Structural dynamic reliability estimation with advanced extremum Kriging method. Issue 5 (January 2021)

• Record Type: Journal Article
• Authors: Zhao, Y J; Lu, C; Li, H; Liu, H T; Fei, C W
• Is Part Of: IOP conference series. Volume 1043:Issue 5(2021)
• Publication Date: 2021-01
• Subjects: Materials science -- Periodicals; 620.1105
• DOI: 10.1088/1757-899X/1043/5/052030
• Languages: English
• ISSNs: 1757-8981

Abstract

Abstract: To enhance the computational performance of the structural dynamic reliability estimation, advanced extremum Kriging method (AEKM) is proposed fusing Kriging surrogate model, multi-population genetic algorithm and response surface method-based extremum thought (ERSM) in this paper. The applicability of the AEKM is verified via the blisk radial deformation dynamic reliability analysis of an aeroengine high-pressure compressor. The results show that the reliability level of the blisk is 0.9956 as the allowable value of radial running deformation 1.75×10 −3 m. Moreover, the developed AEKM is superior to other three surrogate models, including ERSM, K-ERSM and IK-ERSM, in terms of computing precision and analytical efficiency. Besides, the analytical precision of the developed method is basically consistent with the direct simulation, and the analytical time is obviously superior to the direct simulation. Therefore, the AEKM is illustrated to be an effective way for structural dynamic reliability analysis, which enriches complex structural reliability theory.