.png){kind=icon}
返回
Predicting ultra-high-performance concrete compressive strength using gene expression programming method
.png){kind=icon}
OA
AI
DOI：10.1016/j.cscm.2023.e02074.png)
.png){kind=icon}
.png){kind=icon}
分享.png){kind=icon}
收藏
摘要
En 中文
There have been extensive experimental studies available on the composition and characteristics of Ultra-High-Performance concrete (UHPC). However, the relation between UHPC characteris-tics and mixture content, on the other hand, is extremely non-linear and challenging to distin-guish utilizing typical statistical approaches. A comprehensive literature research was carried out for this aim to acquire experimental data on the compressive strength of UHPC. The dataset contains 810 experimental values of compressive strength and 15 most influential parameters that include cement, water, nano-silica, quartz powder, limestone powder, gravel, sand, slag, super-plasticizer, fiber, temperature, age, fly ash, relative humidity, and silica fume, are considered as input. The suggested gene expression programming (GEP) model can estimate the compressive strength of UHPC by using simple mathematical formulations. There is no predetermined function to evaluate in the GEP technique, and it replicates or eliminates numerous combinations of factors to create the formulation that suits the experimental results. For verification and validation of model performance, various statistical measures, SHAP analysis, external validation checks, and comparing with the regression model, are applied. SHAP analysis provided that age, fiber, silica fume, superplasticizer, cement, sand, and water have a high influence on compressive strength while other input parameters have less influence on compressive strength. The model outcomes indicate the robustness and accuracy of the predictive potential of the proposed model. As a result, the GEP model can be used to give practical insights into the mixture design of UHPC for a variety of construction applications, resulting in better predictive capacity at a cheaper cost and in a considerably shorter period. Also, the present study findings can assist the design engineers and builders to understand the significance of each constituent in UHPC.
Keyword:
UHPC
Compressive strength
Machine learning
GEP
Concrete
AI总结
对已上传原文的论文进行重点信息的提取，主要内容包括：简要概述、研究摘要、背景介绍、关键亮点、图文解析、展望与总结。
期刊
IF:
6.6
论文数:
6.3K
被引数:
2.0W
机构
引用论文
Influence of steel fiber content and aspect ratio on the uniaxial tensile and compressive behavior of ultra high performance concrete钢纤维掺量和长径比对超高性能混凝土单轴拉压性能的影响
PREAI
分享收藏Simulation of Depth of Wear of Eco-Friendly Concrete Using Machine Learning Based Computational Approaches.png)
MATERIALSOAAI
MATERIALSIF3.2
OAAI分享收藏
查看更多内容