全球極端氣候日益嚴重，台灣受到颱風與豪雨的影響的規模與強度也更加嚴峻，在豐沛的雨量與降雨時間與空間不均的情形下，如何降低河川淹水災害備受重視。基隆河貫穿首都生活圈北北基三大縣市，為台灣經濟與政治扮演舉足輕重的腳色，本研究主要提供一全新的預報方法，可以減少時間成本並同時提升預報精度，以減少颱風或豪雨事件帶來的損失與衝擊，給予決策者更合理與準確的預測資訊。
近年來，國內外學者紛紛利用各種方法進行河川水位預報。為了追求更高的精度與預報時長，本研究開發新型河川水位預測方法，結合機器學習模型與數值方法（Numerical ML）進行水位預測，並與單純使用機器學習的方法（Original ML）進行比較，以基隆河流域作為研究模擬區域，針對不同測站進行水位預報，作為洪水預警、災前準備、防災應變之參考。
研究成果顯示，在受潮位影響較小的測站，Numerical ML 預測模式比OriginalML 表現更佳。在水位峰值的預報上，Numerical ML 可以明顯提升水位預報的準確性，隨著預報時間增加，Numerical ML 有效降低誤差的程度也大幅上升。除此之外，Numerical ML 在水位預測上所花費的時間更少，能夠節省時間成本。故本研究結果可在颱風或豪雨期間，提供更精準的水位預測資訊。

In this research, a new river stage prediction method that combines machine learning models and numerical methods has been developed (namely the Numerical MLmodel).
For the Numerical ML model, the Runge-Kutta and implicit numerical methods are combined with Multiple Additive Regression Trees (MART) and Support vector regression (SVR) for model training, respectively. The Keelung River Basin was selected as the study area in which river stages are predicted and compared with the ML model without using numerical methods (namely Original ML model).
The research results show that the Numerical ML model performs better than the Original ML at the stations that are less affected by the tide level. The ML model has smaller errors in the prediction of river peaks and time series. In addition, the Numerical ML requires less time in model training. In application, the Numerical ML can be used to improve the accuracy in flood warning during typhoons or heavy rains.

1. Alvisi, S., Mascellani, G., Franchini, M., & Bárdossy, A. Water level forecasting
through fuzzy logic and artificial neural network approaches. Hydrol. Earth Syst.
Sci., 10(1), 1-17. (2006)
2. Amein, M., & Fang, C. S. IMPLICIT FLOOD ROUTING IN NATURAL
CHANNELS. Journal of Hydraulic Engineering, 96, 2481-2500. (1970)
3. Badrzadeh, H., Sarukkalige, R., & Jayawardena, A. W. Hourly runoff forecasting
for flood risk management: Application of various computational intelligence
models. Journal of Hydrology, 529, 1633-1643. (2015)
4. Bechteler, W., Kulisch, H., & Nujic, M. (1992). 2-D dam-break flooding waves
comparison between experimental and calculated results. In Floods and flood
management (pp. 247-260): Springer.
5. Campolo, M., Andreussi, P., & Soldati, A. River flood forecasting with a neural
network model. Water Resources Research, 35(4), 1191-1197. (1999)
6. Ern, A., Piperno, S., & Djadel, K. A well‐balanced Runge–Kutta discontinuous
Galerkin method for the shallow‐water equations with flooding and drying.
International journal for numerical methods in fluids, 58(1), 1-25. (2008)
7. Friedman, J. H., & Meulman, J. J. Multiple additive regression trees with application
in epidemiology. Statistics in Medicine, 22(9), 1365-1381. (2003)
8. 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)
9. Gholinia, M., Hosseinzadeh, K., Mehrzadi, H., Ganji, D. D., & Ranjbar, A. A.
Investigation of MHD Eyring–Powell fluid flow over a rotating disk under effect of
homogeneous–heterogeneous reactions. Case Studies in Thermal Engineering, 13.
(2019)
10. Guo, W.-D., Chen, W.-B., Yeh, S.-H., Chang, C.-H., & Chen, H. Prediction of River
Stage Using Multistep-Ahead Machine Learning Techniques for a Tidal River of
Taiwan. Water, 13(7). (2021)
11. Haddad, K., & Rahman, A. Regional flood frequency analysis in eastern Australia:
Bayesian GLS regression-based methods within fixed region and ROI framework –
Quantile Regression vs. Parameter Regression Technique. Journal of Hydrology,
430-431, 142-161. (2012)
12. Hosseini, F. S., Choubin, B., Mosavi, A., Nabipour, N., Shamshirband, S., Darabi,
H., & Haghighi, A. T. Flash-flood hazard assessment using ensembles and Bayesianbased
machine learning models: Application of the simulated annealing feature
selection method. Sci Total Environ, 711, 135161. (2020)
13. Hosseinzadeh, K., Roghani, S., Mogharrebi, A. R., Asadi, A., Waqas, M., & Ganji,
D. D. Investigation of cross-fluid flow containing motile gyrotactic microorganisms
and nanoparticles over a three-dimensional cylinder. Alexandria Engineering
Journal, 59(5), 3297-3307. (2020)
14. Hsu, M.-H., Lin, S.-H., & Fu, J.-C. Dynamic Wave Models Coupled with ANN for
Stage Prediction - A Case Study of Flood Forecast in Lanyang River. Journal of
Taiwan Agricultural Engineering, 56, 12-31. (2010)
15. Kao, I. F., Zhou, Y., Chang, L.-C., & Chang, F.-J. Exploring a Long Short-Term
Memory based Encoder-Decoder framework for multi-step-ahead flood forecasting.
