Estimation of fresh and hardened properties of self-compacting concrete by optimized radial basis function methods

Document Type : Original Article


1 The King's School, BP1560, Bujumbura, Burundi

2 Univ Modena & Reggio Emilia, Dept Engn Enzo Ferrari, Modena, Italy


Most of the published literature on concrete containing fly ash was limited to predicting the hardened concrete properties. It is understood that exist so restricted studies focusing on forecasting both fresh and hardened properties of self-compacting concrete (SCC). Hence, it is attempted to develop some models to predict the fresh and hardened properties of SCC by the optimized radial basis function neural network (RBFNN) method. This study aims to specify RBFNN method key parameters using arithmetic optimization algorithm (AOA) and grasshopper optimization algorithm (GOA). The considered properties of SCC in the fresh phase are the L-box test, V-funnel test, slump flow, and in the hardened phase compressive strength. The results present powerful workability during the prediction process. It is observed that the developed models have performance evaluation indices in reasonable value in the learning and testing section. All in all, the RBFNN model developed by AOA outperforms others, with R^2 values at 0.9607 (slump flow), 0.9651 (L-box), 0.9905 (V-funnel test), and 0.9934 (compressive strength), which depicts the capability of this algorithm for determining the optimal parameters of the RBFNN, While, it is worth mentioning than the model developed with GOA algorithm is also powerful.