Data-driven stochastic modeling for multi-purpose reservoir simulation. Issue 1 (2nd January 2018)

• Record Type: Journal Article
• Authors: Khadr, Mosaad; Schlenkhoff, Andreas
• Is Part Of: Journal of applied water engineering and research. Volume 6:Issue 1(2018)
• Page Start: 40
• Page End: 47
• Publication Date: 2018-01-02
• Subjects: data-driven modeling -- stochastic modeling -- reservoir operation -- drought -- Standardized Precipitation Index; Water-supply engineering -- Periodicals; Water-supply engineering; Periodicals; 627.05
• DOI: 10.1080/23249676.2016.1209439
• Languages: English
• ISSNs: 2324-9676

Abstract

Abstract : This paper presents an integration of data-driven modeling and stochastic models for simulation of reservoir operation. The simulation model developed in this study was applied to the Ruhr river reservoirs system in Germany. An adaptive neuro-fuzzy inference system, Thomas–Fiering model and hidden Markov model were integrated in a simulation model. The set of model input included the time of the year, reservoir storage, inflow and Standardized Precipitation Index; and the target output was the reservoir release. Predicted and observed release values were evaluated using several common evaluation criteria. Results of model performance showed that the proposed model is capable of simulating reservoir operation and provides reliable reservoir release prediction. Results showed also that the proposed approach could be a good tool at the real-time operation stage to quickly check operational alternatives due to emergency events or planning and real-time incongruence.