簡易檢索 / 詳目顯示

研究生: 馬錦成
Songphon Bunphichayanan
論文名稱: Hybrid Wavelet-LSTM-CNN for water level forecasting in Thailand, Chao Phraya Basin
Hybrid Wavelet-LSTM-CNN for water level forecasting in Thailand, Chao Phraya Basin
指導教授: 呂永和
Yung-Ho Leu
口試委員: 楊維寧
Wei-Ning Yang
陳雲岫
Yun-Shiow Chen
學位類別: 碩士
Master
系所名稱: 管理學院 - 資訊管理系
Department of Information Management
論文出版年: 2021
畢業學年度: 109
語文別: 英文
論文頁數: 74
中文關鍵詞: Water level predictionWavelet transformARIMADeep learningLSTMsTime series
外文關鍵詞: Water level prediction, Wavelet transform, ARIMA, Deep learning, LSTMs, Time series
相關次數: 點閱:282下載:2
分享至:
查詢本校圖書館目錄 查詢臺灣博碩士論文知識加值系統 勘誤回報
  • Flood is a Thailand’s chronic issue, that affected people, social, economic and cause devastating damage especially the great flood in 2011. The accurate forecast model can enhance water management capability and help related agents for making a decision in a critical situation to avoid such a terrible disaster.
    AI advancement in recent years has solved previously unsolvable problems and increase options for a data-driven approach. But in the hydrological, only a few researchers have explored AI potential and develop a novel approach. We develop a data-driven based framework for 72 hours water level prediction at CPY012 station in Chao Phraya River, Ayutthaya.
    In this paper, we conduct experimental to searching for preprocessing that appropriate with the problem and comparing various statistic, machine learning and neural network model on the forecast, including ARIMA, SARIMA, Vector Autoregression, Multi Linear regression, Support Vector Regression, Random Forest, LSTMs, CNNs, and our new proposal hybrid model wavelet-neural network. We found that feature selection and wavelet transform can improve neural network accuracy.
    The result showed that our hybrid wavelet-LSTM-CNN has the best performance with the least error MSE at 0.05, RMSE at 0.22, NSE at 0.91, and R-square 0.92 in whole year evaluation perspective. However, for monsoon season when heavy rain and flood frequency occur, the result slightly less than SARIMA with MSE at 0.03, RMSE at 0.17, NSE at 0.56, and R-square 0.61 but still better than other Machine learning and neural network model.


    Flood is a Thailand’s chronic issue, that affected people, social, economic and cause devastating damage especially the great flood in 2011. The accurate forecast model can enhance water management capability and help related agents for making a decision in a critical situation to avoid such a terrible disaster.
    AI advancement in recent years has solved previously unsolvable problems and increase options for a data-driven approach. But in the hydrological, only a few researchers have explored AI potential and develop a novel approach. We develop a data-driven based framework for 72 hours water level prediction at CPY012 station in Chao Phraya River, Ayutthaya.
    In this paper, we conduct experimental to searching for preprocessing that appropriate with the problem and comparing various statistic, machine learning and neural network model on the forecast, including ARIMA, SARIMA, Vector Autoregression, Multi Linear regression, Support Vector Regression, Random Forest, LSTMs, CNNs, and our new proposal hybrid model wavelet-neural network. We found that feature selection and wavelet transform can improve neural network accuracy.
    The result showed that our hybrid wavelet-LSTM-CNN has the best performance with the least error MSE at 0.05, RMSE at 0.22, NSE at 0.91, and R-square 0.92 in whole year evaluation perspective. However, for monsoon season when heavy rain and flood frequency occur, the result slightly less than SARIMA with MSE at 0.03, RMSE at 0.17, NSE at 0.56, and R-square 0.61 but still better than other Machine learning and neural network model.

