© T. Babu et al., Published by EDP Sciences, 2024

1 Introduction

Renewable energy sources such as wind power are increasingly gaining importance in the fight against climate change and the reduction of greenhouse gas emissions. Wind turbines are a critical technology in this regard, as they are designed to extract energy from the wind and convert it into usable electricity. They consist of three primary components: the rotor, the nacelle, and the tower. The rotor is the part of the wind turbine that captures the energy in the wind and converts it into rotational motion. The nacelle houses the gearbox, generator, and other components that convert the rotational motion of the rotor into electricity. The tower supports the nacelle and rotor and holds them at a sufficient height to capture the wind.

The blade is the most critical component of the rotor, as it is responsible for capturing the energy in the wind and converting it into rotational motion. The design of wind turbine blades is essential to maximize energy extraction from the wind while minimizing the costs of maintenance and repair.

Wind turbine blade is responsible for capturing the energy in the wind and converting it into rotational motion.

The design of wind turbine blades is an important consideration in maximizing energy extraction from the wind. The choice of materials and airfoil profiles can significantly impact the performance of the wind turbine, affecting factors such as power output, efficiency, and maintenance costs. In recent years, there has been a growing interest in the use of advanced materials and most effective airfoil profile to improve the performance of wind turbine blades. The importance of determining the best airfoil profile and material of the wind turbine blade is because it covers 15–20% of the total turbine cost [1]. There are a number of researches carried out to find the best airfoil profile and composite material for wind turbine blade. Most recently, CFD analysis was performed and flow properties around NACA 4412 was experimented by Afshari et al. [2]. The air flow characteristics around NACA 2412 airfoil was measured by Rashid et al. [3]. Alfa/wool woven fabrics reinforced epoxy composite for wind turbine blade was prepared by Lamhour, et al. [4]. Performance studies of hybrid natural fiber-reinforced composites for wind turbine blade using Nacha and Sisal was conducted by Miliket et al. [5]. A newly developed typhoon proof Magnus VAWT type turbine was suggested by Babu et al. [6] for countries with frequent natural calamities like Bangladesh. Despite these researches, as there is a lack in research to determine the optimum airfoil shape after comparing widely used NACA airfoils specially among countries of low wind speed like Bangladesh, a research gap has been found, and this research considers three widely used airfoils NACA 0012, NACA 2412, and NACA 4412 to select the best one to harness the optimum wind energy. Besides, although there are a number of researches carried out to find the best composite material for wind turbine blade among E-Glass, S-Glass, Kevlar, Epoxy carbon UD, Carbon fiber [7] etc.; Carbon nanotube (CNT) has not been yet properly considered among those despite its excellent mechanical properties. Hence, this research also aims to compare CNT with those widely used composite materials to determine which one shows the optimal mechanical properties.

1.1 Objectives

In this research, the underlying objectives have been taken into account:

• To investigate the most efficient airfoil profile for horizontal axis wind turbine among three widely used airfoils NACA 0012, NACA 2412, and NACA 4412.

• To find out numerically the best composite material for making wind turbine blades to generate maximum wind power.

1.2 Methodology

To select the proper material and airfoil profile, simulation has been carried out by following steps:

• At first the CFD analysis has been carried out on some widely used airfoil profiles named NACA 0012, NACA 2412, and NACA 4412 with the help of Ansys Fluent software to find out the best airfoil shape. To conduct the simulation, the average wind speed of Bangladesh (8.1 m/s) has been used as one of the boundary conditions. Finally, the values of coefficient of lift (C_l) has been estimated for various angles of attack (α), and analyzing those values of C_l, the best airfoil profile with the optimum angle of attack have been selected. To justify the simulation, the obtained values have been validated with the verified experimental data of wind tunnel carried out by Abbott and Doenhoff from NASA.

• A simple wind turbine has been designed through Ansys design modeler package and then it has been taken into structural analysis through Ansys Static Structural package to measure following physical properties: total deformation, maximum shear stress, von-Mises stress, strain energy for undermentioned composite materials: E-glass, S-Glass, Kevlar, Epoxy carbon UD, Carbon fiber UD, and Carbon nanotube (CNT). After comparing the obtained values of those properties the best composite material has been chosen.

2 NACA and NACA airfoil

The acronym NACA stands for National Advisory Committee for Aeronautics. It was established on March 3, 1915 as a federal organization in the United States to conduct, encourage, and institutionalize aeronautical research. On October 1, 1958, the organization was disbanded, and its resources and personnel were given to the newly established National Aeronautics and Space Administration (NASA) [8].

