融合滤子技术的拟态物理学算法及其应用
|
摘要
针对复杂约束优化问题,提出一种滤子拟态物理学优化算法(FMAPO)。算法引入滤子技术,将个体适应度值和约束违反度构成滤子对,并共同作为个体评价指标代替罚函数,操作简便。反向学习理论、均匀分布机制和混沌变异操作的引入,保证了算法的全局搜索能力。采用几个标准约束优化测试函数进行性能测试,并将FMAPO算法用于带约束条件的起重机箱型主梁优化。通过几种典型算法进行函数测试仿真及工程优化问题的求解对比,仿真结果表明融合滤子技术的拟态物理学算法能有效求解约束问题且精度更高。
To handle complex constrained optimization problems, the paper proposed a FMAPO agorithm based on filter methods. The fitness and degree of constraint violation composed a filter to act as an evaluating indicator in the algorithm, instead of penalty function, and the operation is easy and simple. Reverse learning principle, uniform distribution and chaotic mutation were applied in the algorithm contributed to the global optimum. Simulation experiments were conducted on the well-known benchmark constrained optimization problems and the optimization of lifting box girder. The simulation results show that the FMAPO algorithm is efficient in the constrained optimization problems and the results are better, compared with several typical algorithms.
To handle complex constrained optimization problems, the paper proposed a FMAPO agorithm based on filter methods. The fitness and degree of constraint violation composed a filter to act as an evaluating indicator in the algorithm, instead of penalty function, and the operation is easy and simple. Reverse learning principle, uniform distribution and chaotic mutation were applied in the algorithm contributed to the global optimum. Simulation experiments were conducted on the well-known benchmark constrained optimization problems and the optimization of lifting box girder. The simulation results show that the FMAPO algorithm is efficient in the constrained optimization problems and the results are better, compared with several typical algorithms.
引文
[1]毕晓君,张磊.基于混合策略的双种群约束优化算法[J].控制与决策,2015,(4):715-720.
[2]贾善坡,伍国军,陈卫忠.基于粒子群算法与混合罚函数法的有限元优化反演模型及应用[J].岩土力学,2011,(S2):598-603.
[3]黄文涛,邓长虹,汪志强,舒征宇,翁毅选.基于外点罚函数法的实时低频切泵策略[J].中国电机工程学报,2013,(16):104-111,12.
[4]司呈勇,兰天,胡俊杰,汪镭,吴启迪.关于惩罚函数中惩罚系数的讨论[J].控制与决策,2014,(9):1707-1710.
[5]高海昌,冯博琴,朱利.智能优化算法求解TSP问题[J].控制与决策,2006,(3):241-247,252.
[6]王少蕾,陈维义,顾雪峰.自适应差分进化算法求解多平台多武器-目标分配问题[J].系统工程与电子技术,2013,(10):2115-2120.
[7]陈昌松,段善旭,蔡涛,刘邦银.基于改进遗传算法的微网能量管理模型[J].电工技术学报,2013,(4):196-201.
[8]唐亮,靖可,何杰.网络化制造模式下基于改进蚁群算法的供应链调度优化研究[J].系统工程理论与实践,2014,(5):1267-1275.
[9]王艳,曾建潮.一种基于拟态物理学优化的多目标优化算法[J].控制与决策,2010,(7):1040-1044.
[10]詹昕,向铁元,陈红坤,余扬波.基于搜索矢量拟态物理学算法的微电网脆弱性评估及重构[J].电工技术学报,2014,(2):74-81,92.
[11]王立国,魏芳洁.结合APO算法的高光谱图像波段选择[J].哈尔滨工业大学学报,2013,(9):100-106.
[12]唐菲,陈丽英,刘秀茹,王君龙,张林基,洪时明.一种以压力一维均匀分布为特征的长条形对顶压砧[J].物理学报,2016,(10):49-55.
[13]范勤勤,吕照民,颜学峰,郭美锦.Chemical process dynamic optimization based on hybrid different ialevolution algorithm integratedwith Alopex[J].Journal of Central South University,2013,(4):950-959.
