Abstract
Recently, anomalies in dam inflows have occurred in Japan and around the world. Owing to the sudden and extreme characteristics of rainfall events, it is very difficult to predict dam inflows and to operate dam outflows. Hence, dam operators prefer faster and more accurate methods to support decision-making to optimise hydroelectric operation. This paper proposes a machine learning method to predict dam inflows. It uses data from rain gauges in the dam regions and upstream river level sensors from previous three hours and predicts the dam inflow in the next three hours. The method can predict the time of rise to the peak, the maximum level at the peak, and the hydrograph shapes to estimate the volume. The paper presents several experiments applied to inflow time series data for 10 years from the Kanto region in Japan, containing 55 floods events and 20,627 time stamped dam in-flow points measured at 15-min intervals. It compares the performance of four regression prediction models: generalised linear model, additive generalised linear model, regression tree and gradient boosting machine and discusses the results.
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
References
UN News Centre: EM-DAT International Disaster Database (2016). www.emdat.be & Reliefweb-reliefweb.int/disaster
Munich, R.: Geo risks research: NatCatSERVICE. https://www.iii.org/fact-statistic/facts-statistics-global-catastrophes. Accessed 31 December 2017
Japan floods: city of Joso hit by ‘unprecedented’ rain BBC, 10 September 2016 http://www.bbc.com/news/world-asia-34205879. Accessed 31 December 2017
Othman, F., et al.: Reservoir inflow forecasting using artificial neural network. Int. J. Phys. Sci. 6(3), 434–440 (2011)
Attygalle, D., et al.: Ensemble forecast for monthly reservoir inflow: a dynamic neural network approach (2016). https://doi.org/10.5176/2251-1938_ors16.22
Shortridge, J.E., et al.: Machine learning methods for empirical streamflow simulation: a comparison of model accuracy, interpretability and uncertainty in seasonal watersheds. Hydrol. Earth Syst. Sci. 20, 2611–2628 (2016)
Li, B., et al.: Comparison of random forests and other statistical methods for the prediction of lake water level: a case study of the Poyang Lake in China. Hydrol Res. 47(S1), 69–83 (2016)
Yang, J.-H., et al.: A time-series water level forecasting model based on imputation and variable selection method. Comput. Intell. Neurosci. 1–11 (2017)
Moeeni, H., et al.: Monthly reservoir inflow forecasting using a new hybrid sARIMA genetic programming approach. J. Earth Syst. Sci. 126, 18 (2017)
Hofmann, M., Klinkenberg, R.: RAPID MINER: Data Mining Use Cases and Business Analytics Applications. Data Mining and knowledge Discovery Series. CRC Press, Cambridge (2014)
Kotu, V., Deshpande, B.: Predictive Analytics and Data Mining: Concepts and Practice with Rapidminer. Morgan Kaufmann, New York (2015)
International Hydropower Association (IHA): cost-benefit and economic performance. Hydropower Sustainability Assessment Protocol. Accessed 16 February 2018
Mattmann, M., et al.: Hydropower externalities: a meta-analysis. Energy Econ. 57, 66–77 (2016)
International Renewable Energy Agency (IRENA): renewable energy technologies: cost analysis series. Hydropower vol. 1: Power Sector (2012)
Domingos, P.: MetaCost: A General framework for making classifiers cost-sensitive. In: ACM Conference on Knowledge Discovery in Databases, pp. 165–174 (1999). 10.1.1.15.7095
Ling, C.X., et al.: Cost-sensitive learning and the class imbalance problem. In: Sammut, C. (ed.) Encyclopedia of machine learning. Springer, Berlin (2008)
Pavlyshenko, B.M.: Linear, machine learning and probabilistic approaches for time series analysis. In: IEEE International Conference on Data Stream Mining and Processing (2016)
Wedderburn, R.W.M.: Quasi-likelihood functions, generalized linear models and the Gauss–Newton method. Biometrika 61(3), 439–447 (1974)
McCullagh, P., Nelder, J.: Generalized Linear Models. Monographs on Statistics and Applied Probability. Chapman & Hall, London (1983)
Hardin, J.W., Hilbe, J.M.: Generalized Linear Models and Extensions, 3rd edn. Stata Press, College Station (2012)
Hastie, T., Tibshirani, R., Friedman, J.: The Elements of Statistical Learning; Data Mining, Inference and Prediction, 2nd edn. Springer, Berlin (2009)
Hastie, T., Tibshirani, R.: Generalize additive models. Stat. Sci. 1, 297–318 (1986)
Brieman, L., Friedman, J. et al.: Classification and Regression Trees. Wadsworth (1984)
Ripley, B.D.: Pattern Recognition and Neural Networks. Cambridge University Press, Cambridge (1996)
Friedman, J., Hastie, T., Tibshirani, R.: Additive logistic regression: a statistical view of boosting (with discussion). Ann. Stat. 28, 307–337 (2000)
Friedman, J.: Greedy function approximation: a gradient boosting machine. Ann. Stat. 29(5), 1189–1232 (2001)
Scholkopf, B., Freund, Y.: Boosting: Foundations and Algorithms. MIT Press, Cambridge (2012)
Efron, B., Hastie, T.: Computer Age Statistical Inference: Algorithms, Evidence and Data Science. Cambridge University Press, Cambridge (2016)
Yee, T.W., Wild, C.J.: Vector generalized additive models. J. R. Stat. Soc. Ser. B 58(3), 481–493 (1996)
Amer, M., Goldstein, M., et al.: Enhancing one-class support vector machines for unsupervised anomaly detection. In: 11th ODD’2013, Chicago (2013)
Acknowledgements
I wish to thank the DaMEMO committee and referees for their time and valuable comments. I also wish to thank Yachiyo Engineering Co., Ltd. for various support based on big data and AI project outcomes since 2012. I am also grateful to Mizuno Takashi for providing development opportunities and Amakata Masazumi for providing domain knowledge in river and dam engineering.
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2018 Springer Nature Switzerland AG
About this paper
Cite this paper
Takato, Y. (2018). Dam Inflow Time Series Regression Models Minimising Loss of Hydropower Opportunities. In: Ganji, M., Rashidi, L., Fung, B., Wang, C. (eds) Trends and Applications in Knowledge Discovery and Data Mining. PAKDD 2018. Lecture Notes in Computer Science(), vol 11154. Springer, Cham. https://doi.org/10.1007/978-3-030-04503-6_34
Download citation
DOI: https://doi.org/10.1007/978-3-030-04503-6_34
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-030-04502-9
Online ISBN: 978-3-030-04503-6
eBook Packages: Computer ScienceComputer Science (R0)