In addition to other innovations, NACA research and development gave rise to the NACA duct, a kind of air intake utilized in contemporary automotive applications, the NACA cowling, and numerous series of NACA airfoils [9], which are still employed in the manufacture of airplanes. Figure 1 depicts the seal and logo of NACA. There are thousands of airfoil shapes named by a number of institutions; among them, NACA series of airfoil is probably the most popular one. The NACA created a series of thoroughly tested airfoils in the late 1920s and early 1930s and came up with a four-digit number that indicated each airfoil section's important geometric characteristics. A cross-section of an airfoil was included as a supplement to the numbering system by the time Langley Research Center (a unit of NASA) devised this approach in 1929. The full catalog of 96 airfoils then appeared in the NACA's annual report for 1933 [10]. All the airfoil profiles of NACA are presented in Figure 2.

Fig. 1 Seal and Logo of NACA (from left) [9].

Fig. 2 All NACA airfoils [11].

2.1 NACA airfoil used in this study

Although, there are 96 number of NACA airfoils, all are not equally efficient to general high lift force. Nevertheless, every airfoil has some specific usages, not all are equally used in the industry. There are some widely used NACA airfoils in the wind turbine and airplane wing, all of which have the ability to produce high lift force. In this study, three representative airfoils have been chosen from the list of most suitable airfoils for wind turbine blade; i.e., NACA 0012, NACA 2412, and NACA 4412.

3 The computational fluid dynamics (CFD) software used in this study

In the present study the CFD software used are Ansys Workbench R2 (Ansys Fluent, and Ansys Static Structural).

3.1 Equations for Reynolds averaged Navier stokes (RANS)

The time-averaged equations of motion for fluid flow are known as the Reynolds-averaged Navier-Stokes (RANS) equations. They are mainly used while handling turbulent flows. The following are the equations in Cartesian tensor form:

$$\frac{\partial \rho }{\partial t}+\frac{\partial }{\partial x_i}\left(\rho u_i\right)=0$$(1)

$$\frac{\partial }{\partial t}\left(\rho u_i\right)+\frac{\partial }{\partial x_i}\left(\rho u_i{u}_j\right)=-\frac{\partial \rho }{\partial x_i}+\frac{\partial }{\partial x_j}\left[\mu \left(\frac{\partial u_i}{\partial x_i}+\frac{\partial u_j}{\partial x_j}-\frac{2}{3}{\delta }_{ij}\frac{\partial u_i}{\partial x_j}\right)\right]+\frac{\partial }{\partial x_i}\left(-\rho u_i{u}_j\right).$$(2)

Equations (1) and (2) are known as Reynolds-Averaged Navier-Stokes equations.

3.2 Model selection

It is important to think about how to adapt these models to work with wall-bounded flows. If the near-wall mesh resolution is sufficient, the Spalart Allmaras and k-ω models may be applied to the whole boundary layer. The Spalart Allmaras and k-ω models were employed in this simulation investigation. However, the k-ε turbulent model is also a useful model and provides good result, but in this experiment the k-ω model and the Spalart Allmaras model provide more accurate value closer to the experimental results obtained in the NASA's experiments. The basic differences of the Spalart Allmaras and k-ω models are given in Table 1.

The Spalart-Allmaras (SA) turbulence model is given as:

$$\frac{\partial {\displaystyle \tilde{v}}}{\partial t}+u_j\frac{\partial {\displaystyle \tilde{v}}}{\partial x_j}=C_{b}1S{\displaystyle \tilde{v}}+\frac{1}{\sigma }\left[\frac{\partial }{\partial x_j}\left(v+{\displaystyle \tilde{v}}\right)\frac{\partial {\displaystyle \tilde{v}}}{\partial x_j}\right]-C_{w}1f_w{\left(\frac{{\displaystyle \tilde{v}}}{d}\right)}^2+\frac{C_{b}2}{\sigma }\frac{\partial {\displaystyle \tilde{v}}}{\partial x_j}\frac{\partial {\displaystyle \tilde{v}}}{\partial x_j}$$(3)

Here, ṽ is modified turbulent kinematic viscosity, t is time, u_j is components of the velocity vector, x_j is spatial coordinates, S is magnitude of vorticity, v is molecular kinematic viscosity, d is distance to the nearest wall, C_b1, C_b2, C_w1, and σ are model constants, f_w is wall-damping function.

The k-ω turbulence model describes turbulence through two transport equations: one for the turbulent kinetic energy (k) and another for the specific dissipation rate (ω).

The turbulent kinetic energy (k) equation is as follows:

$$\frac{\partial k}{\partial t}+U_j\frac{\partial k}{\partial x_j}=P_k-{\beta }^{*}k\omega +\frac{\partial }{\partial x_j}\left[\left(v+{\sigma }_k{v}_t\right)\frac{\partial k}{\partial x_j}\right]$$(4)

k is turbulent kinetic energy, U_j is mean velocity component in direction, P_k is production of turbulent kinetic energy, β^* is model constant, ω is specific dissipation rate, v is molecular kinematic viscosity, v_t is turbulent viscosity, σ_k is turbulent Prandtl number for k.

Specific dissipation rate (ω) equation is as follows:

$$\alpha \frac{\partial \omega }{\partial t}+U_j\frac{\partial \omega }{\partial x_j}=\alpha \frac{\omega }{k}{P}_k-\beta {\omega }^2+\frac{\partial }{\partial x_j}\left[\left(v+{\sigma }_{\omega }{v}_t\right)\frac{\partial \omega }{\partial x_j}\right]$$(5)

ω is specific dissipation rate, α and β are model constant, σ_ω is turbulent Prandtl number for ω.

3.3 Boundary condition

The boundary conditions are given as follows:

Velocity specification method: Magnitude and Direction

Velocity magnitude: 8.1 m/s

Wall motion: Stationary wall

Area: 1 m²

Density: 1.225 kg/m³

Depth: 1 m

Length: 1 m

Pressure: 0 Pascal

Temperature: 288.16 K

Velocity: 8.1 m/s

Viscosity: 1.7894e-05

Y+ for heat transfer coefficient: 300

Reynold's number: 50,000

3.4 Geometry and modeling

The domain models for the fluid and airfoil have been constructed on the XY plane. The co-ordinates from NACA 2412 (Fig. 3), NACA 4412 (Fig. 4), and NACA 0012 (Fig. 5) airfoils have been chosen from the UIUC Airfoil Coordinate Databases. The chord length of the airfoil is roughly 1 m. The Design Modeler of the Ansys workbench has been used to build the geometry. By drawing a horizontal line across the airfoil and a vertical line at the trailing edge of the airfoil, the domain has been divided into four surfaces to provide the mesh generation more control. All 3 airfoils are shown in a position of 15° of angle of attack, as the measured values of required parameters are found optimum in this angle among −15°, −10°, −5°, 0°, 5°, 10°, 15°, and 20° angle of attack. The domain outline specification around the airfoils is: Length = 10 m, height = 10 m, and diameter = 10 m.

Fig. 3 Geometry of NACA 2412 designed by Ansys Design Modeler.

Fig. 4 Geometry of NACA 4412 designed by Ansys design modeler.

Fig. 5 Geometry of NACA 0012 designed by Ansys design modeler.

3.5 Meshing and boundary condition

Meshing is a method for precisely determining the physical shape of an item by dividing its continuous geometric space into thousands or more discrete shapes. Meshing is a crucial component of engineering simulation. The creation of a superior mesh is needed to produce reliable simulation results. Engineering simulations' precision, convergence, and speed are all governed by the mesh. In this work, Ansys fluent meshing was performed because of its specialized meshing approach for producing more effective, smoother, and better meshes. The meshing performed in this study is triangular type meshing. Figures 6–8 represents the triangular meshing of NACA 2412, NACA 4412, and NACA 0012 airfoils respectively.

Figure 9 illustrates the initialization of the fluid domain by demarking the flow field's peripheral conditions. The fluid domain used in the finite element technique is separated into smaller parts depending on a certain form. Since the triangular mesh may provide more precision than the unstructured mesh, it has been used in the fluid domain. Implementing the element size of 1.e⁻⁰⁰³ m and the preference element of 50 has resulted in a fine mesh with an average skewness of 0.04. The number of nodes and elements for NACA 2412, NACA 4412, and NACA 0012 are presented in Table 2.

Fig. 6 Triangular meshing of NACA 2412 airfoil.

Fig. 7 Triangular meshing of NACA 4412 airfoil.

Fig. 8 Triangular meshing of NACA 0012 airfoil.

Fig. 9 Name selection of the fluid domain.

3.6 Mesh quality

To evaluate the mesh quality, several metrics are used. Jacobian ratio, element quality, aspect ratio, skewness, and orthogonal quality are widely used among them. Table 3 depicts the mesh quality of the three considered NACA profiles according to above metrics.

4 Simulation result

4.1 Simulation result for NACA 2412

As the maximum velocity of wind in Bangladesh was found 8.1 m/s, in this simulation, the value of co-efficient of lift have been calculated in Table 4 at following angle of attacks: −15°, −10°, −5°, 0°, 5°, 10°, 15°, 20°.

From Table 4, and Figure 10, it is clear that the co-efficient of lift (C_l) is increasing with the increase of angle of attack. At 15° angle of attack, this lift coefficient is maximum of 1.625.

Hence, it can be deduced from the above table that the efficient and acceptable value of C_l, and α are as follows:

C_l = 1.625

And α = 15°.

Fig. 10 Coefficient of lift at various angles of attack for NACA 2412.

4.2 Simulation result for NACA 4412

The value of co-efficient of lift for NACA 4412 have been calculated at the same wind velocity in Table 5 at following angle of attacks: −15°, −10°, −5°, 0°, 5°, 10°, 15°, 20°.

From Table 5 and Figure 11, it is clear that the co-efficient of lift (C_l) is increasing with the increase of angle of attack. At 15° angle of attack, this lift coefficient is maximum of 1.654. Hence, it can be deduced from the above table that the efficient and acceptable values of C_l, and α are as follows:

C_l = 1.654

And α = 15°.

Fig. 11 Coefficient of lift at various angles of attack for NACA 4412.

4.3 Simulation result for NACA 0012

The value of co-efficient of lift for NACA 0012 have been calculated at the same wind velocity in the Table 6 at following angle of attacks: −15°, −10°, −5°, 0°, 5°, 10°, 15°, 20°.

From Table 6 and Figure 12, it is clear that the co-efficient of lift (C_l) is increasing with the increase of angle of attack. At 15° angle of attack, this lift coefficient is maximum of 1.445. Hence, it can be deduced from the above table that the efficient and acceptable values of C_l and α are as follows:

C_l = 1.445

And α = 15°.

Fig. 12 Coefficient of lift vs. angle of attack for NACA 0012.

4.4 Comparison among three considered airfoil profiles

Table 7 shows the comparison among three considered airfoils for this simulation. From Table 7, it is clear that the coefficient of lift, C_l is maximum at 15° angle of attack for all three airfoils. The rank of three airfoils according to C_l is NACA 4412>NACA 2412>NACA 0012. As the C_l of NACA 4412 among three airfoils is maximum, it can be said that maximum power will be obtained from this airfoil. Hence, NACA 4412 has been chosen from this simulation.

4.5 Pressure and velocity contour of NACA 2412 airfoil

Pressure and velocity contour of NACA 2412 have been shown below.

4.5.1 Pressure contour of NACA 2412

Figure 13 depicts the close view of pressure contour of the airfoil NACA 2412. The maximum pressure 4.096e+01 Pa works at the lower surface of the leading edge of the airfoil.

Fig. 13 Close view of pressure contour of NACA 2412.

4.5.2 Velocity contour of NACA 2412

Figure 14 depicts the close view of velocity contour of the airfoil NACA 2412. The maximum velocity 2.375e+01 m/s works at the upper surface of the leading edge of the airfoil.

Fig. 14 Close view of velocity contour of NACA 2412.

4.6 Pressure and velocity contour of NACA 4412 airfoil

Pressure and velocity contour of NACA 4412 have been shown below.

4.6.1 Pressure contour of NACA 4412

Figure 15 depicts the close view of pressure contour of the airfoil NACA 4412. The maximum pressure 4.109e+01 Pa works at the lower surface of the leading edge of the airfoil.

Fig. 15 Close view of pressure contour of NACA 4412.

4.6.2 Velocity contour of NACA 4412

Figure 16 depicts the close view of velocity contour of the airfoil NACA 2412. The maximum velocity 2.375e+01 m/s works at the upper surface of the leading edge of the airfoil.

Fig. 16 Close view of velocity contour of NACA 4412.

4.7 Pressure and velocity contour of NACA 0012 airfoil

Pressure and velocity contour of NACA 4412 have been shown below.

4.7.1 Pressure contour of NACA 0012

Figure 17 depicts the close view of pressure contour of the airfoil NACA 0012. The maximum pressure 4.059e+01 Pa works at the lower surface of the leading edge of the airfoil.

Fig. 17 Close view of Pressure contour of NACA 0012.

4.7.2 Velocity contour of NACA 0012

Figure 18 depicts the close view of velocity contour of the airfoil NACA 0012. The maximum velocity 2.333e+01 m/s works at the upper surface of the leading edge of the airfoil.

Fig. 18 Close view of velocity contour of NACA 0012.

5 Validation of the simulation

To its reader the book “Theory of Wing Sections: Including a Summary of Airfoil Data”, authored by Ira H. Abbott, and Albert E. Von Doenhoff has been considered as the bible of airfoil and data [12]. First author Ira Abbott worked as a researcher in Langely Research Center, NASA, and Director of Aeronautical and Space Research, NASA. He also worked as an assistant director of research in NACA headquarters [13]. The second author Albert E. Von Doenhoff was a research engineer in NASA Theory of Wing Sections [12].

Since the experimental values for the turbines of all the NACA airfoil profiles found by those authors of that book were measured and tested in the wind tunnels of laboratories of NASA and published in this book, these values are considered as NASA-given standard values.

However, in Table 8, the values obtained from simulation are compared with the experimental values obtained in NASA's wind tunnel experiment and the percentage of error has been calculated for each value. From the table, it is noticed that the maximum percentage of error 4.92% found for NACA 4412 at 10° angle of attack is less than 5%. Since according to the rule of thumb at least 5% percentage of error is easily acceptable, the values of C_l obtained in this simulation is acceptable.

