Your new experience awaits. Try the new design now and help us make it even better

ORIGINAL RESEARCH article

Front. Mater.

Sec. Polymeric and Composite Materials

Damage Evolution Mechanism of FRP-Constrained RC Columns Under Coupled Wind-Sand-Freeze-Thaw Erosion

Provisionally accepted
  • Inner Mongolia University of Technology, Hohhot, China

The final, formatted version of the article will be published soon.

Reinforced concrete (RC) structures in cold, arid regions are subjected to simultaneous wind-sand erosion and freeze–thaw cycling, resulting in progressive deterioration of mechanical performance. To quantify this coupled degradation and assess reinforcement strategies, laboratory simulation tests were conducted on FRP-strengthened RC columns under controlled wind-freeze conditions (26 m/s sand exposure and −20°C to +20°C thermal cycling). After 200 cy-cles, the compressive and flexural strengths of plain concrete decreased by 32.4% and 36.7%, respectively, whereas CFRP-, BFRP-, and GFRP-reinforced specimens retained 87.2%, 84.6%, and 82.1% of their initial strength. Surface abrasion, pore connectivity, and interface deterioration were identified as primary degradation mechanisms. A hybrid prediction framework incorporating physical constraints and Weibull reliability mapping was developed to simulate long-term strength evolution. Compared with conventional models (ANN, SVR, DT, Bagging, RF, AdaBoost, GB, XGBoost), the proposed framework demonstrated superior predictive accuracy, improving R² from 0.9185 to 0.9764 and reducing RMSE from 1.5458 kN to 0.8976 kN. The incorporation of degradation laws ensured monotonic strength decline and eliminated non-physical oscillatory predictions observed in purely data-driven models. Reliability analy-sis further showed that, after 200 cycles, the reliability index of plain RC columns decreased to 0.25, while FRP-reinforced specimens maintained values above 0.65. This study provides experimental evidence and computa-tional insight into the coupled degradation mechanism of FRP-strengthened RC columns and demonstrates that phys-ically constrained hybrid learning offers an effective and interpretable approach for long-term durability assessment under severe environmental exposure.

Keywords: Freeze–thaw coupling, FRP-Confined RC Columns, Hybrid machine learning, Physical constraint embedding, Wind–sand erosion

Received: 31 Oct 2025; Accepted: 02 Jan 2026.

Copyright: © 2026 Ren and A. This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY). The use, distribution or reproduction in other forums is permitted, provided the original author(s) or licensor are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms.

* Correspondence: Wenhao Ren

Disclaimer: All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article or claim that may be made by its manufacturer is not guaranteed or endorsed by the publisher.