1,008 Hits in 6.0 sec

Precipitation Nowcasting with Star-Bridge Networks [article]

Yuan Cao, Qiuying Li, Hongming Shan, Zhizhong Huang, Lei Chen, Leiming Ma, Junping Zhang
2019 arXiv   pre-print
Precipitation nowcasting, which aims to precisely predict the short-term rainfall intensity of a local region, is gaining increasing attention in the artificial intelligence community.  ...  Existing deep learning-based algorithms use a single network to process various rainfall intensities together, compromising the predictive accuracy.  ...  Following the long short-term memory model, ConvLSTM utilizes its recurrent neural network architecture to memorize temporal information in a video sequence and extracts the spatial feature maps by using  ... 
arXiv:1907.08069v2 fatcat:ggklxfd2cfbb3muskesvpuvdwa

Strong Spatiotemporal Radar Echo Nowcasting Combining 3DCNN and Bi-Directional Convolutional LSTM

Suting Chen, Song Zhang, Huantong Geng, Yaodeng Chen, Chuang Zhang, Jinzhong Min
2020 Atmosphere  
The model first constructs dimensions of input data and gets 3D tensor data with spatiotemporal features, extracts local short-term spatiotemporal features of radar echoes through 3D convolution networks  ...  ) combining 3DCNN and bi-directional convolutional long short-term memory.  ...  The proposed model realizes accurate prediction of future short-term echo images.  ... 
doi:10.3390/atmos11060569 fatcat:6eqoh3gpyvbwbn5rzn2hih4flm

Short-Term Intensive Rainfall Forecasting Model Based on a Hierarchical Dynamic Graph Network

Huosheng Xie, Rongyao Zheng, Qing Lin
2022 Atmosphere  
To fully model the correlations among data, the model uses a dynamically constructed graph convolution operator to model the spatial correlation, a recurrent structure to model the time correlation, and  ...  Accurate short-term forecasting of intensive rainfall has high practical value but remains difficult to achieve.  ...  Classical models for time-sequence prediction include Long Short-Term Memory (LSTM) [15] , which is a recurrent neural network with long-and short-term memory cells, and the Deep Belief Network (DBN)  ... 
doi:10.3390/atmos13050703 doaj:a05f4d61e6dd415cacc90d99df722b1f fatcat:g7vfpn3lrjbnnmcnqpwj5nlhie

Rain Fall Prediction using Data Mining Techniques with Modernistic Schemes and Well-Formed Ideas

This paper contains some of the best work done in rain fall prediction using data mining techniques.  ...  There are large number of meteorologist all over the world who are trying their level best to predict the aspects of environment using data mining techniques.  ...  CONCLUSION In this paper, we have explained about some of the best work in the rainfall predictions using data mining techniques. We deeply explained about their significance and agendas of the work.  ... 
doi:10.35940/ijitee.a4011.119119 fatcat:igoerzmvfzfq5b6wzyrcvd7fwy

Rainfall Analysis and Forecasting Using Deep Learning Technique

Pragati Kanchan, Department of Computer Science and Engineering, School of Engineering, MIT ADT University, Pune, India
2021 Journal of Informatics Electrical and Electronics Engineering (JIEEE)  
Three deep learning methods have been used for prediction such as Artificial Neural Network (ANN) - Feed Forward Neural Network, Simple Recurrent Neural Network (RNN), and the Long Short-Term Memory (LSTM  ...  Rainfall forecasting is very challenging due to its uncertain nature and dynamic climate change. It's always been a challenging task for meteorologists.  ...  Rainfall Qiu, M. et al. [9] in this paper, proposed a Multi-Task Convolution Neural Network(MT-CNN) model for rainfall prediction.  ... 
doi:10.54060/jieee/002.02.015 fatcat:4yg6ha6qqngqxf4sgk7bcuhpwm

Multi-step rainfall forecasting using deep learning approach

Sanam Narejo, Muhammad Moazzam Jawaid, Shahnawaz Talpur, Rizwan Baloch, Eros Gian Alessandro Pasero
2021 PeerJ Computer Science  
The results suggest that the temporal DBN model outperforms the conventional Convolutional Neural Network (CNN) specifically on rainfall time series forecasting.  ...  The focus of this work is direct prediction of multistep forecasting, where a separate time series model for each forecasting horizon is considered and forecasts are computed using observed data samples  ...  During our research, it was observed that the parameters required to predict rainfall were enormously complex and subtle even for a short term period.  ... 
doi:10.7717/peerj-cs.514 pmid:34013036 pmcid:PMC8114799 fatcat:7j7noefzmvfgpemldmvocklzce

Short-term Hourly Streamflow Prediction with Graph Convolutional GRU Networks [article]

Muhammed Sit, Bekir Demiray, Ibrahim Demir
2021 arXiv   pre-print
This paper presents a Graph Convolutional GRUs based model to predict the next 36 hours of streamflow for a sensor location using the upstream river network.  ...  short-term streamflow prediction.  ...  ., 2018) proposed a Long short-term memory (LSTM) (Hochreiter & Schmidhuber, 1997 ) model that predicts the hourly streamflow from 1 to 6 hours lead time.  ... 
arXiv:2107.07039v1 fatcat:x7zhkqjxjzetdatd2hj7mfgna4

Convolutional LSTM Architecture for Precipitation Nowcasting Using Satellite Data

