Development of a state-of-the-art Navier-Stokes solver for water flows around moving ships (CWI.4883)
Project nummer:
cwi4883
Omschrijving van het onderzoek
The project aims at the development of a state-of-the-art Navier-Stokes solver for water flows around moving ships in order to essentially improve the reliability and efficiency of the computational modeling technology required for the prediction of ship behavior (performance, stability, control, etc.). The project will be executed in close cooperation with the Maritime Research Institute Netherlands (MARIN). The deliverables will be a new computer program (software) and accompanying technical reports (as well conference proceedings and refereed journal papers). The present research proposal fits in a masterplan, of which a description has been enclosed. (In the masterplan, the present project fits in Project 4). During the project proposed here, independent and on-going joint CWI-MARIN projects on the improvement of MARIN's existing viscous-flow techniques (work on MARIN's PARNASSOS program) will be performed: Projects 1,2 and 3 in the masterplan. (All three projects are/will be fully financed by CWI and MARIN themselves.)
The programs applied nowadays at MARIN are based on a potential-flow and a viscous-flow model. The potential-flow model is fruitfully applied for the computation and optimization of wave drag. Computational technology for potential flows is mature and well-proven at MARIN. However, this is not yet the case for viscous flows. The difference is due to the more complex and difficult mathematics involved in viscous-flow models.
In order to better support the ship building and offshore community, at MARIN the need is felt for an essentially enhanced computational technology. The aim of the present proposal is to essentially improve the reliability and efficiency of MARIN's computational modeling technology for viscous flows. This is the right time to do so, because CWI's involvement in recent developments in Computational Fluid Dynamics (CFD) for aerodynamics applications allows it to adapt these developments for ship hydrodynamics. In this way, CWI expects to assist MARIN in making a major leap forward. MARIN's direct application of the Navier-Stokes code lies in flow-topology studies (flow separation, re-attachment, vortex formation, etc.) around ship hull forms. These studies are required at various sailing conditions (speed, trim, sinkage). So far, MARIN's computational technology for viscous flows is based on and restricted to steady, reduced Navier-Stokes equations, and the corresponding computational domains are restricted to ship-stern flows, and to flat water surfaces only. The research proposed here wants to eliminate these restrictions by the development of a state-of-the-art, full Navier-Stokes solver, suited for accurate and computationally efficient, steady hydrodynamics around full ships (i.e., including both bow and stern) and with free-surface effects (water waves). (Unsteady ship hydrodynamics will not yet be considered in the present project).The method will contain numerical algorithms, which have been recently developed by the applicants for (aircraft) aerodynamics applications and which are in fact lying ready for use in (ship) hydrodynamics applications. The method will be tested on MARIN's own viscous ship hydrodynamics problems.
Resultaten van het onderzoek
Er is nauw samengewerkt met MARIN, waar voor verschillende schepen berekeningen zijn gedaan en zeer goede resultaten zijn verkregen. Bovendien is de vereiste rekentijd een fraktie van wat andere methoden nodig hebben. De ontwikkeling zal door MARIN worden voortgezet en de methode zal daarna in de praktijk worden geintroduceerd. Er is kort geleden meegedaan aan een internationale workshop waar de berekening van het door een schip gegenereerd golfpatroon met Navier-Stokes methoden een prominente plaats innam; de Nederlandse resultaten behoorden tot de beste die daar gepresenteerd werden.
Het onderzoek wordt voortgezet in het BSIK-project BRICKS.
Gebruikers
Er zijn één bedrijf en twee instellingen bij dit project betrokken.
Projectleider
Dr.ir. B. Koren
CWI
Afd. Numerieke Wiskunde
Kruislaan 413
1098 SJ Amsterdam
Status van het project
| Gestart |
: 01-10-1999 |
| Einddatum |
: 01-12-2003 |