logo
CanyamAI summaries for academic research
language
MobileMobile
  • Phone FeatureWeChat Mini Program
EnglishEnglish
noticeNotice
BlogBlog
delete
arrow
Return

A progressive damage model-assisted GW-Gaussian process SHM method for complex composite damage quantification

delete2026-05-06
delete0
PRE
AI
H
Hutao Jing
S
Shenfang Yuan *
Y
Yuanqiang Ren *
陈建 cover
陈建 (Jian Chen)
Y
Ying Wang
DOI：10.1016/j.compositesb.2026.113770delete
deleteOriginal
deleteOriginal request for help
deleteShare
deleteSave
Abstract

Abstract

En 中文
Accurate damage quantification of composites by using structural health monitoring (SHM) is of great significance for the life management of aerospace structures. However, damage mechanism in composite structures is always complicated and progressive due to their heterogeneity and anisotropy, leading to significant dispersion and multi-stage characteristics in damage evolution. To solve this problem, a new progressive damage quantification SHM method for composite structures by combining Gaussian process (GP)-based probabilistic mode transition and progressive damage model assistance is proposed in this paper. The dominant progressive damage mechanism transition is identified by evaluating the statistical consistency between damage features and the GP predictive distribution. By combining the assistance of progressive damage modeling in capturing the entire damage evolution process, a mode-dependent GP probabilistic model is established to quantify the progressive damage. Based on active guided wave (GW)-SHM, the proposed method is validated under quasi-static shear loading using a composite orthogrid stiffened panel with a central cutout, which is derived from an important load-carrying component in reusable launch vehicles (RLV). The validation focuses on skin-stiffener debonding and subsequent skin cracking, which are the dominant progressive damage modes in such structures. Validation results demonstrate the effectiveness of the proposed method in accurately identifying the debonding-to-cracking transition at a load level of 200 kN, as well as in quantifying the debonding area with a maximum absolute error of 8.1%. A sensitivity analysis further confirms the robustness of the method under ±20% variations in key skin–stiffener interface properties.
Keywords:
progressive damage quantification
Gaussian process
structural health monitoring
composite structures
guided wave

Journal

C
composites part b: engineering
IF:
0
Papers:
568
Citations:
0

Organization

No organization information available