Carlos Javier Gamboa-Villafruela, José Carlos Fernández-Alvarez, Maykel Márquez-Mijares, Albenis Pérez-Alarcón, Alfo José Batista-Leyva
2021 Environmental Sciences Proceedings  
In this research, we propose an LSTM convolutional neural network (CNN-LSTM) architecture for immediate prediction of various short-term precipitation events using satellite data.  ...  The trained neural network model is used to predict the sixteenth precipitation data of the corresponding fifteen precipitation sequence and up to a time interval of 180 min.  ...  Recent advances in deep learning, especially recurrent neural network (RNN) and long short-term memory (LSTM) models [8] [9] [10] [11] [12] [13] [14] [15] [16] , provide some useful insights on how to  ... 
doi:10.3390/ecas2021-10340 fatcat:itvmqdx2ibdy3av6xo6ybhmf6i

Cloud Cover Nowcasting with Deep Learning [article]

Léa Berthomier, Bruno Pradel, Lior Perez
2020 arXiv   pre-print
Nowcasting is a field of meteorology which aims at forecasting weather on a short term of up to a few hours.  ...  Following recent deep learning successes on multiple imagery tasks, we applied deep convolutionnal neural networks on Meteosat satellite images for cloud cover nowcasting.  ...  METHODOLOGY Our goal was to forecast the position of clouds on a short time scale of 1h30' from satellite images using deep convolutional networks as our models.  ... 
arXiv:2009.11577v2 fatcat:6yoesbkshvewvoywrs7h7bm2ua

Prediction of Rainfall in Rajasthan, India using Deep and Wide Neural Network [article]

Vikas Bajpai, Anukriti Bansal, Kshitiz Verma, Sanjay Agarwal
2020 arXiv   pre-print
For wide network, instead of using rainfall intensity values directly, we are using features obtained after applying a convolutional layer. For deep part, a multi-layer perceptron (MLP) is used.  ...  Accurate prediction of rainfall intensity is a challenging task and its exact prediction helps in every aspect.  ...  We compare the proposed model with similar advance deep-learning-based models like multilayer perceptron, convolutional neural network and long-short-term-memory-based recurrent neural network.  ... 
arXiv:2010.11787v1 fatcat:wueroov6qvcr5nzwvjfzixeoaa

Survey on the Application of Deep Learning in Extreme Weather Prediction

Wei Fang, Qiongying Xue, Liang Shen, Victor S. Sheng
2021 Atmosphere  
These include the ability to use recurrent neural networks to predict weather phenomena and convolutional neural networks to predict the weather.  ...  They can automatically extract image features of extreme weather phenomena and predict the possibility of extreme weather somewhere by using a deep learning framework.  ...  The Convolutional neural network is a deep neural network that can achieve the latest accuracy and high performance for most computer vision tasks.  ... 
doi:10.3390/atmos12060661 fatcat:viaddv3vvfa6vfzxobybtcdgvy

High-resolution rainfall-runoff modeling using graph neural network [article]

Zhongrun Xiang, Ibrahim Demir
2021 arXiv   pre-print
Time-series modeling has shown great promise in recent studies using the latest deep learning algorithms such as LSTM (Long Short-Term Memory).  ...  In this paper, we propose the GNRRM (Graph Neural Rainfall-Runoff Model), a novel deep learning model that makes full use of spatial information from high-resolution precipitation data, including flow  ...  Because rainfall-runoff modeling is a time-series task, the LSTM (Long Short-Term Memory) advanced deep learning model has been used for hourly [3, 4] , daily [5, 6] , and monthly [7] modeling since  ... 
arXiv:2110.10833v1 fatcat:ztial2jpkrecjebt6zgyuldznu

PTCT: Patches with 3D-Temporal Convolutional Transformer Network for Precipitation Nowcasting [article]

Ziao Yang, Xiangrui Yang, Qifeng Lin
2022 arXiv   pre-print
Precipitation nowcasting is to predict the future rainfall intensity over a short period of time, which mainly relies on the prediction of radar echo sequences.  ...  Though convolutional neural network (CNN) and recurrent neural network (RNN) are widely used to generate radar echo frames, they suffer from inductive bias (i.e., translation invariance and locality) and  ...  The improved predictive recurrent neural network (PredRNN++) proposes a new recurrent structure named Causal LSTM for modeling short-term dependencies, which adds more non-linear layers to recurrent transition  ... 
arXiv:2112.01085v2 fatcat:gsh2kywa2rfa3nzevbuga7sfca

A Multi-task Two-stream Spatiotemporal Convolutional Neural Network for Convective Storm Nowcasting [article]

W. Zhang, H. Liu, P. Li, L. Han
2020 arXiv   pre-print
We integrate two-stream multi-task learning into a single convolutional neural network.  ...  Further, considering the relevance of classification and regression tasks, we develop a multi-task learning strategy that predicts the labels used in such tasks.  ...  Thus, we developed a simple, two-stream multi-task (TSMT) spatiotemporal convolutional neural network (CNN) using pixel-wise data sampling.  ... 
arXiv:2010.14100v1 fatcat:kfekfakgmjg25l6467v65ha7em

Wind speed prediction using multidimensional convolutional neural networks [article]

Kevin Trebing, Siamak Mehrkanoon
2020 arXiv   pre-print
This paper introduces a new model based on convolutional neural networks (CNNs) for wind speed prediction tasks.  ...  The proposed model is compared with traditional 2- and 3-dimensional CNN models, a 2D-CNN model with an attention layer and a 2D-CNN model equipped with upscaling and depthwise separable convolutions.  ...  For weather prediction tasks, convolutions are often used in conjunction with long short-term memory (LSTM).  ... 
arXiv:2007.12567v1 fatcat:rhq4fp4m5nbyhdu7hyfzllepha
« Previous Showing results 1 — 15 out of 1,008 results