Presentation of a Bilevel Formulation for Price based Units Commitment (PBUC)

This paper presents a new model of GenCo strategy for bidding and optimal energy in a centralized market to achieve the greatest profit. Market problem in the modeling of this paper has been considered in a period of time, by considering uncertainties of value prediction and consumption load bidding and uncertainty for GenCo prices’ prediction. Given prediction error and Normal curve-based Monte Carlo method has been used for uncertainty modeling. Due to the problem’s high volume and scenarios’ high number, scenario’s reduction techniques along maintaining uncertainties modeling precision has been used to reduce computation volume and speed up the program implementation. In this paper, PBUC and the market-clearing problem has been modeled simultaneously by a multilevel problem (Bi-Level) with two levels; top level PBUC and low level market clearing. Then, this bi-level problem using Lagrange coefficients and by Karush-Kuhn-Tucker (KKT) conditions has been changed into a single level mixed integer linear problem (MILP) for optimization through typical mathematical techniques. In this paper, GAMS program and CPLEX method have been used in order to solve the MILP problems. Finally, 6-bus network (IEEE) has been examined to prove the efficiency of this new model.