기계학습의 LSTM을 적용한 지상 기상변수 예측모델 개발
Development of Surface Weather Forecast Model by using LSTM Machine Learning Method
2
조회
0
다운로드
2
인용
0
참고문헌
키워드
초록
Numerical weather prediction (NWP) models play an essential role in predicting weather factors, but using them is challenging due to various factors. To overcome the difficulties of NWP models, deep learning models have been deployed in weather forecasting by several recent studies. This study adapts long short-term memory (LSTM), which demonstrates remarkable performance in time-series prediction. The combination of LSTM model input of meteorological features and activation functions have a significant impact on the performance therefore, the results from 5 combinations of input features and 4 activation functions are analyzed in 9 Automated Surface Observing System (ASOS) stations corresponding to cities/ islands/mountains. The optimized LSTM model produces better performance within eight forecast hours than Local Data Assimilation and Prediction System (LDAPS) operated by Korean meteorological administration. Therefore, this study illustrates that this LSTM model can be usefully applied to very short-term weather forecasting, and further studies about CNN-LSTM model with 2-D spatial convolution neural network (CNN) coupled in LSTM are required for improvement.
Abstract
Numerical weather prediction (NWP) models play an essential role in predicting weather factors, but using them is challenging due to various factors. To overcome the difficulties of NWP models, deep learning models have been deployed in weather forecasting by several recent studies. This study adapts long short-term memory (LSTM), which demonstrates remarkable performance in time-series prediction. The combination of LSTM model input of meteorological features and activation functions have a significant impact on the performance therefore, the results from 5 combinations of input features and 4 activation functions are analyzed in 9 Automated Surface Observing System (ASOS) stations corresponding to cities/ islands/mountains. The optimized LSTM model produces better performance within eight forecast hours than Local Data Assimilation and Prediction System (LDAPS) operated by Korean meteorological administration. Therefore, this study illustrates that this LSTM model can be usefully applied to very short-term weather forecasting, and further studies about CNN-LSTM model with 2-D spatial convolution neural network (CNN) coupled in LSTM are required for improvement.
참고문헌 (21)
- Jiandong Chen, "Factor decomposition and prediction of solar energy consumption in the United States", Journal of Cleaner Production, 234, pp.1210~1220 (2019), Elsevier BV[DOI]
- Amir Farzad, "A comparative performance analysis of different activation functions in LSTM networks for classification", Neural Computing and Applications, 31(7), pp.2507~2521 (2019), Springer Science and Business Media LLC[DOI]
- Cong Feng, "A data-driven multi-model methodology with deep feature selection for short-term wind forecasting", Applied Energy, 190, pp.1245~1257 (2017), Elsevier BV[DOI]
- Gal, Y., "A theoretically grounded application of dropout in recurrent neural network", Preprints, 30th Conference on Neural Information Processing Systems, pp.14~ (2016)
- Pradeep Hewage, "Deep learning-based effective fine-grained weather forecasting model", Pattern Analysis and Applications, 24(1), pp.343~366 (2021), Springer Science and Business Media LLC[DOI]
- Holton, J., "An Introduction to Dynamic Meteorology", pp.552~ (2012), Academic Press
- Hochreiter, S., "Long short-term memory", Neural Comput., 9, pp.1735~1780 (1997)
- 홍성유, "Dynamical Downscaling: Fundamental Issues from an NWP Point of View and Recommendations", Asia-Pacific Journal of Atmospheric Sciences, 50(1), pp.83~104 (2014), 한국기상학회
- Jozefowicz, R., "An empirical exploration of recurrent network architectures", Proc., The 32nd International conference on machine learning, 37, pp.2342~2350 (2015), ICML
- Eliyahu Kiperwasser, "Simple and Accurate Dependency Parsing Using Bidirectional LSTM Feature Representations", Transactions of the Association for Computational Linguistics, 4, pp.313~327 (2016), MIT Press - Journals[DOI]
- Validation of Numerical Prediction System (2019), "KMA", Numerical Modeling Center, Korea Meteorological Administration, 297~, pp.11-1360709-000001-10 (2020)
- Lea, C., "Temporal convolutional networks for action segmentation and detection", Proc., 30th IEEE Conference on Computer Vision and Pattern Recognition (CVPR), pp.1003~1012 (2017), Institute of Electrical and Electronics Engineers Inc
- Babita Majhi, "Improved prediction of daily pan evaporation using Deep-LSTM model", Neural Computing and Applications, 32(12), pp.7823~7838 (2020), Springer Science and Business Media LLC[DOI]
- S. Poornima, "Prediction of Rainfall Using Intensified LSTM Based Recurrent Neural Network with Weighted Linear Units", Atmosphere, 10(11), pp.668~ (2019), MDPI AG[DOI]
- Rumelhart, D. E., "Learning representations by back-propagating errors", Nature, 323, pp.533~536 (1986)
- Srivastava, N., "Dropout: a simple way to prevent neural networks from overfitting", J. Machine Learn. Res., 15, pp.1929~1958 (2014)
- Tieleman, T., "Lecture 6.5-rmsprop:Divide the gradient by a running average of its recent magnitude", COURSERA: Neural Networks for Machine Learning, 4, pp.26~31 (2012)
- "On the origin of deep learning" (2017)[URL]
- Bailin Yang, "Traffic flow prediction using LSTM with feature enhancement", Neurocomputing, 332, pp.320~327 (2019), Elsevier BV[DOI]
- Qin Zhang, "Prediction of Sea Surface Temperature Using Long Short-Term Memory", IEEE Geoscience and Remote Sensing Letters, 14(10), pp.1745~1749 (2017), Institute of Electrical and Electronics Engineers (IEEE)[DOI]
- Tianjun Zhang, "Research on Gas Concentration Prediction Models Based on LSTM Multidimensional Time Series", Energies, 12(1), pp.161~ (2019), MDPI AG[DOI]