Prediction of the compressive and tensile strength of HPC concrete with fly ash and micro-silica using hybrid algorithms

Evaluating the impact of fly ash (FA) and micro-silica (MS) on the tensile (TS) and compressive strength (CS) of concrete in different ages provokes to find the effective parameters in predicting the CS and TS, which not only could be usable in the practical works but also is extensible in the future analysis. In this study, in order to evaluate the effective parameters in predicting the CS and TS of concrete containing admixtures and to present a fitted equation, the multivariate adaptive regression splines (MARS) method has been used, which could find a relationship between independent and dependent variables. Next, for optimizing the output equation, hybrid genetic algorithm (GA), particle swarm optimization (PSO), and grey wolf optimization (GWO) methods have been utilized to find the optimal conclusions. It could be concluded that for both predictions of CS and TS, all models have the coefficient of determination (R2) larger than 0.949 and 0.9138, respectively. Furthermore, between three hybrid algorithms, MARS-PSO could be proposed as the best model to obtain the most accuracy in the prediction of CS and TS. The usage of hybrid MARS-PSO techniques causes a noticeable improvement in the prediction procedure.