摘要
研究了一种新颖的客观特征选择方法——蒙特卡罗估计选择(MCES)方法。采用多宇宙并行量子遗传算法对MCES方法进行优化,使其具有更快速的收敛能力和更好的搜索效率,从而能更有效地优化建模。仿真实验中用该方法进行优化布点,优选出的采样点位能够有效地代表和代替原先监测的众多点位,表明该方法用于环境监测优化布点,具有简便、快速、结果合理稳定、易于推广等优点。
This paper studied a novel method of objective characteristic choice：Monte Carlo estimated options （MCES）. Presented a multi-quantum universe parallel genetic algorithm to optimize the MCES way to a more rapid convergence and better search capabilities and efficiency, enabling more effective optimization modeling. Simulation experiments used the method to optimize the layout, the optimum sampling point could effectively represent and replace the original large number of monitoring points. The result indicates that the method used for optimization of environmental monitoring points, is simple, rapid and reasonably stable, easy to promote and so on.
出处
《计算机应用研究》
CSCD
北大核心
2009年第11期4172-4175,共4页
Application Research of Computers
基金
国家"863"计划资助项目(2007AA1Z158)
江南大学青年基金资助项目(2008LQN028)
关键词
特征选择法
优化布点
环境监测
收敛
feature selection method
optimize the layout
environmental monitoring
convergence
