A New Method for Constructing Optimal Design of Computer Experiments
Computer experiments are widely used for the design and development of products. Therefore, a new method for constructing optimal experimental design(SOEDM) is developed in this paper. There arerwo major developments involved in this work. One is on developing a multi-objective optimal criterion by combining correlation and space-filling criteria. The other is on developing an efficient global optimal search algorithm, named as improved enhanced stochastic evolutionary (IESE) algorithm. Several examples are presented to show: the optimal designs are good in terms of both the correlation and distance criteria, the new method can be used in other experimental designs besides Latin hypercube design the number of levels of each factor is equal to the number of the experimental schemes, and the new algorithm is fast.
experimental design computer experiments stochastic evolutionary algorithm optimal criterion
Hong WU Wei-ping WANG Feng YANG
School of Information System and Management, NUDT, Changsha, 410073, China
国际会议
大连
英文
496-504
2011-10-19（万方平台首次上网日期，不代表论文的发表时间）