This June, a workshop organized by ERCOFTAC and Lulea University of Technology (LUT) was held at the Porjus Hydropower Center in Sweden. The purpose of this gathering was to determine the state of the art in the computational modeling of hydraulic turbine draft tubes. LUT and the Porjus Hydropower Center are long-time research partners and are actively examining both experimental and numerical advances in hydraulic turbine technology.

There were a total of 16 participants in this workshop from all over the world, including CFD software providers, universities, and major corporations. Each contributed their CFD results for a hydroturbine draft tube configuration that was provided by the workshop organizers. The experimental data were not available until the workshop, so each contributor was in essence providing “blind test” cases for head-to-head comparison at the workshop.

Fluent contributed two sets of FLUENT results, one using the k-epsilon model, and another using the Reynolds Stress turbulence model. Of the 16 participants, 7 used either FLUENT or FIDAP to get their results.

FLUENT results for the hydroturbine draft tube compared very favorably with pressure data

As it turned out, the workshop organizer did not publish all experimental data at the workshop, but they did have some pressure data available. This enabled the participants to compare results for “pressure recovery factor,” which is practically the most important quantity for draft tube analysis.

Mesh sizes between the various participants varied considerably. Fluent gave one of the best results in terms of the pressure recovery factor, although we used the finest mesh. The experimental data indicated 1.12, and FLUENT predicted 1.13 for the Reynolds Stress turbulence model and 1.15 with the k-epsilon model. Measurements by the software of Fluent’s competitors ranged from .697 to .96.

FluentNEWS