC power, load level, unit output and status of each node; the topology characteristics include the connection relationship between nodes; and the disturbance characteristics include features such as the disturbance type and the location of the disturbance.

[0039] In this embodiment, the constructed multidimensional structural perception graph neural network can jointly model the power grid from multiple structural perspectives, realize the structured modeling and feature extraction of operating mode information, power grid topology information and disturbance space information in sample data, and generate a multidimensional structural feature representation that can comprehensively characterize the operating mode and disturbance scenario of the power grid, providing structured input for subsequent transient voltage stability assessment.

[0040] S103. The importance of the multidimensional structural features from different perspectives is modeled and adaptively weighted and fused in sequence to obtain structural perception enhancement features that reflect the relative importance of the multidimensional structural features.

[0041] In this embodiment, a multi-dimensional structural feature adaptive perception module can be constructed to sequentially model the importance of structural features from different structural perspectives and adaptively assign weights, generating structural perception information that reflects the relative importance of multi-dimensional structural features, and providing this information as constraint information to the subsequent evaluation model. By introducing a structural feature adaptive perception mechanism, the evaluation model can autonomously adjust its attention to different structural features according to different operating modes and disturbance scenarios, thereby enhancing the model's adaptability to topological changes.

[0042] S104. Input the structure-aware enhancement features into the trained transient voltage stability evaluation model and output the transient voltage stability evaluation result.

[0043] In this embodiment, the constructed transient voltage stability assessment model can be a Transformer model with a structure-aware module. By embedding the structure-aware mechanism into the feature association calculation process of the Transformer, the Transformer can explicitly reflect the importance differences of multi-dimensional structural features when modeling the relationships between different features. This allows for global modeling of electrical quantity features while effectively utilizing the multi-dimensional structural feature information of the power grid to more accurately characterize the complex nonlinear relationship between the operating characteristics and transient voltage stability characteristics of a multi-DC hub power grid. The trained transient voltage stability assessment model can output transient voltage stability assessment results online based on a given operating mode and disturbance scenario.

[0044] The technical solution of this invention involves inputting the topological features of a multi-DC hub power grid, as well as the operating mode features and disturbance features of each node in the multi-DC hub power grid, into a pre-constructed multi-dimensional structure-aware graph neural network to generate multi-dimensional structural features characterizing the power grid's structural characteristics and operating state. Importance modeling and adaptive weighted fusion are then performed on the multi-dimensional structural features from different perspectives to obtain enhanced structure-aware features reflecting the relative importance of the multi-dimensional structural features. These enhanced structure-aware features are then input into a trained transient voltage stability assessment model, embedding the adaptive perception information of the structural features into the feature association calculation process. This allows the assessment model to explicitly perceive the topological characteristics of the power grid and spatial information such as disturbance locations while learning electrical quantity features, thereby achieving efficient assessment of the transient voltage stability characteristics of the multi-DC hub power grid.

[0045] In one embodiment, the process of obtaining the topological characteristics of the multi-DC hub power grid, as well as the operating mode characteristics and disturbance characteristics of each node in the multi-DC hub power grid, in step S101 specifically includes: Representing a multi-DC hub power grid as a structured network:

[0046] In the formula, V represents the set of power grid nodes, E represents the set of power grid edges, and A represents the adjacency matrix obtained from the power grid topology connections; the structured network g describes the basic connection relationships of the power grid, and the subsequent structural weight matrix A corresponds to each structural perspective. (s) All are constructed based on adjacency relationships.

[0047] For each node Construct the initial feature vector of the node:

[0048] In the formula, V i P represents the node voltage amplitude. i Q i These are the active and reactive power at the nodes, respectively. The available reactive power regulation capability of a node is used to characterize the reactive power regulation margin that a node can still provide or absorb under the current operating mode. Type i For node type identification, including AC bus nodes, conventional power supply nodes, new energy power supply access nodes, load nodes, etc.; The node-level perturbation embedding is obtained by perturbation feature mapping, where d represents the perturbation feature;

[0049] In the formula, E T Disturbance types include single-phase ground faults, two-phase short-circuit faults, and two-phase short-circuit to ground faults; E L This represents the location where the disturbance occurs. For each disturbance scenario, a disturbance location feature vector is introduced to characterize the disturbance type and its location.

[0050] In this embodiment, the disturbance location is uniformly mapped to the node location for modeling. That is, when the disturbance occurs on the line, it is equivalently mapped to the corresponding line endpoint node, which facilitates unified modeling under the node-based graph neural network framework.

