Title: Evaluation of a hydro-economic forecasting system as a support tool for energy trading
Authors: Nicola Di Marco - Free University of Bozen/Bolzano (Italy) [presenting]
Francesco Ravazzolo - Free University of Bozen-Bolzano (Italy)
Maurizio Righetti - Free University of Bozen-Bolzano (Italy)
Abstract: Real-time balancing of the electricity grid is increasingly challenging due to intermittent supply of energy produced by non-programmable energy sources, such as wind farms, solar, photovoltaic and run-of-the river power plants. Operators on electricity markets have therefore to be careful in scheduling the amount of energy and the bidding price in order to optimize the energy trading as well as to minimize electricity grid unbalancing. The aim is to help trading operators involved in hydro-power energy generation combining forecasts of Northern Italian hourly electricity prices and water inflow to a run-of-the-river power plant, installed in the South-Tyrol region, East-Northern Italy. Prices are predicted with an ARMA model augmented with demand and production information such as renewable energy resources and then combined with hourly water inflow predictions (a proxy for the energy production), which reliability is mostly affected by weather forecast uncertainties. The accuracy of the hydro-economic forecasts are evaluated through a back-test carried out over the period May-October 2019, where hydro-power production data are available. Results show promising performance of the proposed system, even though the accuracy changes over the year and depending on the statistical metric employed.