Optimizing the management of small hydroelectric plants: from the synergetic operation of the turbine system to day-ahead energy forecasting

K.-K. Drakaki, Optimizing the management of small hydroelectric plants: from the synergetic operation of the turbine system to day-ahead energy forecasting, Postgraduate Thesis, 88 pages, Department of Water Resources and Environmental Engineering – National Technical University of Athens, November 2021.



Motivated by the challenges induced by the so-called Target Model and the associated changes to the current structure of the energy market, we revisit two different aspects regarding the everyday management of Small Hydropower Plants (SHPPs) without storage capacity. The first focuses on determining an optimal operational rule for a given turbine system, while the second confronts the problem of day-ahead prediction of energy by looking for a credible forecasting model with minimal data requirements and little complexity. The task of establishing an efficient operational policy is addressed through extended theoretical analysis, in which we investigate alternative configurations of potential turbine combinations. In order to obtain generic conclusions, we provide a dimensionless formulation of the turbine mixing and the power production procedure. The proposed operation policy, next referred to as synergetic, is compared by using as reference a simpler operation rule, the so-called hierarchical. On the other hand, for the day-ahead energy forecasting problem we use as example a typical run-of-river SHPP, in the upper course of river Achelous ,Western Greece. Based on daily hydrological data for a 39-year period, we test alternative forecasting schemes of varying complexity (from regression-based to machine learning) that take advantage of different levels of information. In this respect, we investigate whether it is preferable to use as predictor the known energy production of previous days, or to predict the day-ahead inflows and next estimate the resulting energy production via simulation. Our analyses indicate that the second approach becomes clearly more advantageous when the expert’s knowledge about both the hydrological regime and the technical characteristics of the SHPP is incorporated within the model training procedure. Beyond these, we also focus on the predictive uncertainty that characterize such forecasts, with overarching objective to move beyond the standard, yet risky, point forecasting methods, providing a single expected value of power production. Finally, we discuss the use of the proposed forecasting procedure under uncertainty in the real-world electricity market.

PDF Full text: