Hybrid simulated annealing/genetic algorithm approach to short-term hydro-thermal scheduling with multiple thermal plants

This paper first develops a new short-term hydro-thermal scheduling formulation which takes into consideration of scheduling the thermal units as well as the hydro and thermal generations in a schedule horizon consisting of a number of intervals. Efficient methods, which can avoid unnecessary commitment or decommitment of thermal generators and the employment of hydro generators to regulate the peak system demands, are developed and incorporated into the formulation to reduce the fuel cost. The formulation is then combined with simulated annealing (SA), genetic algorithms (GAs), and two hybrid optimisation techniques to established SA-based, GA-based and hybrid-based algorithms. The equality constraints, inequality and physical constraints of the hydro-thermal systems are fully taken into account in the algorithms. Their performance is validated by applying the algorithms to different test systems.