Study the Effect of Winglet Height Length on the Aerodynamic Performance of Horizontal Axis Wind Turbines Using Computational Investigation

Abstract

Tip vortices are one of the most critical phenomena facing rotary wings such as propellers and wind turbine blades and lead to changes in the aerodynamic parameters of blades. The winglet (WL) device is considered one of the most significant passive flow control devices. It is used to diminish the strength of vortices at the blade tip, enhance the aerodynamic characteristics of turbine rotor blades, and thereby increase the overall turbine efficiency. The main objective of this research is to improve the aerodynamic characteristics of wind turbines by adding a winglet at the blade tip. An optimum turbine blade profile was taken to build the turbine rotor geometry. The turbine has three blades with a radius of 0.36 m, and the NACA4418 airfoil blade sections were used to build the blade profile. The computational domain was created by ANSYS software, and the model was validated for spalart-allmaras and k-ω SST turbulence models with experimental measurements. The computational model was solved for blade shapes without and with tip winglets. Various winglet height lengths per blade radius (WHLR) of 0.008, 0.02, 0.04, 0.05, 0.06, 0.07, and 0.08 were studied for a 90-degree cant-angle and a constant design tip speed ratio of 4.92. Generally, the results illustrate that the performance characteristics of the turbine rotor were improved by using the tip winglet. The lift-to-drag ratio coefficient (C_L/C_D) and power coefficient (C_p) are increasing with increasing WHLR until they reach the highest improvement value, and then they start to decrease gradually. The optimum WHLR is about 0.042, with a percentage improvement in the lift-to-drag ratio (C_L/C_D) and power coefficient (C_p) related to the blade without winglet of about 11.6% and 6.9%, respectively, and an increase in the thrust force of 14.8%. This is mainly caused by decreasing the vortex strength near the tip region and improving the characteristics of stall behaviors.

1. Introduction

There is widespread agreement that wind is a renewable energy source that will never run out. Instead of using fossil fuels to create power, using wind energy can safeguard the environment and conserve traditional energy resources [1,2,3]. The capability of wind turbines to deflect more kinetic energy coming from the wind has increased significantly over the past 20 years because of swiftly developing technology, particularly the blade’s aerodynamic construction in commercial horizontal axis wind turbines (HAWTs) [4,5]. Like aircraft wings, effects in three dimensions can be seen in wind turbine blades, especially near the blade tip, as presented in Figure 1. The pressure dissimilarity on two sides of the blade, which causes a tip vortex flow, affects the flow at the blade tip. As a result, the lift force decreases and induces drag; additional drag is also generated [6,7].

Most of the recent research has fitted winglets to decrease drag and improve HAWT performance [8]. More research studies have been conducted on the advantages of adding winglets to wind turbine rotors than on the effects of doing so on plane wings [9]. The streams near a wing of the NACA 0012 airfoil section with a teeny winglet were experimentally investigated. Many WL of various cant angles of 0°, 55°, 65°, and 75° were studied with different wind speeds of 20, 30, and 40 m/s, the results show that the WL with a cant angle of 55° has the best aerodynamic performance of the blade [10]. The effect of different cant angles of 0, 30, and 45° and different sweep angles of 0, 20, and 40° on the performance of the wings of NACA2412 cross-sectional airfoils with winglets was investigated, and the findings demonstrate that winglets can enhance lift by about 12%. The winglet’s capacity to reduce drag is roughly 23%. During various phases of flight, the wing with WL can improve the C_L/C_D ratio by about 13% [11]. To increase the power coefficient of HAWT (C_p) without adding stress to the blade, a HAWT with a light-weight WL was investigated; according to the obtained results, the power coefficient might be raised by 2.45% [12]. Two alternate winglet configurations with modified cant angles, winglet spans, and blade shapes are studied, and the findings demonstrate that the best performance was achieved by a 45° cant angle, a 15 cm rectangular WL, and an S809 airfoil, which increased power by up to 9.4% [13]. The blade with a winglet for a small turbine was examined using a free-vortex analysis technique; the calculations indicated that the turbine power would rise by 10% at a tip-speed ratio of 4.7 [14]. An artificial neural network (ANN) was used to study and optimize the numerical model for several winglet designs with various lengths and cant angles. Different winglet configurations were examined, and wind turbines performance parameters such as the power coefficient and thrust force coefficient were estimated. The best improvement in the turbine’s performance was achieved when the cant angle of 48.30° and the WL length of 6.32% from the HAWT rotor radius were used. At these conditions, the power coefficients and thrust coefficients were identical, increasing by about 8.787% [15].

