Weighted Fuzzy Risk Priority Number Evaluation of Turbine and Compressor Blades Considering Failure Mode Correlations

Luping Gan, Yan-Feng Li, Shun-Peng Zhu, Yuan-Jian Yang, Hong-Zhong Huang
2014 International journal of turbo & jet-engines  
Failure mode, effects and criticality analysis (FMECA) and Fault tree analysis (FTA) are powerful tools to evaluate reliability of systems. Although single failure mode issue can be efficiently addressed by traditional FMECA, multiple failure modes and component correla tions in complex systems cannot be effectively evaluated. In addition, correlated variables and parameters are often assumed to be precisely known in quantitative analysis. In fact, due to the lack of information, epistemic
more » ... ion, epistemic uncertainty commonly exists in engineering design. To solve these problems, the advantages of FMECA, FTA, fuzzy theory, and Copula theory are integrated into a unified hybrid method called fuzzy probability weighted geometric mean (FPWGM) risk priority number (RPN) method. The epis temic uncertainty of risk variables and parameters are characterized by fuzzy number to obtain fuzzy weighted geometric mean (FWGM) RPN for single failure mode. Multiple failure modes are connected using minimum cut sets (MCS), and Boolean logic is used to combine fuzzy risk priority number (FRPN) of each MCS. Moreover, Copula theory is applied to analyze the correlation of multiple failure modes in order to derive the failure pro babilities of each MCS. Compared to the case where de pendency among multiple failure modes is not consid ered, the Copula modeling approach eliminates the error of reliability analysis. Furthermore, for purpose of quan titative analysis, probabilities importance weight from failure probabilities are assigned to FWGM RPN to reas sess the risk priority, which generalize the definition of probability weight and FRPN, resulting in a more accurate estimation than that of the traditional models. Finally, a basic fatigue analysis case drawn from turbine and com pressor blades in aeroengine is used to demonstrate the effectiveness and robustness of the presented method. The result provides some important insights on fatigue reliability analysis and risk priority assessment of struc tural system under failure correlations.
doi:10.1515/tjj-2013-0038 fatcat:z46c2pjq2zegdiyzdurfyp22je