近年來，高強度降雨頻繁的發生，傳統的防洪設施如排水道、堤防、地下道等工程方法有其限度，超過設計保護標準的洪水，往往導致嚴重的淹水災害。本研究結合淹水感測水深資料、數值模擬、以及機器學習模式建立淹水預報模式，探討設置淹水感測器對於淹水預警之幫助。
本研究以台南市三爺溪集水區為研究區域，蒐集24場歷年降雨資料，帶入數值模式進行淹水模擬，將所產製之模擬淹水深度作為訓練資料庫，以支撐向量迴歸(Support Vector Regression)機器學習模型，建置四種淹水預報模式，分別探討(1)不採用淹水感測水深資料、(2)採用淹水感測水深資料，無預報修正、(3)採用淹水感測水深資料，以數值方法進行預報修正、及(4)採用淹水感測水深資料，以機器學習方法進行誤差修正，四種情況下之淹水預報精度。
本研究以2019年8月13日豪雨事件為驗證對象，比較四種模式預報淹水深度與範圍。結果顯示，考慮淹水感測器水深資料，採用淹水感測器水深資料，以數值方法進行預報修正之模式SVR-FE有最佳預測效果，淹水深度預測RMSE在0.06至0.15m，峰值誤差(ED_p)在1.4%至27%，洪峰時刻誤差(ET_p)在90分鐘以內。此外，在淹水範圍預測方面考慮淹水感測器資料之三種模型皆有80%的準確率。
本研究顯示淹水感測器資料，對於提升淹水預警精度有明顯的助益。結合淹水感測水深所建置之淹水預報機器學習模型，可於洪災期間根據利用淹水感測觀測水深，進行即時淹水深度預報與修正，對於防洪減災操作與緊急疏散有極大的幫助。

In recent years, high-intensity rainfall happened frequently and has caused severe
flooding around the world under the limitation of engineering systems. This study
combines flood sensors, numerical simulation, and machine learning method to
establish a flood forecast model for Sanye River watershed in Taiwan. The 24-
rainfall data from 2013 to 2018 were collected and introduced into a numerical
model for flooding simulation. The simulated flood depth is used as a training
database to support Vector regression (SVR) establish four flood forecasting
models: (1) without using flood sensor data, (2) using flood sensor data, (3) using
flood sensor data and numerical method for bias correction, and (4) using flood
sensor data and machine learning method for bias correction. The overall accuracy
is 80% for the three models that consider the flood sensors depth data in the flood
extent prediction.
This research shows that the data of flood sensors is obviously helpful for
improving the accuracy of flood forecasting. The introduction of flood sensor data
for real-time flooding depth forecast are helpful for flood control, disaster
mitigation, and emergency operation.

1. Berkhahn, S., Fuchs, L., & Neuweiler, I. An ensemble neural network model for real-time prediction of urban floods. Journal of Hydrology, 575, 743-754. (2019).
2. Chang, M.-J., Chang, H.-K., Chen, Y.-C., Lin, G.-F., Chen, P.-A., Lai, J.-S., & Tan, Y.-C. A Support Vector Machine Forecasting Model for Typhoon Flood Inundation Mapping and Early Flood Warning Systems. Water, 10(12). (2018).
3. Chen, A. S., Evans, B., Djordjević, S., & Savić, D. A. A coarse-grid approach to representing building blockage effects in 2D urban flood modelling. Journal of Hydrology, 426-427, 1-16. (2012).
4. Chen, R.-S., Wang, K.-L., Yang, K.-H., & Chang, S.-C. Application of Improved Tank Model on Upland Field Rainfall-Runoff Research at Hillslope. Journal of Chinese Soil and Water Conservation, 40(2), 147-157. (2009).
5. Darabi, H., Choubin, B., Rahmati, O., Torabi Haghighi, A., Pradhan, B., & Kløve, B. Urban flood risk mapping using the GARP and QUEST models: A comparative study of machine learning techniques. Journal of Hydrology, 569, 142-154. (2019).
6. de Oliveira, J. F. L., Pacífico, L. D. S., de Mattos Neto, P. S. G., Barreiros, E. F. S., Rodrigues, C. M. d. O., & Filho, A. T. d. A. A hybrid optimized error correction system for time series forecasting. Applied Soft Computing, 87. (2020).
7. Fu, J.-C., Huang, H.-Y., Jang, J.-H., & Huang, P.-H. River Stage Forecasting Using Multiple Additive Regression Trees. Water Resources Management, 33(13), 4491-4507. (2019).
8. Hong, H., Tsangaratos, P., Ilia, I., Liu, J., Zhu, A. X., & Chen, W. Application of fuzzy weight of evidence and data mining techniques in construction of flood susceptibility map of Poyang County, China. Sci Total Environ, 625, 575-588. (2018).
9. Hosseiny, H., Nazari, F., Smith, V., & Nataraj, C. A Framework for Modeling Flood Depth Using a Hybrid of Hydraulics and Machine Learning. Scientific reports, 10(1), 1-14. (2020).
10. Hostache, R., Matgen, P., Montanari, A., Montanari, M., Hoffmann, L., & Pfister, L. Propagation of uncertainties in coupled hydro-meteorological forecasting systems: A stochastic approach for the assessment of the total predictive uncertainty. Atmospheric Research, 100(2-3), 263-274. (2011).
11. Jang, J.-H. An Advanced Method to Apply Multiple Rainfall Thresholds for Urban Flood Warnings. Water, 7(11), 6056-6078. (2015).