To improve the power of a turbine for both sweeping and straight blades of a HAWT, a parametric study employing computational fluid dynamics (CFD) modeling of a winglet was performed. At both winglet locations (upstream and downstream), the impacts of WL orientation, twist angle, and cant angle were investigated utilizing the computational fluid dynamics code ANSYS Fluent for the desired tip speed ratio. The simulation results indicate that the greatest enhancement in the coefficient of power is about 4.39%. This was accomplished by downstream swept blades with winglets pointing upstream, 40° cant angle, and a 10° twist angle [16]. The impact of WL for HAWT blades on power generation was studied computationally by using a genetic algorithm as the optimization method with the help of an artificial neural network. According to the obtained results, the turbine’s output power was improved by about 9% when compared to a blade without a WL [17]. The aerodynamic characteristics of small HAWT blades with and without winglets were examined using the ANSYS Fluent program, and according to their outcomes, the power coefficient was improved when the length and radius of curvature of the winglet were decreased [18]. The impact of a fusion WL on the power of a HAWT was studied numerically, and the obtained results indicate that inserting winglets into the HAWT blades increased the performance of the turbine by increasing the tip speed ratio [19].

The aerodynamic characteristics of the NACA653218 airfoil section were analyzed using computational investigations [20]. A trailing-edge flap was added to the turbine blades at the outer part of the blade to enhance the aerodynamic characteristics of HAWT; the effect of the flap was examined for various ranges of operation wind speed and different trailing-edge flap deflections using computational investigations [21]. Before constructing the wind farm, the wind conditions and the wake behavior on the site must be evaluated. Generally, knowledge of some problems, like how the wind turbines that operate in a wind farm operate in complex terrain, is still limited. Additional insight into the flow situation occurring around the onshore wind turbines in the presence of different complex terrain types was investigated [22]. A scale model in a 3-D wind-testing chamber was used to display how minor changes in the complex terrain can cause significant variations in the flow behaviors at the turbine height. The obtained results show that these variations not only affect the turbine power generation but also the maintenance and lifetime of the turbines, and hence the feasibility and economy of the wind turbine farms [23]. Accurate analysis of unsteady relations between the wind turbines and distinct boundary layers in the complex terrain is needed to ensure the optimized operation, design, and lifetime of the wind turbines and farms. A laboratory experiment was carried out to discover the interaction between the HAWT wake and boundary layer airflow over the forest-like canopies and the modulation of the forest density in turbulent exchange. All experiments were made in a wind tunnel fully covered with the tree models of height (H/zhub), and particle image velocimetry was used to illustrate the incoming airflow [24].

The novelty in the current work is that no major changes were made to the blade profile of the wind turbine, and the WL body was installed externally at the blade tip. These elements are simple and can be installed on the existing conventional turbine blade profile with minimum cost and maximum benefit of airflow control. According to the above-mentioned literature review, this research aims to study the aerodynamic performance characteristics of HAWT equipped with WL, investigate the potential enhancement of these modification systems on turbine power generation, and discover the possibility of using these systems as a controlling device for turbine power augmentation and load regulation. In this work, the performance of minor HAWT blades with and without WL was investigated. Various WL lengths were studied at a cant angle of 90° for constant rotational speed, a wind speed of 10 m/s, and a tip speed ratio of 4.92. The effect of WL on wind turbine blade characteristics was examined computationally. The power coefficient (C_p), drag coefficient (C_D), lift coefficient (C_L), lift-to-drag coefficient ratio (C_L/C_D), pressure, velocity contours, and velocity vectors were predicted. The percentage improvement in power coefficient (C_p), lift-to-drag coefficient ratio (C_L/C_D) and thrust force (T) were determined and examined; also, the stall behaviors were monitored.