Figures 19–21 represents the graphs of the simulated values and experimental values of C_l. Curves for the simulated values and experimental values have almost coincided with each other. In Figure 19, for the NACA 2412, the experimental and simulated values are almost the same up to 5 degree angle of attack. From the angle of attack 5 to 15 degree, there is a slight fluctuation of the simulated curve from the experimented curve. From 15 degree to 20 degree, those curves converge again. In Figure 20, the similar trend has been observed for NACA 4412. In Figure 21, for NACA 0012 both curves are almost coinciding due to comparatively lesser percentage of errors.

Fig. 19 Values of Cl of NACA 2412 obtained from simulation and experiment of NASA.

Fig. 20 Values of Cl of NACA 4412 obtained from simulation and experiment of NASA.

Fig. 21 Values of Cl of NACA 0012 obtained from simulation and experiment of NASA.

6 Properties of the materials

The goal of this research is to choose the best material for windmill blades. Fiber-reinforced materials are often used in blades. In this simulation, Kevlar 49 (Aramid), Epoxy Carbon UD, Carbon Nanotube, E-Glass, S-Glass, Carbon Fiber, and Carbon Fiber are used. The properties of the utilized materials are listed in Table 9.

6.1 Structural analysis using Ansys

The performance parameters of the blade structure were calculated using Ansys Workbench's Static Structural on the imported wind turbine blade. E-glass, S-glass, carbon fiber, carbon nanotube, kevlar 49 (aramid), and epoxy carbon UD were the materials used.

6.2 Geometry and modeling

Since among three airfoils, NACA 4412 has been proved as the best airfoil profile to build the wind turbine blade in the given boundary condition, this profile has been chosen to design the geometry of our imaginary simple wind turbine with the help of design modeler of Ansys. The length of the wind turbine is 3 m. The chord length of the airfoil at the root of the wind turbine is 1 m, the chord length of the airfoil at the middle of the turbine is 0.5 m, and the chord length at the tip of the airfoil is 0.3 m. The twist angle of the tip from the root is 25 degree. The airfoil NACA 4412 is uniformly maintained over the whole turbine. Figure 22 demonstrates the left hand side view and front view of the designed simple wind turbine.

Fig. 22 L.H.S view (up), Front view (down) of the designed simple wind turbine using airfoil NACA 4412.

6.3 Meshing

The meshing performed in this study is C type meshing. After entering all engineering data and importing the geometry, meshing in Ansys Static Structural has been done, as depicted in Figure 23.

Fig. 23 Meshing on the modeled wind turbine.

6.4 Mesh Quality

To evaluate the mesh quality, metrics mentioned in Table 10 are used.

6.5 Boundary condition

The same boundary condition was applied for all six materials. These are given below:

• Standard Earth Gravity: X component = 0 m/s. Y component = -9.8 m/s, and Z component = 0 m/s

• Rotation velocity of the turbine blade = 2.09 rad/s (assumed rpm = 20)

• Pressure applied by surface effect = 20 Pa

• Fixed support = root of the turbine blade

7 Results and analysis

Finally, the simulation process has been done and maximum and minimum values for total deformation, equivalent stress, maximum shear stress, and strain energy have been measured. The simulation results are given below:

7.1 Simulation result for E-Glass

Figures 24–27 depict the total deformation, strain energy, maximum shear stress, and equivalent (von-Mises) stress of E-Glass, respectively. According to Figure 24, the maximum and minimum total deformation are 0.00048572 m and 0 m, respectively. The maximum total deformation occurs on upper surface of the tip side of the blade, which is significantly smaller than the minimum total deformation that occurs on the upper surface of the root. The gradual distribution of the deformation from the minimum to the maximum value seems to be in bands of various lengths. In Figure 25, the maximum strain energy, exhibited narrowly above the surface of the root, is 3.3192E-05 J, and the minimum strain energy, exhibited along almost the whole of the blade, is 4.4686E-15 J. According to Figure 26, the maximum and minimum values of the maximum shear stress are 9.0796E+05 Pa and 15.981 Pa, respectively. The maximum value of the maximum shear stress works on the surface near the root, and the minimum value of the maximum shear stress works on the surface of the periphery and near the tip of the blade. Figure 27 shows the equivalent (von-Mises) stress as having a maximum value of 1.762E+06 Pa and a minimum value of 28.998 Pa.

Fig. 24 Total deformation of E-glass.

Fig. 25 Strain energy of E-glass.

Fig. 26 Maximum shear stress of E-glass.

Fig. 27 Equivalent (Von-Mises) stress of E-glass.

