Multistream Graph Attention Networks for Wind Speed Forecasting

TL;DR

This paper introduces a novel multistream graph attention network model that combines GAT and LSTM to improve wind speed forecasting by capturing spatial and temporal weather data relationships, providing better accuracy and interpretability.

Contribution

The paper proposes an extended GAT architecture with learnable adjacency and variable-specific attention, integrated with LSTM for enhanced wind speed prediction and interpretability.

Findings

Outperforms previous models in wind speed prediction accuracy.

Learns attention weights indicating key weather variables and locations.

Effectively captures complex spatial-temporal weather data relationships.

Abstract

Reliable and accurate wind speed prediction has significant impact in many industrial sectors such as economic, business and management among others. This paper presents a new model for wind speed prediction based on Graph Attention Networks (GAT). In particular, the proposed model extends GAT architecture by equipping it with a learnable adjacency matrix as well as incorporating a new attention mechanism with the aim of obtaining attention scores per weather variable. The output of the GAT based model is combined with the LSTM layer in order to exploit both the spatial and temporal characteristics of the multivariate multidimensional historical weather data. Real weather data collected from several cities in Denmark and Netherlands are used to conduct the experiments and evaluate the performance of the proposed model. We show that in comparison to previous architectures used for wind speed prediction, the proposed model is able to better learn the complex input-output relationships of the weather data. Furthermore, thanks to the learned attention weights, the model provides an additional insights on the most important weather variables and cities for the studied prediction task.