2. Numerical Validation and Verification

Computational fluid dynamics (CFD) is growing into a tool for developing, sustaining, optimizing, coming up with, verifying, and, most importantly, validating techniques. The steady-state solution method was used in this study, and the Navier–Stokes equation was solved to predict the blade’s performance. Generally, the computational simulation involves three main process steps, as shown in Figure 2. In this research, the main objective is to predict the power coefficient (C_p) of a HAWT with and without winglets (WL) of various lengths at a constant cant angle. Validating and verifying previous stages of the code calibration process is considered by [20]. In this research, HAWT blade dimensions with a 0.72 m blade diameter are investigated with a NACA4418 airfoil for validation. Then, the same HAWT blade geometry with and without winglet (WL) was investigated for various lengths at constant cant angles. The blade geometry configuration and the grid generation were generated at the preprocessing stage. The DM ANSYS software constructed the model’s geometry, and the meshing tools in ANSYS software constructed the model’s grid. The second stage involved a computational simulation using the finite volume approach with the FLUENT solver in ANSYS software. The aerodynamic characteristics of the HAWT blade with the NACA 4418 airfoil are studied in the post-processing stage. The power coefficient (C_p), lift-to-drag ratio (C_L/C_D), thrust force (T), pressure contours, velocity vectors, and velocity contours are illustrated using CFD modeling.

2.1. Model Geometry

The computational investigation of the blade of 3D HAWT designs with diameters of 0.72 m utilizing the NACA4418 airfoil section was done for validation and verification with experimental data [21,22,23,24,25]. The rotor blade was constructed of a tapered, twisted NACA4418 airfoil section, and the rotor has three blades with a rated power of 50 W. The blade length is about 0.36 m, and the span is divided into ten sections with different chords and twist angles. The design modular (DM) tool of the ANSYS software program was used to construct the blade geometry, as illustrated in Figure 3 and Figure 4. Figure 3A illustrates the twist angle distribution against the radial direction. Figure 3B shows the top view of the blade to illustrate the blade angle of twist. Figure 4 shows the chord length distribution against the radial direction and the blade shape with a radius of 0.36 m.

2.2. Computational Domain of Solution

It is investigated in 3D HAWT without winglets for verification and validation, as in the paper [21]. A tapered and twisted HAWT with a NACA4418 airfoil section is investigated. Five criteria are set for the design of a small HAWT blade: The design tip speed ratio is 4.92, the power of the turbine is 50 W, the speed of the wind is 10 m/s, the angle of attack (AoA) is 5.5°, and there are 3 blades in total. The distribution of the angle of the relative wind, which is made up of the pitch angle (${\theta }_p$), twist angle (${\theta }_T$) and angle of attack (α), may be determined by applying BEM theory to one of the blade airfoil sections, as illustrated in Figure 5 [21,25]. Figure 5 further illustrates the forces operating on a section of an airfoil, where T_N represents thrust and T_Q represents torque. The forces of lift (L) and drag (D) are responsible for producing both forces. The aerodynamic forces in the airfoil blade section are obtainable as utilities of the angle of attack, Reynolds number (Re), and tip speed ratio, which are termed:

$$\lambda =\frac{\omega \times R}{V}$$

(1)

$$\mathit{Re}=\frac{\rho V_{rel}{C}_{avr}}{\mu }$$

(2)

$$C_p=\frac{P_{out}}{\frac{1}{2}\rho V^3\left(\pi R^2\right)}=\frac{T_Q\times \omega }{\frac{1}{2}\rho V^3\left(\pi R^2\right)}$$

(3)

2.2.1. Domain Geometry

In the computational domain, it is reasonable to omit the tower, hub, and ground when modeling the HAWT blades. The outer domain is shaped similarly to a cylinder, with five multiples of the HAWT radius upstream and ten multiples of the HAWT radius downstream (based on distance from the axial center). The length of the outer domain is ten times the HAWT radius in the downwind direction and five times the HAWT radius length in the upwind direction. All outer domain dimensions of the computational modeling were settled according to similar research on wind turbine studies [8,15,16,17,21,22,25,26]. The problem can be made simpler by taking advantage of the three-blade’s 120-degree periodicity, as shown in Figure 6.

