航空氣象預測對飛行安全至關重要，不正確的航空氣象預測不僅會給飛航管制員、簽派員和航空器駕駛員帶來困擾，更會增加航空事故發生的風險。為了提高航空氣象預測的準確度，本研究結合深度學習中的卷積神經網路模型(CNN)和遞歸神經網路中的長短期記憶模型(LSTM)以提高機場能見度之預測。本研究將CNN模型應用於1,500張不同大小的天氣圖形作為圖形預測之數據集，並將數據集分成訓練集1,050張、驗證集150張以及測試集300張。實驗結果顯示CNN模型訓練及預測之準確度為訓練集100.00%、驗證集97.33%以及測試集97.67%。多變量數值預測部分本研究採用中央氣象局於2010年至2020年松山機場觀測的10個天氣特徵資料作為數據集，並使用標準化、多元線性迴歸以及皮爾森積差相關係數作為LSTM模型訓練前的資料前處理。本研究經由資料前處理後的溫度預測誤差由RMSE 4.0274降低至2.2215以及MAPE由23.0538%降低至5.0069%。而能見度預測部分經由資料前處理並結合CNN模型預測之結果，將能見度依照其相對應的級別劃分作為一個新的特徵加入LSTM模型訓練可以提高其預測之準確度。本研究結果顯示，此方法能有效的提高能見度預測之準確度，其誤差由RMSE 12.8390降低至2.6798以及MAPE 85.1492% 降低至13.4126 %。

Weather forecasts are essential to aviation safety. Unreliable forecasts not only cause problems to pilots and air traffic controllers, but also lead to aviation accidents and incidents. To enhance the forecast accuracy, a convolutional neural network (CNN) model in deep learning is adopted to classify different weather images as precursor for long short-term memory (LSTM) to increase visibility forecasting. A dataset of 1,500 weather images is applied in preprocessing, resizing and later trained by the CNN model with training, validation, and testing accuracy 100.00%, 97.33%, and 97.67%, respectively. In addition, a numerical dataset of 10 weather features from 2010 to 2020 collected from the Central Weather Bureau of Taiwan is applied to temperature and visibility forecasting by using data standardization and feature selection of linear regression and Pearson’s correlation coefficients as data preprocessing in LSTM model. The temperature forecasting errors can be reduced from RMSE 4.0274 to 2.2215 and MAPE 23.0538% to 5.0069%. By integrating the CNN result and the LSTM, the visibility forecasting can be improved significantly. The results show that the visibility forecasting errors can be reduced from RMSE 12.8390 to 2.6798 and MAPE 85.1492% to 13.4126%. An effective forecast by combining associated weather image and numerical data can be applied to explore aviation weather.

[1] Richardson, L. F. Weather Prediction by Numerical Process, Cambridge University Press, United Kingdom, 2010.
[2] Charney, J. G., Fjörtoft, R. and Neumann, J. V. “Numerical Integration of the Barotropic Vorticity Equation,” Tellus: A Quarterly Journal of Geophysics, 2(4), pp. 237-254, 1950.
[3] Glahn, H. R. and Lowry, D. A. “The Use of Model Output Statistics (MOS) in Objective Weather Forecasting,” Journal of Applied Meteorology and Climatology, 11(8), pp. 1203-1211, 1972.
[4] McCulloch, W. S. and Pitts, W. “A Logical Calculus of the Ideas Immanent in Nervous Activity,” Bulletin of Mathematical Biophysics, 5, pp. 115-133, 1943.
[5] Minsky, M. and Papert, S. A. Perceptrons: An introduction to Computational Geometry, MIT Press, United States, 2017.
[6] Werbos, P. J. The Roots of Backpropagation: From Ordered Derivatives to Neural Networks and Political Forecasting, Wiley-Interscience Press, United States, 1994.
[7] Schizas, C. N., Michaelides, S., Pattichis, C. S. and Livesay, R. R. “Artificial Neural Networks in Forecasting Minimum Temperature (weather),” Artificial Neural Networks, pp. 112-114, 1991.
[8] Oishi, S., Ikebuchi, S. and Kojiri, T. “Severe Rainfall Prediction Method Using Artificial Intelligence,” IEEE-International Conference on Systems, Man, and Cybernetics, pp. 4820-4825, San Diego, 1998.
[9] Qi, H. and Zhang, M. “Rainfall Estimation Using M-PHONN Model,” International Joint Conference on Neural Networks, pp. 1620-1624, Washington, 2001.
