Enhancing the performance of differential evolution using orthogonal design method

Differential evolution (DE) is a simple and efficient global optimization algorithm. It has been successfully applied to solve a wide range of real-world optimization problems. However, DE has been shown to have certain weaknesses, especially if the global optimum should be located using a limited number of function evaluations (NFEs). In this paper, we incorporate the orthogonal design method into DE to accelerate its convergence rate. The orthogonal design method is not only to be used to generate the initial population, but also to be applied to design the crossover operator. In addition, two models of DE method are investigated. Moreover, the self-adaptive parameter control is employed to avoid tuning the parameters of DE. Experiments have been conducted on 25 problems of diverse complexities. And the results indicate that our approach is able to find the optimal or close-to-optimal solutions in all cases. Compared with other state-of-the-art evolutionary algorithms (EAs), our approach performs better, or at least comparably, in terms of the quality and stability of the final solutions.