2.2.2. Boundary Conditions

Using FLUENT software, the flow field, pressure, temperature, and velocity in the analyzed CFD model of the NACA 4418 turbine blade are solved using the sector dimensions and boundary conditions listed below, as shown in Figure 7. With an assumed uniform inlet boundary condition, the domain’s wind speed is 10 m/s and its design AoA is 5.5°. The boundary condition for the top surface of the cylinder is supposed to be uniform with a 10 m/s wind velocity. It is assumed that the two side walls have a periodic boundary condition. The extreme surface of the downwind field is taken to have a pressure outlet condition, and the blade element surface is thought to have a no-slip wall condition. The temperature of the free stream is 288.16 K, the equivalent of the ambient temperature. At the specified temperature, the air has a density of 1.225 kg/m³, a pressure of 101,325 pa, and a viscosity of μ = 1.7894 × 10⁻⁵ kg/m. The interior surface is a collection of interior cell faces. It rotates the computational domain in the opposite y direction with the rotational speed of a wind turbine rotor (ω). The rotational speed has been changed, which relates to the tip speed ratios of 2, 3, 4, 5, and 7, where 4.92 is the tip speed ratio for the design case, as stated in Equation (1), to reflect changes in the experimental results.

2.2.3. Grid Dependency Check and Verification of Numerical Model

To accurately obtain the performance of aerodynamics on a 3D blade of a HAWT, grids close to the 3D blade wind turbine volume should be sufficiently dense, and calculated fields should be sufficiently large to achieve an accuracy-to-calculation-time equilibrium. However, using too many fields and grids will use up too much computational power and slow calculation times. A supercomputer, more time, more mesh cells, and a 3D unstructured tetrahedral mesh are needed to resolve the problem. To solve the issues mentioned above, numerous researchers created numerous techniques and tools. From [27], a variety of grid generation techniques for creating high-quality single- and multi-block structural grids for compound shapes are provided. By building a sphere around the 3D blade volume and performing body inflation, the multi-block unstructured grid method is employed in the current study to increase grids close to the blade volume, as shown in Figure 8D. Figure 8A–C illustrates how to face inflation and face scaling is implemented in 3D HAWT blade volume surfaces to speed up calculation and achieve precise wind turbine aerodynamic efficiency.

As a pre-processor and mesh generator, ANSYS FLUENT MESH is used to create the mesh. Hence, for a sphere around a standard 3D blade volume, it is used as a method of body sizing to adjust a small grid scale of the boundary layers (at blade surfaces), and it is increased by a precise growing rate to the largest grid size extreme away from the boundary. The body sizing investigation investigated procedures that had a growth rate of 1.05, the highest element size of 0.2 m, and the lowest element size of 2.9827 × 10⁻⁵ m at blade surfaces. A tetrahedral-hybrid unstructured grid is the type of grid used. Figure 8 displays the meshed control volume of the block surrounding the 3D blade volume. The quality of meshing with a 0.23371 average skewness is demonstrated. That represents excellent value for mesh quality. There are approximately 3,000,000 cells in the whole grid.

In addition to reducing the number of grid cells while maintaining accuracy, grid independence research was also carried out. In general, utilizing more cells makes a numerical answer more precise, but doing so also increases the amount of computer memory and processing time required. By raising the cell numbers until the mesh is sufficiently fine that additional refining has no effect on the results, enough nodes may be obtained. Twelve various mesh types are created to test the results’ independence from the number of cells. A HAWT power coefficient is shown in Figure 9 in relation to the grid number of cells at a speed of wind inlet of 10 m/s, an optimal AoA of 5.5°, and a rotating speed of 136.6667 rad/sec at a 4.92 tip speed ratio for the design case. It requires about 2,600,000 cells to decrease the time of solution without any effect on results. For all these calculations, an Intel Core i7 processor and 16 GB of RAM were used, and the scaled residual was reduced to a value of 10⁻⁶.

A comparable computational model of the 3D HAWT blade shape is created using NACA 4418 airfoil sections with the same grid size and type as previously indicated for validating the computational model with the experiment model investigated with the same boundary conditions [21,25]. It investigated two different turbulence models, Spalart-Allmaras and k-ω SST. The outcomes of the computational model are compared with those of the experimental, numerical, and BEM theory measurements [21,25]. The rotational speed has been changed, which relates to the tip speed ratios of 2, 3, 4, 5, and 7, where 4.92 is the tip speed ratio for the design case, to contest variations in the experimental data. This was done using equation (1). Figure 10 illustrates the findings, which reveal a power coefficient that is in perfect accord with the measurements found in the experimental, numerical, and BEM theories [21]. The numerical model with Spalart-Allmaras and k-ω SST turbulence models has a maximum error of around 10% and 2%, respectively, related to the experimental measurements. Finally, it is illustrated that the computational model with the k-ω SST turbulence model is more suitable for a HAWT [21,25,28,29].

