Separation control on stern of a ship by vortex generators

Flow separation, which is detachment of a boundary layer from the wall surface due to adverse pressure gradient, is an undesired phenomenon in terms of ship hydrodynamics. Flow separation on marine vessels can cause negative results as well as increasing resistance and propulsion energy losses. It is possible to determine the location of the separation by the aspects of hydrodynamics and aerodynamics scopes on surfaces of aerial and nautical vessels. It can be a significant issue to vanish or delay flow this incident. As an application, vortex generators already has favorable effects on flow separation by generating vortex in order to increase energy level in the boundary layer with a consequence of deriving streamlines towards wall surface. Vortex generators, which are commonly in use in the field of aerodynamics, have potential for hydrodynamics studies. In scope of this thesis, vortex generators are utilized as an implantation on stern of a tanker with severe flow separations to investigate separation control of a marine vessel by vortex generators. The literature review related to vortex generators and its types and parameters are stated in Chapter 1. Committed studies are executed by using Computational Fluid Dynamics (CFD). The methods and the governing equations are briefly mentioned in Chapter 2. Besides, in this chapter, detailed information is given about incompressible Reynolds Averaged Navier Stokes (RANS) based two equation eddy viscosity SST (Shear Stress Transport) k-ω turbulence model and pressure velocity relation algorithm for steady flow is taken SIMPLE (Semi Implicit Method for Pressure Linked Equations) algorithm which denotes pressure-velocity relation for steady flows that are chosen by this thesis. Moreover, in this chapter Law of the Wall is briefly explained. In this study, dimensionless distance from the wall y+ values is attempted lower than 5. Since, the flow separation is investigated properly with y+<5 the viscous sublayer region. In CFD, the validations of techniques and methods are essential in order to ensure for reliability of studies which are applied. In Chapter 3, validations of CFD techniques and methods that are carried out are presented. In validation studies, U-shaped stern formed second version of the KVLCC (KRISO – Korea Research Institute of Ships and Ocean Engineering – Very Large Crude Carrier) by MOERI (Maritime and Ocean Engineering Institute, Korea) tanker KVLCC2 is used. KVLCC2 validation studies are performed with double model approach and infinite depth assumption. Therefore, free surface and wave effects are neglected and trim and sinkage dynamic motions are not taken account into calculations. Furthermore, split from symmetry of the hull model is carried out for CFD calculations. Validations are fulfilled by comparison between experimental data which are provided in literature and obtained CFD results from this study. These experimental results which are mainly compared with CFD results are resistance values, form factor (1+k) and velocity contours taken from particular sections. Due to limited time and power of computer processors, cutting model is mainly preferred. For this preference, an additional validation for cutting model method is performed between cut model uncut model. For validate this method, cut model is generated from KVLCC2 mesh. Resistance values, velocity, pressure, turbulent kinetic energy (k) and specific dissipation rate (ω) contours are compared. According to these comparisons, the resistance values of the named selected regions are satisfyingly near each other and the contours lines of cut and uncut models are reasonably matched. In Chapter 4, vortex generators implementations on stern of a tanker with severe flow separations are included. Before implementation of vortex generators, the mesh dependency studies for bare hull tanker are accomplished. Moreover, cutting model method is used for the ideal mesh, which is provided from mesh dependency studies. This method is also validated for ideal tanker mesh with comparison of the resistance values, velocity, pressure, turbulent kinetic energy and specific dissipation rate contours from cut and uncut models. The inlet profile, which is acquired from cutting model method, is used for all following simulations as inlet condition. Afterwards, for calculating flow near the wall directly, y+ dimensionless distance from the wall value is diminished to 2 according to Law of the Wall. After decreased the value, mesh quality is getting poorer. Therefore, quantity of the nodes on the surface of the hull is increased and the quality of the mesh is increased. By these implantations, the separation region on the hull is determined. Furthermore, the nominal wake region and vorticity contours taken from shaft region of the hull propeller are obtained. In addition, the pressure coefficient distribution of the stern is taken. Effects of vortex generators, which are implemented on stern of the tanker, are examined in terms of not only ability of controlling flow separation but also hull resistance force components. By considering these two issues, parametric studies are executed to determine the most efficient vortex generator configurations. In Chapter 4.4, four vortex generator parameters which are the distance between vortex generators, angle of attack of vortex