Journal of Hydrology, 583. (2020)
16. Kim, G., & Barros, A. P. Quantitative flood forecasting using multisensor data and
neural networks. Journal of Hydrology, 246(1-4), 45-62. (2001)
17. Kourgialas, N. N., Dokou, Z., & Karatzas, G. P. Statistical analysis and ANN
modeling for predicting hydrological extremes under climate change scenarios: the
example of a small Mediterranean agro-watershed. J Environ Manage, 154, 86-101.
(2015)
18. Kroll, C. N., & Vogel, R. M. Probability Distribution of Low Streamflow Series in
the United States. Journal of Hydrologic Engineering, 7(2), 137-146. (2002)
19. Laio, F., Porporato, A., Revelli, R., & Ridolfi, L. A comparison of nonlinear flood
forecasting methods. Water Resources Research, 39(5). (2003)
20. Liang, Q. Flood simulation using a well-balanced shallow flow model. Journal of
Hydraulic Engineering, 136(9), 669-675. (2010)
21. Lin, G.-F., Chen, G.-R., & Huang, P.-Y. Effective typhoon characteristics and their
effects on hourly reservoir inflow forecasting. Advances in Water Resources, 33(8),
887-898. (2010)
22. Lin, G.-F., Chou, Y.-C., & Wu, M.-C. Typhoon flood forecasting using integrated
two-stage support vector machine approach. Journal of Hydrology, 486, 334-342.
(2013)
23. Maity, R., Bhagwat, P. P., & Bhatnagar, A. Potential of support vector regression for
prediction of monthly streamflow using endogenous property. Hydrological
Processes, 24(7), 917-923. (2010)
24. Mukerji, A., Chatterjee, C., & Raghuwanshi, N. S. Flood forecasting using ANN,
neuro-fuzzy, and neuro-GA models. Journal of Hydrologic Engineering, 14(6), 647-
652. (2009)
25. Niu, F., & Chen, L. Forecasting of Landslide Stability Based on Gradient Boosting
Decision Tree Model. International Core Journal of Engineering, 5(11), 42-48.
(2019)
26. Peng, S., Zhang, Z., Liu, E., Liu, W., & Qiao, W. A new hybrid algorithm model for
prediction of internal corrosion rate of multiphase pipeline. Journal of Natural Gas
Science and Engineering, 85. (2021)
27. Pini, M., Scalvini, A., & UsmanLiaqat, M. Evaluation of Machine Learning
Techniques for Inflow Prediction in Lake Como, Italy.
28. Raschendorfer, M., Majewski, D., Förstner, J., Seifert, A., Baldauf, M., & Reinhardt,
T. Operational Convective-Scale Numerical Weather Prediction with the COSMO
Model: Description and Sensitivities. Monthly Weather Review, 139(12), 3887-3905.
(2011)
29. Ribeiro, M. H. D. M., & dos Santos Coelho, L. Ensemble approach based on bagging,
boosting and stacking for short-term prediction in agribusiness time series. Applied
Soft Computing, 86. (2020)
30. Sayama, T., Ozawa, G., Kawakami, T., Nabesaka, S., & Fukami, K. Rainfall–runoff–
inundation analysis of the 2010 Pakistan flood in the Kabul River basin.
Hydrological Sciences Journal, 57(2), 298-312. (2012)
31. Tiwari, M. K., & Chatterjee, C. Development of an accurate and reliable hourly
flood forecasting model using wavelet–bootstrap–ANN (WBANN) hybrid approach.
Journal of Hydrology, 394(3-4), 458-470. (2010a)
32. Tiwari, M. K., & Chatterjee, C. Uncertainty assessment and ensemble flood
forecasting using bootstrap based artificial neural networks (BANNs). Journal of
Hydrology, 382(1-4), 20-33. (2010b)
33. Valipour, M., Banihabib, M. E., & Behbahani, S. M. R. Comparison of the ARMA,
ARIMA, and the autoregressive artificial neural network models in forecasting the
monthly inflow of Dez dam reservoir. Journal of Hydrology, 476, 433-441. (2013)
34. Yoon, H., Jun, S.-C., Hyun, Y., Bae, G.-O., & Lee, K.-K. A comparative study of
artificial neural networks and support vector machines for predicting groundwater
levels in a coastal aquifer. Journal of Hydrology, 396(1-2), 128-138. (2011)
35. Zhang, Y., & Haghani, A. A gradient boosting method to improve travel time
prediction. Transportation Research Part C: Emerging Technologies, 58, 308-324.
(2015)
36. 余思亮. (2012). 河川洪水系集預報模式. 國立臺灣大學, Available from Airiti
AiritiLibrary database. (2012 年)
37. 林欣禾. (2005). 應用率定曲線不確定性分析於洪水即時預報之研究. 國立成
功大學, Available from Airiti AiritiLibrary database. (2005 年)
38. 林洙宏. (2010). 水文即時監測資料應用在河川洪水預報之研究. 國立臺灣大
學, Available from Airiti AiritiLibrary database. (2010 年)
39. 邱啟平. (2009). 基隆河上游集水區含員山子分洪道之出水口水位預報模式.
國立臺灣大學, Available from Airiti AiritiLibrary database. (2009 年)
40. 郭 家 妏 . (2014). 隨機森林在河川水位即時預報之應用. 國 立 成 功 大 學 ,
Available from Airiti AiritiLibrary database. (2014 年)