改进约束控制间距法生成双目标问题的等距代表点
首发时间:2020-09-10
摘要:对于多目标优化问题,寻找一组具有良好性能的离散代表点是近年来许多学者研究的重点之一。基于约束控制间距方法,和结合折衷规划问题,我们提出了一种改进的约束控制间距方法。首先通过理想点求解折衷规划问题,找到最接近理想点的非支配点(弱非支配点)。然后,将该点作为初始迭代点,以同时搜索两个不同的受约束的区域,以找到离散的等距代表点分布在帕累托边界。最后通过数值算例,我们可以得出结论,在大多数情况下,该方法比文献中的方法花费更少的时间。
For information in English, please click here
Generating Equidistant Representative Points by Modified Constraint Controlled-Spacing Method for Biobjective Problem
Abstract:For multi-objective optimization problems, finding a set of discrete representative points with good performance is one of the focuses of many scholars in recent years. Based on the constrained control spacing method and the compromise programming problem, we propose an improved constrained control spacing method. First, by solving the compromise programming problem with ideal points, we can obtain a non-dominated point (weak non dominated point) closest to the ideal point. Then, we use this point as the initial iteration point to search two different constrained regions at the same time, so as to find the discrete equidistant representative points distributed on the Pareto boundary. Finally, through numerical examples, we can conclude that in most cases, this method takes less time than the methods in literatures.?????
Keywords: Biobjectiveprograms representative points epsilon-constraint compromise programming connectivity
引用
No.****
动态公开评议
共计0人参与
勘误表
改进约束控制间距法生成双目标问题的等距代表点
评论
全部评论0/1000