    TABLE OF CONTENT ABSTRACT ..................................................................................................................... i ACKNOWLEDGEMENT............................................................................................... ii TABLE OF CONTENT..................................................................................................iii LIST OF FIGURES......................................................................................................... v LIST OF TABLES......................................................................................................... vii Chapter 1. Introduction .............................................................................................. 1 1.1. Background and motivation.................................................................................... 1 1.2. Objective................................................................................................................. 3 1.3. Research Limitation................................................................................................ 3 1.4. Contribution ............................................................................................................ 4 1.5. Summary................................................................................................................. 4 Chapter 2. Literature Review..................................................................................... 5 2.1. Hydrological model ................................................................................................ 5 2.2. Data-driven approach.............................................................................................. 6 Chapter 3. Theoretical base...................................................................................... 11 3.1. Study area (Chao Phraya river basin) ................................................................... 11 3.2. Multivariate Adaptive Regression Splines (MARS) ............................................ 12 3.3. Principal Component Analysis (PCA).................................................................. 14 3.4. Sliding window..................................................................................................... 14 3.5. Min-Max Normalization ....................................................................................... 15 3.6. Stationary Wavelet Transform.............................................................................. 15 3.7. Machine Learning ................................................................................................. 16 3.7.1. Multivariate linear regression................................................................. 16 3.7.2. ARIMA................................................................................................... 16 3.7.3. Vector autoregression............................................................................. 18 3.7.4. Support vector machines........................................................................ 18 3.7.5. Random Forest (RF)............................................................................... 19 3.8. Deep Learning....................................................................................................... 20 3.8.1. Artificial neural network (ANN)............................................................ 20 3.8.2. Convolutional neural network (CNNs) .................................................. 21 3.8.3. Recurrent Neural Network (RNNs)........................................................ 22 3.8.4. Long Short-Term Memory (LSTMs)..................................................... 23 3.9. Evaluation ............................................................................................................. 25 Chapter 4. Research Methodology........................................................................... 27 4.1. Research setting .................................................................................................... 27 4.2. System Description ............................................................................................... 28 4.2.1 Reading data .............................................................................................. 29 4.2.2 Data Preprocessing .................................................................................... 30 4.2.3 Training model........................................................................................... 34 4.2.4 Model Evaluation....................................................................................... 38 Chapter 5. Experiment and result............................................................................. 40 5.1. Data collection ...................................................................................................... 40 5.2. Data Preprocessing ............................................................................................... 40 5.3. Parameter Tuning.................................................................................................. 44 5.4. Experiment result.................................................................................................. 45 Chapter 6. Conclusion and Future Works................................................................ 53 6.1. Result .................................................................................................................... 53 6.2. Future Works ........................................................................................................ 53 References ..................................................................................................................... 54 APPENDIX ...................................................................................................................... i

    [1] D. Komori, S. Nakamura and M. Kiguci, "Characteristics of the 2011 Chao Phraya River flood in Central Thailand," Hydrological Research Letters, vol. 6, pp. 41-46, 2012.
    [2] Japan International cooperation agency (JICA), "Kingdom of Thailand project fro the comprehensuve flood management plan for the Chao phraya river basin component 3," Foundation of river and basin integrated communication, Japan, 2013.
    [3] M. Haraguchi and U. Lall, "Flood risks and impacts: A case study of Thailand’s floods in 2011 and research questions for supply chain decision making," International Journal of Disaster Risk Reduction, vol. 14, pp. 256-272, 2015.
    [4] P. Promchote, S.-Y. S. Wang and P. G. Johnson, "The 2011 Great Flood in Thailand: Climate Diagnostics and Implications from Climate Change," Journal of Climate , vol. 29, no. 1, p. 367–379., 2016.
    [5] G. K. Devi, G. B. P and D. G. S, "A Review on Hydrological Models," in INTERNATIONAL CONFERENCE ON WATER RESOURCES, COASTAL AND OCEAN ENGINEERING (ICWRCOE 2015), Mangalore, India, 2015.
    [6] K. Pasupa and S. Jungjareantrat, "Water Levels Forecast In Thailand: A Case Study Of Chao Phraya River," in 14th International Conference on Control, Automation, Robotics & Vision, Phuket, Thailand, 2016.
    [7] C. Soomlek, N. Kaewchainam, T. Simano and C. So-In, "Using backpropagation neural networks for flood forecasting in PhraNakhon Si Ayutthaya, Thailand," in International Computer Science and Engineering Conference (ICSEC), Chiang mai, Thailand, 2015.
    [8] T. Tingsanchali, "Forecasting model of Chao Phraya river flood levels at Bangkok," in Proceeding of the International Conference on Chao Phraya Delta, Bangkok, Thailand, 2000.
    [9] G. B. P. D. G. S. Gayathri K Devi, "A Review on Hydrological Models," in INTERNATIONAL CONFERENCE ON WATER RESOURCES, COASTAL AND OCEAN, Mangalore, India, 2015.
