Self-Organized Criticality Identification of Power System Based on SC-GCN Network
ID：97
Submission ID：181 View Protection：ATTENDEE
Updated Time：2023-11-20 13:45:42 Hits：201
Oral Presentation
Start Time：2023-12-10 11:00 （Asia/Shanghai）
Duration：15min
Session：[S8] AI-driven technology » [S8] AI-driven technology
Presentation File
Tips: The file permissions under this presentation are only for participants. You have not logged in yet and cannot view it temporarily.
Abstract
In order to solve the problem that the traditional power system self-organized criticality identification method is weak in dealing with graphical data and nonlinear coupling factors, this paper proposes a power system self-organized criticality identification method based on SC-GCN neural network. First, the self-organized criticality evolution process is simulated based on the OPA model. Secondly, the feature quantities of each day, including bus features, line features and global features, are obtained according to the self-organized criticality evolution process, and the data are processed and classified as graph data. Further, a SC-GCN neural network model is obtained by adding jump connections to the traditional graph convolutional neural network to solve the gradient vanishing problem and accelerate the model convergence. Finally, the dataset is input into the SC-GCN neural network model for training and testing, and the simulation results show that compared with the traditional power system self-organized criticality identification method, the power system self-organized criticality identification method proposed in this paper can be more perfect, simple and rapid to identify the power system self-organized criticality online.
Keywords
Graph Convolutional Neural Networks; Self-Organizing Criticality; OPA Model; Jump Connections;
Speaker
Comment submit