[10] Jaruszewicz, M. and Mandziuk, J. “Application of PCA Method to Weather Prediction Task,” The 9th International Conference on Neural Information Processing, pp. 2359-2363, Singapore, 2002.
[11] Maqsood, I., Khan, M. R. and Abraham, A. “Intelligent Weather Monitoring Systems Using Connectionist Models,” Neural, Parallel & Scientific Computations, 10(2), pp. 157-178, 2002.
[12] Maqsood, I., Khan, M. R. and Abraham, A. Weather Forecasting Models Using Ensembles of Neural Networks, Springer Press: Intelligent Systems Design and Applications, Germany, pp. 33-42, 2003.
[13] Li, K. and Liu, Y. S. “A Rough Set Based Fuzzy Neural Network Algorithm for Weather Prediction,” International Conference on Machine Learning and Cybernetics, pp. 1888-1892, Guangzhou, 2005.
[14] Paras, S. M., Kumar, A. and Chandra, M. “A Feature Based Neural Network Model for Weather Forecasting,” International Journal of Computational Intelligence Systems, 4(3), pp. 209-216, 2008.
[15] Nayak, R., Patheja, P. S. and Waoo, A. A. “An Artificial Neural Network Model for Weather Forecasting in Bhopal,” IEEE-International Conference on Advances in Engineering, Science and Management, pp. 747-749, 2012.
[16] Shi, X., Chen, Z., Wang, H., Yeung, D. Y., Wong, W. K. and Woo, W. C. “Convolutional LSTM Network: A Machine Learning Approach for Precipitation Nowcasting,” Advances in Neural Information Processing Systems, 28, 2015.
[17] Jiang, Z. Y., Jia, Q. S. and Guan, X. H. “Review of Wind Power Forecasting Methods: From Multi-Spatial and Temporal Perspective,” The 36th Chinese Control Conference, pp. 10576-10583, 2017.
[18] Reichstein, M., Camps-Valls, G., Stevens, B., Jung, M., Denzler, J. and Carvalhais, N. “Deep Learning and Process Understanding for Data-Driven Earth System Science,” Nature, 566, pp. 195-204, 2019.
[19] Ravuri, S., Lenc, K., Willson, M., Kangin, D., Lam, R., Mirowski, P., Fitzsimons, M., Athanassiadou, M., Kashem, S., Madge, S., Prudden, R., Mandhane, A., Clark, A., Brock, A., Simonyan, K., Hadsell, R., Robinson, N., Clancy, E., Arribas, A. and Mohamed, S. “Skilful Precipitation Nowcasting Using Deep Generative Models of Radar,” Nature, 597, pp. 672-677, 2021.
[20] Collobert, R. and Weston, J. “A Unified Architecture for Natural Language Processing: Deep Neural Networks with Multitask Learning,” The 25th International Conference on Machine Learning, pp. 160-167, New York, 2008.
[21] Van den Oord, A., Dieleman, S. and Schrauwen, B. “Deep Content-Based Music Recommendation,” Advances in Neural Information Processing Systems, pp. 2643-2651, 2013.
[22] Tsantekidis, A., Passalis, N., Tefas, A., Kanniainen, J., Gabbouj, M. and Iosifidis, A. “Forecasting Stock Prices from the Limit Order Book Using Convolutional Neural Networks,” The 19th Conference on Business Informatics, pp. 7-12, Thessaloniki, 2017.
[23] Avilov, O., Rimbert, S., Popov, A. and Bougrain, L. “Deep Learning Techniques to Improve Intraoperative Awareness Detection from Electroencephalographic Signals,” The 42nd Annual International Conference of the IEEE Engineering in Medicine & Biology Society, pp. 142-145, Montreal, 2020.
[24] Elhoseiny, M., Huang, S. and Elgammal, A. “Weather Classification with Deep Convolutional Neural Networks,” IEEE-International Conference on Image Processing, pp. 3349-3353, 2015.
[25] Al-Haija, Q. A., Smadi, M. A. and Zein-Sabatto, S. “Multi-Class Weather Classification Using ResNet-18 CNN for Autonomous IoT and CPS Applications,” International Conference on Computational Science and Computational Intelligence, pp. 1586-1591, 2020.
[26] Zhao, B., Li, X., Lu, X. and Wang, Z. “A CNN-RNN Architecture for Multi-Label Weather Recognition,” Neurocomputing, 322, pp. 47-57, 2018.
