An Improved Hybrid Genetic Algorithm using Particle Swarm Optimization

Generally, optimization is considered to be a complex problem which requires accurate and fast search methods. Due to slow convergence, traditional Genetic Algorithms (GA) are not eficient enough to solve this problem. Hence, a lot of efforts have been carried out to improve GA performance in terms of convergence rate and accuracy. Similar to Genetic Algorithm, Particle Swarm Optimization (PSO) is an evolutionary computational model which is based on swarm intelligence. Although Particle Swarm Optimization provides faster convergence, however it does not perform well due to the early convergence an d local maxima problem. Moreover, the tradeoff between fast convergence and optimum exploration is unavoidable. In this paper, we propose a new genetic algorithm method using Particle Swarm Optimization of individuals. In this method, all individuals of the so called common population will be promoted via Particle Swarm Optimization, before genetic operations have been accomplished. The experimental results have shown better convergence rate, more stability in dzfferent runs, and also better exploration accuracy compared to the pure search methods.

کلیدواژه ها:

Search Strategies ، Genetic Algorithms ، Particle Swarm Optimization ، Convergence Rate ، Exploration Accuracy