AbstractsBusiness Management & Administration

A stochastic optimization approach for pricing hydropower regulation capacity in electricity markets

by Bjørnar Fjelldal




Institution: Norwegian University of Science and Technology
Department:
Year: 2014
Record ID: 1294030
Full text PDF: http://urn.kb.se/resolve?urn=urn:nbn:no:ntnu:diva-26955


Abstract

With the introduction of an increasing number of regulation markets,hydropower producers face multiple opportunities regarding how toutilize their water optimally. Such regulation markets can increaseprofits drastically compared to a single day-ahead spot market, thus theuse and comprehension of multi-market optimization tools are valuablefor all hydropower producers with storage.This report contains the development of a stochastic multi-stageoptimization model where the effects of committing upon different levelsof regulation obligations in electricity markets are investigated seen froma hydropower producer’s perspective. A mathematical formulation waswritten, and a corresponding optimization code implemented. Ananalysis was further conducted through a case study where a stochasticspot price was constructed and optimized upon by an implementedwatercourse with authentic properties.Comparison between different levels of regulation obligations, the effectconcerning the introduction of new information during the stochasticoptimization period and the benefits of multiple executions (MonteCarlo-simulations) are among the essential output data analyzed.The results indicate such comparisons of regulation obligations to beuseful when deciding whether or not to provide regulation capacity, andif so, at which level. Also, the value of good information should not beunderestimated when participating in markets with uncertaintyregarding the future.The market designs are affecting the choice and design of theoptimization strategy used – in markets with more complex properties,analyzing multiple regulation obligations may prove hard to present andtime consuming to process. Thus, using a forecasted price of regulationmarkets is often preferred, but unfortunately such forecasts are hard toobtain. Nevertheless, decision-making tools regarding optimal levels ofregulation obligations can be utilized by considering factors such as theneeded regulation price to break even from a potential loss in the spotmarket, as well as the marginal cost of increasing a regulation obligation.