Development of numerical tools for performance prediction

​This is the main track of research within the Unit. A reliable tool for velocity prediction is useful for all other CYRU projects.

Traditional Velocity Prediction Program (VPP) codes solve the rigid body equations of motion for the yacht in three or four degrees of freedom. The solution could be either static, where force and moment equilibrium is imposed, or dynamic, where the equations are integrated in time. Hydro and aerodynamic input is obtained from empirical relations, experiments or Computational Fluid Dynamics (CFD).

In our approach we use CFD for the entire VPP, i.e. we solve the fluid and rigid body equations of motion in a coupled manner as a function of time. This development will be taken in several steps described below. The technique developed is named CVPP.






Chalmers Yacht Research Unit – Project overview. Green: finished; Yellow: ongoing; White: planned

Project 1.1. In his Licentiate thesis, Lindstrand Levin simulated the hydrodynamics using a Reynolds-Averaged Navier-Stokes (RANS) solver, but the aerodynamic flow was modelled using empirical data. See Lindstrand Levin and Larsson (2017) in the publications list. This requires a Navier-Stokes based flow solver and a 6 DOF rigid body solver, strongly coupled. Special routines are needed for rudder movement, sail trimming and reefing, as well as for the operation of the program. For stability reasons the DOFs have to be introduced gradually and the time stepping varies, which calls for considerable testing and validations.

Carl Janmark (Janmark 2017) carried out his Master’s thesis within the project and improved the original technique in several respects, for instance by a better rudder control algorithm and a more efficient monitoring of the process by external Java scripting.

We plan to further develop CVPP as follows:

1.2.    Include also the aerodynamic part in the CFD solution. With this update, the entire hydro/aero flow problem will be solved in a coupled manner
1.3.    Introduce incoming waves. With this functionality, motions when sailing in a seaway, as well as speed loss may be simulated under realistic conditions
1.4.    Include body motions of the crew. By moving around the crew can impose motions on a dinghy that can boost speed, particularly when sailing in waves.
1.5.    Include fluid-structure interaction (FSI) of sails and hull/appendages. The shape of the sails determines the pressure distribution, which in turn changes the shape. Taking sail flexibility into account when computing aerodynamic forces is important, particularly under unsteady conditions. FSI will also be considered for the hull/appendages.
1.6.    Include rapid controlled motions of the sails (pumping, flicking). There are several techniques used for artificially increase sail power. All are based on rapid changes of the sail angle of attack. With the above items in place these effects can be studied under realistic sailing conditions.
 

Project 1.7 is a pre-amble to 1.2, where CFD computations of the flow around a jib/mast/main combination are carried out both for steady conditions and in pitch at two frequencies. Only upwind cases are considered. The computations are validated against wind-tunnel experiments. Initial computations are reported in the MSc thesis by Persson (2016). More complete results are provided in Persson, Muggiasca and Larsson (2017). The experience from the numerical modelling will be exploited when introducing CFD for sails in the CVPP (Project 1.2).

Project 1.8 is also related to 1.2 but for downwind sails. This work is carried out by Gustaf Magnander as a 60hp MSc project. Only steady state conditions are studied and validated against wind tunnel data. Note that the flow around a downwind sail is fundamentally different from that on an upwind sail.

Project 1.9 is a pre-amble to Project 1.5, but carried out separately from the CVPP to investigate the Fluid-Structure Interaction of both upwind and downwind cases. It will be particularly important downwind. Only steady conditions will be considered in this initial study.

 

Published: Mon 18 Sep 2017. Modified: Mon 25 Feb 2019