3. Modeling and Simulation of a HAWT Blade with WL

Enhancing the wind turbine blade’s aerodynamic performance is among the most crucial ways to optimize the efficiency of transferring wind kinetic energy into mechanical energy and ultimately electrical energy. In this study, the effect of winglets (WL) with various heights at a constant cant angle of 90° on the aerodynamic performance of a HAWT blade at the design tip speed ratio was investigated. The HAWT blades have a radius of 0.36 m, chord length, and twist angle distributions as shown in Figure 3 and Figure 4. The winglet height length ratio (WHLR) was defined as the winglet height length (WHL) per turbine blade radius (R) and can be determined from Equation (4). The turbine blade with Winglet (WL) with all geometry parameters described was illustrated as shown in Figure 11.

$$\mathrm{W}\mathrm{H}\mathrm{L}\mathrm{R}=\frac{\mathrm{W}\mathrm{i}\mathrm{n}\mathrm{g} \mathrm{l}\mathrm{e}\mathrm{t} \mathrm{h}\mathrm{e}\mathrm{i}\mathrm{g}\mathrm{h}\mathrm{t}}{\mathrm{w}\mathrm{i}\mathrm{n}\mathrm{d} \mathrm{t}\mathrm{u}\mathrm{r}\mathrm{b}\mathrm{i}\mathrm{n}\mathrm{e} \mathrm{r}\mathrm{a}\mathrm{d}\mathrm{i}\mathrm{u}\mathrm{s}}=\frac{\mathrm{W}\mathrm{H}\mathrm{L}}{\mathrm{R}}$$

(4)

The effect of several WHLRs of [0.008, 0.02, 0.04, 0.05, 0.06, 0.07, and 0.08] on the performance of a 3D HAWT blade was investigated at a 10 m/s speed of the wind and a 4.92 tip speed ratio for the design case with a constant rotational speed of 136.6667 rad/s.

Wing Let Height Ratio (WHLR)

As illustrated in Figure 12, the analytic setup consists of seven cases: a 3D HAWT blade with a winglet (WL) at a 90-degree cant angle and seven winglet height length ratios (WHLR) of 0.008, 0.02, 0.04, 0.05, 0.06, 0.07, and 0.08. The 3D HAWT blade has a radius of 0.36 m, and chord length and twist angle distributions are shown in Figure 3 and Figure 4 with the NACA4418 airfoil section. Figure 12 shows the winglet (WL), which has the same airfoil as the HAWT and a sharp edge attachment to the wingtip. The initial conditions are the same in all seven of these cases. The turbine performance of the 3D HAWT with WL is examined with a 10 m/s wind speed and a 4.92 tip speed ratio for the design case with a constant rotational speed of 136.6667 rad/s.

4. Results and Discussions

FLUENT software is applied to solve the velocity and pressure flow fields in the computational model of a 3D HAWT with a NACA 4418 turbine blade airfoil section and with WL. It is illustrated the effect of WHLR variation of [0.008, 0.02, 0.04, 0.05, 0.06, 0.07, and 0.08] with a constant speed of wind of 10 m/s at a constant rotational speed of 136.6667 rad/sec [Tip speed ratio (λ) of 4.92] on a HAWT performance. The temperature of the free stream is 288.16 K, which is identical to the ambient temperature. At the given temperature, the air has a density of 1.225 kg/m³, a pressure of 101,325 pa, and a viscosity of μ = 1.7894×10⁻⁵ kg/m.

4.1. Wind Turbine Blade without WL

4.1.1. Velocity Vectors

The velocity vectors and their magnitudes for HAWT blades without WLs generated by the CFD code are shown in Figure 13, and they are essential for verifying and validating the proposed solution at a 10 m/s speed of wind and a 4.92 tip speed ratio for the design case.

4.1.2. Velocity Contours

For the HAWT blade without WL, Figure 14 depicts the velocity contours of various radial sections at a 10 m/s speed of the wind and a 4.92 tip speed ratio for the design case. In all radial sections of the HAWT blade, it soon becomes clear that the top surfaces have larger velocity values than the bottom surfaces. The velocity differential between the top and bottom surfaces increases from root to tip. The flow separations in the trailing edge of HAWT blade sections are small and decrease from root to tip, which is predicted to decrease drag and increase lift. The velocity variance between the bottom and top surfaces from root to tip is what causes a 3D HAWT blade to produce its power.

