基于時(shí)空?qǐng)D卷積的強(qiáng)對(duì)流降水臨近預(yù)報(bào)研究

doi:10.12306/2024jms.0023

首頁(yè) > 過(guò)刊瀏覽>2024年第44卷第3期 >487-497. DOI:10.12306/2024jms.0023

基于時(shí)空?qǐng)D卷積的強(qiáng)對(duì)流降水臨近預(yù)報(bào)研究
DOI:
                        10.12306/2024jms.0023
                    
CSTR:
                        
                    
作者:
                        
                        
                    
作者單位:
作者簡(jiǎn)介:
通訊作者:
中圖分類號(hào):P412.25
基金項(xiàng)目:國(guó)家自然科學(xué)基金資助項(xiàng)目(42075007);蘇州大學(xué)計(jì)算機(jī)信息處理技術(shù)重點(diǎn)實(shí)驗(yàn)室開(kāi)放項(xiàng)目(KJS2275);南京氣象科技創(chuàng)新研究院北極閣開(kāi)放研究基金資助項(xiàng)目(BJG202306);江蘇省研究生科研與實(shí)踐創(chuàng)新計(jì)劃項(xiàng)目(NO.KYCX23_1388)

A study on the proximity prediction of strong convective precipitation based on the spatio-temporal graph convolution

Author:

Affiliation:

Fund Project:

摘要

圖/表

訪問(wèn)統(tǒng)計(jì)

參考文獻(xiàn)

相似文獻(xiàn)

引證文獻(xiàn)

資源附件

文章評(píng)論

摘要:

降水臨近預(yù)報(bào)對(duì)于強(qiáng)對(duì)流天氣的預(yù)報(bào)具有重要的支撐作用。氣象業(yè)務(wù)中主要采用雷達(dá)回波外推方法解決此問(wèn)題。然而,現(xiàn)有方法通常缺乏從序列雷達(dá)數(shù)據(jù)中有效學(xué)習(xí)的能力,導(dǎo)致預(yù)測(cè)精度不佳。為了解決這一問(wèn)題,本文提出了一種改進(jìn)的時(shí)空?qǐng)D卷積模型ASTGCN(A Spatio-Temporal Graph Convolution Neural Network)用于強(qiáng)對(duì)流降水的臨近預(yù)報(bào)。利用時(shí)空?qǐng)D卷積網(wǎng)絡(luò),有效地捕獲相鄰雷達(dá)幀之間的時(shí)間依賴性。此外,利用注意力機(jī)制和自動(dòng)編碼器來(lái)增強(qiáng)模型捕獲時(shí)空相關(guān)性的能力。結(jié)果表明,該模型可以從數(shù)據(jù)中發(fā)現(xiàn)隱藏的圖結(jié)構(gòu),從而捕獲隱藏的空間關(guān)系。與現(xiàn)有模型(Transformer)相比,該模型的臨界成功指數(shù)(CSI)提高了28%,表明其在強(qiáng)對(duì)流降水臨近預(yù)報(bào)方面具有優(yōu)越的性能。

Abstract:

Precipitation nowcasting plays an important supporting role in forecasting severe convective weather. In meteorological services, the radar echo extrapolation method is mainly used to solve precipitation nowcasting problems. However, existing methods often lack the ability to effectively learn from sequential radar data, resulting in poor prediction accuracy. In order to solve this problem, this paper proposed ASTGCN(A Spatio-Temporal Graph Convolution Neural Network) for nowcasting of severe convective precipitation. Efficiently capture the temporal dependence between adjacent radar frames using a spatio-temporal graph convolutional network. In addition, an attention mechanism and an autoencoder were utilized to enhance the model's ability to capture spatio-temporal correlations. Experimental results show that the model can discover hidden graph structures from data and thereby capture hidden spatial relationships. Compared with the existing model (Transformer), the Critical Success Index (CSI) of this model is improved by 28%, indicating its superior performance in the nowcasting of severe convective precipitation.

參考文獻(xiàn)

相似文獻(xiàn)

引證文獻(xiàn)

引用本文

方巍,薛瓊瑩,陶恩屹,齊媚涵.基于時(shí)空?qǐng)D卷積的強(qiáng)對(duì)流降水臨近預(yù)報(bào)研究.氣象科學(xué),2024,44(3):487-497 FANG Wei, XUE Qiongying, TAO Enyi, QI Meihan. A study on the proximity prediction of strong convective precipitation based on the spatio-temporal graph convolution. Journal of the Meteorological Sciences,2024,44(3):487-497

復(fù)制

文章指標(biāo)

點(diǎn)擊次數(shù):
下載次數(shù):
HTML閱讀次數(shù):
引用次數(shù):

歷史

收稿日期:2023-10-09
最后修改日期:2024-04-03
錄用日期:
在線發(fā)布日期: 2024-07-16
出版日期:

文章二維碼

您是第位訪問(wèn)者
氣象科學(xué) ® 2026 版權(quán)所有
技術(shù)支持：北京勤云科技發(fā)展有限公司

抽查舔水白紧大视频,冲喜娘子超甜哒激情五月,亚洲黄色图片偷拍网站,亚洲三级黄片免费播放,91在线观看喷潮红桃,日韩精品在线观看一区二区,久久久久久久久久久久黄色片子,日产精品99久久久久久久久,欧洲亚洲另类图片小说

引用本文

分享

相關(guān)視頻

文章指標(biāo)

歷史

文章二維碼

抽查舔水白紧大视频,冲喜娘子超甜哒 激情五月,亚洲黄色图片偷拍网站,亚洲三级黄片免费播放,91在线观看喷潮红桃,日韩精品在线观看一区二区,久久久久久久久久久久黄色片子,日产精品99久久久久久久久,欧洲亚洲另类图片小说

引用本文

分享

相關(guān)視頻

文章指標(biāo)

歷史

文章二維碼

抽查舔水白紧大视频,冲喜娘子超甜哒激情五月,亚洲黄色图片偷拍网站,亚洲三级黄片免费播放,91在线观看喷潮红桃,日韩精品在线观看一区二区,久久久久久久久久久久黄色片子,日产精品99久久久久久久久,欧洲亚洲另类图片小说