7.2 Simulation result for S-Glass

Figures 28–31 depict the total deformation, strain energy, maximum shear stress, and equivalent (von-Mises) stress of S-Glass, respectively. According to Figure 28, the maximum and minimum total deformation are 0.00037882 m and 0 m, respectively. The maximum total deformation occurs on upper surface of the tip side of the blade, which is significantly smaller than the minimum total deformation that occurs on the upper surface of the root. The gradual distribution of the deformation from the minimum to the maximum value seems to be in bands of various lengths. In Figure 29, the maximum strain energy, exhibited narrowly above the surface of the root, is 2.4891E-05 J, and the minimum strain energy, exhibited along almost the whole of the blade, is 3.3208E-15 J. According to Figure 30, the maximum and minimum values of the maximum shear stress are 8.7304E+05 Pa and 15.401 Pa, respectively. The maximum value of the maximum shear stress works on the surface near the root, and the minimum value of the maximum shear stress works on the surface of the periphery and near the upper surface of the tip of the blade. Figure 31 shows the equivalent (von-Mises) stress as having a maximum value of 1.6942E+06 Pa and a minimum value of 27.817 Pa.

Fig. 28 Total deformation of S-Glass.

Fig. 29 Strain energy of S-Glass.

Fig. 30 Maximum shear stress of S-Glass.

Fig. 31 Equivalent (von-Mises) stress of S-Glass.

7.3 Simulation result for Kevlar 49 (Aramid)

Figures 32–35 depict the total deformation, strain energy, maximum shear stress, and equivalent (von-Mises) stress of Kevlar 49, respectively. According to Figure 32, the maximum and minimum total deformation are 0.00017197 m and 0 m, respectively. The maximum total deformation occurs on upper surface of the tip side of the blade, which is significantly smaller than the minimum total deformation that occurs on the upper surface of the root. The gradual distribution of the deformation from the minimum to the maximum value seems to be in bands of various lengths. In Figure 33, the maximum strain energy, exhibited narrowly on the surface of the root, is 6.383E-06 J, and the minimum strain energy, exhibited along almost the whole of the blade, is 9.1561E-16 J. According to Figure 34, the maximum and minimum values of the maximum shear stress are 5.6518E+05 Pa and 10.758 Pa, respectively. The maximum value of the maximum shear stress works on the surface near the root, and the minimum value of the maximum shear stress works on the surface of the periphery and near the tip of the blade. Figure 35 shows the equivalent (von-Mises) stress as having a maximum value of 1.0253E+06 Pa and a minimum value of 19.34 Pa.

Fig. 32 Total deformation of Kevlar-49.

Fig. 33 Strain energy of Kevlar-49.

Fig. 34 Maximum shear stress of Kevlar-49.

Fig. 35 Equivalent (von-Mises) Stress of Kevlar-49.

7.4 Simulation result for Epoxy Carbon UD (395 GPa) Prepreg

Figures 36–39 depict the total deformation, strain energy, maximum shear stress, and equivalent (von-Mises) stress of Epoxy Carbon UD (395 GPa) Prepreg., respectively. According to Figure 36, the maximum and minimum total deformation are 0.00222 m and 0 m, respectively. The maximum total deformation occurs on upper surface of the tip side of the blade, which is significantly smaller than the minimum total deformation that occurs on the upper surface of the root. The gradual distribution of the deformation from the minimum to the maximum value seems to be in bands of various lengths. In Figure 37, the maximum strain energy, exhibited narrowly on the surface of the root, is 7.941E-05 J, and the minimum strain energy, exhibited along almost the whole of the blade, is 2.1914E-14 J. According to Figure 38, the maximum and minimum values of the maximum shear stress are 5.0751E+05 Pa and 7.9404 Pa, respectively. The maximum value of the maximum shear stress works on the surface near the root, and the minimum value of the maximum shear stress works on the surface of the periphery and near the tip of the blade. Figure 39 shows the equivalent (von-Mises) stress as having a maximum value of 9.7182E+05 Pa and a minimum value of 14.604 Pa.

Fig. 36 Total deformation of epoxy carbon UD (395 GPa).

Fig. 37 Strain energy of epoxy carbon UD (395 GPa).

Fig. 38 Maximum shear stress of epoxy carbon UD (395 GPa).

Fig. 39 Equivalent (von-Mises) stress of epoxy carbon UD (395 GPa).

7.5 Simulation result for Carbon Fiber (395 GPa)

Figures 40–43 depict the total deformation, strain energy, maximum shear stress, and equivalent (von-Mises) stress of Carbon Fiber (395 GPa), respectively. According to Figure 40, the maximum and minimum total deformation are 0.0040844 m and 0 m, respectively. The maximum total deformation occurs on upper surface of the tip side of the blade, which is significantly smaller than the minimum total deformation that occurs on the upper surface of the root. The gradual distribution of the deformation from the minimum to the maximum value seems to be in bands of various lengths. In Figure 41, the maximum strain energy, exhibited narrowly on the surface of the root, is 1.7453E-04 J, and the minimum strain energy, exhibited along almost the whole of the blade, is 1.8608E-14 J. According to Figure 42, the maximum and minimum values of the maximum shear stress are 6.0174E+05 Pa and 4.9317 Pa, respectively. The maximum value of the maximum shear stress works on the surface near the root, and the minimum value of the maximum shear stress works on the surface of the periphery and near the upper surface of the tip of the blade. Figure 43 shows the equivalent (von-Mises) stress as having a maximum value of 1.1869E+06 Pa and a minimum value of 8.5419 Pa.

