Operator Learning for Power Systems Simulation

2025-10-09

AI Summary

Time-domain simulation of power systems with high renewable energy penetration becomes computationally intractable due to the need to resolve microsecond-scale ultrafast dynamics. Method: This study proposes an operator-learning-based surrogate modeling framework that directly learns the nonlinear, function-to-function mapping governing system state evolution, using three distinct neural operator architectures. Contribution/Results: We introduce and empirically validate the novel concept of “simulation time-step invariance,” enabling zero-shot super-resolution inference—i.e., high-accuracy prediction at unseen fine-grained time scales without retraining. The model demonstrates strong generalization across stable and unstable dynamics, low cross-step prediction error, and excellent scalability. Validated on a simplified test system, it establishes a new paradigm for real-time or quasi-real-time simulation of large-scale renewable-integrated power grids.

Technology Category

• Search and Optimization: Sampling/Simulation-based Search
• Planning, Routing, and Scheduling: Optimization of Spatio-temporal Systems
• Cognitive Modeling & Cognitive Systems: Simulating Human Behavior

Application Category

• User Modeling, Personalization and Recommendation: User modeling and simulation for interactive and conversational systems
• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications
• Economics, Online Markets and Human Computation: Cost models of using LLMs in production systems

Abstract

Time domain simulation, i.e., modeling the system's evolution over time, is a crucial tool for studying and enhancing power system stability and dynamic performance. However, these simulations become computationally intractable for renewable-penetrated grids, due to the small simulation time step required to capture renewable energy resources' ultra-fast dynamic phenomena in the range of 1-50 microseconds. This creates a critical need for solutions that are both fast and scalable, posing a major barrier for the stable integration of renewable energy resources and thus climate change mitigation. This paper explores operator learning, a family of machine learning methods that learn mappings between functions, as a surrogate model for these costly simulations. The paper investigates, for the first time, the fundamental concept of simulation time step-invariance, which enables models trained on coarse time steps to generalize to fine-resolution dynamics. Three operator learning methods are benchmarked on a simple test system that, while not incorporating practical complexities of renewable-penetrated grids, serves as a first proof-of-concept to demonstrate the viability of time step-invariance. Models are evaluated on (i) zero-shot super-resolution, where training is performed on a coarse simulation time step and inference is performed at super-resolution, and (ii) generalization between stable and unstable dynamic regimes. This work addresses a key challenge in the integration of renewable energy for the mitigation of climate change by benchmarking operator learning methods to model physical systems.

Problem

• Developing fast scalable power system simulation methods
• Enabling time step-invariant modeling for renewable grids
• Addressing computational barriers in renewable energy integration

Innovation

• Operator learning replaces costly power system simulations
• Time step-invariance enables coarse-to-fine resolution generalization
• Benchmarked zero-shot super-resolution for renewable grid dynamics