[27] Weyn, J. A., Durran, D. R. and Caruana, R. “Improving Data‐Driven Global Weather Prediction Using Deep Convolutional Neural Networks on a Cubed Sphere,” Journal of Advances in Modeling Earth Systems, 12(9), 2020.
[28] Denby, L. “Discovering the Importance of Mesoscale Cloud Organization Through Unsupervised Classification,” Geophysical Research Letters, 47(1), 2020.
[29] Tan, L., Xuan, D., Xia, J. and Wang, C. “Weather Recognition Based on 3C-CNN,” KSII Transactions on Internet and Information Systems, 14(8), pp. 3567-3582, 2020.
[30] Krizhevsky, A., Sutskever, I. and Hinton, G. E. “Imagenet Classification with Deep Convolutional Neural Networks,” Advances in Neural Information Processing Systems, 25, pp. 1097-1105, 2012.
[31] Simonyan, K. and Zisserman, A. “Very Deep Convolutional Networks for Large-Scale Image Recognition,” International Conference on Learning Representations, 2015.
[32] He, K., Zhang, X., Ren, S. and Sun, J. “Deep Residual Learning for Image Recognition,” IEEE Conference on Computer Vision and Pattern Recognition, pp. 770-778, 2016.
[33] Gorricha, J., Lobo, V. and Costa, A. C. “A Framework for Exploratory Analysis of Extreme Weather Events Using Geostatistical Procedures and 3D Self-Organizing Maps,” International Journal on Advances in Intelligent Systems, 5, pp. 16-26, 2013.
[34] Liu, Y., Racah, E., Correa, J., Khosrowshahi, A., Lavers, D., Kunkel, K., Collins, W. and Wehner, M. “Application of Deep Convolutional Neural Networks for Detecting Extreme Weather in Climate Datasets,” Computer Vision and Pattern Recognition, 2016.
[35] Heinzler, R., Piewak, F., Schindler, P. and Stork, W. “CNN-Based Lidar Point Cloud De-Noising in Adverse Weather,” IEEE Robotics and Automation Letters, 5(2), pp. 2514-2521, 2020.
[36] Zhou, K., Zheng, Y., Li, B., Dong, W. and Zhang, X. “Forecasting Different Types of Convective Weather: A Deep Learning Approach,” Journal of Meteorological Research, 33(5), pp. 797-809, 2019.
[37] Weyn, J. A., Durran, D. R. and Caruana, R. “Can Machines Learn to Predict Weather? Using Deep Learning to Predict Gridded 500‐hPa Geopotential Height from Historical Weather Data,” Journal of Advances in Modeling Earth Systems, 11(8), pp. 2680-2693, 2019.
[38] Kashinath, K., Mudigonda, M., Kim, S., Kapp-Schwoerer, L., Graubner, A., Karaismailoglu, E., Von Kleist, L., Kurth, T., Greiner, A. and Mahesh, A. “ClimateNet: an Expert-Labeled Open Dataset and Deep Learning Architecture for Enabling High-Precision Analyses of Extreme Weather,” Geoscientific Model Development, 14(1), pp. 107-124, 2021.
[39] Hopfield, J. J. “Neural Networks and Physical Systems with Emergent Collective Computational Abilities,” National Academy of Sciences, 79(8), pp. 2554-2558, 1982.
[40] Rumelhart, D. E., Hinton, G. E. and Williams, R. J. “Learning Representations by Back-Propagating Errors,” Nature, 323(6088), pp. 533-536, 1986.
[41] Carrassi, A., Bocquet, M., Bertino, L. and Evensen, G. “Data Assimilation in the Geosciences: An Overview of Methods, Issues, and Perspectives,” Wiley Interdisciplinary Reviews-Climate Change, 9(5), pp, 1-50, 2018.
[42] Vlachas, P. R., Pathak, J., Hunt, B. R., Sapsis, T. P., Girvan, M., Ott, E. and Koumoutsakos, P. “Backpropagation Algorithms and Reservoir Computing in Recurrent Neural Networks for the Forecasting of Complex Spatiotemporal Dynamics,” Neural Networks, 126, pp. 191-217, 2020.
[43] Bocquet, M., Brajard, J., Carrassi, A. and Bertino, L. “Bayesian Inference of Chaotic Dynamics by Merging Data Assimilation, Machine Learning and Expectation-Maximization,” Foundations of Data Science, 2(1), pp. 55-80, 2020.
[44] Geer, A. J. “Learning Earth System Models from Observations: Machine Learning or Data Assimilation?,” Philosophical Transactions of the Royal Society A: Mathematical Physical and Engineering Sciences, 379(2194), 2021.
