基于机器学习回归模型的UHPC抗压强度预测研究

头部文案

发布时间：2020-01-06 00:00:00

全国建材科技期刊
全国中文核心期刊
中国科技论文统计源期刊
万方数据-数字化期刊群入网期刊
中国学术期刊（光盘版）全文收录期刊

华东地区优秀科技期刊
江苏省期刊方阵“双效期刊”
中国期刊网全文收录期刊
中国科技期刊数据库全文收录期刊



基于机器学习回归模型的UHPC抗压强度预测研究

Compressive strength prediction of UHPC based on machine learning regression model

2023年第12期

超高性能混凝土；抗压强度；随机森林回归；支持向量机回归；多层感知机回归

Ultra-high performance concrete(UHPC); Compressive strength; Random Forest Regression; Support Vector Machine Regression; Multilayer Perceptron Regression

2023年第12期

10.19761/j.1000-4637.2023.12.023.06

代柱端1，陈健2，周晓阳1

1.武汉优城科技有限公司，湖北武汉 430080；2.湖北筑选新材料有限公司，湖北武汉 430300

代柱端1，陈健2，周晓阳1

代柱端,陈健,周晓阳.基于机器学习回归模型的UHPC抗压强度预测研究[J].混凝土与水泥制品,2023(12):23-27,33.

DAI Z D,CHEN J,ZHOU X Y,et al.Compressive strength prediction of UHPC based on machine learning regression model[J].China Concrete and Cement Products,2023(12):23-27,33.

浏览量:

1000

摘要：以超高性能混凝土（UHPC）组成原材料中的水泥、矿粉、硅灰、钢纤维、减水剂、消泡剂和水的用量为特征，28 d抗压强度为标签建立了数据集，并采用随机森林回归（RFR）、支持向量机回归（SVR）和多层感知机回归（MLPR）3种机器学习回归模型对数据集进行了训练和预测。结果表明：MLPR模型的拟合优度最高；RFR模型中对UHPC 的28 d抗压强度影响相对较大的3个因素为硅灰、水泥和水的用量；SVR模型和MLPR模型的预测值均落在5%置信区间内，回归效果均较理想。 Abstract: The data set was established with the dosages of cement, mineral powder, silica fume, steel fiber, water reducer, defoamer and water in the constituent raw materials of UHPC as the characteristics, and the 28 d compressive strength as the label, and three machine learning regression models of Random Forest Regression (RFR), Support Vector Machine Regression (SVR) and Multilayer Perceptron Regression(MLPR) were used to train and predict the data set. The results show that the MLPR model has the highest goodness of fit. The three factors in the RFR model that have a relatively large influence on the 28 d compressive strength of UHPC are the dosages of silica fume, cement and water. The predict values of the SVR model and the MLPR model both fall within the 5% confidence interval, and the regression effect of the two models are relatively ideal.

英文名 : Compressive strength prediction of UHPC based on machine learning regression model

刊期 : 2023年第12期

关键词 : 超高性能混凝土；抗压强度；随机森林回归；支持向量机回归；多层感知机回归

Key words : Ultra-high performance concrete(UHPC); Compressive strength; Random Forest Regression; Support Vector Machine Regression; Multilayer Perceptron Regression

刊期 : 2023年第12期

DOI : 10.19761/j.1000-4637.2023.12.023.06

文章编号 :

基金项目 :

作者 : 代柱端1，陈健2，周晓阳1

单位 : 1.武汉优城科技有限公司，湖北武汉 430080；2.湖北筑选新材料有限公司，湖北武汉 430300

代柱端1，陈健2，周晓阳1

代柱端,陈健,周晓阳.基于机器学习回归模型的UHPC抗压强度预测研究[J].混凝土与水泥制品,2023(12):23-27,33.

DAI Z D,CHEN J,ZHOU X Y,et al.Compressive strength prediction of UHPC based on machine learning regression model[J].China Concrete and Cement Products,2023(12):23-27,33.





摘要

参数

结论

参考文献

引用本文

Abstract: The data set was established with the dosages of cement, mineral powder, silica fume, steel fiber, water reducer, defoamer and water in the constituent raw materials of UHPC as the characteristics, and the 28 d compressive strength as the label, and three machine learning regression models of Random Forest Regression (RFR), Support Vector Machine Regression (SVR) and Multilayer Perceptron Regression(MLPR) were used to train and predict the data set. The results show that the MLPR model has the highest goodness of fit. The three factors in the RFR model that have a relatively large influence on the 28 d compressive strength of UHPC are the dosages of silica fume, cement and water. The predict values of the SVR model and the MLPR model both fall within the 5% confidence interval, and the regression effect of the two models are relatively ideal.

扫二维码用手机看

[1] 王晖.超高性能混凝土(UHPC)研究综述[J].混凝土与水泥制品,2022(4):25-28.
[2] 王慧慧,商涛平,龙广成,等.UHPC抗压强度与组成参数之间的关系模型研究[J].铁道科学与工程学报,2020,17(11):2816-2822.
[3] 樊俊江,於林锋,韩建军.配比参数对UHPC流动性及抗压强度的影响试验研究[J].新型建筑材料,2019,46(4):5-8.
[4] 姚武,李杰.高性能混凝土抗拉强度的长期性能研究[C]//中国建筑学会.新世纪海峡两岸高性能混凝土研究与应用学术会议论文集.上海:同济大学出版社,2002,131-135.
[5] 马家淦.纳米材料对混凝土性能的影响研究[J].湖南造纸,2021,50(9):33-35.

[6] 吴贤国,刘鹏程,陈虹宇,等.基于随机森林的高性能混凝土抗压强度预测[J].混凝土,2022(1):17-24.
[7] 徐潇航,胡张莉,刘加平,等.基于机器学习回归模型的三峡大坝混凝土强度预测[J].材料导报,2023,37(2):68-76.
[8] 陈庆,马瑞,蒋正武,等.基于GA-BP神经网络的UHPC抗压强度预测与配合比设计[J].建筑材料学报,2020,23(1):176-191.
[9] CHOU J S,PHAM A D.Smart artificial firefly colony algorithm-based support vector regression for enhanced forecasting in civil engineering[J].Computer-Aided Civil and Infrastructure Engineering,2015,30(9):715-732.
[10] MOZUMDER R A,ROY B,LASKAR A I.Support vector regression approach to predict the strength of FRP confined concrete[J].Arabian Journal for Science and Engineering,2017,42(3):1129-1146.
[11] HAN Q H,GUI C Q,XU J,et a1.A generalized method to predict the compressive strength of high-performance concrete by improved random forest algorithm[J].Construction and Building Materials,2019,226:734-742.
[12] 李航.统计学习方法(第2版)[M].北京:清华大学出版社,2019.
[13] 张浩,朱吉鹏,卓德才,等.基于随机森林和支持向量机的混凝土抗压强度预测模型研究[J].工程与建设,2022,36(6):1784-1815.
[14] 季韬,林挺伟,林旭健.基于人工神经网络的混凝土抗压强度预测方法[J].建筑材料学报,2005,8(6):5-9.

上一个

粉煤灰和石英粉对超高性能混凝土性能的影响

下一个

建筑3D打印胶凝材料工程性能评价体系研究

头部文案

基于机器学习回归模型的UHPC抗压强度预测研究

相关文件