    [10] S. Chuanpongpanich, P. Arlai, M. Koch and K. Tanaka, "Tide prediction for the downstream boundary condition of the chao-phraya river model using harmonic analysis," in Proceeding of the 10th International Conference on Hydroscience and Engineering (ICHE 2012), Orlando, USA, 2012.
    [11] T. Ogata, O. C.SaavedraValeriano, C. Yoshimura, W. Liengcharernsit and Y. Hirabayashi, "Past and future hydrological simulations of Chao Phraya river basin," Journal of Japan Society of Civil Engineers Ser B1 (Hydraulic Engineering) , vol. 68, no. 4, 2012.
    [12] Z. Yu, G. Lei, Z. Jiang and F. Liu, "ARIMA Modelling and Forecasting of Water Level in the Middle Reach of the Yangtze River," in 4th International Conference on Transportation Information and Safety (ICTIS), Banff, Canada, 2017.
    [13] W. M. Wong, S. K. Subramaniam, F. S. Feroz, I. D. Subramaniam and L. A. F. Rose, "Flood Prediction using ARIMA Model in Sungai Melaka,Malaysia," International Journal of Advanced Trends in Computer Science and Engineering, vol. 9, no. 4, p. 5287, 2020.
    [14] T. Hastie, R. Tibshirani and J. Friedman, "MARS: Multivariate Adaptive Regression," in The Elements of Statistical Learning: Data Mining, Inference, and Prediction, Second Edition (Springer Series in Statistics), Springer, 2009, p. 321.
    [15] W. Zhang, A. T. C. Goh and Y. Zhang, "Multivariate Adaptive Regression Splines Application for Multivariate Geotechnical Problems with Big Data," Geotechnical and Geological Engineering, vol. 34, p. pages193–204, 2016.
    [16] T. P. York, L. J. Eaves and E. J. C. G. v. d. Oord, "Multivariate adaptive regression splines: a powerful method for detecting disease-risk relationship differences among subgroups," Stat Med., vol. 25, no. 8, pp. 1355-67, 2006.
    [17] T.-H. Huanga, Y. Leu and W.-T. Pan, "Constructing ZSCORE-Based Financial Crisis Warning Models using Fruit Fly Optimization Algorithm and General Regression Neural Network," Kybernetes, vol. 45, pp. 650-665, 2016.
    [18] T. Hu, F. Wu and X. Zhang, "Rainfall–runoff modeling using principal component analysis," Hydrology Research, vol. 38, no. 3, pp. 35-248, 2007.
    [19] P.-S. Yu, S.-T. Chen and I.-F. Chang, "Support vector regression for real-time flood stage forecasting," Journal of Hydrology, vol. 328, p. 704– 716, 2006.
    [20] J.-H. Yang and C.-H. C.-P. Chan, "A Time-Series Water Level Forecasting Model Based on Imputation and Variable Selection Method," Computational Intelligence and Neuroscience, vol. 2017, pp. 1-11, 2017.
    [21] C. Choi, J. Kim, H. Han, D. Han and H. S. Kim, "Development of Water Level Prediction Models Using Machine Learning in Wetlands: A Case Study of Upo Wetland in South Korea," Water, vol. 12, no. 1, p. 93, 2019.
    [22] H. Hidayat, A. J. F. Hoitink, M. G. Sassi and a. P. J. J. F. Torfs, "Prediction of Discharge in a Tidal River Using Artificial Neural Networks," Journal of Hydrologic Engineering, vol. 19, no. 8, 2019.
    [23] S. Zhang, L. Lu, J. Yu and H. Zhou, "Short-term water level prediction using different artificial intelligent models," in Conference: 2016 5th International Conference on Agro-geoinformatics (Agro-geoinformatics), Tianjin, China, 2016.
    [24] N. T. N. Anh, N. Q. Dat, N. T. Van, N. N. Doanh and N. L. An, "Wavelet-artificial neural network model for water level forecasting," in International Conference on Research in Intelligent and Computing in Engineering (RICE), San Salvador, 2018.
    [25] J. Qiu, B. Wang and C. Zhou, "Forecasting stock prices with long-short term memory neural network based on attention mechanism," PLOS ONE, vol. 15, no. 1, 2020.
