GA/SA-based hybrid techniques for the scheduling of generator maintenance in power systems

This paper proposes the application of a genetic algorithm and simulated annealing based hybrid approach for the scheduling of generator maintenance in power systems using an integer representation. The adapted approach uses the probabilistic acceptance criterion of simulated annealing within the genetic algorithm framework. A case study is formulated in this paper as an integer programming problem using a reliability based objective function and typical problem constraints. The implementation and performance of the proposed solution technique are discussed Results contained in this paper will demonstrate that the technique is more effective than approaches based solely on genetic algorithms or solely on simulated annealing. It therefore proves to be a valid approach for the solution of generator maintenance scheduling problems.