Reliability-based analysis method of fluid dynamics for turbomachinery with interval distribution parameters. Issue 6 (3rd June 2021)

• Record Type: Journal Article
• Authors: XiaoZhang, Q. U.; Yu, Jianghong; Yao, Qishui; Lv, Shuangyan
• Is Part Of: Engineering optimization. Volume 53:Issue 6(2021)
• Page Start: 919
• Page End: 940
• Publication Date: 2021-06-03
• Subjects: Hybrid reliability analysis -- fluid dynamics -- turbomachinery -- probability model -- interval model; Engineering design -- Periodicals; Mathematical optimization -- Periodicals; 620.0042
• DOI: 10.1080/0305215X.2020.1758688
• Languages: English
• ISSNs: 0305-215X

Abstract

Abstract : Machining level and manufacturing accuracy have an important influence on the reliability of turbomachinery flow physics and dynamic performance. This article proposes a hybrid uncertainty analysis method for fluid dynamics problems with manufacturing constraints. The parameters with sufficient and insufficient experimental samples are processed by the stochastic method and interval method, respectively, which can quickly solve the difficulties of epistemic uncertainty modelling caused by the lack of experimental samples. On this basis, an efficient iterative algorithm is proposed to solve the hybrid reliability model based on the update response surface, and the interval of the failure probability of fluid dynamics for turbomachinery is calculated. Two examples of cooling axial flow and centrifugal turbomachinery are given to verify the validity of the proposed method. The results show that the reliability of the turbomachinery flow performance can be significantly improved by increasing the uncertainty level of key parameters.