基于CRITIC-LightGBM算法的混凝土强度预测模型_《混凝土与水泥制品》杂志社

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CRITIC-LightGBM算法的混凝土强度预测模型

Prediction model for compressive strength of concrete based on CRITIC- LightGBM algorithm

2024年第9期

混凝土；抗压强度；预测；模型；机器学习

Concrete; Compressive strength; Prediction; Model; Machine learning

2024年第9期

10.19761/j.1000-4637.2024.09.017.05

戚喜章，吴雨航*，钱大桐

中国铁建投资集团有限公司，广东珠海 519000

戚喜章，吴雨航*，钱大桐

戚喜章,吴雨航,钱大桐.基于CRITIC-LightGBM算法的混凝土强度预测模型[J].混凝土与水泥制品,2024(9):17-20,27.

QI X Z,WU Y H,QIAN D T.Prediction model for compressive strength of concrete based on CRITIC-LightGBM algorithm[J].China Concrete and Cement Products ,2024(9):17-20,27.

浏览量:

1000

摘要：为了更精确地对混凝土抗压强度进行预测，以水泥、炉渣、粉煤灰、水、减水剂、粗骨料、细骨料、龄期作为混凝土强度预测指标，运用CRITIC算法对指标进行了加权处理，在此基础上，引入新的机器学习算法轻量级梯度提升机（LightGBM）对样本进行了计算训练，同时，运用K-10折交叉验证方式对模型参数进行了优化，最终建立了CRITIC-LightGBM混凝土抗压强度预测模型。运用该模型对混凝土的抗压强度进行了预测，并将结果与LightGBM、多层感知机（MLP）和支持向量机回归（SVR）的预测结果进行了对比。结果表明：与较单一的LightGBM、MLP和SVR算法相比，CRITIC-LightGBM模型预测的平均绝对误差和均方根误差更小，准确性更高，R2更接近1，模型的泛化能力明显提高。 Abstract: In order to predict the compressive strength of concrete more accurately, cement, slag, fly ash, water, water reducer, coarse aggregate, fine aggregate and age were used as the prediction indexes of concrete strength. The CRITIC algorithm was used to weight the indexes. On this basis, a new machine learning algorithm Light Gradient Boosting Machine(LightGBM) was introduced to calculate and train the samples. At the same time, the K-10 fold cross-validation method was used to optimize the model parameters. Finally, the CRITIC-LightGBM model for predicting the compressive strength of concrete was established. The model was used to predict the compressive strength of concrete, and the prediction results were compared with those of LightGBM, multi-layer perceptron(MLP) and support vector machine regression(SVR) algorithm. The results show that compared with the relatively single LightGBM, MLP, and SVR algorithms,the CRITIC-LightGBM prediction model has smaller mean absolute error and root mean square error, and has higher accuracy. R2 is closer to 1, and the generalization ability of the model is significantly improved.

英文名 : Prediction model for compressive strength of concrete based on CRITIC- LightGBM algorithm

刊期 : 2024年第9期

关键词 : 混凝土；抗压强度；预测；模型；机器学习

Key words : Concrete; Compressive strength; Prediction; Model; Machine learning

刊期 : 2024年第9期

DOI : 10.19761/j.1000-4637.2024.09.017.05

文章编号 :

基金项目 :

作者 : 戚喜章，吴雨航*，钱大桐

单位 : 中国铁建投资集团有限公司，广东珠海 519000

戚喜章，吴雨航*，钱大桐

戚喜章,吴雨航,钱大桐.基于CRITIC-LightGBM算法的混凝土强度预测模型[J].混凝土与水泥制品,2024(9):17-20,27.

QI X Z,WU Y H,QIAN D T.Prediction model for compressive strength of concrete based on CRITIC-LightGBM algorithm[J].China Concrete and Cement Products ,2024(9):17-20,27.





