一种基于粒子群算法求解约束优化问题的混合算法  

Hybrid algorithm based on particle swarm optimization for solving constrained optimization problems

作　　者：李炳宇[1] 萧蕴诗[1] 吴启迪[1] 

机构地区：[1]同济大学电子与信息工程学院,上海200092

出　　处：《控制与决策》2004年第7期804-807,812,共5页Control and Decision

基　　金：国家自然科学基金资助项目(70271035;60104004);国家973子项目资助(2002CB312202);上海市启明计划(03QG14053).

摘　　要：通过将粒子群算法(PSO)与差别进化算法(DE)相结合,提出一种混合算法PSODE,用于求解约束优化问题.PSODE是在PSO算法中适当引入不可行解,将粒子群拉向约束边界,加强对约束边界的搜索,同时与DE算法结合以加强搜索能力.基于典型高维复杂函数的仿真表明,该算法简单高效,鲁棒性强.A hybrid algorithm, PSODE, is proposed by combining particle swarm optimization (PSO) with different evolution (DE), for solving constrained optimization problems. PSODE is a powerful searching algorithm by keeping properly infeasible solution to attract the swarm to the constraint boundary and by combining with DE to enhance the searching ability. Simulation results on benchmark complex functions with high dimension show that the hybrid algorithm is effective, efficient and fairly robust to initial conditions.

关 键 词：约束优化问题 粒子群优化算法 群体智能 差别进化 

分 类 号：TP301.6[自动化与计算机技术—计算机系统结构]