Clustering for multi-objective thermal generator scheduling

In this paper, a hybrid approach to solve thermal generator scheduling is proposed. The proposed approach combined modified priority list method with 2-opt (MPL-2OPT) heuristic search method to solve the unit commitment problem and the economic load dispatch problem is solved by lambda iteration. The proposed method MPL-2OPT is first used to solve thermal generating unit scheduling with a single objective which is to minimize the total system operation cost taking into consideration the various constraints like unit generation limit, minimum up time of the generator and minimum down time of the generator, the spinning reserve and power balance constraint at each time interval of the whole duration of the scheduling horizon. After testing the proposed algorithm on a single objective, another objective emission cost is taken into account to form a multi-objective thermal generating unit-scheduling problem. MPL-2OPT is used to investigate the tradeoff relationship between the emission cost and the total system operation cost. The proposed algorithm has been applied to different system sizes ranging from 10 units system to 60 units system.