使用深度神經網路預測弱綜觀環境下午後強對流及肇始–以WRF系集資料進行系統建置

以作者查詢圖書館館藏

、以作者查詢臺灣博碩士

、以作者查詢全國書目

、勘誤回報

、線上人數：58

、訪客IP：3.144.119.50

姓名

黃展皇(Chan-Huang Huang) 查詢紙本館藏

畢業系所

大氣科學學系

論文名稱

使用深度神經網路預測弱綜觀環境下午後強對流及肇始–以WRF系集資料進行系統建置
(Using Deep Neural Networks to predict afternoon thunderstorm and initiation under weak synoptic forcing - model development using WRF ensembles)

相關論文

★ 單雷達風場反演—【移動坐標法】的特性分析與應用	★ 由都卜勒風場反演熱動力場的新方法 ——TAMEX IOP#2颮線個案應用分析
★ 利用VAD技術及回波保守方程反演渦度場	★ 利用單都卜勒雷達反演三維風場之研究─以數值模式資料驗證
★ 在地形上由都卜勒風場反演熱動力場	★ 利用Extended-GBVTD方法反求非軸對稱颱風（颶風）風場結構
★ 同化雷達資料對數值預報影響之研究	★ 使用系集卡曼濾波器同化都卜勒雷達資料之研究
★ 以3DVAR同化都卜勒雷達觀測及反演資料對於數值模擬結果的影響	★ 台灣北部初秋豪雨個案之降雨特性研究
★ 同化都卜勒雷達資料改善模式預報之研究	★ 2008年台灣西南部地區TRMM降雨雷達與七股雷達回波觀測比較分析及降雨估計應用研究
★ 使用四維變分同化都卜勒雷達資料以改進短期定量降雨預報	★ 同化多部都卜勒雷達資料以提升降水預報能力之研究-2008 SoWMEX IOP8個案分析
★ 結合VDRAS、WRF與雷達網聯資料，以檢視對台灣地區短期降水預報改善之成效	★ 結合都卜勒雷達觀測及反演氣象變數與COSMIC RO資料以改進模式預報之可行性研究

檔案

[Endnote RIS 格式]

[Bibtex 格式]

[相關文章]

[文章引用]

[完整記錄]

[館藏目錄]

[檢視]

[下載]

本電子論文使用權限為同意立即開放。
已達開放權限電子全文僅授權使用者為學術研究之目的，進行個人非營利性質之檢索、閱讀、列印。
請遵守中華民國著作權法之相關規定，切勿任意重製、散佈、改作、轉貼、播送，以免觸法。

摘要(中)

台灣夏季在弱綜觀情況下，午後常有致災性且難以預報的強對流事件發生，因此快速且準確預測對流事件的肇始(Convection Initiation, CI)相當重要。由於小尺度強對流具有非線性特性，而深度神經網路(Deep Neural Networks, DNN)具備擬合非線性函數的能力，因此本研究針對弱綜觀強對流事件進行WRF (Weather Research and Forecasting)系集模擬，採用OSSE (Observing System Simulation Experiment)架構取得各式變數，並經過詳細的個案分析、預測目標定義與資料處理後再使用DNN進行強對流或CI的一小時預報，並且針對DNN進行多種如SHAP (SHapley Additive exPlanations)分析得知各變數的重要性，並依此展開許多面向的探討。
本研究經過許多討論，可提供DNN系統參數等建置以及資料選用與處理的指南，且使用了許多不同面向的驗證方法，包括客觀分數、繪圖疊圖分析、性能圖、SHAP分析、persisFail實驗、坐標系測試實驗以及敏感度測試實驗，對模型性能、預測傾向以及變數重要性做了一系列討論並得出結果，變數重要性分析上與前人對於CI的統計結果接近，另外還考慮了提升DNN預報能力的方法例如加入空間相關的變數，以及在不同個案上應用的問題以及提出混合訓練的解決方式，最後預測表現雖無法與模糊邏輯做比較但顯著優於外延法，且僅需要以純測站變數組合即可快速應用於實際資料，若以中央氣象局標準將鄉鎮市行政區視為最小預報單位的情形，使用純測站可取得的資料進行變數組合，ETS (Equitable Threat Score)的預報得分對強對流為0.338，對CI為0.333，且ETS與事件發生頻率僅呈現低度正相關。

