Leveraging Graph Neural Networks to Forecast Electricity Consumption

TL;DR

This paper presents a novel graph neural network-based approach for electricity demand forecasting, capturing spatial and relational data in decentralized energy networks to improve accuracy and explainability.

Contribution

It introduces graph inference methods and a framework for evaluating GNN models specifically tailored for electricity consumption prediction.

Findings

GNN models outperform traditional methods in forecasting accuracy.

The proposed framework enhances model interpretability.

Experiments on French regions validate the approach's effectiveness.

Abstract

Accurate electricity demand forecasting is essential for several reasons, especially as the integration of renewable energy sources and the transition to a decentralized network paradigm introduce greater complexity and uncertainty. The proposed methodology leverages graph-based representations to effectively capture the spatial distribution and relational intricacies inherent in this decentralized network structure. This research work offers a novel approach that extends beyond the conventional Generalized Additive Model framework by considering models like Graph Convolutional Networks or Graph SAGE. These graph-based models enable the incorporation of various levels of interconnectedness and information sharing among nodes, where each node corresponds to the combined load (i.e. consumption) of a subset of consumers (e.g. the regions of a country). More specifically, we introduce a range of methods for inferring graphs tailored to consumption forecasting, along with a framework for evaluating the developed models in terms of both performance and explainability. We conduct experiments on electricity forecasting, in both a synthetic and a real framework considering the French mainland regions, and the performance and merits of our approach are discussed.