Prediction of cement mortar strength by replacement of hydrated lime using RSM and ANN: Fid-Bau Portal

Prediction of cement mortar strength by replacement of hydrated lime using RSM and ANN

Nakkeeran, G. / Krishnaraj, L.

This research work focuses on the investigation of the possible roles of hydrated lime on the properties of hardened mortars. In this study, the effect of the hydrated lime mortar has been investigated using response surface method (RSM) and artificial neural network (ANN). This research paper also summarizes the experimental results obtained from the investigation of mortar samples with 10, 20, 30, and 40% of hydrated lime that replaced cement. The mortar mixes have been prepared in a ratio of 1:3 and a w/c ratio of 0.4, which has been taken as the standard mix ratio. In experimental work that has been carried out, the compressive strength is measured post 28 days. RSM method has been used to predict the properties of mortar, which has been taken as standard and the mortars, with the hydrated lime to compare the experimental data. RSM model has shown an accurate result (R² ≥ 0.99), in predicting the mechanical properties of the mortars. The same experimental research design has been used to train the neural network. ANN model has also supported by showing an accurate result (R² ≥ 0.99), in predicting the mechanical properties. Predictions with root-mean-squared error (RMSE), the mean absolute error (MAE), and the model predictive error (MPE) have been carried out to test the ability of both methodologies which have been done and compared. As a result, the RSM and ANN technique has been validated for the use in both response estimation and effective parameter identification. Furthermore, the RSM and ANN technique has also been used to perceive the optimal parameters.