This paper introduces DeepOPF-U, a novel deep neural network model designed to solve alternating-current (AC) optimal power flow (OPF) problems across different power networks. Unlike traditional models, which lack generalizability to varying network topologies and growing distributed energy resources (DERs), DeepOPF-U can handle varying numbers of buses, lines, loads, and DERs. The model features elastic input and output layers to adapt to different network sizes and an incremental training process for expanding networks. Simulations have shown improved performance of DeepOPF-U compared to existing models.
Publication date: 25 Sep 2023
Project Page: Not provided
Paper: https://arxiv.org/pdf/2309.12849
