气候变化领域集成服务门户(CCPortal): Improving ECMWF short-term intensive rainfall forecasts using generative adversarial nets and deep belief networks

CCPortal

DOI	10.1016/j.atmosres.2020.105281
	Improving ECMWF short-term intensive rainfall forecasts using generative adversarial nets and deep belief networks
	Xie H.; Wu L.; Xie W.; Lin Q.; Liu M.; Lin Y.
发表日期	2021
ISSN	0169-8095
卷号	249
英文摘要	Short-term intensive rainfall (3-h rainfall amount > 30 mm) is a destructive weather phenomenon that is poorly predicted using traditional forecasting methods. In this study, we propose a model using European Center for Medium-Range Weather Forecasts (ECMWF) data and a machine learning framework to improve the ability of short-term intensive rainfall forecasting in Fujian Province, China. ECMWF forecast data and ground observation station data (2015–2018) were interpolated using a radial basis function, outliers were processed, and the data were blocked according to the monthly cumulative rainfall and forecast window. Subsequently, the box difference index was used to select features for each data block. As short-term intensive rainfall events are rare, a data processing method based on the K-means and generative adversarial nets was used to address data imbalances in the rainfall distribution. Finally, focal loss object detection was combined with a deep belief network to construct the short-term intensive rainfall classification model. The results show that the data preprocessing method and resampling method used in this study were effective. Furthermore, the classification model was superior to other machine learning methods for predicting short-term intensive rainfall. © 2020 Elsevier B.V.
英文关键词	Deep belief network; ECMWF; Fujian Province; Generative adversarial nets; Machine learning; Short-term intensive rainfall forecast
语种	英语
scopus关键词	Bayesian networks; Data handling; Deep learning; Learning systems; Object detection; Rain; Turing machines; Classification models; Data processing methods; Deep belief networks; European center for medium-range weather forecasts; Ground observations; Machine learning methods; Radial basis functions; Rainfall distribution; Weather forecasting; data processing; ensemble forecasting; machine learning; precipitation intensity; prediction; seasonal variation; weather forecasting; China; Fujian
来源期刊	Atmospheric Research
文献类型	期刊论文
条目标识符	http://gcip.llas.ac.cn/handle/2XKMVOVA/141664
作者单位	College of Mathematics and Computer Science, Fuzhou University, Fuzhou, China; Fujian Meteorological Observatory, Fuzhou, China; Fujian Key Laboratory of Severe Weather, Fuzhou, China
推荐引用方式 GB/T 7714	Xie H.,Wu L.,Xie W.,et al. Improving ECMWF short-term intensive rainfall forecasts using generative adversarial nets and deep belief networks[J],2021,249.
APA	Xie H.,Wu L.,Xie W.,Lin Q.,Liu M.,&Lin Y..(2021).Improving ECMWF short-term intensive rainfall forecasts using generative adversarial nets and deep belief networks.Atmospheric Research,249.
MLA	Xie H.,et al."Improving ECMWF short-term intensive rainfall forecasts using generative adversarial nets and deep belief networks".Atmospheric Research 249(2021).