KonKav II

Correlation of Cavitation Effects Under Consideration of the Wake Field

Motivation

Propeller cavitation is a very unpleasant phenomenon which has a couple of harmful effects: Increased vibration of the structure, noise and erosion. This warrants a broad research work on this theme. The purpose of the research project KonKav II is to develop a more accurate and marketable cavitation prognosis. In the project the focus is put on how scale effects influence the interaction between wake field, propeller cavitation and the resulting pressure fluctuations. A deeper understanding of these processes can help to convert model test results to full scale predictions in a reliable way. In the long term it even will become possible to predict cavitation and pressure fluctuations without model tests.


Objectives and Realisation

The project contains both experimental and computational research. The computational parts are contributed by the Hamburg University of Technology:

  • Improved prediction of the full scale wake field under consideration of the propeller influence and propeller cavitation. The solver coupling functionality to RANS solvers serves as a framework for this task: The main computations are carried out by a RANS solver. To reduce the computational effort of the full scale RANS computations, the propeller is not geometrically existent in the viscous fluid domain but represented by a body force model (click here for more information.)

  • Prediction of full scale pressure fluctuations under consideration of cavitation effects. Again the solver coupling functionality is the base line. In contrast to the foregoing part, the model in panMARE is extended to the whole ship including propeller. With an improved potential-theory-based cavitation model being able to capture both sheet and tip vortex cavitation it is possible to make a fast and accurate prediction of pressure fluctuations.

  • Accurate capturing of hub and tip vortices in full scale RANS computations. This requires a very fine grid resolution. To limit the computational effort at high Reynolds numbers to a marketable level, a combination of vortex-confinement techniques and dynamic local grid refinement is employed.

  • Improved simulation of a full scale wake field in the context of model tests. Model tests still play an important role in cavitation prognosis. It is common practice to use dummy models with attached strainers influencing the flow in a way that the wake field of the full scale version is simulated. A procedure based on an adjoint sensitivity analysis is developed to find an appropriate dummy model geometry and appropriate mesh parameters of the strainers to simulate the full scale wake field, which has been calculated previously.


Cooperation


Personnel


Funding

The work is sponsored by the German Ministry of Economy and Technology (BMWi).