An investigation of mechanical behavior of glass fiber reinforced concrete specimens and modeling with artificial neural network

In this study, porosity, capillary water absorption coefficient, pulse velocity, compressive strength and split tensile strength of the concrete specimens, which have water-cement ratios of 0.45, 0.50 and 0.55, dosages of 300, 350 and 400, and 5, 10, 15, 20 kg/m3 ratio of glass fiber, were tested. Furthermore compressive strength and split tensile strength of the concrete specimens were modeled with artificial neural network. As a result of the study the porosity and capillary water absorption coefficient of the concretes have increased depending on the rise of the glass fiber ratio, As well as the pulse velocity of the concretes have decreased. Although the compressive strength of the concretes small increases in ratios low fiber, showed decreases ratios high-fiber. Similarly, the split tensile strength of the concrete specimens have increased low ratios glass fiber, but the splitting tensile strength of the concrete specimens which is containing 20 kg/m3 glass fiber have decreased. The data which were provided by ANN model were compared with the data obtained from experimental study and a good agreement was determined between the results.Key Words: Concrete, Glass Fiber, Compressive Strength, Split Tensile Strength, Artificial Neural Network