Due to the continuous increase in computing power, Reynolds-averaged Navier Stokes CFD simulations have become an integral part of propeller design. Moreover, advances in computer technology led to the development of modern CAD/CAE systems like CAESES that allow for parametric design and shape optimization by tightly coupling CAD to CFD. The goal of this project is to investigate and validate open-source CFD tools that can be used to evaluate the open water characteristics of a propeller and

more »

... automated in order to support an optimization process. Then, develop a computationally cheap methodology inside CAESES that will eventually be used to optimize a propeller with the limited computational resources available. The efforts are focused on the difficulties of the automation the grid generation process, that still remains a bottleneck in optimization studies with CFD. The search leads to the selection of OpenFOAM to simulate the flow and cfMesh to generate the computational grid. The steady state Moving Reference Frame approximation is used to model the rotation and the flow is solved for one of the blades, using periodic boundaries. The turbulence is modelled with the k-ω SST turbulence model. The method used is validated using the Potsdam Propeller Test Case (PPTC) by comparing simulation results to experimental data. A parametric model for the PPTC is created, CFD automation is achieved and the propeller is optimized, by modifying the radial distributions that define the blade shape. The optimization is performed using the Sobol sequence and the Tangent Search Method that are available in CAESES. Finally, basic geometrical operations that are essential for the connection of a 3D propeller model to the CFD setup are scripted and automated inside CAESES, minimizing the manual effort needed to get a fast and rough calculation of any imported propeller blade. The scripts developed are validated with two powerboat propellers, by comparing the acquired CFD results to Vortex Lattice results.