[0051] In one embodiment, step S102 specifically includes: updating the node features layer by layer through a multi-layer stacked neural network; wherein, let the feature of node i in the l-th layer be represented as... Let l represent the number of network layers; then the initial node representation is obtained from the node feature vectors:

[0052] In the formula, For a multilayer perceptron, it is a function used to map node features to initial feature representations; This embodiment uses multiple structural perspectives to jointly model the power grid, including the operational structure perspective (DC feed-in power, unit status, etc.), the topology perspective (node ​​connection relationships), and the disturbance spatial structure perspective (spatial correlation of disturbance locations to nodes). The specific architecture can be found in [reference needed]. Figure 2 From different structural perspectives Construct the corresponding structure weight matrices respectively. .in, Indicates the first s Nodes from a structural perspective j For nodes i The structural association weights are used to characterize the connection strength or influence between nodes from the perspective of that structure. In the first... l In the layer, the first s Nodes from a structural perspective i The feature aggregation is represented as:

[0053] In the formula, Indicates the first s From a structural perspective, nodes i A set of nodes that are related; Indicates the first l In the layer for the first s Learnable feature transformation matrices are set for each structural perspective to characterize the mapping relationship of node features under different structural perspectives; This is the activation function.

[0054] In each layer update, perturbation features d The feature subspace is affected by the initial feature vectors of nodes and the feature propagation process from various structural perspectives. H (s) The characterization results are used to further influence the importance scores from different structural perspectives through a structural importance scoring function. and the corresponding structure-aware weights ; Fusion of aggregation results from different structural perspectives: By fusing feature information from different perspectives through the output of a multi-layered perceptual graph neural network, a multi-dimensional structural feature representation of each node is finally obtained:

[0055] In the formula, Indicates the first l The first in the layer s Feature fusion weights corresponding to each structural perspective; Represents a node i In the s Updated feature representation from a structural perspective; By stacking multiple layers of the above structure, collaborative modeling of the multidimensional structural characteristics of the power grid is achieved, resulting in a multidimensional structural feature representation of the nodes:

[0056] Through the modeling process of the multidimensional structural perceptual graph neural network, the obtained multidimensional structural feature representation H can characterize the operating state and spatial characteristics of the power grid from multiple structural perspectives.

[0057] In one embodiment, step S103 specifically includes: The importance modeling and adaptive weighted fusion of the multidimensional structural features from different perspectives are performed sequentially to obtain structure-aware enhanced features that reflect the relative importance of the multidimensional structural features, including: Since the impact of different structural perspectives on transient voltage stability varies under different operating modes and disturbance scenarios, it is necessary to model the importance of the multidimensional structural features under different perspectives, including: For the s-th structural perspective, a structural importance scoring function is introduced:

[0058] In the formula, H (s) This represents the feature subspace corresponding to the s-th structural viewpoint. This is a global pooling operation; Then, adaptive weighted fusion is performed to obtain structure-aware enhanced features that reflect the relative importance of multidimensional structural features, including: The importance scores of all structural perspectives are normalized to obtain the structural perception weights:

[0059] In the formula, Indicates the first s The importance normalized weights of each structural perspective are used to characterize the relative contribution of different structural perspectives to the transient voltage stability assessment task.

[0060] Based on structure-aware weights, multi-dimensional structural features are reconstructed using weighted averages.

[0061] In the formula, This is a structure-aware enhancement feature that integrates structural importance information.

[0062] Through the modeling process of the multidimensional structural feature adaptive sensing module described above, a structural sensing enhancement feature that integrates structural importance information is obtained. This feature not only retains the information of the multidimensional structural features, but also explicitly reflects the relative importance of different structural features under the current operating mode and disturbance scenario, providing key input for constructing a structural sensing transient voltage stability assessment model.

[0063] In one embodiment, the transient voltage stability evaluation model in step S104 is a Transformer model with a structure-aware module; the transient voltage stability evaluation model is used to model the relationship between different features and reflect the importance differences of multidimensional structural features.

[0064] Step S104 specifically includes: applying the structure-aware enhancement features Mapped to the input sequence of the Transformer model:

[0065] Structure-aware self-attention mechanism: In the Transformer self-attention calculation process, structure-aware weights are introduced to modulate the attention calculation, and their calculation form is as follows:

[0066] In the formula, Q is the query matrix; K is the key matrix; V is the value matrix; d is the feature dimension; and B is the structure-aware weight. The constructed structural bias matrix is ​​used to reflect the differences in importance of multidimensional structural features; High-level feature representations of the input sequence are extracted using a multi-layer perceptual Transformer encoder. The transient voltage stability evaluation results are obtained by evaluating the output layer:

[0067] In the formula, This indicates the result of transient voltage stability judgment or transient voltage stability margin assessment.