[45] Kashinath, K., Mustafa, M., Albert, A., Wu, J., Jiang, C., Esmaeilzadeh, S., Azizzadenesheli, K., Wang, R., Chattopadyay, A., Singh, A., Manepalli, A., Chirila, D., Yu, R., Walters, R., White, B., Xiao, H., Tchelepi, H. A., Marcus, P., Anandkumar, A. and Hassanzadeh, P. “Physics-Informed Machine Learning: Case Studies for Weather and Climate Modelling,” Philosophical Transactions of the Royal Society A: Mathematical Physical and Engineering Sciences, 379(2194), 2021.
[46] Hochreiter, S. and Schmidhuber, J. “Long Short-Term Memory,” Neural Computation, 9(8), pp. 1735-1780, 1997.
[47] Gomez, P., Nebot, A., Ribeiro, S., Alquézar, R., Mugica, F. and Wotawa, F. “Local Maximum Ozone Concentration Prediction Using Soft Computing Methodologies,” Systems Analysis Modelling Simulation, 43(8), pp. 1011-1031, 2003.
[48] Graves, A., Liwicki, M., Fernández, S., Bertolami, R., Bunke, H. and Schmidhuber, J. “A Novel Connectionist System for Unconstrained Handwriting Recognition,” IEEE Transactions on Pattern Analysis and Machine Intelligence, 31(5), pp. 855-868, 2008.
[49] Fente, D. N. and Singh, D. K. “Weather Forecasting Using Artificial Neural Network,” Inventive Communication and Computational Technologies, pp. 1757-1761, 2018.
[50] Salman, A. G., Heryadi, Y., Abdurahman, E. and Suparta, W. “Single Layer & Multi-Layer Long Short-Term Memory (LSTM) Model with Intermediate Variables for Weather Forecasting,” Procedia Computer Science, 135, pp. 89-98, 2018.
[51] Vlachas, P. R., Byeon, W., Wan, Z. Y., Sapsis, T. P. and Koumoutsakos, P. “Data-Driven Forecasting of High-Dimensional Chaotic Systems with Long Short-Term Memory Networks,” Proceedings of the Royal Society A: Mathematical Physical and Engineering Sciences, 474(2213), 2018.
[52] Sangiorgio, M. and Dercole, F. “Robustness of LSTM Neural Networks for Multi-Step Forecasting of Chaotic Time Series,” Chaos, Solitons and Fractals, 139, 2020.
[53] Hong, S. J., Kim, J. H., Choi, D. S. and Baek, K. H. “Development of Surface Weather for East Model by Using LSTM Machine Learning Method,” Korean Meteorological Society, 31(1), pp. 73-83, 2021.
[54] Xu, Z., Du, J., Wang, J., Jiang, C. and Ren, Y. “Satellite Image Prediction Relying on GAN and LSTM Neural Networks,” IEEE International Conference on Communications, pp. 1-6, 2019.
[55] Zhu, L., Zhu, G., Han, L. and Wang, N. “The Application of Deep Learning in Airport Visibility Forecast,” Atmospheric and Climate Sciences, 7(3), pp. 314-322, 2017.
[56] Zhang, C., Wu, M., Chen, J., Chen, K., Zhang, C., Xie, C., Huang, B. and He, Z. “Weather Visibility Prediction Based on Multimodal Fusion,” IEEE Access, 7, pp. 74776-74786, 2019.
[57] Deng, T., Cheng, A., Han, W. and Lin, H. X. “Visibility Forecast for Airport Operations by LSTM Neural Network,” International Conference on Agents and Artificial Intelligence, pp. 466-473, 2019.
[58] Meng, Y., Qi, F., Zuo, H., Chen, B., Yuan, X. and Xiao, Y. “Multi-step LSTM Prediction Model for Visibility Prediction,” International Joint Conference on Neural Networks, pp. 1-8, 2020.
[59] Fukushima, K. “Neocognitron,” Scholarpedia, 2(1), pp. 1717, 2007.
[60] Hubel, D. H. and Wiesel, T. N. “Receptive Fields and Functional Architecture of Monkey Striate Cortex,” The Journal of Physiology, 195(1), pp. 215-243, 1968.
[61] Fukushima, K. “Neocognitron: A Self-Organizing Neural Network Model for a Mechanism of Pattern Recognition Unaffected by Shift in Position,” Biological Cybernetics, 36(4), pp.193-202, 1980.