Fig. 40 Total deformation of carbon fiber (395 GPa).

Fig. 41 Strain energy of carbon fiber (395 GPa).

Fig. 42 Maximum shear stress of carbon fiber (395 GPa).

Fig. 43 Equivalent (von-Mises) stress of carbon fiber (395 GPa).

7.6 Simulation result for carbon nanotube

Figures 44–47 depict the total deformation, strain energy, maximum shear stress, and equivalent (von-Mises) stress of Carbon Nanotube, respectively. According to Figure 44, the maximum and minimum total deformation are 2.1608E-5 m and 0 m, respectively. The maximum total deformation occurs on upper surface of the tip side of the blade, which is significantly smaller than the minimum total deformation that occurs on the upper surface of the root. The gradual distribution of the deformation from the minimum to the maximum value seems to be in bands of various lengths. In Figure 45, the maximum strain energy, exhibited narrowly on the surface of the root, is 9.127E-07 J, and the minimum strain energy, exhibited along almost the whole of the blade, is 1.24E-16 J. According to Figure 46, the maximum and minimum values of the maximum shear stress are 5.9528E+05 Pa and 5.352 Pa, respectively. The maximum value of the maximum shear stress works on the surface near the root, and the minimum value of the maximum shear stress works on the surface of the periphery and near the upper surface of the tip of the blade. Figure 47 shows the equivalent (von-Mises) stress as having a maximum value of 1.1028E+06 Pa and a minimum value of 10.307 Pa.

Fig. 44 Total deformation of carbon nanotube.

Fig. 45 Strain energy of carbon nanotube.

Fig. 46 Maximum shear stress of carbon nanotube.

Fig. 47 Equivalent (von-Mises) stress of carbon nanotube.

7.7 Tabular and graphical summary of the simulations

To select the most suitable material for wind turbine blade, the above simulation results have been summarized in Table 11: The tabular values are also graphically represented in Figure 48 for better visualization. From Table 11 and Figure 48, it is clear that:

• Carbon nanotube exhibits the least amount of total deformation (2.1608×10⁻⁵m) among the considered composites. With a value of 0.00017197 m, Kevlar 49 shows the second-lowest total deformation.

• In accordance with the maximum shear stress, the Epoxy carbon with a value of 5.0751E×10°⁵ Pa replaces the carbon nanotube from the first lowest position to third lowest (5.9528×10°⁵ Pa). Kevlar 49 still remains the second lowest with a value of 5.6518×10°⁵ Pa.

• In case of maximum von-Mises stress or maximum equivalent stress, the maximum value of the equivalent stress is the lowest for Epoxy carbon (9.7182×10°⁵Pa), which indicates its most suitability in this parameter. Kevlar 49, having value of 1.02553×10°⁶ Pa, and Carbon Nanotube, having value of 1.1028×10°⁶ Pa, remain the second lowest and third lowest respectively.

• According to the maximum strain energy, again Carbon nanotube shows the least strain energy with a value of 9.127×10⁻⁷J. Kevlar 49 shows the second least value of 3.1061×10⁻⁵J.

From Table 11, and Figure 48 it is clear that in case of total deformation, and strain energy, carbon nanotube dominates over other composites by a wide margin. On the other hand, in case of maximum shear stress and von-Mises stress, since epoxy carbon exhibits the lowest value, it is comparatively more suitable. However, the difference of the value of maximum shear stress, and von-Mises stress for epoxy carbon, Kevlar 49, and carbon nanotube is very low, hence negligible. It is interestingly noticed that for all the parameters, Kevlar 49 exhibits the second lowest value. That is why, it can be decided that due to the significantly lowest value of total deformation and strain energy, carbon nanotube reinforced polymer is the most suitable composite material for wind turbine blade. However, because of being a novel material, carbon nanotube is comparatively expensive than other composites. Hence, it may increase the installation cost of a wind farm, but due to better mechanical and physical properties, its longevity is more. However, cases where high installation cost is a serious issue, Kevlar 49 can easily be the second option due to its good mechanical and physical properties closer to Carbon nanotube.

Fig. 48 Graphical representations of ANSYS results for several reinforcements and a composite: (a) maximum total deformation, (b) maximum equivalent stress, (c) maximum shear stress, (d) maximum strain energy.

Fig. 48 Continued.

