Exploring Variational Graph Autoencoders for Distribution Grid Data Generation
This work addresses the lack of public power system data for researchers, though it is incremental as it highlights limitations of existing methods.
The paper tackled the problem of generating synthetic distribution grid data for machine learning research in energy networks by investigating variational graph autoencoders (VGAEs), finding that GCN-based decoders achieved strong fidelity on simpler datasets but struggled with complex ones, producing artifacts like disconnected components.
To address the lack of public power system data for machine learning research in energy networks, we investigate the use of variational graph autoencoders (VGAEs) for synthetic distribution grid generation. Using two open-source datasets, ENGAGE and DINGO, we evaluate four decoder variants and compare generated networks against the original grids using structural and spectral metrics. Results indicate that simple decoders fail to capture realistic topologies, while GCN-based approaches achieve strong fidelity on ENGAGE but struggle on the more complex DINGO dataset, producing artifacts such as disconnected components and repeated motifs. These findings highlight both the promise and limitations of VGAEs for grid synthesis, underscoring the need for more expressive generative models and robust evaluation. We release our models and analysis as open source to support benchmarking and accelerate progress in ML-driven power system research.