A two-stage robust investment model for a risk-averse price-maker power producer

Abstract

Generation investors face uncertainties that can have a considerable effect on the profitability and, thus, on the generation investment plan. To hedge the risks associated with uncertainties in generation investment decision-making, this paper proposes a two-stage bi-level robust model for application in risk-averse price-making generation investment considering participation in the spot electricity market. The investor is immunized in the model against worst-case realization of uncertainties, such as future demand and power produced by non-dispatchable generation units. The level of risk-aversion in the proposed model is controlled using an uncertainty budget. The resulting mixed-integer linear programming model is solved using Benders decomposition algorithm. The performance of the proposed model is demonstrated using the Garver and IEEE reliability test systems (IEEE-RTS) and the results are compared with those obtained in deterministic cases. As the results show, the proposed model is linearized to be globally optimized by existing branch-and-bound solvers with a good accuracy. A 10% uncertainty in forecasted values of demand/non-dispatchable power generation can lead to a decrease of 16.16% and 26.56% in profit in comparison with deterministic cases in the Garver and IEEE-RTS test systems, respectively. Moreover, the more the risk-averse is investor, the less profit is obtained from generation investment.