A Hybrid Differential Evolution for Numerical Optimization

Differential evolution is a well-known optimization technique to deal with nonlinear and complex problems. However, it suffers from some difficulties, such as expensive computation, problem-dependent parameters, etc. In order to tackle these problems, this paper presents a hybrid DE algorithm, called SAODE, by employing opposition-based learning (OBL) and a self-adapting mechanism to adjust parameters. Experimental results on six benchmark functions show that the proposed approach SAODE outperforms opposition-based DE (ODE), self-adapting DE (SADE), classical evolutionary programming (CEP) and fast evolutionary programming (FEP) on most test functions.