4.1.3. Pressure Contours

For the HAWT blades without WLs, Figure 15 illustrates the static pressure contours of various radial sections. In all radial sections of HAWT blades, it becomes apparent that the static pressure values are larger at the bottom surface than at the top surface. The variance in static pressure among the bottom and top surfaces grows from root to tip. This static pressure difference between the top and bottom surfaces from root to tip causes the lift, tangential forces, and increased power of the turbine.

4.2. Wind Turbine Blade WL

4.2.1. Pressure Contours

It is investigated the gauge pressure contours on top and bottom surfaces of the 3D HAWT blades with WL at various WHLRs of 0.008, 0.02, 0.04, and 0.08 at a constant cant angle of 90°, a tip speed ratio of 4.92 for the design case, and a speed of wind of 10 m/s, as illustrated in Figure 16. Figure 16A shows top gauge pressure contours for a 3D HAWT blade with WL at various WHLRs of 0.008, 0.02, 0.04, and 0.08 at a cant angle of 90°. It is illustrated that by increasing WHLR, the pressure on the top surface decreases (the blue area increases on the top surface). From all HAWT blades with WL at various WHLRs, it is shown that the low-pressure region on the root of the HAWT blade is higher than the low-pressure region at the tip of the HAWT blade. The lower pressure lowers from the blade’s root to its tip and along WL to the atmosphere’s pressure, which results in fewer vortices at the wing tip and improves the aerodynamic performance of wind turbines. From the color map in Figure 16A, the low gauge pressure value at the top surface is increasing (static pressure at the top surface is decreasing) with increasing WHLR until a certain value (WHLR = 0.04) and then increasing. It is predicted to improve wind turbine aerodynamic performance by increasing WHLR to a certain value and then reducing it. Figure 16B shows bottom static pressure contours for a 3D HAWT blade with WL at various WHLRs of 0.008, 0.02, 0.04, and 0.08 at a cant angle of 90°. It is illustrated that increasing WHLR that is increasing the pressure on the bottom surface (red area increasing on the bottom surface) until a certain value of WHLR of 0.04 then the pressure in the bottom surface is decreasing (red area decreasing on the bottom surface). It is predicted that wind turbine aerodynamic performance will improve by increasing WHLR to a certain value and then reducing it.

It is investigated the gauge pressure contours at different radial sections of [r/R = 0.93, 0.95, 0.97, and 0.99] for the 3D HAWT blades with and without WL at various WHLR of 0.008, 0.02, 0.04, and 0.08 at a constant cant angle of 90°, a 10 m/s speed of the wind, and a 4.92 tip speed ratio for the design case, as illustrated in Figure 17. When WL length is increased, the suction pressure over the entire top surface of the main airfoil decreases while the positive pressure on the bottom surface increases. This is demonstrated by comparing the gauge pressure contours for a 3D HAWT blade without WL (Figure 16A) and with WL (Figure 17B–E) at the same angle of attack. It is indicated that WL improves the wind turbine’s aerodynamic performance. Figure 17B illustrates the pressure contours at a WHLR of 0.008 at a velocity of 10 m/s. The pressure on the bottom surface increases (the red area increases more than without WL on the bottom surface) and the pressure on the top surface decreases (the blue area increases more than without WL on the top surface) compared to the blade without WL due to an improvement in the total C_L/C_D ratio by decreasing induced drag, which indicates an increase in the C_L/C_D ratio and C_p coefficient of HAWT. Figure 17C,D It is illustrated by the pressure contours at WHLR of 0.02 and 0.04 at a velocity of 10 m/s. The pressure on the bottom surface is still increasing, and the pressure on the top surface is still decreasing, indicating an expected increase in the lift-to-drag ratio and power coefficient of HAWT. Figure 17E is illustrated by the pressure contours at a WHLR of 0.08 at a velocity of 10 m/s. The pressure on the bottom surface is decreasing compared to other HAWT blades with WL (red area decreasing compared to WHLR of 0.008, 0.02, and 0.04 on the bottom surface), and the pressure on the top surface is increasing compared to other HAWT blades with WL (blue area decreasing compared to WHLR of 0.008, 0.02, and 0.04 on the top surface) due to an increase in total drag by increasing in form drag, which is indicated by a decrease in the C_L/C_D ratio and C_p coefficient. For a specific wind velocity, it is concluded that as the WHLR increases, the HAWT aerodynamic performance increases until a specific value of the WHLR is reached, and then the HAWT aerodynamic performance drops.

