Revealing drivers and risks for power grid frequency stability with explainable AI

TL;DR

This paper demonstrates how explainable AI can accurately predict power grid frequency stability and reveal key factors influencing stability, addressing the limitations of traditional models and black-box AI methods.

Contribution

It introduces an explainable AI approach for predicting and understanding power grid frequency stability, highlighting critical features affecting system security.

Findings

Accurately predicts frequency stability indicators like RoCoF and Nadir.

Identifies load ramps, generation ramps, prices, and forecast errors as key factors.

Provides insights into power grid stability for European synchronous areas.

Abstract

Stable operation of the electrical power system requires the power grid frequency to stay within strict operational limits. With millions of consumers and thousands of generators connected to a power grid, detailed human-build models can no longer capture the full dynamics of this complex system. Modern machine learning algorithms provide a powerful alternative for system modelling and prediction, but the intrinsic black-box character of many models impedes scientific insights and poses severe security risks. Here, we show how eXplainable AI (XAI) alleviates these problems by revealing critical dependencies and influences on the power grid frequency. We accurately predict frequency stability indicators (such as RoCoF and Nadir) for three major European synchronous areas and identify key features that determine the power grid stability. Load ramps, specific generation ramps but also prices and forecast errors are central to understand and stabilize the power grid.