3 results
Filters
Settings
Now showing1 - 3 of 3
- Some of the metrics are blocked by yourconsent settings
Item type:Publication, Access status: without files , Disequilibrium wall function in RANSE computation applied to flow conditions around ship’s hull marine 2019(2019-05); Through years of scaled physical model tests to predict ship induced wave loads on rip-rap and bank protection in channels and waterways at Federal Waterways Engineering and Research Institute, a comprehensive collection of squat measurements had grown and became foundation of some principal and systematic conclusions on the squat effect in shallow and restricted waters [1]. There was the observation of a significant increase of the trim angle when water depth to draft ratio is decreased to less than = 1.3. A change in the flow regime was suspected to be responsible for this effect [2]. To prove the assumption of different flow regimes the boundary layers in the gap between the hull’s bottom and the ground were further investigated. All the investigations so far have been in the scale 1:40 in accordance with Froude’s Law. Boundary layers and gap flow conditions heavily depend on the Reynolds’ Number and the turbulence degree. Unfortunately it is impossible to satisfy Reynolds’ and Froude’s Law for this scale and this experimental set-up at the same time. It was decided to bypass this quandary with an extension of the analysis by numerical simulation of the flow around the hull at full scale. In the scaled model tests, the occurrence of a coalescence of the two boundary layers, at the hull and the ground, in the aft third of the hull was observed. In terms of time and computing cost, computational fluid dynamics simulation of flow around a ship’s hull at full scale are not reasonably feasible without application of wall functions today. The flow condition in the boundary layer violates the basic assumption of local equilibrium which is common for the regular wall functions implemented in the CFD-Codes solving Reynolds Averaged Navier-Stokes Equations with turbulence models to close the system of equations for solving. To cope with the non-equilibrium condition in the boundary layer at the hull, the “Generalized Wall Functions” [3] were implemented in OpenFOAM [4] and applied to the numerical analysis of the flow conditions around the ship hull.Publicationtype: Conference Paper45 - Some of the metrics are blocked by yourconsent settings
Item type:Publication, Access status: without files , 72 - Some of the metrics are blocked by yourconsent settings
Item type:Publication, Access status: with files , Validation of a GPU-accelerated LBM based numerical ice tank for a Voith Schneider propulsion system(2019-05-13); ; ; This contribution outlines the capabilities of an enhanced GPU-accelerated Lattice Boltzmann method (LBM) for the simulation of ice-going Voith Schneider propelled (VSP) ships. Reported results cover a collaboration between an academic and an industrial partner, i.e. TUHH and Voith, within a joint research project to assess the ice-induced loads for cycloidal propellers. The goal is to establish a simulation framework which supports extensive parameter studies in a competitive computational time. Simulation objectives are an insight into the ice clearing capabilities of a hull, the ice’s influence on the propulsion efficiency as well as the additional ice-induced blade loads. Investigated influences refer to the ice thickness, the ice floe geometry, the ship speed and/or the VSP’s parameters. The experimental and numerical ice tanks are restricted to pre-broken ice floes and ice-breaking is not considered. The simulated dynamics of the multiple rigid bodies system, i.e. the. ice floes, the hull and the VSP, are obtained from a monolithic coupling of the flow solver and a contact-dynamics Physics Engine (Open Dynamics Engine; ODE). For the coupling of the explicit LBM and the motion solver, a bidirectional and explicit coupling approach is used. Computations were performed on moderate LBM grids featuring about 20 million nodes using local grid refinement. Prior to the assessment of ice loads, hydrodynamic validation in open-water conditions of the LBM against established FV-simulations will be reported. Subsequently, the comparison of simulated and measured ice loads experienced by a 5-bladed VSP attached to a generic hull geometry will be analyzed. A comparison of the ice floe dynamics displays reasonable agreement and force assessments confirm that the simulation is able to capture all important effects. Final applications are concerned with a tug geometry - propelled by 2 5-bladed VSPs - in pack ice conditions.Publicationtype: Conference Presentation198 193