Closed-Form Equation for Estimating Unconfined Compressive Strength of Granite from Three Non-destructive Tests Using Soft Computing Models

Abstract The use of three artificial neural network (ANN)-based models for the prediction unconfined compressive strength (UCS) granite using non-destructive test indicators, namely pulse velocity, Schmidt hammer rebound number, and effective porosity, has been investigated in this study. For purpose, a sum 274 datasets was compiled used to train validate ANN including constructed Levenberg–Marquardt algorithm (ANN-LM), combination particle swarm optimization (ANN-PSO), imperialist competitive (ANN-ICA). ANN-LM model proven be most accurate based on experimental findings. In validation phase, achieved best predictive performance with R = 0.9607 RMSE 14.8272. Experimental results show that developed outperforms number existing available literature. Furthermore, Graphical User Interface (GUI) which can readily estimate UCS through model. GUI is made as supplementary material.