Abstract
This study concerns CFD-informed methods to obtain the derivatives of lumped parameters maneuvering models for underwater vehicles. The uncertainty in maneuvering predictions due to the method of obtaining hydrodynamic derivatives is discussed and assessed. Hydrodynamic derivatives are typically obtained from planar motion mechanism tests in both planes. Unsteady RANSE solvers offer the possibility to simulate the same captive steady and unsteady maneuvers to obtain these derivatives by identification procedures. Upon comparing some general representations of such identification procedures, it is found that the values of the estimated coefficient vary due to the method used. The uncertainty affecting the derivatives is propagated onto standard maneuvering predictions. This article discusses the level of uncertainty related to each method and offers insight based on the type of derivatives that vary as well as the type of maneuvers that are affected through the application on two different underwater vehicles. It is observed that there is a recognizable difference in the accuracy and applicability of these methods and a possible way forward is proposed.