[0068] This embodiment realizes the representation of multi-dimensional structural features, adaptive modeling of the importance of structural features, and structurally aware feature correlation calculation under a unified framework, forming a complete transient voltage stability assessment process that can quickly output transient voltage stability assessment results based on a given operating mode and disturbance scenario.

[0069] In one embodiment, the method further includes: acquiring a large amount of historical topology data of multi-DC hub power grids, operation mode data and disturbance data of each node, and constructing a dataset; A transient voltage stability assessment model for a multi-DC hub power grid is trained based on a dataset, resulting in a well-trained transient voltage stability assessment model for a multi-DC hub power grid. The transient voltage stability assessment model for a multi-DC hub power grid includes a multi-dimensional structure-aware graph neural network and a Transformer transient voltage stability assessment model with a structure-aware module.

[0070] It should be noted that a large amount of historical topology data, operation mode data, and disturbance data of multi-DC hub power grids can be obtained, and transient voltage stability results can be labeled to construct a dataset. Based on the dataset, a multi-DC hub power grid transient voltage stability assessment model is trained. The multi-DC hub power grid transient voltage stability assessment model includes a multi-dimensional structure-aware graph neural network and a Transformer transient voltage stability assessment model with a structure-aware module. Training stops when a preset number of iterations is reached, resulting in a well-trained multi-DC hub power grid transient voltage stability assessment model.

[0071] The technical solution of this invention constructs a multi-dimensional structure-aware graph neural network (MSA-GNN) to perform unified structured modeling of a multi-DC hub power grid from multiple structural perspectives, including operating mode characteristics, topological characteristics, and spatial characteristics of disturbance locations. This achieves a coordinated representation of the multi-dimensional structural characteristics of the power grid. Compared to data-driven methods that model solely based on numerical features or a single topological structure, this invention can more comprehensively characterize the structural characteristics of a multi-DC hub power grid under different operating modes and disturbance scenarios, thereby improving the adaptability of transient voltage stability assessment to changes in power grid topology and disturbance location. Furthermore, by constructing a multi-dimensional structural feature adaptive perception module, the importance of structural features from different structural perspectives is modeled and adaptively weighted, enabling the assessment model to dynamically adjust the degree of attention given to various structural features according to different operating modes and disturbance scenarios. This mechanism effectively avoids the problem of traditional data-driven models treating different structural information equally, allowing the model to highlight structural features that have a more significant impact on transient voltage stability, thereby improving the accuracy and robustness of the assessment results. Finally, by constructing a Transformer transient voltage stability assessment model with a structure-aware module and embedding multi-dimensional structural feature perception information into the Transformer's feature correlation calculation process, the Transformer can explicitly perceive the topological characteristics and disturbance spatial characteristics of the power grid while modeling the global relationships between features. Compared to the traditional Transformer model that relies solely on feature numerical similarity for correlation calculation, the structure-aware Transformer model constructed in this invention can more accurately learn the complex nonlinear relationship between the operating characteristics of multi-DC hub power grids and transient voltage stability characteristics, thereby improving the overall performance of transient voltage stability assessment.

[0072] Example 2 Figure 3 This is a schematic diagram of a transient voltage stability assessment device for a multi-DC hub power grid provided in Embodiment 2 of the present invention. Figure 3 As shown, the device includes: The multidimensional feature extraction unit 301 is used to input the topological features of the multi-DC hub power grid, as well as the operating mode features and disturbance features of each node in the multi-DC hub power grid, into the constructed multidimensional structure perception graph neural network to generate multidimensional structural features that characterize the power grid structure characteristics and operating status. The structure-aware enhancement unit 302 is used to sequentially perform importance modeling and adaptive weighted fusion on the multidimensional structural features from different perspectives to obtain structure-aware enhancement features that reflect the relative importance of the multidimensional structural features. The evaluation unit 303 is used to input the structure-aware enhancement features into the trained transient voltage stability evaluation model and output the transient voltage stability evaluation result.

[0073] The transient voltage stability assessment device for multi-DC hub power grids provided in this embodiment of the invention can execute the transient voltage stability assessment method for multi-DC hub power grids provided in any embodiment of the invention, and has the corresponding functional modules and beneficial effects of the method.

