A mean-variance optimization approach to the development of portfolios of renewable generation in transmission-constrained systems

We propose a general modeling framework that, under a set of assumptions, allows the representation of problems involving the construction of portfolios of renewable generators, with explicit modeling of the effects of intermittency and variability of generation over the loading of transmission facilities, as mean-variance portfolio optimization problems. The proposed formulation can be solved with classical mathematical programming techniques with little computational effort and may be used for rapid assessment and screening of renewable portfolio options and of operating scenarios, and for the computation of Pareto frontiers of efficient portfolios.