| 摘要 |
<p>According to an embodiment of the present invention, a method for sequential approximate optimization design comprises the steps of: generating a multi-fidelity meta-model (MF) based on high-fidelity data and low-fidelity data; and determining an optimal solution using a generalized expected improvement (GEI) value calculated through a predicted value and standard error of the generated MF. Thus, a global optimal solution may be obtained by using the MF and a sequential sampling technique based on a Bayesian approach.</p> |
