Predictability of Power Grid Frequency

Abstract

The power grid frequency is the central observable in power system control, as it measures the balance of electrical supply and demand. A reliable frequency forecast can facilitate rapid control actions and may thus greatly improve power system stability. Here, we develop a weighted-nearestneighbour (WNN) predictor to investigate how predictable the frequency trajectories are. Our forecasts for up to one hour are more precise than averaged daily profiles and could increase the efficiency of frequency control actions. Furthermore, we gain an increased understanding of the specific properties of different synchronous areas by interpreting the optimal prediction parameters (number of nearest neighbours, the prediction horizon, etc.) in terms of the physical system. Finally, prediction errors indicate the occurrence of exceptional external perturbations. Overall, we provide a diagnostics tool and an accurate predictor of the power grid frequency time series, allowing better understanding of the underlying dynamics.

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

“…The daily profile, i.e., the daily average evolution of a target, is the most important recurring pattern of frequency dynamics. 61 Predicting the stability indicators based on their daily profiles thus represents an important null model. Our GTB model consistently outperformed the daily profile for all areas and indicators (see supplemental experimental procedures S5 for a detailed performance evaluation).…”

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

“…The daily profile, i.e. the daily average evolution of a target, is the most important recurring pattern of frequency dynamics [48]. Predicting the stability indicators based on their daily profiles thus represents an important null model.…”

Exploring deterministic frequency deviations with explainable AI

“…Methods from Machine Learning (ML) are excellent candidates to model the complex effects of multiple features on power grid frequency fluctuations. Due to control measures, frequency dynamics exhibit complex non-linear dependencies [1] and measurement errors further modify many publicly available grid frequency measurements [10]. Moreover, drivers of frequency deviations such as load or generation ramps are strongly correlated [9].…”