In recent years, the use of artificial intelligence methods to predict the properties of cement based mortar and concrete has become very attractive with the aim of preventing expensive laboratory testing. Since the most properties of mortar depend on the cement used for their preparation and also the properties of the cement are driven by the materials used for their production, properties of mineral raw materials have a significant impact on the characteristics of cement produced. This study considered different percentages of raw materials entering the kiln (Sio2, Al2O3, Fe2o3, CaO, MgO, SO3, K2O and Na2O) to produce Portland cement, and using a neural network algorithm to estimate 2-day compressive strength of cement mortar. Result showed that the prediction of compressive strength is accurate enough, so it is possible to use this very inexpensive method to replace, reduce or complement existing costly and continuous laboratory testing in cement factories.

Keywords