Hybrid Pso/ Self-adaptive Evolutionary Programs for Economic Dispatch with Nonsmooth Cost Function

This paper investigates into hybridization between PSO and self-adaptive evolutionary programming techniques for solving economic dispatch (ED) problem with non-smooth cost curves where conventional gradient based methods are in-applicable. The convergence capability of evolutionary programming technique is enhanced with hybridization of self-adaptive evolutionary programming technique with PSO intelligence. Three types of hybridization in between PSO and self-adaptive classical EP (CEP)) i.e. PSO-CEP, CEP-PSO and CEP-PSO′ are examined The performances of the hybrid algorithms are demonstrated on a moderately large power system with 40 units and comparison is drawn among floating point GA (GAF), CEP, PSO, PSO-CEP, CEP-PSO and CEP-PSO′ methods in terms of the solution quality and computation efficiency. The simulation results show that CEP-PSO′ method is the most efficient in finding higher quality solutions in non-convex ED problems.