Unsteady computational fluid dynamics simulations of six degrees-of-freedom submarine manoeuvres
University of New Brunswick
Submarine hydrodynamicists are continually striving to improve submarine manoeuvring predictions in order to better understand operational limitations. Traditional manoeuvring simulations use semi-empirical force estimation methods or captive model experiments to predict hydrodynamic forces and moments on the manoeuvring submarine. Semi-empirical methods are inexpensive and fast but not consistently accurate. Coefficient-based models derived from captive model experiments are more accurate but much more expensive. Time-history effects are also difficult to incorporate in coefficient-based modelling and are thus often neglected. The effect of the seabed and other passing vessels are also not included in traditional models. In order to address these issues, an unsteady, six degrees-of-freedom (DOF), Reynolds Averaged Navier-Stokes (RANS) submarine simulation approach has been developed and applied to simulating the emergency rising manoeuvre and the interaction between a submarine and a passing tanker. In this method, the 6 DOF motion of the full-scale submarine is integrated simultaneously with the RANS solution for fluid forces. Coefficient-based modelling is used for the propeller, control-surface deflections, and ballast blowing in order to achieve economical solutions while accounting for the important viscous effects. The submarine motion is integrated implicitly using a robust predictor-corrector approach that uses a novel application of semiempirical and coefficient-based models to accelerate convergence at each timestep. A combination of sliding interfaces and mesh deformation is used to account for the relative motion between the tanker and submarine. Simulation results reproduce the roll-instability observed for buoyantly rising submarines and provide insight on the magnitude of the disturbance forces and submarine motions induced by a large tanker passing over a diesel-electric submarine.