The estimation of PV production has been widely investigated previously, where many empirical models have been proposed to account for wind and soiling effects for specific locations. However, the performance of these models varies among the investigated sites. Hence, it is vital to assess and evaluate the performance of these models and benchmark them against the common PV estimation model that accounts only for the ambient temperature. Therefore, this study aims to evaluate the accuracy and performance of four empirical wind models considering the soiling effect, and compare them to the standard model for a 103 MW PV plant in Jordan. Moreover, the study investigates the effect of cleaning frequency on the annual energy production and the plant’s levelized cost of electricity (LCOE). The results indicate almost identical performance for the adopted models when comparing the actual energy production with R2 and RMSE (root mean square error) ranges of 0.93–0.98 and 0.93–1.56 MWh for both sub-plants, with a slight superiority of the models that incorporate wind effect. Finally, it is recommended in this study to clean the PV panels every two weeks instead of every three months, which would increase annual energy production by 4%, and decrease the LCOE by 5% of the two PV sub-plants.
Abstract. 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
the problem of day-ahead prediction of power production from Small
Hydropower Plants (SHPPs) without storage capacity. Using as an example a
typical run-of-river SHPP in Western Greece, we test alternative forecasting
schemes (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
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.