一种改进的自适应差分演化算法

An Improved Adaptive Differential Evolution Algorithm

差分演化是一种简单、有效的全局数值优化算法,相关研究表明,参数的自适应能够有效提高算法的性能.提出了一种集成的混合参数自适应差分演化算法,并巧妙利用一种自适应选择机制来选择算法池中的算法,通过对25个国际标准测试函数进行测试,实验结果表明,该方法在最优解质量、稳定性、收敛速度优于其它被比较的算法.

Differential evolution （DE） is a simple and efficient global optimization algorithm for numerical optimization. Parameter adaptation is proved to be effective to improve the performance of the classical DE algo- rithm. Different parameter adaptation techniques have been proposed in the DE literature. In this paper, a hy- brid parameter adaptation method is proposed, and the performance of the proposed technique is compared with two adaptive DE variants, namely, jDE and JADE. Based on the analysis of the results, different parameter ad- aptation techniques are combined and an improved adaptive selection technique is used to choose the most suit- able one while solving a specific problem. Experimental results indicate that the improved algorithm outperforms the compared state-of-the-art DE variants.