面向组合优化问题的智能优化算法解质量评价方法.doc

文档介绍：
面向组合优化问题的智能优化算法解质量
评价方法#
刘珍宝1,2,冯祖仁1,2,常洪浩1,2**
5
10
15
20
25
30
35
40
(1. 西安交通大学系统工程研究所,西安 710049;
2. 西安交通大学机械制造系统工程国家重点实验室,西安 710049)
摘要:如何评价智能优化算法在有限时间内所得解的质量是计算机智能基础研究和工程实践
中都亟待解决的问题,受序优化思想启发,本文针对组合优化问题,以蚁群算法求解旅行
商问题为例,提出了一种解质量的评价方法:首先讨论了在适应度空间上分析问题的可行
性,然后对智能优化算法采集的样本在适应度值空间进行前端分区域等概率采样,最后利
用二项式算法计算出的对准概率及最优解在总体中的序作为评价指标,实验结果表明了该
方法的可行性及有效性。
关键词:系统工程;解质量;智能优化算法;组合优化
中图分类号:TP18
Solution Quality Evaluation Method of Intelligent
Optimization Algorithm binatorial Optimization
Problem
LIU Zhenbao1,2, FENG Zuren1,2, CHANG Honghao1,2
(1. System Engineeering Institute, Xi'an Jiaotong University, Xi'an 710049;
2. State Key Laboratory for Manufacturing System Engineering, Xi'an 710049)
Abstract: In the domain putational intelligence and its application, how to evaluate the
quality of solutions obtained from intelligent optimization algorithm in finite time is an important
problem to be solved. binatorial optimization problem, an approach employing ACO to
solve the TSP is proposed. Firstly, the feasibility to analyse the problem in fitness space is
dicussed. Then, Frontier uiform sampling is applied to the samples collected from intelligent
optimization algrithm. Finally, based on the binomial algrithm, alignment probability and the
order of optimal solution are calculated. The simulation results show the effectiveness and
feasibility of the prop