2018144-1
研究与开发
基于对标学习的智能优化算法
谢安世
(浙江工业大学,浙江 杭州 310014)
摘 要:科研、工程和管理中的很多问题都可以转化为优化问题。应用于这些优化问题的各种方法本身就是
各种模型,设计不同的方法即设计不同的模型。将标杆管理理念建模成为一种用于单目标优化问题的元启发
式搜索方法。基于奥卡姆剃刀原则,摒弃了复杂的操作算子的概率调优规则,用一个简单的框架来组织核心
算子,从而达到许多组合算法的搜索效果。
关键词:智能优化算法;探索性与开发性;全局搜索与局部优化;标杆管理
中图分类号:N940,TP202
文献标识码:A
doi:10.11959/j.issn.1000−0801.2018144
Intelligent optimization algorithm based on benchmarking
XIE Anshi
Zhejiang University of Technology, Hangzhou 310014, China
Abstract: Many of the issues in scientific research, engineeringand management can be transformed into optimiza-
tion problems. The various methods applied to these problems were a variety of models. Designing different methods
was designing different models. The theme was to model the benchmarking philosophy in business management as a
meta-heuristic search method for single objective bound-constrained real-parameter optimization problems. Accord-
ing to the principle of Occam’s Razor, many complicated operators and their probability tuning rules were abandoned
and a simple framework was used to organize the core operators to achieve the effect of many composition algorithms.
Key words: intelligent optimization algorithm, exploration and exploitation, global search and local optimization,
benchmarking management
1 引言
元启发式优化/搜索算法的研究,从横向看,
可以分为算法自身设计研究(即设计不同的优化
搜索算法)、算法自身的理论研究(包括复杂度、
收敛性等)和算法在各学科领域的具体应用研究
收稿日期:2017−08−01;修回日期:2018−04−05
基金项目:浙江省哲学社会科学重点研究基地技术创新与企业国际化研究中心项目;浙江工业大学中小微企业转型升级协同创
新中心项目;教育部人文社会科学基金资助项目(No.17YJC630011)
Foundation Items:Research Fund from “Zhejiang Provincial Key Research Base of Philosophy and Social Sciences-Research Centre
for Technology Innovation and Enterprise Internationalization”, Research Fund from “Collaborative Innovation Center for Transfor-
mation and Upgrading of Micro, Small and Medium Enterprises, Zhejiang University of Technology”, The Ministry of Education o
Humanities and Social Science Project (No.17YJC630011)