A new genetic approach for solving the unit commitment problem

This paper presents a new genetic approach for solving the thermal Unit Commitment (UC) problem. A parallel system model has been developed to handle the infeasibility problem in a structured way and thus to provide an effective search. Typical constraints like minimum up and down times, start up and shutdown ramps, must run and must not run have been considered. A number of important parameters related to UC problem have been identified. These parameters have been utilized to develop some intelligent mutation operators as well as two special schemes. Tests have been performed on 10, 16 and 20 unit systems over a scheduling period of 24 hours. The results are found to be better than those obtained through two genetic schemes proposed in earlier literature.