Concrete utilization is increased with the rapidly growing construction industry, compaction is the main exertion that arisen in the concrete. Self-compacting concrete (SCC) is a flowable concrete that can flow under its own weight in the congested reinforcements without any need for external vibration. As the lesser usage of aggregates leads to the decrease in stiffness of SCC which may cause the earlier formation of cracks, adding fibers increase the stiffness of SCC, and also it has a lot of consequences for finding out fresh and mechanical properties. This study mainly focuses on the application of Artificial Neural Networks (ANN) to predict the fresh and mechanical properties of steel fiber reinforced SCC. In the proposed model nine input parameters and seven output parameters are considered for modeling. For training and testing of the data along with regression analysis was performed using MATLAB using the ANN tool. It is used for the complete modeling and one hidden layer and ten neurons and 1000 epochs. The model performance was evaluated based on three metrics sets which includes correlation coefficient (R²), root mean square error (RMSE) and mean absolute percentage error (MAPE). The obtained correlation coefficient value will be between 0.9 and 1, which implies good accuracy of prediction.

PDF (Download Full text)

Download Citation (XML)