Ehsan Rasa - BSc. Student of Civil engineering
Hamed Ketabchi - BSc. Student of Civil engineering
Mohammad Hadi Afshar - Associate Professor of Civil engineering Department

An artificial neural network of the feed-forward back-propagation type has been applied for predicting density and compressive strength properties of cement paste portion of concrete mixtures. Artificial neural networks (ANNs) have recently been introduced as an efficient artificial intelligence modeling technique for applications incorporating a large number of variables. Mechanical properties of concrete are highly influenced by density and compressive strength of concrete cement paste. Density and compressive strength of concrete cement paste are affected by several parameters, viz. water-cementitious materials ratio, silica fume unit contents, percentage of super-plasticizer, curing, cement type and etc. The 28-day compressive strength and saturated surface dry (SSD) density values are considered as the aim of the prediction. A total of 600 specimens were selected. The system was trained based on 350 training pairs chosen randomly from the data set, and tested using remaining 250 examples. Results indicate that density and compressive strength of concrete cement paste can be predicted much more accurately using ANN method compared to conventional models (Traditional regression analysis, statistical methods and etc.).

Cement Paste, Compressive Strength, Density, Neural Network, Silica Fume