Comparative research on genetic algorithm, particle swarm optimization and hybrid GA-PSO

Genetic algorithm (GA) has been proved to be efficient for optimization problems. It contains four operators including coding, selection, crossover and mutation. It is based on `survival of the fittest´ theory of Charles Darwin. Due to some drawbacks, it cannot be applied on all optimization problems. Several experiments have been done to improve the quality of GA. In this paper, a hybrid form of GA is presented with particle swarm optimization algorithm which is an iteration based algorithm. This hybrid algorithm has been tested on 5 global optimization test functions (beale, booth, matyas, levy, schaffer,). The simulation results shows that hybrid GA performs better than simple GA. This is by far the first paper in which a comparison table among GA, PSO and hybrid GA-PSO is presented and the testing is performed on 5 global optimization functions.