<p>Many researchers have experimentally studied the mechanical properties of carbon fiber reinforced concrete, a common construction material. However, there has been a complete lack of analytical research on this topic. This study is the first to develop predictive models for estimating the concrete compressive strength of carbon fiber reinforced concrete CFRC, using a large dataset of 171 data points collected from various previous experimental studies. The models include linear, non-linear, and multi-linear regression models, developed with 120 data points from the dataset. They were then evaluated with an additional 35 data points in the testing phase and validated using 16 data points in the validation phase. The models were also assessed for prediction errors and error tendencies. The results indicate that the multi-linear regression model performs exceptionally well and is reliable for use in the construction and structural sectors, achieving the highest coefficient of determination at around 0.88 compared to the other models in predicting concrete strength in carbon fiber reinforced concrete.</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

Prediction of concrete compressive strength of carbon fiber reinforced concrete: soft computing models

  • Bedar Rauf Hassan,
  • Hawkar Ali Haji,
  • Fkrat Latif Hamid

摘要

Many researchers have experimentally studied the mechanical properties of carbon fiber reinforced concrete, a common construction material. However, there has been a complete lack of analytical research on this topic. This study is the first to develop predictive models for estimating the concrete compressive strength of carbon fiber reinforced concrete CFRC, using a large dataset of 171 data points collected from various previous experimental studies. The models include linear, non-linear, and multi-linear regression models, developed with 120 data points from the dataset. They were then evaluated with an additional 35 data points in the testing phase and validated using 16 data points in the validation phase. The models were also assessed for prediction errors and error tendencies. The results indicate that the multi-linear regression model performs exceptionally well and is reliable for use in the construction and structural sectors, achieving the highest coefficient of determination at around 0.88 compared to the other models in predicting concrete strength in carbon fiber reinforced concrete.