TY - GEN
T1 - Degradation Modeling of Compressive Mechanical Properties of Delamination-Defective Composites Considering Heteroscedastic Dispersion
AU - Chen, Qian
AU - Wang, Ling
AU - Wang, Han
AU - Ma, Xiaobing
AU - Ye, Kewei
AU - Liu, Yujie
N1 - Publisher Copyright:
© 2025 IEEE.
PY - 2025
Y1 - 2025
N2 - Due to the limitations of composite manufacturing processes, the mechanical properties of composite materials are significantly affected by defects such as delamination and porosity. Current degradation analyses of composite materials under defect conditions are mostly based on average or minimum performance, neglecting dispersion analysis, which restricts the generalization ability of the models. This paper takes ZT7H/5428 carbon fiber composite materials as the research object and proposes a compressive performance degradation prediction model that considers the size and location of delamination defects. By combining the heteroscedastic characteristics of the degradation process, the quantitative analysis of the uncertainty in degradation is conducted. The proposed model has been well validated with experimental data, achieving prediction R2 of 0.917, Root Mean Square Error (RMSE) of 364.96, and a prediction interval coverage probability (PICP) of 97.14% under 95% confidence. The proposed model provides guidance for the design, maintenance, and production process control of composite material structures.
AB - Due to the limitations of composite manufacturing processes, the mechanical properties of composite materials are significantly affected by defects such as delamination and porosity. Current degradation analyses of composite materials under defect conditions are mostly based on average or minimum performance, neglecting dispersion analysis, which restricts the generalization ability of the models. This paper takes ZT7H/5428 carbon fiber composite materials as the research object and proposes a compressive performance degradation prediction model that considers the size and location of delamination defects. By combining the heteroscedastic characteristics of the degradation process, the quantitative analysis of the uncertainty in degradation is conducted. The proposed model has been well validated with experimental data, achieving prediction R2 of 0.917, Root Mean Square Error (RMSE) of 364.96, and a prediction interval coverage probability (PICP) of 97.14% under 95% confidence. The proposed model provides guidance for the design, maintenance, and production process control of composite material structures.
KW - composite
KW - compressive properties
KW - degradation modeling
KW - delamination defects
KW - heteroscedastic dispersion
UR - https://www.scopus.com/pages/publications/105030066032
U2 - 10.1109/ICRMS65480.2025.00089
DO - 10.1109/ICRMS65480.2025.00089
M3 - 会议稿件
AN - SCOPUS:105030066032
T3 - Proceedings - 2025 16th International Conference on Reliability, Maintainability and Safety, ICRMS 2025
SP - 484
EP - 488
BT - Proceedings - 2025 16th International Conference on Reliability, Maintainability and Safety, ICRMS 2025
PB - Institute of Electrical and Electronics Engineers Inc.
T2 - 16th International Conference on Reliability, Maintainability and Safety, ICRMS 2025
Y2 - 27 July 2025 through 30 July 2025
ER -