[0074] Example 3 Figure 4 A schematic diagram of an electronic device 10, which can be used to implement embodiments of the present invention, is shown. The electronic device is intended to represent various forms of digital computers, such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device can also represent various forms of mobile devices, such as personal digital processors, cellular phones, smartphones, wearable devices (e.g., helmets, glasses, watches, etc.), and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely illustrative and are not intended to limit the implementation of the invention described and / or claimed herein.

[0075] like Figure 4As shown, the electronic device 10 includes at least one processor 11 and a memory, such as a read-only memory (ROM) 12 or a random access memory (RAM) 13, communicatively connected to the at least one processor 11. The memory stores computer programs executable by the at least one processor. The processor 11 can perform various appropriate actions and processes based on the computer program stored in the ROM 12 or loaded from storage unit 18 into the RAM 13. The RAM 13 can also store various programs and data required for the operation of the electronic device 10. The processor 11, ROM 12, and RAM 13 are interconnected via a bus 14. An input / output (I / O) interface 15 is also connected to the bus 14.

[0076] Multiple components in electronic device 10 are connected to I / O interface 15, including: input unit 16, such as keyboard, mouse, etc.; output unit 17, such as various types of displays, speakers, etc.; storage unit 18, such as disk, optical disk, etc.; and communication unit 19, such as network card, modem, wireless transceiver, etc. Communication unit 19 allows electronic device 10 to exchange information / data with other devices through computer networks such as the Internet and / or various telecommunications networks.

[0077] Processor 11 can be a variety of general-purpose and / or special-purpose processing components with processing and computing capabilities. Some examples of processor 11 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various special-purpose artificial intelligence (AI) computing chips, various processors running machine learning model algorithms, a digital signal processor (DSP), and any suitable processor, controller, microcontroller, etc. Processor 11 performs the various methods and processes described above, such as a method for assessing transient voltage stability in a multi-DC hub power grid.

[0078] In some embodiments, a method for assessing the transient voltage stability of a multi-DC hub power grid can be implemented as a computer program tangibly contained in a computer-readable storage medium, such as storage unit 18. In some embodiments, part or all of the computer program can be loaded and / or installed on electronic device 10 via ROM 12 and / or communication unit 19. When the computer program is loaded into RAM 13 and executed by processor 11, one or more steps of the multi-DC hub power grid transient voltage stability assessment method described above can be performed. Alternatively, in other embodiments, processor 11 can be configured to perform a multi-DC hub power grid transient voltage stability assessment method by any other suitable means (e.g., by means of firmware).

[0079] Various embodiments of the systems and techniques described above herein can be implemented in digital electronic circuit systems, integrated circuit systems, field-programmable gate arrays (FPGAs), application-specific integrated circuits (ASICs), application-specific standard products (ASSPs), systems-on-a-chip (SoCs), payload-programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various embodiments may include implementations in one or more computer programs that can be executed and / or interpreted on a programmable system including at least one programmable processor, which may be a dedicated or general-purpose programmable processor, capable of receiving data and instructions from a storage system, at least one input device, and at least one output device, and transmitting data and instructions to the storage system, the at least one input device, and the at least one output device.

[0080] Computer programs used to implement the methods of the present invention may be written in any combination of one or more programming languages. These computer programs may be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing device, such that when executed by the processor, the computer programs cause the functions / operations specified in the flowcharts and / or block diagrams to be performed. The computer programs may be executed entirely on a machine, partially on a machine, or as a standalone software package, partially on a machine and partially on a remote machine, or entirely on a remote machine or server.

[0081] In the context of this invention, a computer-readable storage medium can be a tangible medium that may contain or store a computer program for use by or in conjunction with an instruction execution system, apparatus, or device. A computer-readable storage medium may include, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatus, or devices, or any suitable combination thereof. Alternatively, a computer-readable storage medium may be a machine-readable signal medium. More specific examples of machine-readable storage media include electrical connections based on one or more wires, portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fibers, portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination thereof.

[0082] To provide interaction with a user, the systems and techniques described herein can be implemented on an electronic device having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user; and a keyboard and pointing device (e.g., a mouse or trackball) through which the user provides input to the electronic device. Other types of devices can also be used to provide interaction with the user; for example, feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form (including sound input, voice input, or tactile input).

