Predicting shrinkage of alkali-activated blast furnace-fly ash mortars using artificial neural network (ANN)

Drying shrinkage of alkali-activated binders are recognized as one of the most important properties towards quality assurance of the binders. In this study, results of experimental studies and predictive models developed to determine the drying shrinkage of alkali -activated blast furnace-fly ash mortars are presented and discussed. Different parameters were altered in the experimental study such as the content of GGBFS, FA, activator modulus (Ms), and curing temperature. Their effects on the drying shrinkage of the mortars were then evaluated. Artificial neural network (ANN) and Multiple Linear Regression (MLR) models were built to predict the drying shrinkage at 28 days using the above-mentioned parameters as inputs. The experimental results and ANN model predictions showed strong correlations. The prediction of 28-days drying shrinkage for the alkali-activated GGBFS-FA was more accurate using ANN than MLR.