Thermal generator scheduling using hybrid genetic/simulated-annealing approach

A hybrid genetic/simulated annealing approach is established for solving the thermal generator scheduling problem. It develops a method for encoding generator schedules in the hybrid approach. The method leads to a large chance of producing feasible schedules in the solution process. For the infeasible schedules produced, methods are developed to restore their feasibility. The methods developed are incorporated into the hybrid algorithm GAA2 for the determination of the most economical schedules. The original GAA2 developed previously is employed to access the cost of the generator schedules. The hybrid approach has the ability to deal with the nonconvexity of the scheduling problem. Its usefulness is demonstrated by its application to a real-life power system consisting of 13 generators