An artificial intelligence model for ballistic performance of thin plates

Abstract

This work presents the influence of ballistic testing variables on residual velocity of projectile and absorbed energy of aluminum 1100-H12 using design of experiments (DOE) and artificial neural network (ANN) approach. Simulations have been carried out with three process variables: Projectile nose shape, impact velocity and target thickness. Taguchi’s technique has been employed for experimental investigation. Trials were planned using an L 18 (61x33) orthogonal array with 18 combinations of testing variables was selected to assess the influence of various factors. Optimum testing variable combination was achieved by using analysis of signal to noise (S/N) ratio. Simulations were performed using Ansys Autodyn 3 D code. Simulated and experimental results were compared with each other and found well. The predictions of the ANN model, simulation results were in good agreement with the experimental data taken from literature.