4.2.2. Velocity Contours

It is investigated the velocity contours at different radial sections of [r/R = 0.93, 0.95, 0.97, and 0.99] for the 3D HAWT blades with WL at WHLR of 0.008, 0.02, 0.04, and 0.08 at a constant cant angle of 90°, a 4.92 tip speed ratio for the design case, and a 10 m/s speed of the wind, as illustrated in Figure 18. It is comparing the velocity contours of a 3D HAWT with WL at various WHLR (Figure 18) against a 3D HAWT blade without WL (Figure 14). From Figure 18A–C, the velocity contours are investigated for the HAWT blade with WL at WHLR of 0.008, 0.02, and 0.04. In all radial sections of the HAWT blade, it is shown that the velocities’ values at the top surface are greater than those at the bottom surface. From root to tip, there is an increase in the velocity variance among the top and bottom surfaces. The flow separations in the trailing edge of blade sections are small and increase imperceptibly from root to tip, which is expected to increase lift and decrease drag. Figure 18D, it is investigated the 3D HAWT blade with WL at a WHLR of 0.08. It is illustrated that the velocity values at the top surface are higher than those at the bottom surface in all radial sections of the HAWT blade. The difference between top and bottom surface velocities decreases from root to tip. The flow separations in the trailing edge of blade sections are increasing from root to tip (they clear at radial sections [r/R = 0.99]) of the blade, which is expected to decrease lift force and increase drag force (due to increasing form drag). For a specific wind velocity, it is concluded that the WHLR increases, the HAWT aerodynamic performance increases until a specific value of the WHLR is reached, and then the HAWT aerodynamic performance drops.

4.2.3. Velocity Vectors

It is investigated velocity vectors at the section front of the 3D HAWT blades without and with WL at various WHLRs of 0.008, 0.02, 0.04, and 0.08 at a constant cant angle of 90-°, a tip speed ratio of 4.92 for the design case, and a speed of wind of 10 m/s, as shown in Figure 19. The vortices at the tip of the HAWT blade without WL are caused by the dissimilar pressure between the top and bottom surfaces. The tip vortices are increasing drag, which is decreasing wind turbine aerodynamic performance. The velocity vectors in the front section of a HAWT blade can readily be used to deduce the rotation sense. Figure 19A represents the velocity vector’s view of flow over what is considered a 3D HAWT without WL. These velocity vectors are concentrated at the blade tip, where trailing vortices occur greatly, which decreases wind turbine aerodynamic performance. Figure 19B–E represents the velocity vector views of flow over the considered 3D HAWT with WL at various WHLRs of 0.008, 0.02, 0.04, and 0.08 at a 90-degree cant angle. The winglet dissipated the tip vortex and was taken away from the HAWT blade. The small tip vortices are established in the blade’s winglet tip. When the WHLR is increasing, the vortices at the tip of the HAWT blade are decreasing, which is expected to improve wind turbine aerodynamic performance.

4.2.4. Lift to Drag Ratio (C_L/C_D)

The C_L/C_D ratio is investigated for the 3D HAWT blades with WL at various WHLRs of 0.008, 0.02, 0.04, 0.05, 0.06, 0.07, and 0.08 at a constant cant angle of 90°, a 4.92 tip speed ratio for the design case, and a 10 m/s speed of the wind, as illustrated in Figure 20. From the previous discussion, it is found that the C_L/C_D ratio increases with increasing WHLR until a certain optimum value of WHLR is reached, at which point the C_L/C_D ratio decreases, as shown in Figure 20A. Figure 20B investigates the improving percentage of a 3D HAWT blade with WL at various WHLRs against a 3D HAWT blade without WL. From Figure 20, it is illustrated that the percentage improving ratios in the C_L/C_D are about 1.2%, 2.3%, 11.5%, 10.4%, 8.5%, 6.48%, and 5% for WHLR of 0.008, 0.02, 0.04, 0.05, 0.06, 0.07, and 0.08, respectively. Finally, at wind speed v = 10 m/s and a cant angle of 90°, the optimum value of WHLR is about 0.042, with an improving C_L/C_D ratio of about 11.6% related to a 3D HAWT blade without WL.

