Improved differential evolution algorithm and its application in complex function optimization

Abstract

When solving complex function optimization problem, Differential evolution(DE) algorithms may suffer from low convergence rate. In this paper, we propose an improved differential evolution algorithm named n-IDE. Our algorithm uses Gaussian sequence to dynamically generate zoom factors and applies an improved hybrid mutation strategy to individuals in order to improve the overall performance. We compare n-IDE with existing DE approaches using benchmark functions and the experimental result shows that n-IDE has significant improvement on the convergence rate.