Gaussian-Process-Based Demand Forecasting for Predictive Control of Drinking Water Networks
This paper focuses on water demand forecasting for predictive control of Drinking Water Networks (DWN) in the short term by using Gaussian Process (GP). For the predictive control strategy, system states in a finite horizon are generated by a DWN model and demands are regarded as system disturbances. The goal is to provide a demand estimation within a given confidence interval. For the sake of obtaining a desired forecasting performance, the forecasting process is carried out in two parts: the expected part is forecasted by Double-Seasonal Holt-Winters (DSHW) method and the stochastic part is forecasted by GP method. The mean value of water demand is firstly estimated by DSHW while GP provides estimations within a confidence interval. GP is applied with random inputs to propagate uncertainty at each step. Results of the application of the proposed approach to a real case study based on the Barcelona DWN have shown that the general goal has been successfully reached.
KeywordsGaussian process Water demand forecasting Drinking Water Networks Double-Seasonal Holt-Winters Predictive control
This work is partially supported by the research projects CICYT SHERECS DPI-2011-26243 and ECOCIS DPI-2013-48243-C2-1-R, both of the Spanish Ministry of Education, by EFFINET grant FP7-ICT-2012-318556 of the European Commission and by AGAUR Doctorat Industrial 2013-DI-041. Ye Wang also thanks China Scholarship Council for providing postgraduate scholarship.
- 2.Blanch, J., Quevedo, J., Saludes, J., Puig, V.: Short-term demand forecasting for operational control of thebarcelona water transport network. In: Conferencia Nacional de Jóvenes Profesionales del Agua deEspaña, pp. 1–10. Barcelona (2010)Google Scholar
- 4.Deisenroth, M.P.: Efficient reinforcement learning using Gaussian processes. Ph.D. thesis, Karlsruhe Institute of Technology (2010)Google Scholar
- 5.Fung, Y., Rao Tummala, V.: Forecasting of electricity consumption: a comparative analysis ofregression and artificial neural network models. In: 2nd International Conference on Advances in Power SystemControl, Operation and Management, vol. 2, pp. 782–787 (1993)Google Scholar
- 8.Hayati, M., Shirvany, Y.: Artificial neural network approach for short term load forecasting for illam region. World Acad. Sci. Eng. Technol. 22, 280–284 (2007)Google Scholar
- 10.Maciejowski, J.M., Yang, X.: Fault tolerant control using gaussian processes and model predictivecontrol. In: 2013 Conference on Control and Fault-Tolerant Systems (SysTol), pp. 1–12. Nice (2013)Google Scholar
- 11.Msiza, I.S., Nelwamondo, F.V., Marwala, T.: Water demand prediction using artificial neural networks and support vector regression. Digital Intell. 3(11), 1–8 (2008)Google Scholar
- 13.Pawlowski, A., Guzman, J.L., Rodriguez, F., Berenguel, M., Normey-Rico, J.E.: Predictive control with disturbance forecasting for greenhousediurnal temperature control. In: The 18th IFAC World Congress, pp. 1779–1784. Milano (2011)Google Scholar
- 14.Quinonero-Candela, J., Girard, A., Rasmussen, C.E.: Prediction at an uncertain input for gaussian processes and relevancevector machines application to multiple-step ahead time-series forecasting. Technical report, University of Glasgow, Department of ComputingScience (2002)Google Scholar
- 15.Rasmussen, C.E., Williams, C.K.I.: Gaussian Processes for Machine Learning. ISBN 026218253X. the MIT Press, Massachusetts Institute of Technology (2006)Google Scholar
- 16.Samarasinghe, M., Al-Hawani, W.: Short-term forecasting of electricity consumption using gaussian processes. Master’s thesis, University of Agder (2012)Google Scholar
- 18.Wang, Y.: Model predictive control for drinking water networks based on gaussian processes. Master’s thesis, Technical University of Catalonia (2014)Google Scholar