[0083] The systems and technologies described herein can be implemented in computing systems that include backend components (e.g., as data servers), or middleware components (e.g., application servers), or frontend components (e.g., user computers with graphical user interfaces or web browsers through which users can interact with implementations of the systems and technologies described herein), or any combination of such backend, middleware, or frontend components. The components of the system can be interconnected via digital data communication of any form or medium (e.g., communication networks). Examples of communication networks include local area networks (LANs), wide area networks (WANs), blockchain networks, and the Internet.

[0084] A computing system can include clients and servers. Clients and servers are generally located far apart and typically interact through communication networks. The client-server relationship is created by computer programs running on the respective computers and having a client-server relationship with each other. The server can be a cloud server, also known as a cloud computing server or cloud host, which is a hosting product within the cloud computing service system to address the shortcomings of traditional physical hosts and VPS services, such as high management difficulty and weak business scalability.

[0085] It should be understood that the various forms of processes shown above can be used, with steps reordered, added, or deleted. For example, the steps described in this invention can be executed in parallel, sequentially, or in different orders, as long as the desired result of the technical solution of this invention can be achieved, and no limitation is imposed herein.

[0086] The specific embodiments described above do not constitute a limitation on the scope of protection of this invention. Those skilled in the art should understand that various modifications, combinations, sub-combinations, and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of this invention should be included within the scope of protection of this invention.

Claims

1. A method for evaluating transient voltage stability in a multi-DC hub power grid, characterized in that, include: The topological characteristics of the multi-DC hub power grid, as well as the operating mode characteristics and disturbance characteristics of each node in the multi-DC hub power grid, are input into the constructed multi-dimensional structure perception graph neural network to generate multi-dimensional structural features that characterize the power grid structure characteristics and operating status. The importance of the multidimensional structural features from different perspectives is modeled and adaptively weighted and fused in sequence to obtain structure-aware enhanced features that reflect the relative importance of the multidimensional structural features. The structure-aware enhanced features are input into the trained transient voltage stability evaluation model, and the transient voltage stability evaluation results are output.

2. The method for evaluating transient voltage stability of a multi-DC hub power grid according to claim 1, characterized in that, The topological characteristics of the multi-DC hub power grid, as well as the operating mode characteristics and disturbance characteristics of each node in the multi-DC hub power grid, include: Representing a multi-DC hub power grid as a structured network: In the formula, V represents the set of power grid nodes, E represents the set of power grid edges, and A represents the adjacency matrix obtained from the power grid topology connection relationship; For each node Construct the initial feature vector of the node: In the formula, V i P represents the node voltage amplitude. i Q i These are the active and reactive power at the nodes, respectively. The available reactive power regulation capability of a node is used to characterize the reactive power regulation margin that a node can still provide or absorb under the current operating mode. Type i For node type identification, including AC bus nodes, conventional power supply nodes, new energy power supply access nodes, load nodes, etc.; The node-level perturbation embedding is obtained by perturbation feature mapping, where d represents the perturbation feature; In the formula, E T Disturbance types include single-phase ground faults, two-phase short-circuit faults, and two-phase short-circuit to ground faults; E L This indicates the location where the disturbance occurred.

3. The method for evaluating transient voltage stability of a multi-DC hub power grid according to claim 2, characterized in that, The process involves inputting the topological characteristics of the multi-DC hub power grid, as well as the operational and disturbance characteristics of each node within the multi-DC hub power grid, into a pre-constructed multi-dimensional structure-aware graph neural network to generate multi-dimensional structural features characterizing the power grid's structural properties and operational status. These features include: The node features are updated layer by layer using a multi-layer stacked neural network; where the feature of node i in the l-th layer is represented as... l represents the network layer number; the initial node representation is obtained from the node feature vectors: In the formula, For a multilayer perceptron, it is a function used to map node features to initial feature representations; Different structural perspectives Construct the corresponding structure weight matrices respectively. ;in, Indicates the first s Nodes from a structural perspective j For nodes i The structural association weights are used to characterize the connection strength or influence between nodes from the perspective of that structure; in the... l In the layer, the first s Nodes from a structural perspective i The feature aggregation is represented as: In the formula, Indicates the first s From a structural perspective, nodes i A set of nodes that are related; Indicates the first l In the layer for the first s Learnable feature transformation matrices are set for each structural perspective to characterize the mapping relationship of node features under different structural perspectives; For activation functions; Fusion of aggregation results from different structural perspectives: By fusing feature information from different perspectives through the output of a multi-layered perceptual graph neural network, a multi-dimensional structural feature representation of each node is finally obtained: In the formula, Indicates the first l The first in the layer s Feature fusion weights corresponding to each structural perspective; Represents a node i In the s Updated feature representation from a structural perspective; The multidimensional structural feature representation of the node is obtained: Through the modeling process of the multidimensional structural perceptual graph neural network, the obtained multidimensional structural feature representation H can characterize the operating state and spatial characteristics of the power grid from multiple structural perspectives.