摘要(英)

During Taiwan′s weak synoptic summer, there are often unpredictable and disastrous strong convective events occurring in the afternoon, making it important to rapidly and accurately predict the initiation of convection (CI). Due to the non-linear characteristics of small-scale strong convection, and the ability of Deep Neural Networks (DNN) to fit non-linear functions, this study focuses on weak synoptic strong convection events using the Weather Research and Forecasting (WRF) system and adopting an Observing System Simulation Experiment (OSSE) framework to obtain various variables. After detailed case analysis, prediction target definition, and data processing, DNN is used to forecast strong convection or CI for one hour, and various analyses like SHAP (SHapley Additive exPlanations) are conducted on DNN to determine the importance of each variable, leading to exploration in many different angle.
After extensive discussions, this study provides guidelines for the establishment of DNN system parameters, data selection and processing, and uses various validation methods, including objective scores, graph overlay analysis, performance diagram, SHAP analysis, persisFail experiments, coordinate system testing experiments, and sensitivity testing experiments. A series of discussions were conducted on the model performance, prediction tendencies, and variable importance. The analysis of variable importance is consistent with previous statistical results for the CI, and methods to enhance DNN forecasting ability, such as adding spatially-related variables and solving problems with mixed training, were also considered. Although the prediction performance of the DNN cannot be compared with fuzzy logic, the results show a significant improvement over extrapolation methods, and can be quickly applied to actual data using only pure station variables. When using central weather bureau standards and considering the smallest forecast units as townships, the ETS (Equitable Threat Score) forecast scores for strong convection and the CI are 0.338 and 0.333, respectively, and the ETS and event occurrence frequency only weak positive correlation.

關鍵字(中)

★ 對流肇始
★ 深度神經網路

關鍵字(英)

★ Convection Initiation
★ Deep Neural Network

論文目次

摘要 I
ABSTRACT II
誌謝 IV
目錄 V
表目錄 VII
圖目錄 VII
第一章緒論 1
1-1. 對流肇始(CONVECTION INITIATION, CI) 1
1-2. 深度神經網路(DEEP NEURAL NETWORKS, DNN) 2
1-3. 研究動機與目標 2
第二章資料 4
2-1. 實驗架構 4
2-2. 真實與模擬個案 4
2-3. 資料處理 5
2-4. 預測目標定義 7
第三章方法 9
3-1. DNN原理 9
3-2. DNN架構設定 11
第四章驗證 13
4-1. 預測比較與驗證 13
4-1-1. 預測pattern疊圖比較 13
4-1-2. 公平預兆得分(Equitable Threat Score, ETS) 13
4-1-3. 性能圖(performance diagram) 14
4-1-4. 接收者操作特徵曲線(Receiver operating characteristic curve, ROC) 錯誤! 尚未定義書籤。
4-1-5. SHAP分析(SHapley Additive exPlanations analysis) 錯誤! 尚未定義書籤。
4-2. PERSISFAIL測試實驗 16
第五章結果 17
5-1. 變數相關測試 17
5-1-1. 變數座標與預測目標選擇 17
5-1-2. 變數敏感性測試 18
5-1-3 加入時空間資訊效益 20
5-1-4. SHAP分析 21
5-2. 實際應用與比較 21
5-2-1. 與其他預測法比較 21
5-2-2. 跨個案預測效能 23
5-2-3. 鄉鎮市標準ETS 24
5-2-4. 其他測試 25
第六章結論與討論 27
6-1. 討論 27
6-2. 結論 27
6-3. 未來展望 28
參考資料 30
附表 32
附圖 37

