Research on Two Stage Evolutionary Modeling Based on Gene Expression Programming

Gene expression programming is presented here for two stage evolutionary modeling. It uses character linear chromosomes composed of genes which encode expression trees. This feature which is different from existing algorithms allows the algorithm to perform with high efficiency when dealing with the same problem. And then, the analysis of convergence of this algorithm is mentioned. Finally, a numeric experiment is given to verify the efficiency of this algorithm. The results show that multiple highly precise ordinary differential equations (ODEs) model can be found out, and its predict values surprisingly coincide with the exact solutions.