12. Jang, J.-H., Yu, P.-S., Yeh, S.-H., Fu, J.-C., & Huang, C.-J. A probabilistic model for real-time flood warning based on deterministic flood inundation mapping. Hydrological Processes, 26(7), 1079-1089. (2012).
13. Kia, M. B., Pirasteh, S., Pradhan, B., Mahmud, A. R., Sulaiman, W. N. A., & Moradi, A. An artificial neural network model for flood simulation using GIS: Johor River Basin, Malaysia. Environmental Earth Sciences, 67(1), 251-264. (2011).
14. Kim, H. I., & Han, K. Y. Linking Hydraulic Modeling with a Machine Learning Approach for Extreme Flood Prediction and Response. Atmosphere, 11(9), 987. (2020).
15. Lamichhane, N., & Sharma, S. Development of Flood Warning System and Flood Inundation Mapping Using Field Survey and LiDAR Data for the Grand River near the City of Painesville, Ohio. Hydrology, 4(2). (2017).
16. Li, X., Yan, D., Wang, K., Weng, B., Qin, T., & Liu, S. Flood Risk Assessment of Global Watersheds Based on Multiple Machine Learning Models. Water, 11(8). (2019).
17. Ostfeld, A., & Solomatine, D. P. Data-driven modelling: some past experiences and new approaches. Journal of Hydroinformatics, 10(1), 3-22. (2008).
18. Pan, J., Yin, Y., Xiong, J., Luo, W., Gui, G., & Sari, H. Deep Learning-Based Unmanned Surveillance Systems for Observing Water Levels. IEEE Access, 6, 73561-73571. (2018).
19. Razavi Termeh, S. V., Kornejady, A., Pourghasemi, H. R., & Keesstra, S. Flood susceptibility mapping using novel ensembles of adaptive neuro fuzzy inference system and metaheuristic algorithms. Sci Total Environ, 615, 438-451. (2018).
20. Sadler, J. M., Goodall, J. L., Morsy, M. M., & Spencer, K. Modeling urban coastal flood severity from crowd-sourced flood reports using Poisson regression and Random Forest. Journal of Hydrology, 559, 43-55. (2018).
21. Sahai, A. K., Borah, N., Chattopadhyay, R., Joseph, S., & Abhilash, S. A bias-correction and downscaling technique for operational extended range forecasts based on self organizing map. Climate Dynamics, 48(7-8), 2437-2451. (2016).
22. Shen, J.-C., Chang, C.-H., Wu, S.-J., Hsu, C.-T., & Lien, H.-C. Real-time correction of water stage forecast using combination of forecasted errors by time series models and Kalman filter method. Stochastic Environmental Research and Risk Assessment, 29(7), 1903-1920. (2015).
23. Sun, A. Y., & Scanlon, B. R. How can Big Data and machine learning benefit environment and water management: a survey of methods, applications, and future directions. Environmental Research Letters, 14(7). (2019).
24. Sun, K., Hu, L., Guo, J., Yang, Z., Zhai, Y., & Zhang, S. Enhancing the understanding of hydrological responses induced by ecological water replenishment using improved machine learning models: A case study in Yongding River. Science of The Total Environment, 768, 145489. (2021).
25. Tehrany, M. S., Lee, M.-J., Pradhan, B., Jebur, M. N., & Lee, S. Flood susceptibility mapping using integrated bivariate and multivariate statistical models. Environmental Earth Sciences, 72(10), 4001-4015. (2014a).
26. Tehrany, M. S., Pradhan, B., & Jebur, M. N. Flood susceptibility mapping using a novel ensemble weights-of-evidence and support vector machine models in GIS. Journal of Hydrology, 512, 332-343. (2014b).
27. Tien Bui, D., & Hoang, N.-D. A Bayesian framework based on a Gaussian mixture model and radial-basis-function Fisher discriminant analysis (BayGmmKda V1.1) for spatial prediction of floods. Geoscientific Model Development, 10(9), 3391-3409. (2017).
28. Toda, L. L., Yokingco, J. C. E., Paringit, E. C., & Lasco, R. D. A LiDAR-based flood modelling approach for mapping rice cultivation areas in Apalit, Pampanga. Applied Geography, 80, 34-47. (2017).
29. Woznicki, S. A., Baynes, J., Panlasigui, S., Mehaffey, M., & Neale, A. Development of a spatially complete floodplain map of the conterminous United States using random forest. Sci Total Environ, 647, 942-953. (2019).
30. Wu, S.-J., Lien, H.-C., Chang, C.-H., & Shen, J.-C. Real-time correction of water stage forecast during rainstorm events using combination of forecast errors. Stochastic Environmental Research and Risk Assessment, 26(4), 519-531. (2011).
31. Zhang, S., & Pan, B. An urban storm-inundation simulation method based on GIS. Journal of Hydrology, 517, 260-268. (2014).
32. Zhao, G., Pang, B., Xu, Z., Peng, D., & Xu, L. Assessment of urban flood susceptibility using semi-supervised machine learning model. Sci Total Environ, 659, 940-949. (2019).
33. 國家災害防救科技中心. 2020年初印尼雅加達洪災事件. (2020)
34. 經濟部水利署. 『易淹水地區水患治理計畫』 三爺溪排水系統整治及環境營造規劃報告. (2008)
35. 經濟部水利署. 0823熱帶低氣壓豪雨淹水檢討報告. (2018)
36. 黎哲緯. 運用淹水感測器物聯網與機器學習方法進行淹水推估. (2020)