Ballistic response of armour plates using Generative Adversarial Networks

It is important to understand how ballistic materials respond impact from projectiles such that informed decisions can be made in the design process of protective armour systems. Ballistic testing a standards-based where are tested determine whether they meet protection, safety and performance criteria. For V50 test, fired at different velocities key parameter known as limit velocity (BLV), above which perforate target. These tests, however, destructive by nature there considerable associated costs, especially when studying complex This study proposes unique solution problem using recent class machine learning system Generative Adversarial Network (GAN). The GAN used generate new samples opposed performing additional experiments. A network architecture trained on three data sets, their compared. networks were able successfully produce curves with an overall RMSE between 10 20 % predicted BLV each case error less than 5 %. results demonstrate it possible train generative limited number use many representative was on. paper spotlights benefits bring applications provides alternative expensive during early stages process.