Probabilistic-based combined high and low cycle fatigue assessment for turbine blades using a substructure-based kriging surrogate model

By Haifeng Gao, Enrico Zio, Anjenq Wang, Guangchen Bai, C. Fei. 2020

79 citations recorded by the source corpus.

Abstract Fatigue assessment for gas turbine blades under combined high and low fatigue cyclic loading is very difficult. In this paper, we propose a probabilistic approach based on a substructure-based kriging surrogate model (SKM) that embeds substructure simulation into a kriging surrogate model (KSM). A distributed collaborative SKM (DCSKM) approach is, then, proposed based on the combination of a distributed collaborative strategy (DC) with SKM. Low-cycle fatigue (LCF) life and high-cycle fatigue (HCF) life of turbine blades are predicted with DCSKM. Based on the simulation data, the combined high and low cycle fatigue (CCF) life and damage assessment are performed with respect to the linear cumulative damage by DCSKM. Further, the relationships between the number of applied cycles and CCF reliability R with survival probabilities P = 0.5 , 0.9 and 0.95, for a confidence level of 0.95, are fitted. Finally, the DCSKM is compared with the Monte Carlo method (MCM) and response surface method (RSM). It is found that (1) the CCF reliability of turbine blades decreases with increasing survival probability for the same applied cycle and decreases with increasing applied cycles under the same survival probability; (2) LCF holds a significant influence on the CCF damage of gas turbine blades; (3) the proposed DCSKM is found to be an available probabilistic analysis approach for the CCF assessment of turbine blades.

Subfields: Aerospace Engineering, Mechanics of Materials, Statistics and Probability

View Paper: https://www.semanticscholar.org/paper/af46ebe494cc2c8dea5df896f9dad9a88bed2ed3