期刊
APPLIED SOFT COMPUTING
卷 24, 期 -, 页码 985-993出版社
ELSEVIER SCIENCE BV
DOI: 10.1016/j.asoc.2014.08.044
关键词
Evolutionary polynomial regression; Optimisation; Hybrid techniques; Sulphuric acid attack; Corrosion; Sewer pipes
资金
- UK Engineering and Physical Sciences Research Council (EPSRC) [EP/1032150/1]
Concrete corrosion due to sulphuric acid attack is known to be one of the main contributory factors for degradation of concrete sewer pipes. This article proposes to use a novel data mining technique, namely, evolutionary polynomial regression (EPR), to predict degradation of concrete subject to sulphuric acid attack. A comprehensive dataset from literature is collected to train and develop an EPR model for this purpose. The results show that the EPR model can successfully predict mass loss of concrete specimens exposed to sulphuric acid. Parametric studies show that the proposed model is capable of representing the degree to which individual contributing parameters can affect the degradation of concrete. The developed EPR model is compared with a model based on artificial neural network (ANN) and the advantageous of the EPR approach over ANN is highlighted. In addition, based on the developed EPR model and using an optimisation technique, the optimum concrete mixture to provide maximum resistance against sulphuric acid attack has been identified. (C) 2014 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