Power System Network Topology Identification Based on Knowledge Graph and Graph Neural Network

Abstract

The automatic identification of the topology of power networks is important for the data-driven and situation-aware operation of power grids. Traditional methods of topology identification lack a data-tolerant mechanism, and the accuracy of their performance in terms of identification is thus affected by the quality of data. Topology identification is related to the link prediction problem. The graph neural network can be used to predict the state of unlabeled nodes (lines) through training on features of labeled nodes (lines) with fault tolerance. Inspired by the characteristics of the graph neural network, we applied it to topology identification in this study. We propose a method to identify the topology of a power network based on a knowledge graph and the graph neural network. Traditional knowledge graphs can quickly mine possible connections between entities and generate graph structure data, but in the case of errors or informational conflicts in the data, they cannot accurately determine whether the relationships between the entities really exist. The graph neural network can use data mining to determine whether a connection obtained between entities is based on their eigenvalues, and has a fault tolerance mechanism to adapt to errors and informational conflicts in the graph data, but needs the graph data as database. The combination of the knowledge graph and the graph neural network can compensate for the deficiency of the single knowledge graph method. We tested the proposed method by using the IEEE 118-bus system and a provincial network system. The results showed that our approach is feasible and highly fault tolerant. It can accurately identify network topology even in the presence of conflicting and missing measurement-related information.

Comprehensive Mapping of Continuous/Switching Circuits in CCM and DCM to Machine Learning Domain Using Homogeneous Graph Neural Networks

“…In [68], [69] the provided network learns to solve load flow problem on random power grids whose size range from 10 to 110 buses. A method to identify the topology of a PS network is proposed in [70] based on GNN, avoiding errors in Traditional knowledge graphs in the case of errors or informational conflicts in the data. All previously mentioned research empirically transform the PS network into graph without following a circuit-laws-consistent formulation.…”

Recommendation Method of Power Knowledge Retrieval Based on Graph Neural Network

“…By this end-to-end modeling and learning approach, the graph structure data can be effectively embedded, thus improving the link prediction results. Especially in the field of electricity, the topology of the grid is important for data retrieval; thus, Wang [33] et al used graph neural networks to mine the connections between entities. In the field of film and television, Yu [34] et al proposed a graph convolutional network (OR-GCN) based on object relations; this method analyzes directed graphs by building a graph convolutional network and achieves better results in film classification.…”

A Review of Graph Neural Networks and Their Applications in Power Systems

“…The automatic topology identification of distribution networks is crucial for the data-driven safe operation of power grids. In [101], the input data are represented as the graphstructured data, and then the spectral-based GCNs are employed to identify power network topology. In [102], the novel hybrid forms of GNNs are designed to test whether medium-voltage distribution networks satisfy the safe property of the topology or not.…”