模拟生物理想自由分布模型的萤火虫算法  

Artificial glowworm swarm optimization algorithm mimicking biological ideal free distribution model

作　　者：莫愿斌[1,2] 刘付永[1] 马彦追[1] 

机构地区：[1]广西民族大学,广西南宁530006 [2]广西混杂计算与集成电路设计分析重点实验室,广西南宁530006

出　　处：《计算机与应用化学》2014年第2期153-160,共8页Computers and Applied Chemistry

基　　金：中国博士后基金(2012M511711);广西混杂计算与集成电路设计分析重点实验室开放基金(2012HCI08);广西教育厅项目(201204LX082);广西民族大学项目(2011MDYB030)

摘　　要：通过分析生物在觅食行为中选择资源斑块的理想自由分布模型,提出1种模拟生物理想自由分布模型的萤火虫算法(IFDGSO)。该算法将萤火虫群中几个不重叠的个体最优位置的适应度视为资源斑块的食物数量,根据理想自由分布模型随机分配相应数量的萤火虫到每个资源斑块中,间隔一定的迭代次数,将各资源斑块的萤火虫重新组合,并重新随机分配。标准测试函数的仿真结果表明,改进后的IFDGSO算法比基本GSO算法有更优的性能。将IFDGSO算法用于解决伸缩绳设计和焊接条设计这2个典型的工程约束优化问题,结果表明,该方法具有收敛速度快、优化精度高、稳定性好的特点,具有较好的全局寻优能力。A novel glowworm swarm optimization （GSO） algorithm, ideal free distribution GSO, is proposed based on the analysis of ideal free distribution （IFD） model, in which, several non-overlapping personal best positions of the glowworms are selected, and their fitness values are regarded as food amount of resource patch. Glowworms are randomly assigned to each resource patch according to ideal free distribution model. After a certain number of iterations, all sub-population glowworms are regrouped and randomly assigned. The test results of standard test functions show that the IFDGSO algorithm than the basic GSO algorithm has better performance. It shows, from the simulation results of two typical engineering optimization functions, that the algorithm has the performance of rapid convergence rate, high precision of optimization and good stability. It indicates that the algorithm has better performance in global optimization.

关 键 词：萤火虫算法 理想自由分布模型 资源斑块 伸缩绳 焊接条 

分 类 号：TP18[自动化与计算机技术—控制理论与控制工程]