Investigation of marine propeller performance characteristic with nonlinear blade element momentum theory and computational fluid dynamics methods

Determining the hydrodynamic characteristics of marine propellers has always been a very challenging problem in fluid mechanics. With the developing computer technology, computational fluid dynamics method has become available to solve the problems in vast areas of fluid dynamics. Despite of the day by day evolution for numerical methods in computational fluid dynamics method, it is difficult to generate computational fluid dynamics grids to achieve accurate and converged results for rotating propellers and turbine blades. The calculations are very sensitive to the computational fluid dynamics grids and boundary conditions. For this reason, when the difficulty of grid generation and long computational times are taken into consideration, computational fluid dynamics method is still impractical in the preliminary design of the propeller. Nowadays, low-order accuracy methods are still preferred in the conceptual design phase. The use of these methods in the conceptual design phase allows the use of computational fluid dynamics method with higher order accuracy in the later stages of design and allows model testing to be carried out, thus greatly shortens the design process. Blade element momentum theory is one of the low order accuracy methods. This theory is a combination of momentum and blade element theories. Momentum theory provides that the velocity and pressure values across the propeller disk and flow direction, considered as an infinite number of propeller blades in the stream tube, can be calculated based on the principles of conservation of energy and axial momentum. According to this theory, flow is considered to be • Inviscid • Irrotational • Incompressible. For this reason, energy losses are neglected. The momentum theory, however, does not take the torsional effects into account and thus ignores tangential effects. As a result, thrust effects appear to be uniformly distributed over the propeller disk. An important result is obtained despite all the approximations and neglected effects of the momentum theory; the induced velocity at the propeller disk equals to the half of the induced velocity at the far downstream of the propeller. Blade element theory accepts the lift and thrust coefficients of the radial sections of the propeller as an input and calculates open water performance values of the propeller corresponding to relevant advance ratio, thrust and torque coefficients. In this theory, the resultant velocity acting on the radial sections of the propeller are utilized. For this reason, it is necessary to correctly determine the pitch angle, hydrodynamic pitch angle and angle of attack. The blade element theory has been developed based on the principle of conservation of angular momentum. In order to compensate the shortcomings of neglected effects during the mathematical modeling of these two theories, blade element momentum theory has been proposed by combining blade element theory and momentum theory. The closed form of the Linear Blade Element Momentum Theory equation is presented by McCormick. McCormick assumed that the drag has little effect on the induced angles of attack; the induced angle of attack are therefore small at all sections along the blade. With these approaches, the closed form of the Blade Element Momentum Theorem is obtained. However, it is known that this approach gives unacceptable results in cases with high advanced ratios. In blade element momentum theory, the lift (CL) and drag (CD) coefficients of the blade sections must be determined in order to calculate the performance values of the propeller. For this purpose, nonlinear blade moment theory program developed in Python takes these values from the XFOIL program in the Reynolds number of each section. In order to examine the results of XFOIL program under different initial conditions in this thesis study, flows around NACA0012 airfoil was calculated with the developed source-doublet method program, XFOIL and computational fluid dynamics, and the results have been compared with the experimental results. In this thesis, the open water performance of DTMB 4119 propeller, the standard test propeller, has been investigated by using non-linear blade element momentum theory program developed to avoids mentioned innacurate small angle assumptions, it has been compared with analytical blade element momentum theory, computational fluid dynamics method and open water propeller test results. Following the DTMB4119 propeller, skewed DTMB4382 propeller was investigated by using developed non-linear blade element momentum theory program and the results have been compared with the experimental results. Since the skew can not be modeled with blade element momentum theory method, the skew effect on propeller performance has been examined. For computational fluid dynamic analysis of NACA0012 aerofoil, 3 different mesh have been generated and Grid Convergence Index (GCI) procedure has been carried out. The ANSYS Fluent 17.2 program was used for the RANS solution. Turbulence model has been used as k-ε turbulence model. y+ value has been kept between 30 and 300. The second order upwind scheme has been used for the momentum and turbulence terms, and the simple algorithm for velocity pressure interaction has been preferred. According to these results, it is observed that the potential theory cannot properly model the stall situation. Given that the potential theory does not modeled the stall situation, the CL and CD values of the sections required for the blade element momentum theory method can be obtained by computational fluid Dynamic method instead of the XFOIL program, resulting in more precise results. However, it is possible to say that this method to be followed will significantly increase the solution time. Finally,the open water performance values of the DTMB4119 propeller has been calculated using the computational fluid dynamics method and the open water performance values results have been compared with the experimental results. Three different mesh have been generated for computational fluid dynamic analysis and Grid Convergence Index (GCI) procedure has been carried out. The ANSYS Fluent 17.2 program has been used for the RANS solution. As the turbulence model, k-ε turbulence model has been used. y + value has been kept between 30-300. The second order upwind scheme has been used for the momentum and turbulence terms, and the simple algorithm for velocity pressure interaction has been preferred. It has been observed that the relative error has been increased when advance ratio increase because the k-ε turbulence model and the wall function can not accurately modeled the viscous effects in high Reynolds number. It is thought that use of k-ω SST turbulence model instead of k-ε turbulence model and wall function may give more accuare results in high advanced ratio.