An algorithm for thermal unit maintenance scheduling through combined use of GA, SA and TS

This paper describes a method of solving a large scale long-term thermal generating unit maintenance scheduling problem. In the solution algorithm, the genetic algorithm (GA), simulated annealing (SA) and the tabu search (TS) method are used cooperatively. The solution algorithm keeps the advantages of the individual algorithms and shows a reasonable combination of local and global searches. The method takes the maintenance class and several consecutive years scheduling into consideration. Several real-scale numerical examples demonstrate the effectiveness of the proposed method.