[62] Matusugu, M., Mori, K., Mitari, Y. and Kaneda, Y. “Subject Independent Facial Expression Recognition with Robust Face Detection Using a Convolutional Neural Network,” Neural Networks, 16(5), pp. 555-559, 2003.
[63] Convolutional Neural Networks (LeNet)-Deep Learning 0.1 Documentation, Retrieved from LISA Lab, 2013.
[64] Aghdam, H. H. and Heravi, E. J. Guide to Convolutional Neural Networks: A Practical Application to Traffic-Sign Detection and Classification, Springer, United States, 2017.
[65] Ciresan, D. C., Meier, U., Masci, J., Gambardella, L. M. and Schmidhuber, J. “Flexible, High Performance Convolutional Neural Networks for Image Classification,” The 22nd International Joint Conference on Artificial Intelligence, pp. 1237-1242, Barcelona, 2011.
[66] Yamaguchi, K., Sakamoto, K., Akabane, T. and Fujimoto, Y. “A Neural Network for Speaker-Independent Isolated Word Recognition,” The 1st International Conference on Spoken Language Processing, pp. 1077-1080, Kobe, 1990.
[67] Ciregan, D., Meier, U. and Schmidhuber, J. “Multi-Column Deep Neural Networks for Image Classification,” IEEE Conference on Computer Vision and Pattern Recognition, pp. 3642-3649, Providence, 2012.
[68] Mittal, S. “A Survey of FPGA-Based Accelerators for Convolutional Neural Networks,” Neural Computing and Applications, 32(4), pp. 1109-1139, 2018.
[69] Chincholkar, S. and Rajapandy, M. “Fog Image Classification and Visibility Detection Using CNN,” International Conference on Intelligent Computing, Information and Control Systems, pp. 249-257, 2019.
[70] Bengio, Y., Simard, P. and Frasconi, P. “Learning Long-Term Dependencies with Gradient Descent Is Difficult,” IEEE Transactions on Neural Networks, 5(2), pp. 157-166, 1994.
[71] Li, S., Wang, S. L. and Xie, D. B. “Research on Radar Clutter Recognition Method Based on LSTM,” The Institution of Engineering and Technology, 2019(19), pp. 6247-6251, 2019.
[72] Huffman, D. A. “A Method for the Construction of Minimum-Redundancy Codes,” Iconic Research and Engineering Journals, 40(9), pp. 1098-1101, 1952.
[73] Srivastava, R. K., Greff, K. and Schmidhuber, J. “Highway Networks,” International Conference on Machine Learning, 2015.
[74] Ripley, B. D. Pattern Recognition and Neural Networks, Cambridge University Press, United Kingdom, 2007.
[75] Nair, V. and Hinton, G. E. “Rectified Linear Units Improve Restricted Boltzmann Machines,” International Conference on Machine Learning, 2010.
[76] Murphy, K. P. Machine Learning: A Probabilistic Perspective, MIT Press, United Kingdom, 2012.
[77] Shore, J. and Johnson, R. “Axiomatic Derivation of the Principle of Maximum Entropy and the Principle of Minimum Cross-Entropy,” IEEE Transactions on Information Theory, 26(1), pp. 26-37, 1980.
[78] Stehman, S. V. “Selecting and Interpreting Measures of Thematic Classification Accuracy,” Remote Sensing of Environment, 62(1), pp. 77-89, 1997.
[79] Lewis, C. D. Industrial and Business Forecasting Methods: A Practical Guide to Exponential Smoothing and Curve Fitting, Butterworth Scientific, London, 1982.
[80] Pascanu, R., Mikolov, T. and Bengio, Y. “On the Difficulty of Training Recurrent Neural Networks,” Proceedings of the 30th International Conference on Machine Learning, 28(3), pp. 1310-1318, 2013.
[81] Kingma, D. P. and Ba, J. “Adam: A Method for Stochastic Optimization,” International Conference for Learning Representations, San Diego, 2015.
[82] Wasserstein, R. L. and Lazar, N. A. “The ASA Statement on P-values: Context, Process, and Purpose,” Taylor & Francis, 70(2), pp. 129-133, 2016.
[83] Pearson, K. “LIII. On Lines and Planes of Closest Fit to Systems of Points in Space,” The London, Edinburgh, and Dublin Philosophical Magazine and Journal of Science, 2(11), pp 559-572, 1901.
[84] Cohen, J. Statistical Power Analysis for the Behavioral Sciences, Routledge Press, United Kingdom, 2013.