generators, profile height of vortex generators and vortex rotation directory of vortex generators are systematically applied on stern of the tanker. The effects of each parameter which are applied are investigated by the aspects of resistance and flow separation. Distance between the vortex generators S are made dimensionless by dividing length of the vortex generators and symbolized as S/L. α is angle of attack of the vortex generators. According to profile height, vortex generators are categorized as conventional and low profile. In respect to vortex rotation directory, vortex generators are classified as co-rotating and counter-rotating vortex generators. Parametric study cases are designated in accordance with derivate rules. In all cases, case names starts with GY letters. Third and fourth characters are represented by S/L distance ratios, fifth and sixth characters represent α, and seventh letter is K for conventional vortex generator cases or D for low profile vortex generators. The last letter is E for co-rotating vortex generator cases or K for counter-rotating vortex generator cases. Bare hull ship case is designated as M1. By these changing vortex generator parameters, 18 vortex generator cases are investigated systematically for finding out the best vortex generator configuration for the hull. Broader information about coding cases and samples are included in Chapter 4.4. S/L takes place as the first changing parameter, which is included in Chapter 4.4.1. In these cases, the other constant parameters are chosen as conventional, co-rotating and 160 angle of attack. In the range of S/L parameters, S/L = 2.5 configured case GY2516KE, which includes 6 number of the vortex generator gives the best results in efficiency. In comparison between GY2516KE and M1, it can be seen that the flow separation is almost vanished and resistance components are decreased by 4.6%. The most inefficient results are taken from GY1016KE and GY3516KE. In the changes of S/L parameter, except GY3516KE case, which includes only three vortex generators, all cases, are successful in controlling flow separations. By the prospection of GY1016KE case which includes 15 number of the vortex generators, it can be seen that vortexes which are generated by the vortex generators reach towards transom stern of the tanker according to the surface pressure coefficient contours and vorticity contours at designated stations. Investigated second parameter α is objected in Chapter 4.4.2. Additional to the 160 angle of attack, 100 and 230 angle of attacks are examined. Three successful S/L ratios in Chapter 4.3.1 are implemented with these two angles of attack. There are 2.5, 2 and 1.5. Examined angles of attack with all these S/L values which are 100 and 230 give no results better than α = 160 angle of attack at any S/L ratios. The reason is figured out that stronger vortexes which are generated by the α = 100 and α = 230 angles of attack vortex generators reach towards transom stern of the tanker. Examined third parameter, which is profile height, is investigated in Chapter 4.3.2. Two kind of vortex generators which are conventional and low profile are exist in terms of profile height in this study. Low profile vortex generator heights are taken 0.2 times boundary layer of the hull (δ). Moreover, the length of the vortex generators are taken same as the conventional vortex generators for making reasonable comparisons with conventional vortex generator cases. Until Chapter 4.3.2, conventional vortex generator applications cases are included. For comparison low profile vortex generators 2.5, 2, 1.5 and 1.0 S/L ratios with 160 angle of attack examined. Along all results of low profile vortex generator cases, GY1016DE with S/L = 1.0 is the most efficient configuration in terms of flow separation control and low hull resistance. The reduction of hull resistance is about 4.21% in configuration GY1016DE. For only S/L = 1.0 low profile vortex generators with 100 angle of attack is tested in GY1010DE case. However, it can be seen that GY1010DE is less efficient than GY1016DE likewise conventional vortex generator cases. In comparison between low profile and conventional vortex generator cases, low profile vortex generators are not successful as conventional vortex generators in terms of separation control on stern of the tanker. Despite of that, low profile vortex generator cases give almost as favorable as conventional vortex generator cases in terms of reduction of hull resistance. This can be occurred due to vortexes which are generated low profile vortex generators are weakened towards transom stern of the tanker unlikely conventional vortex generators. The last parameter that is examined, which is vortex rotation directory of vortex generators, is investigated in Chapter 4.4.4. This parameter leads to two kind of vortex generators. They are co-rotating and counter-rotating vortex generators. Regarding the results of cases until Chapter 4.4.4, the most efficient configurations with conventional profile, S/L = 2.5 and α = 100, 160 are investigated in counter-rotating variations. In contrast of co-rotating vortex generator cases, GY2510KK which has α = 100 gives a result more efficient than GY2516KK which has α = 160. However, it is determined that the two counter-rotating cases does not give better results than S/L = 2.5, angle of attack α = 100 co-rotating conventional GY2516KE case.