Research Article
A Kriging-Based Active Learning Algorithm for Mechanical Reliability Analysis with Time-Consuming and Nonlinear Response
Table 5
Comparison of different methods.
| Method | | | | CPU time |
| MCS | 1 × 105 | 1.406 × 10−2 (δ = 0.0264) | — | 7.264 × 105 s | FORM [1] | 45 | 4.339 × 10−4 (β = 3.33) | 0.9693 | 4.731 × 102 s | RSM + MCS [5] | 500 | 1.650 × 10−3 | 0.8832 | 4.150 × 103 s | HPF+ | 303 | 1.404 × 10−2 | 0.0014 | 2.412 × 103 s | HPF+ | 416 | 1.407 × 10−2 | 0.0007 | 3.455 × 103 s | HPF + improved condition (the proposed method) | 240 | 1.409 × 10−2 | 0.0060 | 1.827 × 103 s |
|
|