Search

This paper presents a new constraint-handling genetic approach based on preserving feasibility of solutions of the economic dispatch problem. A real-coded genetic algorithm is implemented to minimize the active power generation cost while satisfying both power balance and generating limit constraints automatically during the optimization process. This is achieved by introducing a novel strategy which generates solutions within the hyperspace bounded by the problem constraints, such that individuals are only produced on the feasible part of the search space that fulfill all constraints. Computer simulations on two benchmark electrical power systems show the prowess of the proposed approach whose results are similar to those reported by other authors using different methods. [ABSTRACT FROM AUTHOR]