Spatio-Temporal Wind Speed Forecasting using Graph Networks and Novel Transformer Architectures

TL;DR

This paper introduces novel spatio-temporal wind speed forecasting models using graph neural networks combined with Transformer architectures, including new variants like FFTransformer, demonstrating superior accuracy over traditional models.

Contribution

It is the first to apply LogSparse Transformer and Autoformer in wind forecasting within a spatio-temporal GNN framework, and proposes the innovative FFTransformer architecture.

Findings

FFTransformer outperforms all models for 4-hour forecasts.

Transformer-based models outperform LSTM and MLP in accuracy.

Autoformer and FFTransformer achieve the best results for short-term forecasts.

Abstract

This study focuses on multi-step spatio-temporal wind speed forecasting for the Norwegian continental shelf. The study aims to leverage spatial dependencies through the relative physical location of different measurement stations to improve local wind forecasts. Our multi-step forecasting models produce either 10-minute, 1- or 4-hour forecasts, with 10-minute resolution, meaning that the models produce more informative time series for predicted future trends. A graph neural network (GNN) architecture was used to extract spatial dependencies, with different update functions to learn temporal correlations. These update functions were implemented using different neural network architectures. One such architecture, the Transformer, has become increasingly popular for sequence modelling in recent years. Various alterations have been proposed to better facilitate time series forecasting, of which this study focused on the Informer, LogSparse Transformer and Autoformer. This is the first time the LogSparse Transformer and Autoformer have been applied to wind forecasting and the first time any of these or the Informer have been formulated in a spatio-temporal setting for wind forecasting. By comparing against spatio-temporal Long Short-Term Memory (LSTM) and Multi-Layer Perceptron (MLP) models, the study showed that the models using the altered Transformer architectures as update functions in GNNs were able to outperform these. Furthermore, we propose the Fast Fourier Transformer (FFTransformer), which is a novel Transformer architecture based on signal decomposition and consists of two separate streams that analyse the trend and periodic components separately. The FFTransformer and Autoformer were found to achieve superior results for the 10-minute and 1-hour ahead forecasts, with the FFTransformer significantly outperforming all other models for the 4-hour ahead forecasts.