摘要
针对EGO(Efficient Global Optimization)算法在求解昂贵优化问题中需要大量真实评价获得最优解的问题,提出一种基于并行填充准则的EGO算法。首先,设置了离群度量因子,提高分布稀疏区域样本点被选择的几率,从而提高算法优化效率;其次,引入了影响函数,依据已选择填充点对后续待选填充点的影响,构造新的EI(Expected Improvement,简称EI)函数依次选择多个填充点,并对这些点并行计算,从而减少了计算成本。在14个测试函数上对所提算法进行仿真实验,与其它典型代理模型辅助的优化算法进行测试对比,实验结果表明所提算法在有限的的评价次数下拥有更快的收敛速度。
In Efficient Global Optimization(EGO)algorithm,it requires a large number of real evaluations in solving expensive optimization problems.To solve this problem,an EGO algorithm based on parallel filling criterion is proposed in this paper.Firstly,the outlier measure factor is set to improve the optimization efficiency of the algorithm.Secondly,an influence function is introduced to construct a new Expected Improvement(EI)function that selects multiple filling points,thus reducing the calculation cost.The proposed algorithm was tested on 14 test functions,and compared with other optimization algorithms assisted by typical agent models.The experimental results show that the proposed algorithm has better optimization performance in fewer times.
作者
王凤梅
何小娟
孙超利
狄亚坤
WANG Feng-mei;HE Xiao-juan;SUN Chao-li;DI Ya-kun(School of Applied Science,Taiyuan University of Science and Technology,Taiyuan 030024,China;School of computer Science and Technology,Taiyuan University of Science and Technology,Taiyuan 030024,China)
出处
《太原科技大学学报》
2024年第6期543-548,共6页
Journal of Taiyuan University of Science and Technology
基金
国家自然科学基金(61876123)
山西省重点研发计划项目(2021020201010002)。