參考文獻

蘇木春，類神經網路課程講義第三章多層感知機原理，2021，中央大學資訊
工程系。
陳鈞澤（2021年10月）。弱綜觀午後雷暴事件即時預報-機器學習方法。110年
天氣分析與預報研討會，交通部中央氣象局。
Bi, K., Xie, L., Zhang, H., Chen, X., Gu, X., and Tian, Q, 2023. Accurate medium-range global weather forecasting with 3D neural networks. Nature 619, 533–538.
Chang, H.-L., Brown, B.G., Chu, P.-S., Liou, Y.-C., and Wang, W.-H., 2017. Nowcast guidance of afternoon convection initiation for Taiwan. Weather and Forecasting, 32(5): 1801-1817.
Cybenko, G., 1989, Approximation by superpositions of a sigmoidal function. Math. Control Signal Systems, 2: 303–314.
Espeholt, L., Agrawal, S., Sønderby, C., Kumar, M., Heek, J., Bromberg, C., Gazen, C., Carver, R., Andrychowicz, M., Hickey, J., Bell, A., and Kalchbrenner, N., 2022, Deep learning for twelve hour precipitation forecasts. Nature, 13: 5145.
Hinton, G.E., Osindero, S., and The, Y.-W., 2006, A fast learning algorithm for deep belief nets. Neural Comput, 18(7):1527-54.
Hornik, K., 1991, Approximation capabilities of multilayer feedforward networks. Neural Netw, 4(2): 251-257.
Lin, P.-F., Chang, P.-L., Jou, B.J.-D., Wilson, J.W. and Roberts, R.D., 2011. Warm season afternoon thunderstorm characteristics under weak synoptic-scale forcing over Taiwan Island. Weather and forecasting, 26(1): 44-60.
Lin, P.-F., Chang, P.-L., Jou, B.J.-D., Wilson, J.W. and Roberts, R.D., 2012. Objective Prediction of Warm Season Afternoon Thunderstorms in Northern Taiwan Using a Fuzzy Logic Approach. Weather and Forecasting, 27(5): 1178-1197.
Madaus, L. E., and Hakim G. J., 2016. Observable Surface Anomalies Preceding Simulated Isolated Convective Initiation. Monthly Weather Review, 144(6), 2265–2284.
Madaus, L. E., and Hakim G. J., 2017, Constraining Ensemble Forecasts of Discrete Convective Initiation with Surface Observations. Monthly Weather Review, 145(7), 2597–2610.
Mecikalski, J. R., and Bedka K. M., 2006. Forecasting Convective Initiation by Monitoring the Evolution of Moving Cumulus in Daytime GOES Imagery. Monthly Weather Review, 134(1), 49–78.
Pulkkinen, S., Nerini, D., Pérez Hortal, A.A., Velasco-Forero, C., Seed, A., Germann, U., and Foresti, L., 2019. Pysteps: an open-source Python library for probabilistic precipitation nowcasting (v1.0). Geoscientific Model Development, 12(10): 4185-4219.
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., 2021. Skilful precipitation nowcasting using deep generative models of radar. Nature, 597: 672–677.
Roberts, R. D., and Steven, R., 2003. Nowcasting storm initiation and growth using GOES-8 and WSR-88D data. Weather and Forecasting, 18(4) , 562–584.
Scott, M.L., Lee, S.I., 2017, A Unified Approach to Interpreting Model Predictions. Advances in neural information processing systems, 30: 4765-4774.
SHI, X., Chen, Z., Wang, H., Yeung, D.-Y., Wong, W.-K., and WOO, W.-C., 2015. Convolutional LSTM network: A machine learning approach for precipitation nowcasting. Advances in neural information processing systems, 28: 802-810.
Wang, C., 2014, On the Calculation and Correction of Equitable Threat Score for Model Quantitative Precipitation Forecasts for Small Verification Areas: The Example of Taiwan. Wea. Forecasting, 29, 788–798.
Shrikumar, A., Greenside, P., and Kundaje, A., 2017. Learning important features through propagating activation differences. In Proceedings of the 34th International Conference on Machine Learning, 70, 3145–3153.
Zhang, Y., Long, M., Chen, K., Xing, L, Jin, R, Jordan, M.I., and Wag, J, 2023. Skilful nowcasting of extreme precipitation with NowcastNet. Nature 619, 526–532.

指導教授

廖宇慶(Yu-Chieng Liou)

審核日期

2023-7-26

推文