An Improved Particle Swarm Optimization Algorithm based on Adaptive Genetic Strategy for Global Numerical Optimal

Particle swarm optimization, which has attracted a great deal of attention as a global optimization method in recent years, has the drawback that continuous search based on the excellent dynamic characteristics cannot perform well with higher dimension of particles, especially in real world problems. On the contrary, the strong ability of selection, crossover, and mutation in genetic strategies can realize the double goals of maintaining diversity of population and sustaining the convergence capacity. Thus, in this paper, a hybrid particle swarm optimization model with adaptive genetic strategy based on Euclidean distance is proposed. We consider seven PSO models studied by other researchers, and apply these models to eight well known multi-peaked benchmark functions with dimension of 30, 50, and 100 for comparisons. Numerical simulation results demonstrate the stronger ability of the hybrid model to find the global optimum solutions.