The Application Of Genetic Algorithms In Numerical Optimization

This paper consists of two parts.The first section is the introduction on GAs' theory and its applications on limit value of function optimization. Firstly, the paper finds out some important effective factors on the GAs stability and convergence speed by comparing several TSP numerical experiments with different parameters. Then, the paper proposes several novel techniques on improving the convergence speed of GAs.In the second section, the paper also proposes a new Hierarchic Genetic Algorithms (HGA), which overcome some drawbacks on simple Genetic Algorithms (SGA). The paper compares HGA and SGA by several function numerical experiments. The simulation results show that the HGA is better than SGA.