Economic Load Dispatch is one of the optimization problems in power systems. In this paper Genetic Algorithm (GA) based approach has been applied to solve the Economic Load Dispatch with valve-point effect. GA is a robust and powerful adaptive search technique that derives models from the genetic processes of biological organisms based on evolution theory. Furthermore, the GA parameters were set optimally to yield smoother and faster fitness convergence. Binary GA and Continuous GA have been implemented and their performances are compared. The solutions of the economic load dispatch with valve point effect for three cases are solved by genetic algorithm. Test results are given and compared. It will be shown that decimal coding is more effective than binary coding in evaluating the optimal solution. It is suggested that decimal coding provides the best results in terms of convergence and accuracy.