[14]黄宇,刘玉峰,彭志敏,丁艳军.基于量子并行粒子群优化算法的分数阶混沌系统参数估计[J].物理学报,2015,(3):238-245.
[15]N I M Gould,Y Loh,D P Robinson.A Nonmonotone Filter SQP Method:Local Convergence and Numerical Results[J].SIAM Journal on Optimization,2015,25(3):1885-1911.
[16]汤俊杰.求解约束优化问题的滤子遗传算法及其在批量问题上的应用[D].华东理工大学,2015.
[17]T P Runarsson,X Yao.Stochastic ranking for constrained evolutionary optimization[J].IEEE trans on Evolutionary Computation,2000,4(3):284-294.
[18]肖赤心,蔡自兴,王勇,周经野.一种基于佳点集原理的约束优化进化算法[J].控制与决策,2009,(2):249-253,258.
[19]张延萍,杜新峰,卢青波.正交Alopex算法在机械优化设计中的应用[J].现代制造工程,2012,(7):70-73.
[2]贾善坡,伍国军,陈卫忠.基于粒子群算法与混合罚函数法的有限元优化反演模型及应用[J].岩土力学,2011,(S2):598-603.
[3]黄文涛,邓长虹,汪志强,舒征宇,翁毅选.基于外点罚函数法的实时低频切泵策略[J].中国电机工程学报,2013,(16):104-111,12.
[4]司呈勇,兰天,胡俊杰,汪镭,吴启迪.关于惩罚函数中惩罚系数的讨论[J].控制与决策,2014,(9):1707-1710.
[5]高海昌,冯博琴,朱利.智能优化算法求解TSP问题[J].控制与决策,2006,(3):241-247,252.
[6]王少蕾,陈维义,顾雪峰.自适应差分进化算法求解多平台多武器-目标分配问题[J].系统工程与电子技术,2013,(10):2115-2120.
[7]陈昌松,段善旭,蔡涛,刘邦银.基于改进遗传算法的微网能量管理模型[J].电工技术学报,2013,(4):196-201.
[8]唐亮,靖可,何杰.网络化制造模式下基于改进蚁群算法的供应链调度优化研究[J].系统工程理论与实践,2014,(5):1267-1275.
[9]王艳,曾建潮.一种基于拟态物理学优化的多目标优化算法[J].控制与决策,2010,(7):1040-1044.
[10]詹昕,向铁元,陈红坤,余扬波.基于搜索矢量拟态物理学算法的微电网脆弱性评估及重构[J].电工技术学报,2014,(2):74-81,92.
[11]王立国,魏芳洁.结合APO算法的高光谱图像波段选择[J].哈尔滨工业大学学报,2013,(9):100-106.
[12]唐菲,陈丽英,刘秀茹,王君龙,张林基,洪时明.一种以压力一维均匀分布为特征的长条形对顶压砧[J].物理学报,2016,(10):49-55.
[13]范勤勤,吕照民,颜学峰,郭美锦.Chemical process dynamic optimization based on hybrid different ialevolution algorithm integratedwith Alopex[J].Journal of Central South University,2013,(4):950-959.
[14]黄宇,刘玉峰,彭志敏,丁艳军.基于量子并行粒子群优化算法的分数阶混沌系统参数估计[J].物理学报,2015,(3):238-245.
[15]N I M Gould,Y Loh,D P Robinson.A Nonmonotone Filter SQP Method:Local Convergence and Numerical Results[J].SIAM Journal on Optimization,2015,25(3):1885-1911.
[16]汤俊杰.求解约束优化问题的滤子遗传算法及其在批量问题上的应用[D].华东理工大学,2015.
[17]T P Runarsson,X Yao.Stochastic ranking for constrained evolutionary optimization[J].IEEE trans on Evolutionary Computation,2000,4(3):284-294.
[18]肖赤心,蔡自兴,王勇,周经野.一种基于佳点集原理的约束优化进化算法[J].控制与决策,2009,(2):249-253,258.
[19]张延萍,杜新峰,卢青波.正交Alopex算法在机械优化设计中的应用[J].现代制造工程,2012,(7):70-73.