4. The method for evaluating transient voltage stability of a multi-DC hub power grid according to claim 1, characterized in that, The importance modeling and adaptive weighted fusion of the multidimensional structural features from different perspectives are performed sequentially to obtain structure-aware enhanced features that reflect the relative importance of the multidimensional structural features, including: Importance modeling of the multidimensional structural features from different perspectives includes: For the s-th structural perspective, a structural importance scoring function is introduced: In the formula, H (s) This represents the feature subspace corresponding to the s-th structural viewpoint. This is a global pooling operation; Then, adaptive weighted fusion is performed to obtain structure-aware enhanced features that reflect the relative importance of multidimensional structural features, including: The importance scores of all structural perspectives are normalized to obtain the structural perception weights: In the formula, Indicates the first s The importance normalized weights of each structural perspective are used to characterize the relative contribution of different structural perspectives to the transient voltage stability assessment task. Based on structure-aware weights, multi-dimensional structural features are reconstructed using weighted averages. In the formula, This is a structure-aware enhancement feature that integrates structural importance information.

5. The method for evaluating transient voltage stability of a multi-DC hub power grid according to claim 4, characterized in that, The transient voltage stability evaluation model is a Transformer model with a structure-aware module. The transient voltage stability assessment model is used to model the relationship between different features and reflect the differences in importance of multidimensional structural features.

6. The method for evaluating transient voltage stability of a multi-DC hub power grid according to claim 5, characterized in that, The step of inputting the structure-aware enhanced features into the trained transient voltage stability assessment model and outputting transient voltage stability assessment results includes: The structure-aware enhancement features Mapped to the input sequence of the Transformer model: Structure-aware self-attention mechanism: In the Transformer self-attention calculation process, structure-aware weights are introduced to modulate the attention calculation, and their calculation form is as follows: In the formula, Q is the query matrix; K is the key matrix; V is the value matrix; d is the feature dimension; and B is the structure-aware weight. The constructed structural bias matrix is ​​used to reflect the differences in importance of multidimensional structural features; High-level feature representations of the input sequence are extracted using a multi-layer perceptual Transformer encoder. The transient voltage stability evaluation results are obtained by evaluating the output layer: In the formula, This indicates the result of transient voltage stability judgment or transient voltage stability margin assessment.

7. The method for evaluating transient voltage stability of a multi-DC hub power grid according to claim 1, characterized in that, Also includes: Acquire a large amount of historical topology data, operation mode data and disturbance data of each node of the multi-DC hub power grid, and construct a dataset; A transient voltage stability assessment model for a multi-DC hub power grid is trained based on a dataset, resulting in a well-trained transient voltage stability assessment model for a multi-DC hub power grid. The transient voltage stability assessment model for a multi-DC hub power grid includes a multi-dimensional structure-aware graph neural network and a Transformer transient voltage stability assessment model with a structure-aware module.

8. A transient voltage stability assessment device for a multi-DC hub power grid, characterized in that, include: The multidimensional feature extraction unit is used to input the topological features of the multi-DC hub power grid, as well as the operating mode features and disturbance features of each node in the multi-DC hub power grid, into the constructed multidimensional structure perception graph neural network to generate multidimensional structural features that characterize the power grid structure characteristics and operating status. The structure-aware enhancement unit is used to sequentially perform importance modeling and adaptive weighted fusion on the multidimensional structural features from different perspectives to obtain structure-aware enhancement features that reflect the relative importance of the multidimensional structural features. The evaluation unit is used to input the structure-aware enhanced features into the trained transient voltage stability evaluation model and output the transient voltage stability evaluation result.

9. An electronic device, characterized in that, The electronic device includes: At least one processor; and A memory communicatively connected to the at least one processor; wherein, The memory stores a computer program that can be executed by the at least one processor, the computer program being executed by the at least one processor to enable the at least one processor to perform the multi-DC hub power grid transient voltage stability assessment method according to any one of claims 1-7.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer instructions that cause a processor to execute the transient voltage stability assessment method for a multi-DC hub power grid as described in any one of claims 1-7.

Images