日常生活中總是離不開大自然的恩惠諸如陽光、空氣與水等資源。而水資源為大自然生態體系運作之重要因素，更是與日常生活習習相關，而隨著近年科技發展迅速，物聯網技術與生活更加密切，透過佈建各式感測裝置，針對水資源進行管理，期望能更有效運用水資源與預防其所帶來之災害。本研究基於物聯網相關技術結合水利專家專業知識，結合城市下水道模擬系統進行城市淹水之預測，目的在於建立一智慧系統加速水患之防範。
本研究建構一物聯網架構包含終端設備感測層、通訊層、邊緣層與應用層，在終端設備感測層中使用了低功耗之室外感測裝置，本研究在感測器感測上提出Dynamic perception adjustment機制，其為根據需求調整感測的頻率，能降低平時傳輸次數達到節能的效益，亦能於災害時提高資料的即時性。通訊層採用Wi-Fi搭配HTTP協議進行資料傳輸。邊緣層則包含計算與預測，透過感測器回傳之資料計算各節點進水量，進行時序標記，透過這些帶有時序標記的資料可預測目標節點水位高度，並依據預測結果進行相應措施，已達水患之防範功用。
本研究依照實際地形、水利設施參數架設一模擬場域，並佈建物聯網感測進行降雨排水模擬，於城市下水道模擬軟體中預測淹水結果RMSE為0.443，而根據本研究所提出的公式結合物聯網感測回傳之資訊系統預測結果為0.367。由實驗結果可得知，本研究提出之方法較模擬軟體更為準確，能提供決策者作為防災參考的依據。

Daily life is always inseparable from nature, such as sunshine, air, and water. Water resources are an important factor in the operation of the natural ecosystem, and it's closely related to people's daily life. With the rapid development of technology in recent years, the Internet of Things technology and life is more closely related. Through the construction of various sensing devices, management for water resources. Expect to use water resources more effectively and prevent disasters caused by them. This research is based on the IoT technologies, combined with the expertise of water conservancy experts, combined with urban flooded predictions in the city sewer simulation system, to establish a smart system to accelerate the prevention of flood.
This study constructs an IoT architecture, which includes terminal device sensing layer, communication layer, edge layer, and application layer. In the terminal device sensing layer, outdoor sensing devices with low power consumption are used. In this study, a dynamic perception adjustment mechanism is proposed in the sensor sensing, It adjusts the frequency of sensing according to the needs, which can reduce the usual transmission times to achieve energy-saving benefits, and can improve the immediacy of data during disasters. The communication layer uses Wi-Fi and HTTP protocol for data transmission. The edge layer includes calculation and prediction. The water inflow of each node is calculated through the data returned by the sensor and performed time-series marking. Through these data with time series marking, the water level height of the target node can be predicted. Corresponding measures are taken according to the forecast results to prevent flooding.
In this study, a simulation field is set up according to the actual terrain and parameters of water conservancy facilities, and IoT sensing is deployed to simulate rainfall and drainage. The RMSE of the flooding result is predicted in the urban sewer simulation software to be 0.443. In this study, the system prediction result of combining the proposed formula and IoT sensing return information is 0.367. The experimental results can prove that the method proposed in this study is more accurate than the simulation software, and can provide decision-makers as a reference for disaster prevention.

[1] D. Wu, “A big data analytics framework for forecasting rare customer complaints: A use case of predicting MA members' complaints to CMS,” Proceedings of IEEE International Conference on Big Data, pp. 3965-3967, 2017.
[2] A. Galletta, L. Carnevale, A. Buzachis, A. Celesti and M. Villari, “A microservices-based platform for efficiently managing oceanographic data,” Proceedings of 4th International Conference on Big Data Innovations and Applications, pp. 25-29, 2018.
[3] B. Puyun and L. Miao, “Research on analysis system of city price based on big data,” Proceedings of IEEE International Conference on Big Data Analysis, pp. 1-4, 2016.
[4] L. Yang and J.J. Zhang, “Realistic plight of enterprise decision-making management under big data background and coping strategies,” Proceedings of IEEE 2nd International Conference on Big Data Analysis, pp. 402-405, 2017.
[5] Q. Hu and Y. Zhang, “An effective selecting approach for social media big data analysis — Taking commercial hotspot exploration with Weibo check-in data as an example,” Proceedings of IEEE 3rd International Conference on Big Data Analysis, pp. 28-32, 2018.
[6] C.H. Wang, H.L. Yu & Y.C. Liang, J.Y. Wang and T.J. Chang. “Rapid flood inundation simulation of dual drainage system between streets and sewers,” Journal of Taiwan Agricultural Engineering. 65. 1-17.
[7] J. Gironás, L. Roesner, L. Rossman, J. Davis. ” A new applications manual for the Storm Water Management Model (SWMM),” Environmental Modelling & Software. 25. 813-814. 10.1016/j.envsoft.2009.11.009.
[8] S.D. Seyoum, Z. Vojinovic, R.K. Price, S. Weesakul and Ph.D. "Coupled 1D and Non-Inertial 2D Flood Inundation Model for Simulation of Urban Flooding, " Journal of Hydraulic Engineering Vol. 138, Issue 1, January 2012
[9] S. Puttinaovarat and P. Horkaew, "Flood Forecasting System Based on Integrated Big and Crowdsource Data by Using Machine Learning Techniques," in IEEE Access, vol. 8, pp. 5885-5905, 2020, doi: 10.1109/ACCESS.2019.2963819.
[10] S. Miau and W. -H. Hung, "River Flooding Forecasting and Anomaly Detection Based on Deep Learning," in IEEE Access, vol. 8, pp. 198384-198402, 2020, doi: 10.1109/ACCESS.2020.3034875.
[11] J. Pan, Y. Yin, J. Xiong, W. Luo, G. Gui and H. Sari, "Deep Learning-Based Unmanned Surveillance Systems for Observing Water Levels," in IEEE Access, vol. 6, pp. 73561-73571, 2018, doi: 10.1109/ACCESS.2018.2883702.
[12] T. Perumal, M. N. Sulaiman and C. Y. Leong, "Internet of Things (IoT) enabled water monitoring system," 2015 IEEE 4th Global Conference on Consumer Electronics (GCCE), 2015, pp. 86-87, doi: 10.1109/GCCE.2015.7398710.