Scholarlink

This research investigates the incorporation of nano-titanium dioxide (NT) into mortar to enhance its resistance to chloride ion penetration and extend the service life of marine structures. Mortars were prepared with varying NT content (0–3% by weight of cement), and the effective chloride diffusivity (Deff) and migration coefficient (Dmig) were experimentally determined. Fuzzy C-Means (FCM) clustering combined with hybrid algorithms was conducted to model and predict these parameters. Results demonstrated that 3% NT reduced chloride ion diffusivity by up to 90% compared to the control, with significant decreases in Deff and Dmig, and improved service life by up to 35%. FCM effectively clustered data with a silhouette score of 0.87. Particle Swarm Optimization-based FCM models achieved the highest accuracy with an R2 of 0.9887 and VAF of 99.63%. The study highlights the potential of data-driven models to replace extensive laboratory experimentation, offering significant reductions in time and cost with high scalability and adaptability for future research on chloride resistance in marine concrete structures.

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Fuzzy Clustering Method-Based Hybrid Metaheuristic Models of Chloride Diffusivity in Mortar Containing Nano-titanium Dioxide

  • 会议论文

  • 发表日期：

    2026年

查看全文
  • Prashant Tiwari,
  • Yogesh Iyer Murthy,
  • Abhishek Verma

摘要

This research investigates the incorporation of nano-titanium dioxide (NT) into mortar to enhance its resistance to chloride ion penetration and extend the service life of marine structures. Mortars were prepared with varying NT content (0–3% by weight of cement), and the effective chloride diffusivity (Deff) and migration coefficient (Dmig) were experimentally determined. Fuzzy C-Means (FCM) clustering combined with hybrid algorithms was conducted to model and predict these parameters. Results demonstrated that 3% NT reduced chloride ion diffusivity by up to 90% compared to the control, with significant decreases in Deff and Dmig, and improved service life by up to 35%. FCM effectively clustered data with a silhouette score of 0.87. Particle Swarm Optimization-based FCM models achieved the highest accuracy with an R2 of 0.9887 and VAF of 99.63%. The study highlights the potential of data-driven models to replace extensive laboratory experimentation, offering significant reductions in time and cost with high scalability and adaptability for future research on chloride resistance in marine concrete structures.