Prediction of Extreme Precipitation Events Based on LSTM-Self Attention Model
Authors: 
Guilan Luo, Anqi Cao, Xin Ma, Anshun Hu, Caikui WangAuthors Info & Claims
Pages 91 - 97
Abstract
In view of the problems of large errors and low efficiency in the current prediction methods of extreme precipitation events, this paper combines the self-attention mechanism with the LSTM neural network and proposes the LSTM-SelfAttention model for prediction of extreme precipitation events. This model captures long-term dependencies in time series data through the LSTM neural network and uses the self-attention mechanism to identify key features in the data series, fully combining the advantages of the two. Finally, based on Kunming's precipitation data from 1961 to 2020, a simulation comparison test was conducted between this model and the LSTM model and BP neural network model. The results showed that the RMSE and MAE of the LSTM-SelfAttention model were lower than other models, and the prediction accuracy was compared to the comparison. The LSTM model with better results in the experiment improved by 28%, has higher accuracy and reliability, faster convergence speed, and better performance. Therefore, the model proposed in this article can accurately predict extreme precipitation events and provide strong support and help for disaster prevention and reduction.
References
[1]
He Shengbing, Zhu Yunliang. Analysis of social adaptation strategies of disaster immigrants under the background of extreme climate change [J]. Water Conservancy Economics, 2019, 37(05): 73-76+80
[2]
Zhai Panmao, Liu Jing. Extreme weather and climate events and disaster prevention and mitigation in the context of climate warming [J]. Chinese Engineering Science, 2012, 14(09): 55-63+84.
[3]
Xu Yinlong, Zhao Yuncheng, Zhai Panmao. New understanding and enlightenment of IPCC Special Report SRCCL on climate change and food security[J]. Progress in Climate Change Research, 2020,16(01): 37-49.
[4]
Sun Peng, Xiao Mingzhong, Zhang Qiang, Research progress on hydrometeorological extreme events [J]. Journal of Wuhan University (Science Edition), 2018, 64(01): 28-36.
[5]
Li J, Wang B. Predictability of summer extreme precipitation days over eastern China [J]. Climate Dyn, 2018, 51(11): 4543-4554.
[6]
Chen Husheng. Analysis and prediction of precipitation characteristics in Anhui Province based on wavelet analysis [D]. Hefei University of Technology, 2019.
[7]
Wang Xun, Wan Dingsheng, Tang Juan. Extreme precipitation prediction method based on combination model [J]. Computer Simulation, 2015, 32(12): 354-359
[8]
Burlando P, Rosso R, Cadavid L G, Forecasting of short-term rainfall using ARMA models[J]. Journal of Hydrology, 1993,144(1-4):193-211.
[9]
Feng Jun, Pan Fei. An LSTM-BP multi-model combination hydrological forecasting method [J]. Computers and Modernization, 2018, 34(7): 82-85.
[10]
Liu Xin, Zhao Ning, Guo Jinyun, Guo Bin. Monthly precipitation prediction on the Tibetan Plateau based on LSTM neural network [J]. Journal of Earth Information Science, 2020, 22(08): 1617-1629.
[11]
Zhao Yixin, Li Wei, Zhu Jiaming. Quantitative analysis and prediction of extreme precipitation based on GA-BP neural network [J]. Journal of Inner Mongolia Normal University (Natural Science Chinese Edition), 2022, 51(06): 576-581.
[12]
Cui Wei, Gu Ranhao, Chen Benyue, Comparison of the application of BP and LSTM neural networks in hydrological forecasting of small watersheds in Fujian [J]. People's Pearl River, 2020, 41 (2): 74-84.
[13]
Zhang Jixue, Wang Peng, Zhang Lin, Research on the application of artificial neural network in short-term precipitation prediction [J]. Science and Technology Wind, 2016(17): 123-124
[14]
Li Yu. Bitcoin price prediction analysis of LSTM neural network under the deep learning framework [D]. Shanxi University of Finance and Economics, 2023.
[15]
ZHENG H F, LIN F, FENG X X, A hybrid deep learning model with attention-based Conv-LSTM networks for shortterm traffic flow prediction[J]. IEEE Transactions on Intelligent Transportation Systems, 2021, 22(11) :6910-6920.
