Finite-time performance analysis for genetic algorithms  

作　　者：WANG Ling and ZHENG Dazhong(Department of Automation, Tsinghua University, Beijing 100084, China) 

出　　处：《Progress in Natural Science:Materials International》2002年第12期940-944,共5页自然科学进展·国际材料（英文版）

基　　金：the National Natural Science Foundation of China (Grant Nos. 60074012 and 60174022);Basic Research Program of Tsinghua University.

摘　　要：Finite-time performance of genetic algorithm with elitist operator in finite solution space is studied, and the relationship between evolution generation and the quality of the solution found best so far is analyzed. The estimating formulations of the expectation value as well as upper bound and lower bound for the evolution generation earliest achieving specific performance are provided.Finite-time performance of genetic algorithm with elitist operator in finite solution space is studied, and the relationship between evolution generation and the quality of the solution found best so far is analyzed. The estimating formulations of the expectation value as well as upper bound and lower bound for the evolution generation earliest achieving specific performance are provided.

关 键 词：GENETIC algorithm  FINITE-TIME performance  earliest achieving time. 

分 类 号：TP18[自动化与计算机技术—控制理论与控制工程]