Differential Evolution with Landscape-Based Operator Selection for Solving Numerical Optimization Problems

In this paper, a new differential evolution framework is proposed. In it, the best-performing differential evolution mutation strategy, from a given set, is dynamically determined based on a problem's landscape, as well as the performance history of each operator. The performance of the proposed algorithm has been tested on a set of 30 unconstrained single objective real-parameter optimization problems. The experimental results show that the proposed algorithm is capable of producing good solutions that are clearly better than those obtained from a set of considered state-of-the-art algorithms.