Thunderstorm Prediction During Pre-Tactical Air-Traffic-Flow Management Using Convolutional Neural Networks

31 Pages Posted: 3 Oct 2022

See all articles by Aniel Jardines

Aniel Jardines

affiliation not provided to SSRN

Hamidreza Eivazi

affiliation not provided to SSRN

Elías Zea

affiliation not provided to SSRN

Juan Simarro

affiliation not provided to SSRN

Javier García-Heras

Charles III University of Madrid - Departamento de Bioingeniería e Ingeniería Aeroespacial

Manuel Soler

affiliation not provided to SSRN

Evelyn Otero

affiliation not provided to SSRN

Ricardo Vinuesa

affiliation not provided to SSRN

Abstract

Thunderstorms can be a large source of disruption for European air-traffic management causing a chaotic state of operation within the airspace system. In current practice, air-traffic managers are provided with imprecise forecasts which limit their ability to plan strategically. As a result, weather mitigation is performed using tactical measures with a time horizon of three hours. Increasing the lead time of thunderstorm predictions to the day before operations could help air-traffic managers plan around weather and improve the efficiency of air-traffic-management operations. Emerging techniques based on machine learning have provided promising results, partly attributed to reduced human bias and improved capacity in predicting thunderstorms purely from numerical weather prediction data. In this paper, we expand on our previous work on thunderstorm forecasting, by applying convolutional neural networks (CNNs) to exploit the spatial characteristics embedded in the weather data. The learning task of predicting convection is formulated as a binary-classification problem based on satellite data. The performance of multiple CNN-based architectures, including a fully-convolutional neural network (FCN), a CNN-based encoder-decoder, a U-Net, and a pyramid-scene parsing network (PSPNet) are compared against a multi-layer-perceptron (MLP) network. Our work indicates that CNN-based architectures improve the performance of point-prediction models, with a fully-convolutional neural-network architecture having the best performance. Results show that CNN-based architectures can be used to increase the prediction lead time of thunderstorms. Lastly, a case study illustrating the applications of convection-prediction models in an air-traffic-management setting is presented.

Keywords: Thunderstorms, Air Traffic Management, Weather Data, Numerical Weather Prediction, Satellite Images, Convolutional Neural Network, Machine Learning

Suggested Citation

Jardines, Aniel and Eivazi, Hamidreza and Zea, Elías and Simarro, Juan and García-Heras, Javier and Soler, Manuel and Otero, Evelyn and Vinuesa, Ricardo, Thunderstorm Prediction During Pre-Tactical Air-Traffic-Flow Management Using Convolutional Neural Networks. Available at SSRN: https://ssrn.com/abstract=4236016 or http://dx.doi.org/10.2139/ssrn.4236016

Aniel Jardines (Contact Author)

affiliation not provided to SSRN ( email )

No Address Available

Hamidreza Eivazi

affiliation not provided to SSRN ( email )

No Address Available

Elías Zea

affiliation not provided to SSRN ( email )

No Address Available

Juan Simarro

affiliation not provided to SSRN ( email )

No Address Available

Javier García-Heras

Charles III University of Madrid - Departamento de Bioingeniería e Ingeniería Aeroespacial ( email )

Spain

Manuel Soler

affiliation not provided to SSRN ( email )

No Address Available

Evelyn Otero

affiliation not provided to SSRN ( email )

No Address Available

Ricardo Vinuesa

affiliation not provided to SSRN ( email )

No Address Available

Do you have negative results from your research you’d like to share?

Paper statistics

Downloads
101
Abstract Views
360
Rank
479,512
PlumX Metrics