A GNN-Based Generative Model for Generating Synthetic Cyber-Physical Power System Topology

Abstract

Synthetic networks aim at generating realistic projections of real-world networks while concealing the actual system information. This paper proposes a scalable and effective approach based on graph neural networks (GNN) to generate synthetic topologies of Cyber-Physical power Systems (CPS) with realistic network feature distribution. In order to comprehensively capture the characteristics of real CPS networks, we propose a generative model, namely Graph-CPS, based on graph variational autoencoder and graph recurrent neural networks. The method hides the sensitive topological information while maintaining the similar feature distribution of the real networks. We used multiple power and communication networks to prove and assess the effectiveness of the proposed method with experimental results.

Application of improved graph convolutional networks in daily-ahead carbon emission prediction

“…Therefore, the graph convolution network with longand short-term network (TGNN) is constructed for the carbon emission prediction. Recently, several studies have evaluated the applications of GNNs in power grids (Boyaci et al, 2022;Liu et al, 2022;Hansen et al, 2023;Liu et al, 2023). The Bayesian optimization utilizes Gaussian process regression to construct a selection function for obtaining the next sample point.…”

Electricity consumption optimization of power users driven by a dynamic electric carbon factor

“…There are usually conflicting or noncomparable objectives in a multi-objective problem, so it is not possible to simply find a solution that maximizes or minimizes all the objectives at the same time. The aim of multi-objective optimization is to find a set of solutions that form a nondominated set, called a "Pareto frontier," which outperforms other solutions on all objectives (Boyaci et al, 2022;Liu et al, 2022;Zeng et al, 2023b;Liu et al, 2023).…”