Wind speed prediction using multidimensional convolutional neural networks

TL;DR

This paper presents a novel multidimensional CNN model that effectively captures the complex spatio-temporal relationships in weather data for improved wind speed prediction, outperforming traditional CNN approaches.

Contribution

The paper introduces a new multidimensional CNN architecture that better models the spatio-temporal evolution of wind data for forecasting tasks.

Findings

The proposed model outperforms classical CNN models on real-world datasets.

Multidimensional CNNs better capture complex input-output relationships in weather data.

Experimental results demonstrate improved wind speed prediction accuracy.

Abstract

Accurate wind speed forecasting is of great importance for many economic, business and management sectors. This paper introduces a new model based on convolutional neural networks (CNNs) for wind speed prediction tasks. In particular, we show that compared to classical CNN-based models, the proposed model is able to better characterise the spatio-temporal evolution of the wind data by learning the underlying complex input-output relationships from multiple dimensions (views) of the input data. The proposed model exploits the spatio-temporal multivariate multidimensional historical weather data for learning new representations used for wind forecasting. We conduct experiments on two real-life weather datasets. The datasets are measurements from cities in Denmark and in the Netherlands. The proposed model is compared with traditional 2- and 3-dimensional CNN models, a 2D-CNN model with an attention layer and a 2D-CNN model equipped with upscaling and depthwise separable convolutions.