PDF(Pubmed)
Abstract：
The use of hybrid fibre-reinforced Self-compacting concrete (HFR-SCC) has escalated recently due to its significant advantages in contrast to normal concrete such as increased ductility, crack resistance, and eliminating the need for compaction etc. The process of determining residual strength properties of HFR-SCC after a fire event requires rigorous experimental work and extensive resources. Thus, this study presents a novel approach to develop equations for reliable prediction of compressive strength (cs) and flexural strength (fs) of HFR-SCC using gene expression programming (GEP) algorithm. The models were developed using data obtained from internationally published literature having eight inputs including water-cement ratio, temperature, fibre content etc. and two output parameters i.e., cs and fs. Also, different statistical error metrices like mean absolute error (MAE), coefficient of determination ( R 2 ) and objective function (OF) etc. were employed to assess the accuracy of developed equations. The error evaluation and external validation both approved the suitability of developed models to predict residual strengths. Also, sensitivity analysis was performed on the equations which revealed that temperature, water-cement ratio, and superplasticizer are some of the main contributors to predict residual compressive and flexural strength.
摘要：
混合纤维增强自密实混凝土（HFR-SCC）的使用最近由于其与普通混凝土相比具有显着的优势而升级，例如延展性增加。抗裂性,并消除了压实等的需要。确定火灾事件后HFR-SCC的残余强度特性的过程需要严格的实验工作和大量的资源。因此，这项研究提出了一种新的方法来开发使用基因表达式编程（GEP）算法可靠预测HFR-SCC的抗压强度（cs）和抗弯强度（fs）的方程。这些模型是使用从国际出版的文献中获得的数据开发的，这些文献有八个输入，包括水灰比，温度,纤维含量等.和两个输出参数，即，cs和fs。此外，不同的统计误差度量，如平均绝对误差(MAE)，决定系数(R2)和目标函数(OF)等。用于评估已开发方程的准确性。误差评估和外部验证都批准了开发模型预测剩余强度的适用性。此外，对方程进行了灵敏度分析，揭示了温度，水灰比，和高效减水剂是一些预测残余压缩和弯曲强度的主要贡献者。
