Ship hull drag simulation

[mm7]

Hi,rnrnThank you, scientists and developers, very much for the method and software I hope can help to resolve my problem.rnrnI am experienced Linux admin and quite a bit familiar with C++, but I not an expert at all in CFD flow simulations.rnrnAs a programmer I am maintaining an amateur open source project “”FreeShip in Lazarus”” that is for modeling hulls (ships, boats etc). It has methods to analyse various hull’s hydrostatics and hydrodynamics, including thrust/drag using different methodologies. Choice of particular methodology depends very much on shape of a hull, and mode of operation, is it displacement, semi-displacement, semi-planing, or planing… Results vary very much as well.rnrnI hope OpenLB could help to analyze drag, buoyancy, planing etc, of same hull in continuously changing modes, on different speeds, i.e. while it accelerates from displacement to planing.rnAlso a hull may have complex shape or have complex attachments, that are not considered in many methodologies – stepped hulls, cutouts, attached lifting hydrofoils, etc…rnrnSo it would be nice that it could be possible to run a hull like in real conditions, probably even with waves and seeing forces during simulation. It does not need to be real-time or realistically looking simulations.rnrnFreeShip can export to STL file. rnI am looking into OpenLB examples and cannot find what would be most suitable one I should start from.rnAlso, what lattice and turbulence model to choose? rnPlease help me to make first steps in this direction.rnrnThanks,rnMarkrn

[mathias]

Dear Mark,rnrn You can start from cylinder3d or aorta3d. In both cases an stl-file is imported. You may also start from thy venturi3d example. There it is shown how to combine geometry primetive like cones, shperes or cuboids to set up a flow domain. Currently, we do not have a free surface solver. rnrnYou may also consider to come to our spring school in march http://optilb.org/openlb/spring-school-2017 which would help you to get started.rnrnBestrnMathias

[mm7]

Thank you Mathias.rnYou’ve mentioned that OpenLB does not have a free surface solver.rnTo me it sounds that it cannot be used for my task. Or I am wrong?rnI believed that if two “”liquids”” with different densities (water and air), and gravity are specified, then surface will appear automatically. And a hard body of intermediate density will automatically float. Like on this video.rnhttps://www.youtube.com/watch?v=5F-qHsPIrYErnIs it not the case with OpenLB?rnAm I too naive, and it is not that simple? :)rnrn

[mathias]

Hi,rnrnYou could implement it in OpenLB. A simple version of a FreeSurface method should not be so complicated to be implemented.But to get physical accurate results in general can be very tricky. There are PhD thesises on it. rnrnBestrnMathias

[mm7]

Thank you Mathias,rnI’ll try to run simple examples first…

[Author]

Posts