关键词: 差分进化, 自适应, 高斯分布, 反向学习

Abstract: In the study of differential evolutionary algorithms, there are many researches on the dynamic adjustment of control parameters, which aims at solving the problem of premature convergence and low searching ability. In this paper, an adaptive differential evolution algorithm based on opposition-based learning is proposed, which enhances the local search ability of the population and obtains more accurate optimal individuals through the opposition-based elite learning. At the same time, the Gaussian distribution is used to improve the exploitation ability of each individual and increased the diversity of the population, which avoids premature convergence of the algorithm. With the help of opposition-based learning and Gaussian distribution, the algorithm which is proposed in this paper achieves the balance of the global exploitation and local exploitation. In this paper, 12 test functions in CEC 2014 are used to simulate the experiment, and compared with other differential evolution algorithms. The experimental results showed that the proposed algorithm outperforms the other adaptive differential evolution in terms of the convergence speed, solution accuracy and reliability.

Key words: differential evolution, self-adaptive, Gaussian distribution, opposition-based learning