基于Python人工神经网络的再生混凝土碳化深度预测  

Prediction of carbonization depth of recycled concrete based on artificial neural network by Python

作　　者：余海玲 郑建岚[1,2] YU Hailing;ZHENG Jianlan(College of Civil Engineering,Fuzhou University,Fuzhou 350116,China;School of Engineering,Fujian Jiangxia University,Fuzhou 350108,China)

机构地区：[1]福州大学土木工程学院,福建福州350116 [2]福建江夏学院工程学院,福建福州350108

出　　处：《混凝土》2020年第9期52-55,共4页Concrete

基　　金：国家自然科学基金海峡联合基金重点支持项目(U1605242)。

摘　　要：在Python语言的环境下,通过对再生混凝土碳化影响因素的分析,建立人工神经网络模型预测不同再生粗骨料取代率、再生细骨料取代率以及加速碳化天数下的再生混凝土碳化深度。结果表明,在合理的网络参数设置下,模型训练拟合较好,预测结果较理想,平均误差百分比为2.65%。因此用人工神经网络模型较准确地快速预测再生混凝土碳化深度是可行的。By analyzing the factors affecting the carbonation of recycled concrete,an artificial neural network model was established to predict the carbonization depth of recycled concrete of different replacement rate of recycled coarse aggregates,replacement rate of recycled fine aggregates and accelerated carbonization days in the context of Python language.After reasonable network parameters setting,the model training results fit better,and the prediction result is good.The average error percentage is 2.65%.Therefore,it is feasible to use the artificial neural network model to predict the carbonation depth of recycled concrete in a relatively precise and quick way.

关 键 词：人工神经网络 再生混凝土 碳化深度 PYTHON 

分 类 号：TU528.01[建筑科学—建筑技术科学]