Performance of ANN in Predicting the Tensile and Shear Strength of Al-Steel Explosive Clads

Abstract

In this study, an artificial neural network (ANN) model is created to predict aluminium-stainless steel explosive clads' tensile and shear strengths. The parameters for the explosive cladding process, such as the loading ratio (mass ratio of the explosive and the flyer, 0.6-1.0), standoff distance (5-9 mm), preset angle (0°-10°), and groove in the base plate (V/Dovetail), were altered. The ANN algorithm was trained in Python using the tensile and shear strengths gathered from 80% of the experiments (60), trials, and prior results. The constructed model was evaluated utilizing the remaining experimental results. The ANN model successfully predicts the tensile and shear strengths with an accuracy of less than 10% deviation from the experimental result.