Neural Operators for Power Systems: A Physics-Informed Framework for Modeling Power System Components

TL;DR

This paper introduces a neural-operator framework using DeepONets and Physics-Informed DeepONets for fast, accurate, and scalable power system component modeling, significantly outperforming traditional solvers in speed and stability.

Contribution

The paper presents a novel physics-informed neural-operator approach for power systems, improving generalization, efficiency, and stability over existing methods.

Findings

Achieves over 30 times speedup compared to traditional ODE solvers.

Demonstrates superior stability and scalability over PINNs.

Provides accurate predictions across diverse power system scenarios.

Abstract

Modern power systems require fast and accurate dynamic simulations for stability assessment, digital twins, and real-time control, but classical ODE solvers are often too slow for large-scale or online applications. We propose a neural-operator framework for surrogate modeling of power system components, using Deep Operator Networks (DeepONets) to learn mappings from system states and time-varying inputs to full trajectories without step-by-step integration. To enhance generalization and data efficiency, we introduce Physics-Informed DeepONets (PI-DeepONets), which embed the residuals of governing equations into the training loss. Our results show that DeepONets, and especially PI-DeepONets, achieve accurate predictions under diverse scenarios, providing over 30 times speedup compared to high-order ODE solvers. Benchmarking against Physics-Informed Neural Networks (PINNs) highlights superior stability and scalability. Our results demonstrate neural operators as a promising path toward real-time, physics-aware simulation of power system dynamics.