关键词: 人工智能, 优化, 混合优化算法, 早熟, 收敛性

Abstract: A real-coded hybrid optimization algorithm, IGPSOA, was proposed, which was based on genetic algorithm, particle swarm optimization and immune algorithm. The nonlinear competition and selection methods among several crossover offsprings operator and particle swarm optimization operator were proposed to increase the efficiency of the algorithm. Immune selection and new member creation operators were used to retain the diversity of the population and to avoid premature and local convergence. The convergence and computing complexity were theoretically analyzed. The robustness of the algorithm and the range of the parameters were tested by numerical experiments. Testing experiments on 7 benchmark functions show that IGPSOA not only improves the global optimization performance and quickens the convergence speed but also obtains robust solution with better quality, indicating that it is a promising approach for global optimizations.

Key words: artificial intelligence, optimization, hybrid optimization algorithm, premature, convergence