一种改进的多目标粒子群优化算法及其应用.pdf

第27卷第9期 Vol. 27 No. 9 控制与决策 Control and Decision 2012年9月 Sep. 2012 一种改进的多目标粒子群优化算法及其应用文章编号:1001-0920 (2012) 09-1313-07 冯琳,毛志忠,袁平(东北大学信息科学与工程学院,沈阳110819) 摘 要:针对多目标粒子群优化算法在求解约束优化问题时存在难以兼顾收敛性能和求解质量这一问题,,同时给出了速度迁移策略、,达到了减少电量消耗、缩短冶炼时间、延长炉衬使用寿命的目的,同时表明了该算法的有效性. 关键词:粒子群算法;多目标约束优化;速度迁移;自适应变异;聚类免疫网络;供电策略中图分类号: 文献标志码:A An improved multi-objective particle swarm optimization algorithm and its application FENG Lin, MAO Zhi-zhong, YUAN Ping (School of Information Science and Engineering,Northeastern University,Shenyang 110819,: FENG Lin,E-mail:******@.) Abstract:::Considering that the multi-objective particle swarm optimization(MOPSO) algorithm can not give simultaneously attention to convergence performance and solutions quality when it deals with constrained optimization problems, an improved MOPSO algorithm based on work(IN-MOPSO) is proposed. In IN-MOPSO, the information of populations exchange through work in IN-MOPSO in order to achieve cooperative search of both MOPSO and arti?cial work(AIN) for solution space. Meanwhile, an improved migration method of particle velocity, an improved adaptive variance mutation method and clustering work are proposed in order to enhance the function of MOPSO and AIN. The global convergence properties and converge