摘要
将遗传算法与局部优化方法相结合,提出了一种用于多目标优化的混合Pareto遗传算法(HPGA)。针对遗传算法局部优化性能较差的缺点,引入直接搜索策略以增强算法的局部搜索能力。HPGA首先运行Pareto遗传算法,以得到近似的Pareto最优解;然后启动直接搜索对其进行进一步优化。仿真结果表明HPGA兼具有良好的全局优化性能和较强的局部搜索能力。与Pareto遗传算法相比,HPGA不仅提高了优化搜索的效率,而且能够保证收敛到多目标优化问题的Pareto最优前沿面。
Combining genetic search with local search, a hybrid Pareto genetic algorithm (HPGA) for multi-objective optimization is proposed. HPGA introduces local search as a means of acceleration and refinement of the solutions of genetic search. It first runs the Pareto genetic algorithm in order to obtain approximative Pareto optimal solutions. Once the Pareto genetic algorithm is over, local search is then run with each previously obtained solution to find a better solution. Simulation results in section 5 show that HPGA, compared with the known Pareto genetic algorithm (PGA), can improve efficiency of optimization and ensure a better convergence to the true Pareto optimal front.
出处
《系统仿真学报》
CAS
CSCD
2004年第5期1038-1040,共3页
Journal of System Simulation
基金
国家自然科学基金项目(69931040)
关键词
多目标优化
遗传算法
局部搜索
PARETO最优解
multi-objective optimization
genetic algorithm
local search
Pareto optimal solution
