Projectile shape optimization using numerical twin with computational fluid dynamics and design of experiment methods

Abstract

Projectile shape optimization is crucial for enhancing aerodynamic performance and cost-effectiveness in ammunition development. This paper presents an integrated methodology leveraging the numerical twin concept, combining computational fluid dynamics (CFD) with design of experiments (DOE) methods to optimize projectile external geometry. The CFD model, serving as the core of the numerical twin, enables accurate prediction of the aerodynamic drag coefficient, validated against experimental data with a deviation of less than 6.5% at Mach 2.73. A structured experimental plan was established to determine the relationship between key geometric parameters and the drag coefficient for Mach numbers ranging from 2 to 2.75. Based on these results, a response function approximating the drag coefficient was formulated and integrated into the trajectory model, showing agreement with PRODAS software (less than 1.26% difference in terminal velocity and range). A second DOE plan was then used to optimize the shape of a prospective projectile for maximum terminal velocity at 300 meters, achieving 753 m/s. The resulting optimization tool provides time-effective estimations of terminal ballistics improvement for slight geometrical modifications. This holistic CFD-DOE approach streamlines and reduces the cost of projectile shape optimization, lessening the need for extensive prototyping and accelerating the design cycle.