文献类型：期刊文献

英文题名：Update-based evolution control: A new fitness approximation method for evolutionary algorithms

作者：Ma, Haiping[1,2];Fei, Minrui[1];Simon, Dan[3];Mo, Hongwei[4]

机构：[1]Shanghai Univ, Sch Mechatron Engn & Automat, Shanghai Key Lab Power Stn Automat Technol, Shanghai, Peoples R China;[2]Shaoxing Univ, Dept Elect Engn, Shaoxing, Peoples R China;[3]Cleveland State Univ, Dept Elect & Comp Engn, Cleveland, OH 44115 USA;[4]Harbin Engn Univ, Automat Coll, Harbin, Peoples R China

年份：2015

卷号：47

期号：9

起止页码：1177

外文期刊名：ENGINEERING OPTIMIZATION

收录：SCI-EXPANDED(收录号：WOS:000355676700002)、、EI(收录号：20152400930090)、Scopus(收录号：2-s2.0-84930583320)、WOS

基金：This work is based on work supported by the National Science Foundation [grant no. 0826124], the National Natural Science Foundation of China [grant nos 61305078, 61074032 and 61179041] and the Shaoxing City Public Technology Applied Research Project [grant no. 2013B70004].

语种：英文

外文关键词：noisy optimization; design and analysis of computer experiments (DACE); fitness function approximation; constrained optimization; evolutionary algorithm

外文摘要：Evolutionary algorithms are robust optimization methods that have been used in many engineering applications. However, real-world fitness evaluations can be computationally expensive, so it may be necessary to estimate the fitness with an approximate model. This article reviews design and analysis of computer experiments (DACE) as an approximation method that combines a global polynomial with a local Gaussian model to estimate continuous fitness functions. The article incorporates DACE in various evolutionary algorithms, to test unconstrained and constrained benchmarks, both with and without fitness function evaluation noise. The article also introduces a new evolution control strategy called update-based control that estimates the fitness of certain individuals of each generation based on the exact fitness values of other individuals during that same generation. The results show that update-based evolution control outperforms other strategies on noise-free, noisy, constrained and unconstrained benchmarks. The results also show that update-based evolution control can compensate for fitness evaluation noise.