[16]
Hochreiter S, Schmidhuber J. Long short-term memory [J]. Neural Computation, 1997, 9(8):1735-1780.
[17]
Liang Hongtao, Liu Shuo, Du Junwei, A review of research on deep learning applied to time series prediction [J]. Computer Science and Exploration, 2023, 17(06): 1285-1300.
[18]
Huang Jianqiang, Qin Liangxi. Research on correlation prediction of temperature and rainfall using ALSTM model [J]. Journal of Guangxi University (Natural Science Edition), 2021, 46(04): 1024-1035.
[19]
Li Zhanwu, Zhang Shuai, Qiao Yingfeng, Air combat target maneuvering trajectory prediction based on self-attention mechanism and CNN-LSTM [J]. Transactions of Ordnance and Equipment Engineering, 2023, 44(07): 209-216.
[20]
VASWANI A, SHAZEER N, PARMAR N, Attention is all you need [C]// Proceedings of the 31st International Conference on Neural Information Processing Systems, Curran Associates Inc. 2017: 6000–6010.
[21]
Hu Yanli, Tong Tanqian, Zhang Xiaoyu, Deep learning sentiment analysis method integrating self-attention mechanism [J]. Computer Science, 2022, 49(1): 252-258
Index Terms
- Prediction of Extreme Precipitation Events Based on LSTM-Self Attention Model
Index terms have been assigned to the content through auto-classification.
Recommendations
Forecasting Monthly Maximum 5-Day Precipitation Using Artificial Neural Networks with Initial Lags
ISCID '13: Proceedings of the 2013 Sixth International Symposium on Computational Intelligence and Design - Volume 02Successive days of precipitation are known to cause flood in monsoon-type countries. Forecasting of daily precipitation helps to predict the occurrences of rainfall and number of wet days while with a maximum 5-day precipitation, we can predict the ...
Prediction based on traditional network prediction model and LSTM deep neural network model
CNIOT '23: Proceedings of the 2023 4th International Conference on Computing, Networks and Internet of ThingsThe total retail sales of consumer goods is an important index to measure a country’s economic development. For time series data sets, most traditional prediction models often do not make full use of the potential spatial correlation between variable ...
Two-stream convolutional LSTM for precipitation nowcasting
AbstractReliable precipitation nowcasting is essential to many fields, which can guide people to reasonably carry out production activities and respond to rainstorm disasters. However, precipitation nowcasting is a very challenging task because of ...
Comments
Please enable JavaScript to view thecomments powered by Disqus.Information & Contributors
Information
Published In
January 2024
297 pages
Copyright © 2024 ACM.
Permission to make digital or hard copies of all or part of this work for personal or classroom use is granted without fee provided that copies are not made or distributed for profit or commercial advantage and that copies bear this notice and the full citation on the first page. Copyrights for components of this work owned by others than the author(s) must be honored. Abstracting with credit is permitted. To copy otherwise, or republish, to post on servers or to redistribute to lists, requires prior specific permission and/or a fee. Request permissions from [email protected].
Publisher
Association for Computing Machinery
New York, NY, United States
Publication History
Published: 08 March 2024
Check for updates
Author Tags
Qualifiers
- Research-article
- Research
- Refereed limited
Funding Sources
- The National Natural Science Fund project: Research on the Network Ecological Behavior of the Universal Habitat of Erhai Wetland Insect Community (No.61661001)
- Academician Wang Jingxiu Workstation Project of Yunnan Province
Conference
CCEAI 2024
CCEAI 2024: 2024 8th International Conference on Control Engineering and Artificial Intelligence
January 26 - 28, 2024
Shanghai, China
Contributors
Other Metrics
Bibliometrics & Citations
Bibliometrics
Article Metrics
- 0Total Citations
- 61Total Downloads
- Downloads (Last 12 months)61
- Downloads (Last 6 weeks)5
Reflects downloads up to 16 Feb 2025
Other Metrics
Citations
View Options
Login options
Check if you have access through your login credentials or your institution to get full access on this article.
Sign inFull Access
View options
View or Download as a PDF file.
PDFeReader
View online with eReader.
eReaderHTML Format
View this article in HTML Format.
HTML Format