Prediction of the hydrodynamic disturbance characteristics for two projectiles launched successively underwater based on the radial basis function neural network

When two projectiles are successively launched under different launch parameters, the motion of the first projectile affects the hydrodynamic characteristics of the second projectile. To predict and study such disturbances, a radial basis function (RBF) neural network model is established in this paper. Compared with the underwater launch of a single projectile, the hydrodynamic loads for two projectiles successively launched are more complex and severe. When the first projectile is launched, it will affect the forces and moments of subsequent projectiles, leading to launch failure. Thus, we apply a numerical simulation method that is verified through experiments to simulate two projectiles successively launched underwater. Then, we use the generated data to train the RBF neural network. The results show that vortices will form at the tail of the first projectile after launch due to viscous effects, which is the main reason for the hydrodynamic disturbance that affects the second projectile. Compared with numerical simulations and experimental methods, the RBF neural network model can more effectively predict the disturbance of the hydrodynamic characteristic variables of the first projectile to the second projectile. This disturbance can be reduced by increasing the spatial distance of the two projectiles, increasing the time interval between launches, and reducing the platform velocity. However, the launch time interval is the most sensitive factor affecting the hydrodynamic characteristics of projectiles.