期刊论文

Experiment and its prediction artificial neural networks model study on the compressive strength of coal gangue powder concrete

中文

混凝土

1002-3550

2018

12

31-33

10.3969/j.issn.1002-3550.2018.12.009

国家自然科学基金项目(51378247) 湖南省教育厅一般项目(16C1381).

本校为第一机构

为建立煤矸石粉混凝土抗压强度神经网络预测模型,对煤矸石粉混凝土7、28、60、120、180d抗压强度进行试验。煤矸石粉掺量分别为0、10%、15%、20%、25%、30%、35%,煤矸石粉均取代水泥,作为掺合料掺入混凝土中。在试验的基础上,利用试验数据训练、建立神经网络预测模型。预测结果表明:神经网络预测模型有较高的预测精度,可以用来预测煤矸石粉混凝土的抗压强度。

A investigation carried out to build a prediction artificial neural networks model of coal gangue powder concrete.The compressive strength of coal gangue concretes were test at age 7 days,28 days,60 days,120 days,180 days.The optimal replacement level of cement by coal gangue powder is 0,10%,15%,20%,25%,30%,35% respectively.Building a artificial neural networks prediction model of coal gangue powder concrete based on the test data.The results show that artificial neural network model has highly prediction precision and can be us...