Print Email Facebook Twitter Prediction of the performance of ducted propellers with BEM and hybrid RANS-BEM methods Title Prediction of the performance of ducted propellers with BEM and hybrid RANS-BEM methods Author Negrato, C. Contributor Van Terwisga, T.J.C. (mentor) Faculty Mechanical, Maritime and Materials Engineering Department Maritime and Transport Technology Programme Ship Hydromechanics Date 2015-11-13 Abstract A ducted propeller configuration consists of a fixed annular nozzle surrounding the propeller. The nozzle has an airfoil shape which depends on the required performance of the system as well as the operating conditions. The flow-accelerating ducts provide a positive contribution to the thrust of the propulsor and they are used to increase the performance in heavy loading conditions, such as in the cases of tug boats or often for azimuthal thrusters. The performance of the propeller can be assessed with model tests or numerical simulations. As concerns the numerical simulations, boundary element methods (BEM) are used daily in the design stage for open propeller configurations, but the use for ducted propellers is still under development: viscous flow effects become important at the duct surface and the accuracy of the BEM method decreases. Alternatively, Reynolds averaged Navier-Stokes (RANS) simulations are possible but they require large CPU time so they cannot be used at the design stage routinely. In addition, a hybrid RANS-BEM method was developed at the Maritime Research Institute of the Netherlands (MARIN). The hybrid approach couples the viscous flow solution with the boundary element method: the propeller is not physically present in the RANS simulations, where it is substituted with a body force distribution whose strength is given by the BEM. This research has the objective to determine the accuracy and limitations of the BEM and the hybrid RANS-BEM approaches to predict the performance of ducted propellers in open water condition. Two test cases are considered. The propeller is the same for both cases (Ka4-70 propeller) whereas the duct is different (namely duct 19A and duct 37). The BEM method requires a modification of the duct trailing edge geometry to enforce the modelling of the wake sheet attached to the duct. An new iterative scheme that automatically computes the change in geometry based on a pressure-equality condition is developed; as a result, the boundary element method computations provide 2% to 7% accuracy in the prediction of the open water efficiency for the design condition and the high loading conditions. However, for the light loading conditions the BEM method is limited by the occurrence of flow separation at the duct outer surface. The RANS-BEM method improves the prediction at the large advance ratios for the first test case. However, for the second test case there is a constant over-prediction of the propeller thrust and torque. The reason for this over-prediction is likely related to the extent of flow separation occurring at the duct inner surface for all operating conditions, even though the lack of validation data for the second geometry tested does not allow to confirm this hypothesis. Finally, another objective of this study is to provide insight on the flow behaviour at the gap between the propeller and the nozzle. The ultimate goal is to give guidelines on the modelling of the flow in that region in a potential flow context. Former CFD calculations are analyzed in detail with focus on the gap flow. It is observed that the gap flow is dominated by the detachment of a vortex at small chord-wise positions from the blade pressure side to the blade suction side. This is the so-called tip leakage vortex. The vortex is seen to obstruct the gap at the mid-chord positions and it strongly interacts with the tip trailing edge vortex. Subject RANS-BEMDucted PropellersCFD To reference this document use: http://resolver.tudelft.nl/uuid:5dbf8251-33a8-41f8-a454-2b1b55cbf1c7 Embargo date 2018-12-08 Part of collection Student theses Document type master thesis Rights (c) 2015 Negrato, C. Files PDF MSc_thesis_C_Negrato_4321 ... sitory.pdf 31.75 MB Close viewer /islandora/object/uuid:5dbf8251-33a8-41f8-a454-2b1b55cbf1c7/datastream/OBJ/view