Hybrid RBF-based prediction algorithms for evaluation of the compressive strength of HPC enhanced with blast furnace slag and fly ash

The compressive strength (CS) of high-performance concrete (HPC) amended with blast furnace slag (BFS) and fly ash (FA) is estimated using a unique method in this study using a radial basis function (RBF) neural network. Herein, RBF was paired with the Chimp Optimization Algorithm (ChOA) and the Equilibrium Optimizer (EO) to determine the most important parameters of the RBF method that might be optimized, abbreviated as ChRB and EORB. The suggested algorithms were evaluated concerning 1030 experimental samples, eight input variables (HPC age, water, blast furnace slag, cement content, superplasticizer, fly ash, and fine and coarse aggregates), and the target CS as the forecasting. Accurate forecasting of CS enables construction sites to monitor and verify that concrete mixes satisfy strength specifications prior to pouring, therefore mitigating structural problems. This not only lowers carbon emissions but also promotes the use of industrial byproducts in building, in accordance with sustainable construction objectives. Precise CS forecasts facilitate the design of robust and resilient buildings, particularly for infrastructure projects exposed to significant loads or severe conditions. The results indicate that both hybrid ChRB and EORB analyses perform well in productivity for train and test portions. The EORB included lower index values than the ChRB. Based on the comparison results with the existing literature, it is obvious that the produced EORB outperforms all existing literature. In summary, the suggested approach is the EORB framework designed for estimating the CS of HPC enhanced with FA, and BFS.