The Design and Analysis of an Improved Parallel Genetic Algorithm Based on Distributed System
Genetic Algorithm (GA) is a powerful global optimization search algorithm imitating natural selection and genetic mechanism,but it has low search efficiency in the late evolving period.Parallel genetic algorithm (PGA) can improve computational efficiency and accuracy greatly,so it has become one of the main research fields of GA.This paper introduces the procedure of PGA in detail,analyses the migration limitations of traditional PGA,and puts forward an improved coarsegrained PGA based on distributed system,which adopts adaptive migration strategy to evolve.This implementation can fully tap the computing capability of distributed system to improve the convergence speed and ameliorate the population diversity in order to restrain premature convergence.The experiments show that this algorithm not only has faster convergent speed but also has more accurate calculation precision as well as higher parallel speedup.
genetic algorithm parallel genetic algorithm distributed system migration adaptive migration strategy
Yan Chen Zhimei Li
Department of Information Engineering Guilin University of Aerospace Technology,Guilin,China
国际会议
沈阳
英文
11-15
2012-09-26（万方平台首次上网日期，不代表论文的发表时间）