Scholarlink

<p>This study revolves around the integration of multiwalled carbon nanotube into high volume fly ash concrete mixtures to significantly enhance the compressive strength. of high-volume fly ash (HVFA) concrete. Tree based (CHAID, Exhaustive CHAID and CART) machine learning models were also developed to predict compressive strength of HVFA concrete. Among the predictive models developed, the CHAID regression tree consistently outperformed others, offering the most accurate and reliable predictions, as indicated by its superior statistical and graphical performance metrics. Exhaustive CHAID also showed strong results, while CART delivered moderate accuracy. Linear Regression model was also developed but was the least effective, likely due to its inability to capture non-linear interactions within the dataset. The Taylor diagram analysis and monotonicity assessment validated the consistency and interpretability of the CHAID model, making it a robust choice for predictive modelling of compressive strength of HVFA concrete. The study underscores the efficacy of tree-based algorithms, especially CHAID, for predicting mechanical properties in advanced cementitious materials.</p>

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

Tree-based machine learning models for compressive strength prediction of high-volume fly ash concrete reinforced with multiwalled carbon nanotubes

  • 期刊论文

  • 发表日期：

    2025年5月13日

查看全文
  • Sameer Sen,
  • Sanjeev Sinha

摘要

This study revolves around the integration of multiwalled carbon nanotube into high volume fly ash concrete mixtures to significantly enhance the compressive strength. of high-volume fly ash (HVFA) concrete. Tree based (CHAID, Exhaustive CHAID and CART) machine learning models were also developed to predict compressive strength of HVFA concrete. Among the predictive models developed, the CHAID regression tree consistently outperformed others, offering the most accurate and reliable predictions, as indicated by its superior statistical and graphical performance metrics. Exhaustive CHAID also showed strong results, while CART delivered moderate accuracy. Linear Regression model was also developed but was the least effective, likely due to its inability to capture non-linear interactions within the dataset. The Taylor diagram analysis and monotonicity assessment validated the consistency and interpretability of the CHAID model, making it a robust choice for predictive modelling of compressive strength of HVFA concrete. The study underscores the efficacy of tree-based algorithms, especially CHAID, for predicting mechanical properties in advanced cementitious materials.