Mathematical Modelling to Control the Chemical Composition of Blast Furnace Slag Using Artificial Neural Networks and Empirical Correlation

Abstract Portland cement additions have been used for many years with the main objective of reducing amount clinker. Among additions, blast furnace slag, resulting from production pig iron, that is, reusing this by-product, reduces emission carbon dioxide as well decreases exploitation natural limestone and clay reserves, which are raw materials In order to reduce these emissions increase availability materials, research has directed study clinker-free binders, is case activated alkali cements supersulfated cements. way, alkali-activated can only involve reuse industry by-products do not require calcination material, thus polluting gases into atmosphere. Supersulfated composed up 90% in addition 10 20% calcium sulfate. One most important characteristics slag ratio content CaO SiO 2 , also known simplified basicity index (B2). This paper proposes mathematical modeling an artificial neural network predict final chemical composition be produced based on operational parameters aiming its use special such The high values (R) associated low (RMSE) show good statistical performance ANN demonstrating model efficient carry out forecast slag.