摘要

参数

结论

参考文献

引用本文

Abstract: In order to predict the compressive strength of concrete more accurately, cement, slag, fly ash, water, water reducer, coarse aggregate, fine aggregate and age were used as the prediction indexes of concrete strength. The CRITIC algorithm was used to weight the indexes. On this basis, a new machine learning algorithm Light Gradient Boosting Machine(LightGBM) was introduced to calculate and train the samples. At the same time, the K-10 fold cross-validation method was used to optimize the model parameters. Finally, the CRITIC-LightGBM model for predicting the compressive strength of concrete was established. The model was used to predict the compressive strength of concrete, and the prediction results were compared with those of LightGBM, multi-layer perceptron(MLP) and support vector machine regression(SVR) algorithm. The results show that compared with the relatively single LightGBM, MLP, and SVR algorithms,the CRITIC-LightGBM prediction model has smaller mean absolute error and root mean square error, and has higher accuracy. R2 is closer to 1, and the generalization ability of the model is significantly improved.

扫二维码用手机看

[1] 吕业锋,吕学义,莫冠旺,等.非标准混凝土试样力学性能试验研究[J].公路,2022,67(9):357-364.
[2] 刘德胜.基于机器学习算法的混凝土28 d抗压强度预测[J].混凝土与水泥制品,2022,317(9):20-24
[3] 焦楚杰,谭思琪,崔力仕,等.基于神经网络的植生型多孔混凝土抗压强度预测模型[J].混凝土,2022(1):7-10,16.
[4] 赵俊梅.双掺磨细矿渣粉和粉煤灰大流动性混凝土配合比的多元线性回归分析[J].混凝土,2007(12):32-34.
[5] 张静,刘向东.混沌粒子群算法优化最小二乘支持向量机的混凝土强度预测[J].吉林大学学报(工学版),2016,46(4):1097-1102.
[6] 陈洪根,龙蔚莹,李昕,等.基于BP神经网络的粉煤灰混凝土抗压强度预测研究[J].建筑结构,2021,51(增刊2):1041-1045.
[7] 赵明亮,水中和,周华新,等.中低强度等级混凝土抗压强度的BP神经网络模型预测研究[J].混凝土,2021(3):35-38.
[8] TSAI H C.Modeling concrete strength with high-order neural networks[J].Neural computing and applications,2016,27(8):2465-2473.
[9] SHAQADAD A.Prediction of concrete strength using support vector machines algorithm[J].Materials Science Forum,2020,986(1):9-17.
[10] GONG Z,ZHANG Y,HU Y,et al.Cubic meter compressive strength prediction of concrete[J].Journal of Wuhan University of Technology(Materials Science),2016,31(3):590-593.
[11] 胥明,许杰,万友生,等.基于CRITIC法与TOPSIS地铁车站施工安全评价[J].自然灾害学报,2022,31(3):157-166.
[12] 谢学斌,李德玄,孔令燕,等.基于CRITIC-XGB算法的岩爆倾向等级预测模型[J].岩石力学与工程学报,2020,39(10):1975-1982.
[13] ZHANG J T,ZHANG X Y,GUI K,et al.Robust prediction of hourly PM_(2.5) from meteorological data using LightGBM[J].National Science Review,2021,8(10):115-126.
[14] 李占山,姚鑫,刘兆赓,等.基于LightGBM的特征选择算法[J].东北大学学报(自然科学版),2021,42(12):1688-1695.
[15] 李克庆,胡亚飞,韩斌,等.低温下湿喷混凝土强度预测与强度损伤[J].中南大学学报(自然科学版),2021,52(5):1581-1590.
[16] 曹斐,周彧,王春晓,等.一种改进的支持向量回归的混凝土强度预测方法[J].硅酸盐通报,2021,40(1):90-97.

上一个

材料组份对高强环氧灌浆料性能影响研究

下一个

碱激发粉煤灰对淡化海砂海水混凝土性能的影响

头部文案

基于CRITIC-LightGBM算法的混凝土强度 预测模型

相关文件

基于CRITIC-LightGBM算法的混凝土强度预测模型