Research on 3D bump prediction in southwest China based on deep learning neural network
Deng Xiaoguang1,2, Luo Ya1,2, Huang Jinquan2, Chen Yiyi1,2, Xu Wenwen1, Li Junyao2
1. Key Laboratory of Aviation Meteorology of China Meteorogical Administration, Civil Aviation Flight University of China, Guanghan 618300; 2. Guizhou Branch of Southwest Regional Air Traffic Management Bureau, Civil Aviation Administration of China, Guiyang 550000
Abstract：This study conducts a systematic research on the prediction of aircraft turbulence in the southwest region based on deep learning neural networks.By integrating aircraft turbulence report data from March 2008 to March 2024 and ERA5 reanalysis data,a complete dataset containing 5 236 turbulence events and 15 708 nonturbulence events was constructed.The research adopted a deep learning neural network model to construct five hidden layers and ReLU and Sigmoid activation functions,achieving a digital expression of the joint prediction of turbulence intensity and height.Research has found that turbulence events are mainly concentrated in the medium to high altitude (4 200 to 6 300 meters) and moderate intensity range,and their distribution is closely related to meteorological dynamic processes such as jet stream axes and front areas.The results show that the model has a good prediction effect on medium and highlevel and moderate turbulence,with an accuracy rate of 77.35%,and the loss function converges stably to 0.15.When distinguishing between "with bumps" and "without bumps",the model's accuracy rate is as high as 99.93%.However,there is still room for improvement in the prediction of extreme intensity or high turbulence.This study provides an effective method for aviation turbulence prediction in the southwest region and points out the future optimization direction.
邓小光, 罗娅, 黄金全, 陈义义, 徐雯雯, 李俊瑶. 基于深度学习神经网络对西南地区三维颠簸预报的研究[J]. 气象水文海洋仪器, 2025, 42(5): 47-53.
Deng Xiaoguang, Luo Ya, Huang Jinquan, Chen Yiyi, Xu Wenwen, Li Junyao. Research on 3D bump prediction in southwest China based on deep learning neural network. Meteorological Hydrological and Marine Instrument, 2025, 42(5): 47-53.
STORER L N,WILLIAMS P D,GILL P G.Aviation turbulence:dynamics,forecasting,and response to climate change[J].Pure and applied geophysics,2019(5),doi:10.1007/S0002401818220.
KIM J H,CHUN H Y,SHARMAN R D,et al.Evaluations of upperlevel turbulence diagnostics performance using the graphical turbulence guidance (GTG) system and pilot reports (PIREPs) over east asia[J].Journal of applied meteorology and climatology,2011,50(9):19361951.
[7]
KIM J H,SHARMAN R,STRAHAN M,et al.Improvements in nonconvective aviation turbulence prediction for the world area forecast system (WAFS)[J].Bulletin of the american meteorological society,2018,99.
PEARSON J M,SHARMAN R D.Prediction of energy dissipation rates for aviation turbulence.Part Ⅱ:nowcasting convective and nonconvective turbulence[J].Journal of applied meteorology and climatology,2017,56(2):339351.
[10]
SHARMAN R D,PEARSON J M.2017.Prediction of energy dissipation rates for aviation turbulence.Part Ⅰ:forecasting nonconvective turbulence [J].Journal of applied meteorology and climatdogy,2017,56(2):317337.
[11]
STORER L N,WILLIAMS P D,GILL P G.Aviation turbulence:dynamics,forecasting,and response to climate change[J].Pure and applied geophysics,2019,176:20812095.
SHARMAN R,LANE T.An integrated approach to midlevel turbulence forecasting[J].Weather and forecasting,2012,27(1):268287.
[16]
REICHSTEIN M,CAMPSVALLS G,STERENS B,et al.Deep learning and process understanding for datadriven earth system science[J].Nature,2019,566:195204.
[17]
ABERNETHY J,SHARMAN R,BRADLEY E.An artificial intelligence approach to operational aviation turbulence forecasting[J].Journal of applied meterology and climatology,2020,59(11):18831899.
[18]
WILLIAMS,JOHN K.Using random forests to diagnose aviation turbulence[J].Machine learning,2014,95(1):5170.
[19]
HON K K,NG C W,CHAN P W.Machine learning based multiindex prediction of aviation turbulence over the AsiaPacific[J].Machine learning with applications,2020,2:18.
[20]
Emara M,Santos M D,Chartier N,et al.MACHINE LEARNING ENABLED TURBULENCE PREDICTION USING FLIGHT DATA FOR SAFETY ANALYSIS[C]//International Council of the Aeronautical Sciences,2021.
2025, Vol. 42 
