摘要
再生骨料的表面存在微裂缝和残余砂浆,导致由其制备的再生混凝土强度较差,因此需要对其进行改性处理。通过碱激发材料浆液对再生骨料进行裹浆改性处理,研究碱激发材料浆液中不同组分交互作用对再生混凝土强度的影响,并对浆液配比进行优化。通过响应面设计方法(RSM)中Box-Behnken优化试验设计,建立以碱激发材料浆液中矿渣粉与粉煤灰比值、水泥熟料掺量、氯盐掺量为变量,再生混凝土强度为响应的多元非线性回归模型,基于响应面数据库通过BP神经网络和遗传算法构建一种BP-GA高精度预测优化模型。研究表明:再生骨料裹浆改性用碱激发材料浆液中水泥熟料掺量和氯盐掺量交互作用对再生混凝土性能影响最显著;响应面优化浆液配比时,相对误差精度控制为4.00%;利用BP-GA模型进行预测优化时,相对误差精度控制为0.78%,实现了对裹浆改性用碱激发材料浆液配比高精度优化。
There are micro-cracks and residual mortar on the surface of recycled aggregate,which leads to poor strength of recycled concrete prepared from it,so it needs to be modified.In this paper,the alkali-activated material slurry is applied to modify the recycled aggregate,and the influence of the interaction of different components in the alkali-activated material slurry on the strength of recycled concrete is studied,and the mix ratio of the slurry is optimized.By optimizing the design of experiments with Box Behnken in the response surface design method(RSM),a multivariate nonlinear regression model was established with the ratio of slag powder to fly ash,the amount of cement clinker,and the amount of chloride salt as variables,and the strength of recycled concrete as the response.Based on the response surface database,a BP-GA high-precision prediction optimization model is constructed using BP neural network and genetic algorithm.The results are as follows.The interaction between the amount of cement clinker and the amount of chloride salt in the alkali activated material slurry used for the modification of recycled aggregate slurry has the most significant impact on the performance of recycled concrete.When optimizing the slurry mix ratio using response surface methodology,the relative error accuracy is controlled to 4.00%.When the BP-GA model is used for prediction and optimization,the relative error accuracy is controlled to 0.78%,and the high-precision optimization of the slurry mix ratio of the alkali-activated material for slurry modification is realized.
作者
李克亮
弓晋伟
申翔宇
孙作正
杜晓蒙
李宁宁
LI Keliang;GONG Jinwei;SHEN Xiangyu;SUN Zuozheng;DU Xiaomeng;LI Ningning(School of Civil Engineering and Transportation,North China University of Water Resources and Electric Power,Zhengzhou 450045,China;School of Mechanical Engineering,Tongji University,Shanghai 200092,China;Zhengzhou DingSheng Engineering Technology Co.,Ltd.,Zhengzhou 450001,China)
出处
《华北水利水电大学学报（自然科学版）》
北大核心
2024年第1期73-82,共10页
Journal of North China University of Water Resources and Electric Power：Natural Science Edition
基金
国家自然科学基金项目(52179133)
河南省科技攻关项目(222102320131)
华北水利水电大学研究生教育创新计划基金项目(YK2021-19)。
关键词
再生骨料
裹浆改性
碱激发材料浆液
响应面
BP神经网络
遗传算法
recycled aggregate
slurry modification
alkali-activated material slurry
response surface
BP neural network
genetic algorithm
