Hybrid genetic algorithm-based unit commitment

In the unit commitment problem, the objective is to determine the unit availability and generations that minimize the thermal system cost subject to practical constraints. The generations and availabilities are typically found hourly over a day, a week, or a month. The commitment is such that the total cost, which involves both production costs and costs associated with startup and shutdown of units, is minimal. Recently, artificial intelligence based techniques have been applied to the solution of the unit commitment problem. When implemented as stand-alone systems, they suffer from computational time limitations, especially when the systems are scaled up. In this article, a hybrid genetic algorithm incorporating a priority list order scheme is proposed for the unit commitment problem. The proposed algorithm is tested on 5-, 10-, and 26-unit systems.