Risk shaping of optimal electricity portfolios in the stochastic LCOE theory. (August 2018)

• Record Type: Journal Article
• Authors: Lucheroni, Carlo; Mari, Carlo
• Is Part Of: Computers & operations research. Volume 96(2018)
• Page Start: 374
• Page End: 385
• Publication Date: 2018-08
• Subjects: Operations research -- Periodicals; Electronic digital computers -- Periodicals; 004.05
• DOI: 10.1016/j.cor.2018.02.011
• Languages: English
• ISSNs: 0305-0548

Abstract

Highlights: Stochastic LCOE theory is a powerful tool to study generation portfolios in their asset composition. Stochastic LCOE theory can be used for generation portfolio optimization under different risk measures. CVaR Deviation is a risk measure particularly appropriate for this problem to account for asymmetric tail risk. The three risk sources taken into account in the model are coal, natural gas and CO2 market prices. In the model, under these three risk sources, a nuclear asset can be used as a risk free asset. Abstract: In this paper we review and extend the stochastic LCOE portfolio theory, a mean-risk analysis of electricity generation investment portfolios, focusing on the distinction between risk and deviation risk measures in terms of risk distribution shaping. Using standard and more advanced stochastic optimization risk measures, we derive optimal portfolios in the case of fossil fuels only, and in the case which includes the nuclear asset, interpreted as a risk free asset useful to hedge and reduce LCOE dispersion around its mean, in a US market case study. Four CO2 price volatility scenarios are used to illustrate how the theory handles the impact of indirect correlation among different fuel technologies induced by CO2 costs on the determination of optimal portfolios.