Optimal Power Grid Operations with Foundation Models

TL;DR

This work argues for using AI Foundation Models to tackle the growing complexity and uncertainty in power-grid operations caused by the energy transition. It proposes self-supervised learning of power-flow dynamics via a GraphMAE-based pretraining task, incorporating physics-informed losses to align with $S_i$ and $S_{i,j}$ relations, and aims to generalize across varied grid topologies with Graph Neural Networks. The approach targets speeding up critical tasks such as contingency analysis and economic dispatch, potentially enabling large-scale stochastic simulations and a grid digital twin. If successful, this could yield significant operational efficiency, lower emissions, and substantial cost savings, demonstrated by references to real-world gains like improved line throughput in large grids.

Abstract

The energy transition, crucial for tackling the climate crisis, demands integrating numerous distributed, renewable energy sources into existing grids. Along with climate change and consumer behavioral changes, this leads to changes and variability in generation and load patterns, introducing significant complexity and uncertainty into grid planning and operations. While the industry has already started to exploit AI to overcome computational challenges of established grid simulation tools, we propose the use of AI Foundation Models (FMs) and advances in Graph Neural Networks to efficiently exploit poorly available grid data for different downstream tasks, enhancing grid operations. For capturing the grid's underlying physics, we believe that building a self-supervised model learning the power flow dynamics is a critical first step towards developing an FM for the power grid. We show how this approach may close the gap between the industry needs and current grid analysis capabilities, to bring the industry closer to optimal grid operation and planning.

Figures (1)

• Figure 1: Masking and reconstruction steps for grid FM training. 1. Masking: Given a graph representation of the transmission grid $\mathcal{G}$, the function $m_\alpha$ randomly masks node variables (independently of the bus type) with masking probability $\alpha$. The resulting masked graph is $\mathcal{G}'$. 2. Reconstruction: We assume the existence of a function $f$ that, given a masked graph $\mathcal{G}'$, returns the original graph $\mathcal{G}$. Our model is trained to approximate $f$.