4.2.5. Power Coefficient of HAWT (C_p)

It is investigated that the power coefficient (C_p) of a 3D HAWT blade without and with WL at various WHLRs of 0.008, 0.02, 0.04, 0.05, 0.06, 0.07, and 0.08 at a constant cant angle of 90-°, a tip speed ratio of 4.92 for the design case, and a speed of wind of 10 m/s, as shown in Figure 21. From the previous discussion, it is found that the power coefficient (C_p) increases with increasing WHLR until a certain optimum value of WHLR is reached, and then the power coefficient (C_p) decreases, as shown in Figure 21A. Figure 21B investigates the improving percentage of power coefficient (C_p) for a 3D HAWT blade with WL at various WHLRs compared to a 3D HAWT blade without WL. From Figure 20, it is illustrated that the percentage improving ratios in the power coefficient (C_p) are about 1.18%, 2.27%, 6.6%, 5.54%, 3.4%, 1.14%, −0.2% for WHLR of 0.008, 0.02, 0.04, 0.05, 0.06, 0.07, and 0.08, respectively. Finally, at wind speed v = 10 m/s and a cant angle of 90°, the optimum value of WHLR is about 0.042, with an improvement in power coefficient (C_p) of about 6.9% from a 3D HAWT blade without WL.

4.2.6. Thrust Force (T)

It is unfair to evaluate the performance enhancement brought on by WL on the HAWT blade without considering its impact on the thrust coefficient. A 3D HAWT blade’s thrust on the rotor is increased by the addition of WL, while the pressure distribution at the tip is enhanced. Figure 22 illustrates the increase in thrust brought on using WL, which various authors have confirmed [26]. The thrust force (T) for a 3D HAWT blade with WL at various WHLRs of 0.008, 0.02, 0.04, 0.05, 0.06, 0.07, and 0.08 at a constant cant angle of 90°, a tip speed ratio of 4.92 for the design case, and a speed of wind of 10 m/s is investigated, as shown in Figure 22. It is found that thrust force (T) increases with increasing WHLR, as shown in Figure 22A. Figure 22B investigates the increasing percentage of thrust force (T) for a 3D HAWT blade with WL at various WHLRs compared to the HAWT blade without WL. From Figure 20, it is illustrated that the percentage increasing ratios in the thrust force are about 4.1%, 8.5%, 14.1%, 16.42%, 18.5%, 20.7%, and 22.4% for WHLR of 0.008, 0.02, 0.04, 0.05, 0.06, 0.07, and 0.08, respectively. Finally, at wind speed v = 10 m/s and cant angle of 90°, the optimum WHLR is about 0.042 with improved C_L/C_D ratio and power coefficient (C_p) by about 11.6% and 6.9%, respectively, from the HAWT blade without winglet, but it has the drawback of increasing thrust force by about 14.8%.

5. Conclusions

In this study, a 3D HAWT blade was investigated numerically to predict the effect of WL on the wind turbine’s aerodynamic performance. The computational model was adapted and validated for different models of turbulence and compared with similar BEM results and similar experimental data for wind turbines with the same operating conditions. The aerodynamic performance of HAWT blades with winglets (WL) was investigated for various winglet height length ratios (WHLR), and the results can be summarized as follows:

• Generally, the performance characteristics of turbine rotors were improved with the tip winglet (WL) due to decreasing the vortex strength near the tip region, decreasing the induced drag, and improving the characteristics of stall behaviors.
• The lift-to-drag ratio coefficient (C_L/C_D) and power coefficient (C_p) are increasing with increasing WHLR until they reach the highest improvement value, and then they start to decrease gradually, but the thrust force (T) is increasing with increasing WHLR.
• The percentage improving ratios in the C_L/C_D and the power coefficient (C_p) are about 1.2%, 2.3%, 11.5%, 10.4%, 8.5%, 6.48%, 5%, and (1.18%, 2.27%, 6.6%, 5.54%, 3.4%, 1.14%, −0.2%) for WHLR of 0.008, 0.02, 0.04, 0.05, 0.06, 0.07, and 0.08, respectively.
• The optimum WHLR is about 0.042, with a percentage improvement in the lift-to-drag ratio (C_L/C_D) and power coefficient (C_p) related to the blade without winglet of about 11.6% and 6.9%, respectively, and an increase in the thrust force of 14.8%.
• Finally, we would like to draw attention to the fact that increasing the thrust force (T) is considered a disadvantage of using the WL as a flow control device, but we will take it into account in future works.