http://chineseinput.net/에서 pinyin(병음)방식으로 중국어를 변환할 수 있습니다.
변환된 중국어를 복사하여 사용하시면 됩니다.
Pei Dong Wang(왕베동),최동훈 대한기계학회 2017 대한기계학회 춘추학술대회 Vol.2017 No.5
For sequential approximate global optimization (SAGO), use of an appropriate metamodel is important for efficiently finding the global minimum. In this study, we adopted a radial basis function (RBF) as a metamodel and constrained optimization by radial basis function interpolation (COBRA) proposed by Regis in 2014 as a SAGO method. Difficulty of selecting an appropriate RBF, i.e. selecting appropriate type and associated parameter (if any) of an RBF, lies in it is different from problem to problem. To cope with this intrinsic difficulty, we chose seven mathematical problems representing various kinds of problems, assessed performances of various RBFs using these example problems, and recommended the types and associated parameters (if any) of radial basis functions whose performances were generally good for most of the example problems. the types of RBFs examined were Gaussian, Multi-quadric, and thin plate spline, and associated parameter values (if any) examined were in a range. Also, the number of initial sample points was varied from NDV+1 to 50*NDV, where NDV denotes the number of design variables, to consider its influence on performance. Analyzing all these test results, we recommended the types and parameters of RBFs that are expected to perform well for the SAGO of many problems.