Security-constrained unit commitment using Mixed-Integer Programming with Benders Decomposition

The objective of this paper is to determine the 24-hour unit commitment with minimum total generation cost subjecting to power flow constraints in both normal operating state and contingency state, which is considered the security-constrained unit commitment (SCUC). In this paper, the Benders´ Decomposition technique has been applied by decomposing the problem into two parts: the master problem; and the subproblems. The master problem which is the unit commitment problem contains integer variables as the on-off status of the generating units, and therefore could be solved as Mixed-Integer Programming problem. The branch-and-cut method has been chosen to solve the master mixed-integer problem. There are two subproblems considered in this research: the power flow mismatch equations; and the operating limits of power systems. If the conditions in any subproblems are violated, the Benders Cuts corresponding to the violated constraints are added to the master problem and the mixed-integer unit commitment problem could be resolved. The proposed algorithm is tested on the IEEE 57-bus system with 7 thermal generating units. The 24-hour load data is made available along with the additional data for generating units. The results of typical unit commitment, security-constrained unit commitment under normal operating states and SCUC with contingency states are compared. The results show that the total generating costs of SCUC with contingency are higher than the others. However the committed units of that SCUC ensure that the system could still serve load when contingency occurs.