dc.contributor.advisor | Nielsen, Torbjørn Kristian | nb_NO |
dc.contributor.advisor | Solemslie, Bjørn Winther | nb_NO |
dc.contributor.author | Rygg, Jone Rivrud | nb_NO |
dc.date.accessioned | 2014-12-19T11:50:56Z | |
dc.date.available | 2014-12-19T11:50:56Z | |
dc.date.created | 2013-09-19 | nb_NO |
dc.date.issued | 2013 | nb_NO |
dc.identifier | 649748 | nb_NO |
dc.identifier | ntnudaim:10032 | nb_NO |
dc.identifier.uri | http://hdl.handle.net/11250/235212 | |
dc.description.abstract | During the spring of 2012, Lorentz Fjellanger Barstad developed a method for modelling the flow in a Pelton turbine subject to a high-speed water jet using the Computational Fluid Dynamics (CFD) software ANSYS CFX. The torque measurement was validated against experimental data. The aim of this master's thesis has been to develop a similar method with the Open Source tool OpenFOAM and to compare the two models.A method has been created using the OpenFOAM solver interDyMFoam, capable of handling two-phase flow together with mesh motion. The approach has been to use both a stationary and a rotating mesh domain to allow for the relative motion between the high-speed jet and the turbine buckets. The Arbitrary Mesh Interface (AMI) was used as a boundary condition for the patches between the two domains to allow simulation between them. Meshing was done both with the built-in tool snappyHexMesh and with ANSYS Meshing. The latter gave the best control over mesh refinement and the mesh quality.The results achieved from the method were unfortunately not as desired. Much water seems to accumulate between the buckets, giving severe backwash. The measured torque was significantly larger than both the experimental torque and the torque measured with the ANSYS CFX method. Additionally, the torque measurement curve from OpenFOAM contained instabilities and did not coincide well with the one generated in ANSYS CFX. The measured maximum torque of the method seemed to go towards the actual solution when the density of the mesh increased, but at the same time it gave more noise in the output, and made smoothing of the results necessary. The computational time needed for the simulations has been problematic, being almost thirty times that of ANSYS CFX. | nb_NO |
dc.language | eng | nb_NO |
dc.publisher | Institutt for energi- og prosessteknikk | nb_NO |
dc.title | CFD Analysis of a Pelton Turbine in OpenFOAM | nb_NO |
dc.type | Master thesis | nb_NO |
dc.source.pagenumber | 117 | nb_NO |
dc.contributor.department | Norges teknisk-naturvitenskapelige universitet, Fakultet for ingeniørvitenskap og teknologi, Institutt for energi- og prosessteknikk | nb_NO |