论文部分内容阅读
针对实值分布估计算法(RECEDA)求解单模和多模优化问题存在的搜索空间有限和种群多样性保持能力较差等问题,提出了一种自适应实值分布估计算法(ARECEDA),该算法根据种群多样性性能指标自适应改变新种群采样过程中的方差值,使改进算法有效地提高种群多样性。实验结果表明,自适应实值分布估计算法比原有算法在优化精度上有很大提高。
In order to solve the problems that the real value distribution estimation algorithm (RECEDA) solves the single-mode and multi-mode optimization problems, such as limited search space and poor population diversity retention, an adaptive real-valued distribution estimation algorithm (ARECEDA) According to the index of population diversity, the variance value in the sampling process of new population was adaptively changed so that the improved algorithm could effectively improve the population diversity. Experimental results show that the adaptive real-valued distribution estimation algorithm greatly improves the optimization accuracy compared with the original algorithm.