A parallel tabu search approach to unit commitment in power systems

In this paper, a parallel tabu search (PTS) approach is developed to solve the unit commitment problem in power systems. It is a job shop scheduling combinatorial problem. The objective of the unit commitment is to operate appropriate generators in an economical sense so that the generation cost is minimized while satisfying the constraints. At present, utilities are quite keen on saving operational costs. Thus, it is of importance to calculate better solutions in the unit commitment problem. The conventional methods have made use of mathematical programming to solve the problem. However, the technique is not necessarily effective for the unit commitment problem due to the computational efficiency and accuracy in the sense of global minima. To overcome the problem, meta-heuristics are attractive because they focus on the solution accuracy. The idea of meta-heuristics is to obtain better solutions near a global minimum by repeating some rules or heuristics.