A Kriging-Based Active Learning Algorithm for Mechanical Reliability Analysis with Time-Consuming and Nonlinear Response

Table 3

Comparison of the with different methods.

Reliability method			(10^−2)
Learning function	Stopping criterion
	MCS	1 × 10^5	2.873	0.01839	—
HPF		58	2.872	0.01839	0.00035
	P(C) ≥ 0.99	65	2.873	0.01839	0.00000
	Improved criterion	36	2874	0.01838	0.00035
EFF		64	2.870	0.01840	0.00104
	P(C) ≥ 0.99	73	2.873	0.01839	0.00000
	Improved criterion	42	2.892	0.01832	0.00661
ERF		61	2.871	0.01839	0.00070
	P(C) ≥ 0.99	69	2.873	0.01839	0.00000
	Improved criterion	41	2.847	0.01847	0.00905
U		68	2.873	0.01839	0.00000
	P(C) ≥ 0.99	68	2.873	0.01839	0.00000
	Improved criterion	40	2.873	0.01839	0.00000