文献类型：期刊文献

英文题名：Differential Evolution With Event-Triggered Impulsive Control

作者：Du, Wei[1,2];Leung, Sunney Yung Sun[2];Tang, Yang[1];Vasilakos, Athanasios V.[3]

机构：[1]East China Univ Sci & Technol, Key Lab Adv Control & Optimizat Chem Proc, Minist Educ, Shanghai 200237, Peoples R China;[2]Hong Kong Polytech Univ, Inst Text & Clothing, Hong Kong, Hong Kong, Peoples R China;[3]Lulea Univ Technol, Dept Comp Sci Elect & Space Engn, S-97187 Lulea, Sweden

年份：2017

卷号：47

期号：1

起止页码：244

外文期刊名：IEEE TRANSACTIONS ON CYBERNETICS

收录：;EI(收录号：20160401850211);WOS:【SCI-EXPANDED(收录号:WOS:000391481400021)】；

基金：This work was supported in part by the National Natural Science Foundation of China under Grant 61590923, 61333010, 61422303, 61305081, in part by the Fundamental Research Funds for the Central Universities of China under Grant 222201514328, and in part by the Recruitment Program for Young Professionals (Thousand Youth Talents Plan).

语种：英文

外文关键词：Optimization - Computation theory - Benchmarking

摘要：Differential evolution (DE) is a simple but powerful evolutionary algorithm, which has been widely and successfully used in various areas. In this paper, an event-triggered impulsive (ETI) control scheme is introduced to improve the performance of DE. Impulsive control (IPC), the concept of which derives from control theory, aims at regulating the states of a network by instantly adjusting the states of a fraction of nodes at certain instants, and these instants are determined by event-triggered mechanism (ETM). By introducing IPC and ETM into DE, we hope to change the search performance of the population in a positive way after revising the positions of some individuals at certain moments. At the end of each generation, the IPC operation is triggered when the update rate of the population declines or equals to zero. In detail, inspired by the concepts of IPC, two types of impulses are presented within the framework of DE in this paper: 1) stabilizing impulses and 2) destabilizing impulses. Stabilizing impulses help the individuals with lower rankings instantly move to a desired state determined by the individuals with better fitness values. Destabilizing impulses randomly alter the positions of inferior individuals within the range of the current population. By means of intelligently modifying the positions of a part of individuals with these two kinds of impulses, both exploitation and exploration abilities of the whole population can be meliorated. In addition, the proposed ETI is flexible to be incorporated into several state-of-the-art DE variants. Experimental results over the IEEE Congress on Evolutionary Computation (CEC) 2014 benchmark functions exhibit that the developed scheme is simple yet effective, which significantly improves the performance of the considered DE algorithms.