Long-Short Term Memory for an Effective Short-Term Weather Forecasting Model Using Surface Weather Data

Research output: Contribution to journalConference proceeding article (ISSN)

17 Downloads (Pure)

Abstract

Numerical Weather Prediction (NWP) requires considerable computer power to solve complex mathematical equations to obtain a forecast based on current weather conditions. In this article, we propose a lightweight data-driven weather forecasting model by exploring state-of-the-art deep learning techniques based on Artificial Neural Network (ANN). Weather information is captured by time-series data and thus, we explore the latest Long Short-Term Memory (LSTM) layered model, which is a specialised form of Recurrent Neural Network (RNN) for weather prediction. The aim of this research is to develop and evaluate a short-term weather forecasting model using the LSTM and evaluate the accuracy compared to the well-established Weather Research and Forecasting (WRF) NWP model. The proposed deep model consists of stacked LSTM layers that uses surface weather parameters over a given period of time for weather forecasting. The model is experimented with different number of LSTM layers, optimisers, and learning rates and optimised for effective short-term weather predictions. Our experiment shows that the proposed lightweight model produces better results compared to the well-known and complex WRF model, demonstrating its potential for efficient and accurate short-term weather forecasting.
Original languageEnglish
Pages (from-to)382-390
Number of pages9
JournalIFIP Advances in Information and Communication Technology
Volume559
DOIs
Publication statusPublished - 12 May 2019

Fingerprint

Weather forecasting
Long short-term memory
Weather
Recurrent neural networks
Time series
Neural networks

Keywords

  • Long Short-Term Memory
  • Numerical Weather Prediction
  • WRF
  • Surface Weather Parameters
  • time-series data analysis

Cite this

@article{eec06ecb97bc471b938cf8e0ed075483,
title = "Long-Short Term Memory for an Effective Short-Term Weather Forecasting Model Using Surface Weather Data",
abstract = "Numerical Weather Prediction (NWP) requires considerable computer power to solve complex mathematical equations to obtain a forecast based on current weather conditions. In this article, we propose a lightweight data-driven weather forecasting model by exploring state-of-the-art deep learning techniques based on Artificial Neural Network (ANN). Weather information is captured by time-series data and thus, we explore the latest Long Short-Term Memory (LSTM) layered model, which is a specialised form of Recurrent Neural Network (RNN) for weather prediction. The aim of this research is to develop and evaluate a short-term weather forecasting model using the LSTM and evaluate the accuracy compared to the well-established Weather Research and Forecasting (WRF) NWP model. The proposed deep model consists of stacked LSTM layers that uses surface weather parameters over a given period of time for weather forecasting. The model is experimented with different number of LSTM layers, optimisers, and learning rates and optimised for effective short-term weather predictions. Our experiment shows that the proposed lightweight model produces better results compared to the well-known and complex WRF model, demonstrating its potential for efficient and accurate short-term weather forecasting.",
keywords = "Long Short-Term Memory, Numerical Weather Prediction, WRF, Surface Weather Parameters, time-series data analysis",
author = "{GALBOKKA HEWAGE}, {PRADEEP RUWAN PADMASIRI} and ARDHENDU BEHERA and MARCELLO TROVATI and ELLA PEREIRA",
year = "2019",
month = "5",
day = "12",
doi = "10.1007/978-3-030-19823-7_32",
language = "English",
volume = "559",
pages = "382--390",
journal = "IFIP Advances in Information and Communication Technology",
issn = "1868-4238",
publisher = "Springer New York",

}

TY - JOUR

T1 - Long-Short Term Memory for an Effective Short-Term Weather Forecasting Model Using Surface Weather Data

AU - GALBOKKA HEWAGE, PRADEEP RUWAN PADMASIRI

AU - BEHERA, ARDHENDU

AU - TROVATI, MARCELLO

AU - PEREIRA, ELLA

PY - 2019/5/12

Y1 - 2019/5/12

N2 - Numerical Weather Prediction (NWP) requires considerable computer power to solve complex mathematical equations to obtain a forecast based on current weather conditions. In this article, we propose a lightweight data-driven weather forecasting model by exploring state-of-the-art deep learning techniques based on Artificial Neural Network (ANN). Weather information is captured by time-series data and thus, we explore the latest Long Short-Term Memory (LSTM) layered model, which is a specialised form of Recurrent Neural Network (RNN) for weather prediction. The aim of this research is to develop and evaluate a short-term weather forecasting model using the LSTM and evaluate the accuracy compared to the well-established Weather Research and Forecasting (WRF) NWP model. The proposed deep model consists of stacked LSTM layers that uses surface weather parameters over a given period of time for weather forecasting. The model is experimented with different number of LSTM layers, optimisers, and learning rates and optimised for effective short-term weather predictions. Our experiment shows that the proposed lightweight model produces better results compared to the well-known and complex WRF model, demonstrating its potential for efficient and accurate short-term weather forecasting.

AB - Numerical Weather Prediction (NWP) requires considerable computer power to solve complex mathematical equations to obtain a forecast based on current weather conditions. In this article, we propose a lightweight data-driven weather forecasting model by exploring state-of-the-art deep learning techniques based on Artificial Neural Network (ANN). Weather information is captured by time-series data and thus, we explore the latest Long Short-Term Memory (LSTM) layered model, which is a specialised form of Recurrent Neural Network (RNN) for weather prediction. The aim of this research is to develop and evaluate a short-term weather forecasting model using the LSTM and evaluate the accuracy compared to the well-established Weather Research and Forecasting (WRF) NWP model. The proposed deep model consists of stacked LSTM layers that uses surface weather parameters over a given period of time for weather forecasting. The model is experimented with different number of LSTM layers, optimisers, and learning rates and optimised for effective short-term weather predictions. Our experiment shows that the proposed lightweight model produces better results compared to the well-known and complex WRF model, demonstrating its potential for efficient and accurate short-term weather forecasting.

KW - Long Short-Term Memory

KW - Numerical Weather Prediction

KW - WRF

KW - Surface Weather Parameters

KW - time-series data analysis

U2 - 10.1007/978-3-030-19823-7_32

DO - 10.1007/978-3-030-19823-7_32

M3 - Conference proceeding article (ISSN)

VL - 559

SP - 382

EP - 390

JO - IFIP Advances in Information and Communication Technology

JF - IFIP Advances in Information and Communication Technology

SN - 1868-4238

ER -