Artificial neural network for the analysis of color expression of GBFS mortar using carbon amino silica black

This study was to investigate the effects of granulated blast furnace slag (GBFS) on the color expression of black colored mortar. For this purpose, color evaluation and was carried out on white Portland cement (WPC) mortar mixed with of carbon amino silica black (CASB) by changing the proportion of GBFS. Each mortar was measured at five locations on the surface and averaged. Color can be evaluated by measurements of tristimulus values L∗, a∗ and b∗, represented in the chromatic space CIELAB. Artificial Neural Networks (ANN) model is constructed, trained and tested using these data. The data used in the ANN model are arranged in a format of 3 input parameters that cover the WPC, GBFS, days and, an output parameter which is the color parameters of the black colored mortar. The results showed that ANN can be an alternative approach for the analyzing the color parameters using mortar ingredients as input parameters. And one-step secant back-propagation as the final training algorithm is the most suitable.