Skip to main navigation Skip to search Skip to main content

Autogenous healing of cracked ultra-high performance seawater sea sand concrete under multi-factor environments: Experimental mechanisms and interpretable prediction

  • Zaixian Chen
  • , Xinghao Liu*
  • , Hao Fu
  • , Pang Chen
  • *Corresponding author for this work
  • Harbin Institute of Technology Weihai
  • Key Lab of Civil Engineering Structure and Disaster Prevention in Universities of Shandong
  • Hebei University of Technology

Research output: Contribution to journalArticlepeer-review

Abstract

Ultra-high performance seawater sea sand concrete (UHPSSC) contains a high volume of unhydrated particles, conferring superior autogenous self-healing potential. This study quantitatively evaluates the healing kinetics of UHPSSC under the coupled influences of crack width, environmental humidity, temperature, and diverse aqueous environments. Micro-analytical evidence obtained via XRD and SEM elucidates the underlying healing mechanisms of UHPSSC in freshwater and seawater environments. Results indicate that 100 μm cracks closed completely within 28 days under freshwater immersion. While elevated temperatures accelerate initial healing kinetics, a 60 °C environment eventually limits late-stage hydration. Notably, UHPSSC demonstrates enhanced healing efficiency compared to traditional UHPC across all environmental conditions. The self-healing products primarily consist of CaCO3, ettringite, Mg(OH)2, and C-S-H gel, with the morphologies influenced by environmental ion concentrations. Additionally, five machine learning models were established and compared, with the artificial neural network (ANN) demonstrating optimal predictive accuracy (R2 = 0.956). Furthermore, an interpretable ANN-SHAP framework was implemented to decouple the complex multi-factor interactions, identifying initial crack width as the dominant geometric constraint. This research provides an interpretable data-driven framework for assessing the self-healing performance of UHPSSC in complex marine environments.

Original languageEnglish
Article number116266
JournalJournal of Building Engineering
Volume126
DOIs
StatePublished - 15 May 2026
Externally publishedYes

UN SDGs

This output contributes to the following UN Sustainable Development Goals (SDGs)

  1. SDG 14 - Life Below Water
    SDG 14 Life Below Water

Keywords

  • Machine learning
  • Rehydration
  • SHAP analysis
  • Self-healing
  • Ultra-high performance seawater sea sand concrete

Fingerprint

Dive into the research topics of 'Autogenous healing of cracked ultra-high performance seawater sea sand concrete under multi-factor environments: Experimental mechanisms and interpretable prediction'. Together they form a unique fingerprint.

Cite this