Generation maintenance scheduling via robust optimization

It is essential to develop versatile decision-making tools for GENCOs by which the most efficient maintenance schedules in highly competitive electricity markets are devised. The key of preventive Generation Maintenance Scheduling (GMS), or Outage Planning, is to determine the optimal time period for which generating units of a GENCO should be taken off-line over the course of a one-year planning horizon. In practice, GMS is mathematically modeled as an optimization problem with a multitude of variables and uncertain parameters which need to be determined for optimal commercial decisions. This context explores a robust optimization-based procedure to evaluate the impact of price uncertainty on GENCOs' profits and their desired maintenance schedules. In contrast with the probabilistic and stochastic methods, the proposed approach is more practical because it only requires an uncertainty interval, rather than a probability distribution of the uncertain data. To demonstrate the tremendous performance of the proposed approach, a GENCO with three generation units is investigated.