    [26] S. Kiranyaz, T. Ince, R. Hamila and M. Gabbouj, "Convolutional Neural Networks for patient-specific ECG classification," in 37th Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC), Milan, 2015.
    [27] S. K. L. E. M. A. a. M. G. T. Ince, "Real-Time Motor Fault Detection by 1-D Convolutional Neural Networks," in IEEE Transactions on Industrial Electronics, 2016.
    [28] I. Sutskever, O. Vinyals and Q. V. Le, "Sequence to Sequence Learning with Neural Networks," in Advances in neural information processing systems, 2014, pp. 3104 - 3112.
    [29] T. Mikolov, A. Joulin, S. Chopra, M. Mathieu and M. Ranzato, "Learning Longer Memory in Recurrent Neural Networks," arXiv, no. arXiv:1412.7753, 2014.
    [30] D. M. Q. Nelson, A. C. M. Pereira and R. A. d. Oliveira, "Stock market's price movement prediction with LSTM neural networks," in International Joint Conference on Neural Networks (IJCNN), Anchorage, AK, USA, 2017.
    [31] F. Qian and X. Chen, "Stock Prediction Based on LSTM under Different Stability," in IEEE 4th International Conference on Cloud Computing and Big Data Analysis (ICCCBDA), Chengdu, China, China, 2019.
    [32] Z. Xiang, j. yan and I. Demir, "A Rainfall‐Runoff Model With LSTM‐Based Sequence‐to‐Sequence Learning," Water Resources Research, vol. 56, 03 01 2020.
    [33] C. Hu, Q. Wu, H. Li, S. Jian and N. Li, "Deep Learning with a Long Short-Term Memory Networks Approach for Rainfall-Runoff Simulation," Water, vol. 10, no. 11, pp. 1543 - 1559, 2018.
    [34] W. Li, A. Kiaghadi and C. N. Dawso, "High Temporal Resolution Rainfall Runoff Modelling Using Long-short-term-memory (LSTM) Networks," arXiv, 2020.
    [35] X.-H. Le, H. V. Ho, G. Lee and S. Jung, "Application of Long Short-Term Memory (LSTM) Neural Network for Flood Forecasting," Water, vol. 11, p. 1387, July 2019.
    [36] K. YAN, W. LI and Z. J. , "A Hybrid LSTM Neural Network for Energy Consumption Forecasting of Individual Households," {IEEE Access, vol. 7, pp. 157633-157642, 2019.
    [37] T. V. a. J. Threenat, "Factors Affecting the Accuracy of Water Level Forecasting at M.7 Gauge Station Using Artificial Neural Network Model," in Proceeding of the 2nd STOU Graduate Research Conference, Nonthaburi, Thailand, 2012.
    [38] S. Phuphong and C. Surussavadee, "An Artificial Neural Network Based Runoff Forecasting Model in the Absence of Precipitation Data: A Case Study of Khlong U-Tapao River Basin, Songkhla Province, Thailand," in 4th International Conference on Intelligent Systems, Modelling and Simulation, Bangkok, Thailand, 2013.
    [39] S. Chuanpongpanich, P. Arlai, M. Koch, K. Tanaka and T. Tingsanchali, "Inflow prediction for the upstream boundary condition of the Chao Phraya river model using an artificial neural network," in Proceeding of the 10th International Conference on Hydroscience and Engineering, Orlando, 2012.
    [40] World Water Assessment Programme, UN World Water Development Report 1: Water for People, Water for Life, Paris: Berghahn Books, 2003, pp. 387-400.
    [41] H. Bond, K. Cross, R. Glotzbach and B. Richaud, "Flood and Drought Management Tools Case Study," K-water, Deajeon, Korea, 2018.
    [42] N. Singhrattna, B. Rajagopalan, K. K. Kumar and M. Clark, "Interannual and Interdecadal Variability of Thailand Summer Monsoon Season," Jornal of climate, vol. 34, no. 1, p. 1697–1708, 2009.
    [43] J. H. Friedman, "Multivariate Adaptive Regression Splines," The Annals of Statistics, vol. 19, no. 1, pp. 1-67, 1991.
    [44] T. G. Dietterich, "Machine Learning for Sequential Data: A Review," in Joint IAPR International Workshops on Statistical Techniques in Pattern Recognition (SPR) and Structural and Syntactic Pattern Recognition (SSPR), Windsor, ON, Canada, 2002.
