A Virus Co-evolution Genetic Algorithm Based on Niche Technology
In order to avoid premature convergence and improve global search efficiency of the standard genetic algorithm (SGA) a novel virus co-evolution genetic algorithm based on niche technology (NVEGA) is developed. The virus co-evolution algorithm raises the efficiency of global convergence and accelerates local search by vertical search between child and parent generations and horizontal search among the same generation respectively, where virus populations vertically inherit excellent patterns from the outstanding family of their host populations, subsequently they horizontally pass them on to the child host populations so as to preserve and propagate good patterns from the parent host populations. As a result in the course of evolution of the host populations virus populations evolve themselves by importing the excellent patterns from their host populations and in turn accelerate evolution of host populations. Furthermore the niche technology is adopted to avoid premature convergence, enrich the variety of the population, and enhance capacity of global search. Finally a function optimization sample is given to illustrate the performance of this hybrid genetic algorithm.
genetic algorithm virus co-evolution algorithm niche technology premature convergence
Han Zhao Xiansheng Gao Lingyun Zhu
School of Machinery and Automobile Engineering Hefei University of Technology Hefei, Anhui Province, China
国际会议
2006 IEEE International Conference on Information Acquisition
山东威海
英文
894-899
2006-08-20（万方平台首次上网日期，不代表论文的发表时间）