TY - JOUR
T1 - CT-image-based finite element modeling with gray-level-driven material mapping for failure analysis of SiC/SiC composite turbine disks
AU - Yan, Wenzhe
AU - Fu, Qiang
AU - Liang, Shifei
AU - Liang, Yu
AU - Zhu, Lei
AU - Ai, Shigang
N1 - Publisher Copyright:
© 2026 Elsevier Ltd
PY - 2026/5/15
Y1 - 2026/5/15
N2 - An image-based finite element modeling framework is developed for structural-level analysis of SiC/SiC composite turbine disks, aiming to incorporate manufacturing-induced defect effects into failure assessment. Industrial X-ray computed tomography (CT) data are used to reconstruct the three-dimensional external geometry of the disk and to characterize internal gray-level distributions associated with defects, avoiding reliance on idealized CAD models. Defect-induced material heterogeneity is represented by quantitatively mapping CT grayscale information onto the finite element mesh through image–mesh registration and grayscale quantization, enabling region-dependent assignment of effective material properties rather than explicit geometric modeling of individual defects. Based on experimentally calibrated defect–property relationships, spatially varying elastic and strength parameters are introduced into the structural model. The resulting image-informed finite element model is applied to simulate the mechanical response and failure evolution of the turbine disk under increasing rotational loading. The results show that the proposed framework captures the coupled influence of centrifugal loading and defect-driven material non-uniformity on damage initiation and propagation at the component scale. This work establishes a consistent link between CT-derived defect characteristics, heterogeneous material representation, and structural failure behavior, providing a practical approach for strength assessment of SiC/SiC composite turbine disks.
AB - An image-based finite element modeling framework is developed for structural-level analysis of SiC/SiC composite turbine disks, aiming to incorporate manufacturing-induced defect effects into failure assessment. Industrial X-ray computed tomography (CT) data are used to reconstruct the three-dimensional external geometry of the disk and to characterize internal gray-level distributions associated with defects, avoiding reliance on idealized CAD models. Defect-induced material heterogeneity is represented by quantitatively mapping CT grayscale information onto the finite element mesh through image–mesh registration and grayscale quantization, enabling region-dependent assignment of effective material properties rather than explicit geometric modeling of individual defects. Based on experimentally calibrated defect–property relationships, spatially varying elastic and strength parameters are introduced into the structural model. The resulting image-informed finite element model is applied to simulate the mechanical response and failure evolution of the turbine disk under increasing rotational loading. The results show that the proposed framework captures the coupled influence of centrifugal loading and defect-driven material non-uniformity on damage initiation and propagation at the component scale. This work establishes a consistent link between CT-derived defect characteristics, heterogeneous material representation, and structural failure behavior, providing a practical approach for strength assessment of SiC/SiC composite turbine disks.
KW - CT grayscale–based material mapping
KW - Image-based finite element modeling
KW - Material heterogeneity
KW - Rotational failure
KW - SiC/SiC composite turbine disk
UR - https://www.scopus.com/pages/publications/105032385064
U2 - 10.1016/j.compositesb.2026.113586
DO - 10.1016/j.compositesb.2026.113586
M3 - Article
AN - SCOPUS:105032385064
SN - 1359-8368
VL - 317
JO - Composites Part B: Engineering
JF - Composites Part B: Engineering
M1 - 113586
ER -