    [45] I. D. Nicolae, P. M. Nicolae, C. M. Marinescu and R. F. Marinescu, "Evaluating the Performances of Electric Waveforms," in 8th International Conference on Modern Power Systems (MPS), Cluj Napoca, Romania, Romania, 2019.
    [46] Athanasopoulos, R. J. Hyndman and George, Forecasting: principles and practice, 2 ed., Melbourne: OTexts, 2018.
    [47] H.-M. Krolzig, Markov-Switching Vector Autoregressions: Modelling, Statistical Inference, and Application to Business Cycle Analysis, Springer-Verlag Berlin Heidelberg, 1997.
    [48] Y. LeCun, Y. Bengio and G. Hinton, "Deep learning," Nature, vol. 521, p. 436 – 444, 2015.
    [49] I. Goodfellow, Y. Bengio and A. Courville, Deep Learning, The MIT Press, 2015.
    [50] S. J. Russell and P. Norvig, "Learning in Neural and Belief Networks," in Artificial Intelligence A Modern Approach, Prentice Hall, 1995, pp. 564-570.
    [51] S. Kiranyaz, O. Avci, O. Abdeljaber, T. Ince, M. Gabbouj and D. J. Inman, "1D Convolutional Neural Networks and Applications – A Survey," arXiv, 2019.
    [52] A. Mittal, "Understanding RNN and LSTM," Medium, [Online]. Available: https://towardsdatascience.com/understanding-rnn-and-lstm-f7cdf6dfc14e. [Accessed 19 Aug 2020].
    [53] Y. Bengio, P. Simard and P. Frasconi, "Learning long-term dependencies with gradient descent is difficult," IEEE Transactions on Neural Networks, vol. 5, no. 2, pp. 157 - 166, March 1994.
    [54] S. Hochreiter and J. Schmidhuber, "Long short-term memory," Neural Computer, vol. 9, pp. 1735-1780, 1997.
    [55] J.E.Nash and J.V.Sutcliffe, "River flow forecasting through conceptual models part I — A discussion of principles," Journal of Hydrology, vol. 10, no. 3, pp. 282-290, 1970.
    [56] P. Krause, D. P. Boyle and F. Base, "Comparison of different efficiency criteria for hydrological model assessment," Advances in Geosciences, vol. 5, p. 89–97, 2005.
    [57] G. v. Rossum, Python tutorial, Amsterdam, Netherlands: CWI (Centre for Mathematics and Computer Science), 1995.
    [58] S. v. d. Walt, S. C. Colbert and G. Varoquaux, "The NumPy Array: A Structure for Efficient Numerical Computation," Computing in Science & Engineering, vol. 13, no. 2, 2011.
    [59] W. McKinney, "Data Structures for Statistical Computing in Python," in Proceedings of the 9th Python in Science Conference, S. v. d. W. a. J. Millman, Ed., 2010, pp. 56 - 61.
    [60] J. D. Hunter, "Matplotlib: A 2D graphics environment," Computing in Science & Engineering, vol. 9, no. 3, pp. 90-95, 2007.
    [61] F. Pedregosa, G. Varoquaux, A. a. M. V. Gramfort, B. Thirion, O. Grisel and M. Blondel, "Scikit-learn: Machine Learning in Python," Journal of Machine Learning Research, vol. 12, pp. 2825-2830, 2011.
    [62] G. R. Lee, R. Gommers, F. Wasilewski, K. Wohlfahrt and A. O’Leary, "PyWavelets: A Python package for wavelet analysis," Journal of Open Source Software, vol. 36, no. 4, p. 1237, 2019.
    [63] M. Abadi, A. Agarwal, P. Barham, E. Brevdo, Z. Chen, C. Citro, G. Corrado, A. Davis and J. Dean, "TensorFlow: Large-scale machine learning on heterogeneous distributed systems," arXiv, 2015.
    [64] M. Misiti, Y. Misiti, G. Oppenheim and J.-M. Poggi, Wavelet Toolbox™ User's Guide, Massachusetts: The MathWorks, Inc., 2015.
    [65] K. Fan, H. Wang and Y.-J. Choi, "A physically-based statistical forecast model for the middle-lower reaches of the Yangtze River Valley summer rainfall," Chinese Science Bulletin, vol. 53, p. 602–609, 2008.

    QR CODE