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

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

29 Aug 2022arXiv:2208.13585archive 2025-07-28

Lars Ødegaard Bentsen, Narada Dilp Warakagoda, Roy Stenbro, Paal Engelstad

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.

Paper PDF Code

Code

larsbentsen/fftransformer officialmentioned in paperpytorch report

Repository list and official/mentioned flags are the archive's, frozen 2025-07-28. Reachability, where shown, is from one Syntology probe window (2026-09-16 to 2026-09-18); repositories not probed show nothing. GitHub stars are not tracked.

Code Syntology ran Syntology

Not run by Syntology. Nothing on this page verifies that the listed code works.

Tasks

Graph Neural Network Multivariate Time Series Forecasting Spatio-Temporal Forecasting Time Series Time Series Analysis Time Series Forecasting Weather Forecasting

Results from the paper archive 2025-07-28

No leaderboard rows for this paper in the archive.

Methods

Adam Attention BPE CGNN Dense Connections Dropout Graph Neural Network Label Smoothing Layer Normalization Linear Layer MPNN Multi-Head Attention Position-Wise Feed-Forward Layer Residual Connection SPEED Softmax Transformer

Report a problem or propose a change · a person checks every report against the paper or source before anything changes; decisions are listed on /corrections