A Research On Population Size Impaction On The Performance Of Genetic Algorithm

Genetic algorithm is one of the global optimal random search algorithms which simulate the evolution method in nature. It has been used in many fields. But the traditional genetic algorithm's population size is fixed, and nobody knows how much the population size should be taken. In addition, the research on population size is few. This paper studies mainly population size how to impact on convergence rate, convergence time and global search capability of the genetic algorithm for the typical benchmark functions, and furthermore propose a criterion to evaluate the performance of genetic algorithm. This paper also gives a method to choose suitable population size according the decision variable numbers, according to this theory the best population size between 4n and 6n.