Prediction of Precipitation Based on Recurrent Neural Networks in Jingdezhen, Jiangxi Province, China
Precipitation is a critical input for hydrologic simulation and prediction, and is widely used for agriculture, water resources management, and prediction of flood and drought, among other activities. Traditional precipitation prediction researches often established one or more probability models of...
Main Authors: | , , , , , |
---|---|
Format: | Article |
Language: | English |
Published: |
MDPI AG
2020-02-01
|
Series: | Atmosphere |
Subjects: | |
Online Access: | https://www.mdpi.com/2073-4433/11/3/246 |
id |
doaj-f2249adaa26e461a99760726ba48688e |
---|---|
record_format |
Article |
spelling |
doaj-f2249adaa26e461a99760726ba48688e2020-11-25T02:15:06ZengMDPI AGAtmosphere2073-44332020-02-0111324610.3390/atmos11030246atmos11030246Prediction of Precipitation Based on Recurrent Neural Networks in Jingdezhen, Jiangxi Province, ChinaJinle Kang0Huimin Wang1Feifei Yuan2Zhiqiang Wang3Jing Huang4Tian Qiu5State Key Laboratory of Hydrology-Water Resources and Hydraulic Engineering, Hohai University, Nanjing 210098, ChinaState Key Laboratory of Hydrology-Water Resources and Hydraulic Engineering, Hohai University, Nanjing 210098, ChinaState Key Laboratory of Hydrology-Water Resources and Hydraulic Engineering, Hohai University, Nanjing 210098, ChinaState Key Laboratory of Hydrology-Water Resources and Hydraulic Engineering, Hohai University, Nanjing 210098, ChinaState Key Laboratory of Hydrology-Water Resources and Hydraulic Engineering, Hohai University, Nanjing 210098, ChinaResearch Institute of Management Science, Business School, Hohai University, Nanjing, 211100, ChinaPrecipitation is a critical input for hydrologic simulation and prediction, and is widely used for agriculture, water resources management, and prediction of flood and drought, among other activities. Traditional precipitation prediction researches often established one or more probability models of historical data based on the statistical prediction methods and machine learning techniques. However, few studies have been attempted deep learning methods such as the state-of-the-art for Recurrent Neural Networks (RNNs) networks in meteorological sequence time series predictions. We deployed Long Short-Term Memory (LSTM) network models for predicting the precipitation based on meteorological data from 2008 to 2018 in Jingdezhen City. After identifying the correlation between meteorological variables and the precipitation, nine significant input variables were selected to construct the LSTM model. Then, the selected meteorological variables were refined by the relative importance of input variables to reconstruct the LSTM model. Finally, the LSTM model with final selected input variables is used to predict the precipitation and the performance is compared with other classical statistical algorithms and the machine learning algorithms. The experimental results show that the LSTM is suitable for precipitation prediction. The RNN models, combined with meteorological variables, could predict the precipitation accurately in Jingdezhen City and provide sufficient time to prepare strategies against potential related disasters.https://www.mdpi.com/2073-4433/11/3/246precipitation predictionmeteorological variableslstm |
collection |
DOAJ |
language |
English |
format |
Article |
sources |
DOAJ |
author |
Jinle Kang Huimin Wang Feifei Yuan Zhiqiang Wang Jing Huang Tian Qiu |
spellingShingle |
Jinle Kang Huimin Wang Feifei Yuan Zhiqiang Wang Jing Huang Tian Qiu Prediction of Precipitation Based on Recurrent Neural Networks in Jingdezhen, Jiangxi Province, China Atmosphere precipitation prediction meteorological variables lstm |
author_facet |
Jinle Kang Huimin Wang Feifei Yuan Zhiqiang Wang Jing Huang Tian Qiu |
author_sort |
Jinle Kang |
title |
Prediction of Precipitation Based on Recurrent Neural Networks in Jingdezhen, Jiangxi Province, China |
title_short |
Prediction of Precipitation Based on Recurrent Neural Networks in Jingdezhen, Jiangxi Province, China |
title_full |
Prediction of Precipitation Based on Recurrent Neural Networks in Jingdezhen, Jiangxi Province, China |
title_fullStr |
Prediction of Precipitation Based on Recurrent Neural Networks in Jingdezhen, Jiangxi Province, China |
title_full_unstemmed |
Prediction of Precipitation Based on Recurrent Neural Networks in Jingdezhen, Jiangxi Province, China |
title_sort |
prediction of precipitation based on recurrent neural networks in jingdezhen, jiangxi province, china |
publisher |
MDPI AG |
series |
Atmosphere |
issn |
2073-4433 |
publishDate |
2020-02-01 |
description |
Precipitation is a critical input for hydrologic simulation and prediction, and is widely used for agriculture, water resources management, and prediction of flood and drought, among other activities. Traditional precipitation prediction researches often established one or more probability models of historical data based on the statistical prediction methods and machine learning techniques. However, few studies have been attempted deep learning methods such as the state-of-the-art for Recurrent Neural Networks (RNNs) networks in meteorological sequence time series predictions. We deployed Long Short-Term Memory (LSTM) network models for predicting the precipitation based on meteorological data from 2008 to 2018 in Jingdezhen City. After identifying the correlation between meteorological variables and the precipitation, nine significant input variables were selected to construct the LSTM model. Then, the selected meteorological variables were refined by the relative importance of input variables to reconstruct the LSTM model. Finally, the LSTM model with final selected input variables is used to predict the precipitation and the performance is compared with other classical statistical algorithms and the machine learning algorithms. The experimental results show that the LSTM is suitable for precipitation prediction. The RNN models, combined with meteorological variables, could predict the precipitation accurately in Jingdezhen City and provide sufficient time to prepare strategies against potential related disasters. |
topic |
precipitation prediction meteorological variables lstm |
url |
https://www.mdpi.com/2073-4433/11/3/246 |
work_keys_str_mv |
AT jinlekang predictionofprecipitationbasedonrecurrentneuralnetworksinjingdezhenjiangxiprovincechina AT huiminwang predictionofprecipitationbasedonrecurrentneuralnetworksinjingdezhenjiangxiprovincechina AT feifeiyuan predictionofprecipitationbasedonrecurrentneuralnetworksinjingdezhenjiangxiprovincechina AT zhiqiangwang predictionofprecipitationbasedonrecurrentneuralnetworksinjingdezhenjiangxiprovincechina AT jinghuang predictionofprecipitationbasedonrecurrentneuralnetworksinjingdezhenjiangxiprovincechina AT tianqiu predictionofprecipitationbasedonrecurrentneuralnetworksinjingdezhenjiangxiprovincechina |
_version_ |
1724897786262978560 |