An improved evolutionary algorithm for handling many-objective optimization problems

详细信息    查看全文

• 作者：S. Mohammadi ; MSc^a ; ^{samane_mohammadi20@yahoo.com} ; M.A.S. Monfared^a ; ^{mas_monfared@alzahra.ac.ir} (Associate Professor) ; M. Bashiri^b ; ^{bashiri.m@gmail.com} (Associate Professor)
• 关键词：Many-objective optimization ; Reference point ; Evolutionary algorithm ; NSGA-II
• 刊名：Applied Soft Computing
• 出版年：2017
• 出版时间：March 2017
• 年：2017
• 卷：52
• 期：Complete
• 页码：1239-1252
• 全文大小：4538 K
• 卷排序：52

文摘

A new composite fitness evaluation function is proposed for many-objective cases. The convergence, diversity and user preferences are considered in this function. The proposed fitness function guides the search to the preferred region of the Pareto front. Using the proposed function in an EMO algorithm led to better results. The algorithm has comparable performance with a reasonable amount of computing.