5 results
Now showing1 - 5 of 5
- Some of the metrics are blocked by yourconsent settings
Item type:Publication, Access status: without files , A boundary element-vortex particle hybrid method with inviscid shedding scheme(2018-05-30); ; This work presents a hybrid method combining the low order boundary element method and the vortex particle method, in the sense that inviscid wake panels are shed from the trailing edge and then converted to vortex particles for viscous simulation of the slipstream flow. The way to combine these two solution techniques are described in detail as well as the methodology to convert wake panels to vortex particles. Two elliptical hydrofoils in the steady flow and an oscillating rectangular hydrofoil are investigated with the hybrid method. It is demonstrated that, to obtain a physical vorticity distribution, the inter-particle spacing cannot be chose arbitrarily, but should be correlated to the boundary layer thickness. The hybrid method shows a good capability to conserve the axial vorticity. The obtained tangential velocity profile across the tip vortex also exhibits an acceptable correlation with experimental measurement. Also presented are the obtained hydrodynamic forces, their convergence with regards to panel size and time step, and the comparison with results of RANS solver and boundary element method.Publicationtype: Journal Article208 2 - Some of the metrics are blocked by yourconsent settings
Item type:Publication, Access status: with files , Simulating marine propellers with vortex particle method(American Institute of Physics (AIP), 2017-01-31); ; The vortex particle method is applied to compute the open water characteristics of marine propellers. It is based on the large-eddy simulation technique, and the Smagorinsky-Lilly sub-grid scale model is implemented for the eddy viscosity. The vortex particle method is combined with the boundary element method, in the sense that the body is modelled with boundary elements and the slipstream is modelled with vortex particles. Rotational periodic boundaries are adopted, which leads to a cylindrical sector domain for the slipstream. The particle redistribution scheme and the fast multipole method are modified to consider the rotational periodic boundaries. Open water characteristics of three propellers with different skew angles are calculated with the proposed method. The results are compared with the ones obtained with boundary element method and experiments. It is found that the proposed method predicts the open water characteristics more accurately than the boundary element method, especially for high loading condition and high skew propeller. The influence of the Smagorinsky constant is also studied, which shows the results have a low sensitivity to it.Publicationtype: Journal ArticleScopus© Citations 13 208 442 - Some of the metrics are blocked by yourconsent settings
Item type:Publication, Access status: without files , A fast method to realize the pressure Kutta condition in boundary element method for lifting bodiesThe Newton–Raphson method is always employed to realize the zero pressure jump at the trailing edge of lifting bodies, which is required by pressure Kutta condition. This paper describes an analytical method to evaluate the Jacobian matrix during this procedure. With the proposed method, the computation time of the Jacobian matrix is less than that with the conventional method by five orders of magnitude. This allows for calculating the Jacobian matrix in every iteration of the Newton–Raphson method, which accelerates the convergence rate. In addition, a fast method, called the dipole increment method, to update the unknown singularities on the body surface during the Newton–Raphson procedure is also presented. Combination of these two methods makes the time consumed by the nonlinear pressure Kutta condition negligible (less than 2% of the total simulation time) in the boundary element method for both steady and unsteady flows.Publicationtype: Journal Article108 - Some of the metrics are blocked by yourconsent settings
Item type:Publication, Access status: without files , Convergence of different wake alignment methods in a panel code for steady-state flows(2016-12-01); ; Wake alignment models are always included in the modern panel codes for marine propeller analysis. The wake alignment algorithms influence directly the rate of convergence and the accuracy of calculations. In the present work, firstly, four different numerical methods to implement the wake alignment algorithms for the steady calculation are described. They perform quite differently in terms of convergence history and convergence rate. The comparison with the other methods shows that the direct application of the unsteady method leads to a much slower convergence rate. Secondly, high-order numerical methods including second-order and fourth-order Runge–Kutta methods are introduced into the wake alignment, which results in high-order wake alignment algorithms. The analysis of the results shows that the high-order methods generate a different wake geometry from the low-order method. The thrust coefficient and torque coefficient have also been compared.Publicationtype: Journal Article108 2 - Some of the metrics are blocked by yourconsent settings
Item type:Publication, Access status: without files , Prediction of tip vortex cavitation inception with low-order panel method(2016-10-01); ; ; Tip vortex cavitation is the first type of cavitation to take place in most cases. Because it always produces radiated noise, avoiding or at least delaying tip vortex cavitation is crucial for achieving low radiated noise emissions. The present study aims to predict the tip vortex cavitation inception with the low-order panel method, which is widely used at the early propeller design stage. First, a velocity smoothing parameter varying with the age of vortex is adopted to align the wake. This leads to a wake shape that coincides very well with the wake pattern from a RANS solver. A consistent formula for evaluating the smoothing parameter for different cases is also achieved with the fitting method. Second, the tip vortex circulation is calculated based on the converged wake shape and doublet distribution. This procedure is found to be independent from the panel size. Finally, based on the minimum pressure distribution along the tip vortex, cavitation inception numbers are obtained for the Potsdam Propeller Test Case (PPTC) and the propeller 4383 developed in NSRDC. The results show a good agreement with the experimental measurements.Publicationtype: Journal Article162 1