期刊
AIN SHAMS ENGINEERING JOURNAL
卷 4, 期 4, 页码 651-660出版社
ELSEVIER SCIENCE BV
DOI: 10.1016/j.asej.2013.02.007
关键词
Sulfate attack; Cement type; Fly ash; Silica fume; USBR4908 test method; Artificial neural networks (ANNs)
One of the available tests that can be used to evaluate concrete sulfate resistance is USBR4908. However, there are deficiencies in this test method. This study focuses on the ANN as an alternative approach to evaluate the sulfate expansion. Three types of cement combined with FA or SF, along with variable W/B were study by USBR4908. ANN model were developed by five input parameters, W/B, cement content, FA or SF, C(3)A, and exposure duration; output parameter is determined as expansion. Back propagation algorithm was employed for the ANN training; a Tansig function was used as the nonlinear transfer function. It was clear that the ANN models give high prediction accuracy. In addition, The engineer can avoid the use of the borderline 2.5-5% C(3)A content in severe sulfate environments and borderline 6-8% C(3)A content in moderate sulfate environments, specially with W/B ratio greater than 0.45. (C) 2013 Ain Shams University. Production and hosting by Elsevier B.V. All rights reserved.
作者
我是这篇论文的作者
点击您的名字以认领此论文并将其添加到您的个人资料中。
推荐
暂无数据