An ANN to Predict Sorptivity and Water Absorption of Concrete Made with Industrial Waste and Artificial Sand

Document Type : Original Article

Authors

1 Associate Professor, Department of Civil Engineering, D.Y.Patil College of Engineering and Technology, Kolhapur, Maharashtra, India

2 Assistant Professor, Department of Civil Engineering , D.Y.Patil College of Engineering and Technology, Kolhapur, Maharashtra, India

10.22115/cepm.2023.402893.1239

Abstract

Large-scale cement and natural fine aggregate use in construction has negative environmental effects. In light of this issue, crushed sand is a good material, and wastes from industries such as metakaolin, Furness ash, blast furnace slag, and micro silica be able to substitute for cement. To estimate the water absorption and sorptivity of concrete made by cementitious waste ingredients and in which the natural fine aggregate is moderately replaced by artificial sand Matlab software model was created in this work.150 mm cube specimen was tested for water absorption and sorptivity. An Artificial neural network (ANN) model was created using the experiment. 330 results in total were used for model design, 20% for neural network model testing check, and 80% results for the Artificial neural network model training phase. The 28-day water absorption and sorptivity of concrete mixed by partially substituting cement with pozzolan and partially substituting natural fine aggregate with artificial sand were calculated using the product of 25 input data. The Artificial neural network model's results offer a precise elastic prediction of the water absorption and sorptivity ability of concrete mixed with substituting industrial cementitious waste for cement and artificial sand with naturally occurring fine aggregate.

Keywords

Main Subjects