A New Reliability Analysis Method Combining Adaptive Kriging With Weight Index Monte Carlo Simulation

Jiangang Ma, Ziyan Ren, Guoxin Zhao, Yanli Zhang, Chang Seop Koh

IEEE Transactions on Magnetics · 2017 · 15 citations · 10 references

Concepts

Abstract

In order to extend the application of reliability-based design optimization to engineering electromagnetic problems, the main contribution of this paper is to propose an efficient reliability analysis method: adaptive Kriging-assisted weight index Monte Carlo Simulation (MCS). The Kriging model is constructed to substitute for expensive finite-element analysis of engineering problem during optimization and reliability analysis. To improve the interpolation accuracy of Kriging, a learning function is adopted in adaptive sampling process. A weight index form of MCS method is proposed to enhance the efficiency of conventional MCS method. Finally, taking the result of conventional MCS as a reference, proposed reliability analysis method is compared with reliability index approach and sensitivity-assisted MCS methods.

References

10