Maintenance scheduling of generation companies in electricity market based on two-stage stochastic programming approach under the risk of power unit’s disruption

Since the past two decades, the electricity industry has begun to change from monopolized and under the direct supervision of the state to privatization. Therefore, the power unit’s outage scheduling in the electricity market has become one of the most important issues that affect system reliability. The independent system operator, as a public authority, is responsible for maintaining the desired level of network reliability concerning the proposed arrangements of outages. In most classical models assumed that units will always operate as schedules, while in the real world, units are under the risk of disruption. In this paper, a generation maintenance scheduling problem based on a two-stage stochastic programming approach under the risk of the units’ failure is considered. A scenario is considered as an event where some of the generation units have failed due to the disruption and, the others can provide service. In the first stage, maintenance scheduling decisions are set. Then, generation amounts are made in the second stage. Then the independent operator system determines the corrective signals based on the power units’ contribution in decreasing the reliability index. The generation company reviews and modifies its maintenance scheduling and sends them back to the independent system operator. This iterative procedure will continue until the optimal reliability level reach. To solve the proposed model, a combination method consisting of particle swarm optimization and simplex is proposed. The capability of the proposed algorithm is evaluated on an IEEE reliability test system and the results are reported.