Inductive Spatiotemporal Graph Convolutional Networks for Short-term Quantitative Precipitation Forecasting | |
Yajing, Wu; Xuebing, Yang; Yongqiang, Tang; Chenyang, Zhang; Guoping, Zhang; Wensheng, Zhang | |
刊名 | IEEE Transactions on Geoscience and Remote Sensing |
2022 | |
卷号 | 0期号:0页码:0 |
关键词 | Quantitative precipitation forecasting graph convolutional networks (GCN) spatiotemporal model radar-rain gauge data merging |
DOI | 10.1109/TGRS.2022.3159530 |
英文摘要 | Short-term Quantitative Precipitation Forecasting (SQPF) using weather radar is an important but challenging problem as one must cope with inherent nonlinearity and spatiotemporal correlation in the data. In this paper, we propose a novel deep learning model, named Inductive spatiotemporal Graph Convolutional Networks (InstGCN), to overcome these issues in SQPF. The proposed InstGCN can learn a nonlinear mapping from historical radar reflectivity to future rainfall amounts, and extract informative spatiotemporal representations simultaneously. Specifically, we first provide a formal definition for formulating the SQPF problem from a graph perspective. Then, based on radar reflectivity and rain gauge observation, we propose a novel graph construction approach which utilizes a special elliptic structure to model the spatial dependency of precipitation area. Additionally, a new Node level Differential Block (Node-DB) is introduced to tackle the non-stationary temporal dependency. To execute inductive graph learning for unseen nodes, we design to decompose a whole graph into sub-graphs. We conduct extensive experiments on three real-world datasets in East China and a public weather radar dataset in the south-eastern parts of France. The experimental results confirm the advantages of InstGCN compared with several state-of-the-arts. |
语种 | 英语 |
内容类型 | 期刊论文 |
源URL | [http://ir.ia.ac.cn/handle/173211/47443] |
专题 | 精密感知与控制研究中心_人工智能与机器学习 |
通讯作者 | Xuebing, Yang; Yongqiang, Tang |
作者单位 | 1.the Research Center of Precision Sensing and Control, Institute of Automation, Chinese Academy of Sciences 2.University of Chinese Academy of Sciences 3.the Public Meteorological Service Center of CMA 4.the State Key Laboratory of Management and Control for Complex Systems, Institute of Automation, Chinese Academy of Sciences |
推荐引用方式 GB/T 7714 | Yajing, Wu,Xuebing, Yang,Yongqiang, Tang,et al. Inductive Spatiotemporal Graph Convolutional Networks for Short-term Quantitative Precipitation Forecasting[J]. IEEE Transactions on Geoscience and Remote Sensing,2022,0(0):0. |
APA | Yajing, Wu,Xuebing, Yang,Yongqiang, Tang,Chenyang, Zhang,Guoping, Zhang,&Wensheng, Zhang.(2022).Inductive Spatiotemporal Graph Convolutional Networks for Short-term Quantitative Precipitation Forecasting.IEEE Transactions on Geoscience and Remote Sensing,0(0),0. |
MLA | Yajing, Wu,et al."Inductive Spatiotemporal Graph Convolutional Networks for Short-term Quantitative Precipitation Forecasting".IEEE Transactions on Geoscience and Remote Sensing 0.0(2022):0. |
个性服务 |
查看访问统计 |
相关权益政策 |
暂无数据 |
收藏/分享 |
除非特别说明,本系统中所有内容都受版权保护,并保留所有权利。
修改评论