摘要
讨论了一种基于分布式并行模型的并行克隆选择算法,并在4核CPU的计算机上进行了验证。该并行算法中,多个子种群代替了原来单一的种群,每个子种群独立地进化,在完成一次进化后每个子种群中最好的个体将取代其他种群最坏的个体。并行算法不仅克服了能量值较早收敛的缺点,而且能有效地寻找到全局最优能量值。实验结果显示,改进后的算法性能有了显著提高。
A parallel clonal selection algorithm ( CSA), which was implemented on OpenMP based distributed computing model in a four - core computer, was proposed. In the algorithm, several sub - populations replaced the original single population, each sub - population evolved independently, and the current best individual was distributed into all the sub - populations. The parallel al- gorithm overcame premature convergence and found global optima efficiently. According to experiment, the proposed algorithm gains better performance.
出处
《武汉理工大学学报(信息与管理工程版)》
CAS
2011年第6期920-923,共4页
Journal of Wuhan University of Technology:Information & Management Engineering
基金
国家自然科学基金资助项目(60803160)
中国博士后科学基金资助项目(20060400275)
湖北省自然科学基金重点资助项目(2009CDA136
2009CDA034)
湖北省教育厅科研基金资助项目(Q20101110
D2009110)
武汉市科技攻关计划基金资助项目(201110821225)