Development of Artificial Neural Network Model for Permeability of High Performance Concrete

VAISHALI., G. GHORPADE and BEULAH., M and H. SUDARSANA, RAO (2016) Development of Artificial Neural Network Model for Permeability of High Performance Concrete. In: Fourth International Conference On Advances in Civil, Structural and Mechanical Engineering -ACSM 2016, 07-08 May 2016, Bangkok, Thailand.

20160516_094022.pdf - Published Version

Download (693kB) | Preview
Official URL:


High performance concrete (HPC) is an engineered concrete possessing the most desirable properties during fresh as well as hardened concrete stages. Permeability is one of the most important parameters to quantify the durability of high-performance concrete. This research was to study the chloride ion permeability of high performance concrete with different mineral admixtures like Fly ash, Silicafume and Metakaolin of different percentages, with varying aggregate-binder ratios (2, 2.5). In addition, on the basis of the experimental data an artificial neural network (ANN) technique is executed to demonstrate the possibilities of artificial neural network formulation for the prediction of chloride permeability as a function of four input parameters : water-cement ratio (0.3, 0.325, 0.35, 0.375, 0.4, 0.425, 0.45, 0.475, 0.5), aggregate binder ratio (2,2.5), type of mineral admixtures, percentage replacement of mineral admixtures i.e Fly ash, Silicafume and Metakaolin(0,10,20,30%) as input parameters. Genetic algorithm has been used to extract the weights of ANN model. The developed model is able to predict the permeability within 5% error.

Item Type: Conference or Workshop Item (Paper)
Uncontrolled Keywords: HPC, Genetic Algorithm, ANN, Chloride Permeability
Depositing User: Mr. John Steve
Date Deposited: 23 Mar 2019 12:11
Last Modified: 23 Mar 2019 12:11

Actions (login required)

View Item View Item