Application of Data-driven Deep Learning Model in Global Precipitation Forecasting

Wan Liu, Wan Liu, Yongqiang Wang, Yongqiang Wang

2022

Abstract

With the improvement of data acquisition ability and the rapid increase of computer storage capacity and transmission rate, it is possible to solve the problem of precipitation prediction by using big data and deep learning. In this paper, the three most advanced deep learning models, namely Convolution model, ConvLSTM model and ConvGRU model, are applied to the study of precipitation prediction, and analyze the prediction ability of this method for global short-term precipitation. The experimental results show that the deep learning method can effectively predict global precipitation, and the correlation coefficient of precipitation prediction for the next 6 h is more than 0.75. The performance of convolution model is better when the prediction period is less than 12 h, Otherwise ConvLSTM model and ConvGRU model are more efficient. However, it is difficult to predict precipitation over northern Africa, the west coast of South America, the eastern coast of the South Pacific and the South Atlantic.

Paper Citation

in Harvard Style

Liu W. and Wang Y. (2022). Application of Data-driven Deep Learning Model in Global Precipitation Forecasting. In Proceedings of the International Conference on Big Data Economy and Digital Management - Volume 1: BDEDM, ISBN 978-989-758-593-7, pages 318-324. DOI: 10.5220/0011176400003440

in Bibtex Style

@conference{bdedm22,
author={Wan Liu and Yongqiang Wang},
title={Application of Data-driven Deep Learning Model in Global Precipitation Forecasting},
booktitle={Proceedings of the International Conference on Big Data Economy and Digital Management - Volume 1: BDEDM,},
year={2022},
pages={318-324},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0011176400003440},
isbn={978-989-758-593-7},
}

in EndNote Style

TY - CONF

JO - Proceedings of the International Conference on Big Data Economy and Digital Management - Volume 1: BDEDM,
TI - Application of Data-driven Deep Learning Model in Global Precipitation Forecasting
SN - 978-989-758-593-7
AU - Liu W.
AU - Wang Y.
PY - 2022
SP - 318
EP - 324
DO - 10.5220/0011176400003440