8 Conclusion

In fine, from this research, it can simply be concluded that among several renewable energy options, wind energy is the most pivotal and prospective for the near future from the perspective of Bangladesh. Considering the wind speed scenario of Bangladesh, an investigation has been conducted to find the best airfoil profile and best composite material for a horizontal axis wind turbine. Eventually, following findings have been found:

• Through the ANSYS Workbench program, after simulating three widely used airfoils NACA 2412, NACA 4412, and NACA 0012, the value of coefficient of lift, C_l = 1.654 at 15° angle of attack (α) has been found as maximum for NACA 4412 among three airfoils, it can be considered as the best airfoil profile obtained in this research.

• Also the result obtained through the simulation has been validated with the value obtained by Abbott and Doenhoff from NASA and the percentage of error estimated is 1.47%.

• A wind turbine has been designed with the help of design modeler and its static structural analysis has also been carried out through ANSYS. From the simulation result of ANSYS, since Carbon nanotube exhibits lowest total deformation and strain energy of 2.1608E-05 m, and 9.127E-07 J, respectively by a wide margin among considered five reinforcement materials and a composite (E-Glass, S-Glass, Kevlar 49, Carbon fiber UD, and Carbon nanotube, Epoxy carbon UD), Carbon nanotube is recommended as the most suitable reinforcement material for wind turbine blade production.

• However, as Kevlar 49 exhibits the second best physical properties after Carbon nanotube, and it is cheaper than Carbon nanotube, Carbon nanotube reinforced polymer also can be used where installation cost is an issue.

In this research, only reinforcement and composite materials were chosen for consideration thanks to their longevity, better physical and mechanical properties and most importantly biodegradability.

8.1 Limitation and recommendation

The main limitation of this research is that the values of total deformation, maximum shear stress, equivalent (von-Mises) stress, and strain energy obtained from the simulations are not experimentally validated. Hence, it is recommended to conduct another research to validate these values experimentally. It has a scope to find out the percentage of error, and thusly accuracy of the simulations through ANSYS Static Structural software will be measured.

Funding

N/A.

Conflicts of interest

On behalf of all authors, the corresponding author states that there is no conflict of interest.

Data availability statement

This article has no associated data.

Author contribution statement

Tasruzzaman Babu: Conducted the research, performed data analysis, and drafted the manuscript. Dr. Arefin Kowser: Supervised the research, provided conceptual guidance, and contributed to manuscript revision. Dr. A.N.M. Mominul Islam Mukut: Provided additional revisions and addressed reviewers' comments to improve the manuscript quality.

References

All Tables

All Figures

	Fig. 1 Seal and Logo of NACA (from left) [9].
In the text

	Fig. 2 All NACA airfoils [11].
In the text

	Fig. 3 Geometry of NACA 2412 designed by Ansys Design Modeler.
In the text

	Fig. 4 Geometry of NACA 4412 designed by Ansys design modeler.
In the text

	Fig. 5 Geometry of NACA 0012 designed by Ansys design modeler.
In the text

	Fig. 6 Triangular meshing of NACA 2412 airfoil.
In the text

	Fig. 7 Triangular meshing of NACA 4412 airfoil.
In the text

	Fig. 8 Triangular meshing of NACA 0012 airfoil.
In the text

	Fig. 9 Name selection of the fluid domain.
In the text

	Fig. 10 Coefficient of lift at various angles of attack for NACA 2412.
In the text

	Fig. 11 Coefficient of lift at various angles of attack for NACA 4412.
In the text

	Fig. 12 Coefficient of lift vs. angle of attack for NACA 0012.
In the text

	Fig. 13 Close view of pressure contour of NACA 2412.
In the text

	Fig. 14 Close view of velocity contour of NACA 2412.
In the text

	Fig. 15 Close view of pressure contour of NACA 4412.
In the text

	Fig. 16 Close view of velocity contour of NACA 4412.
In the text

	Fig. 17 Close view of Pressure contour of NACA 0012.
In the text

	Fig. 18 Close view of velocity contour of NACA 0012.
In the text

	Fig. 19 Values of Cl of NACA 2412 obtained from simulation and experiment of NASA.
In the text

	Fig. 20 Values of Cl of NACA 4412 obtained from simulation and experiment of NASA.
In the text

	Fig. 21 Values of Cl of NACA 0012 obtained from simulation and experiment of NASA.
In the text

	Fig. 22 L.H.S view (up), Front view (down) of the designed simple wind turbine using airfoil NACA 4412.
In the text

	Fig. 23 Meshing on the modeled wind turbine.
In the text

	Fig. 24 Total deformation of E-glass.
In the text

	Fig. 25 Strain energy of E-glass.
In the text

	Fig. 26 Maximum shear stress of E-glass.
In the text

	Fig. 27 Equivalent (Von-Mises) stress of E-glass.
In the text

	Fig. 28 Total deformation of S-Glass.
In the text

	Fig. 29 Strain energy of S-Glass.
In the text

	Fig. 30 Maximum shear stress of S-Glass.
In the text

	Fig. 31 Equivalent (von-Mises) stress of S-Glass.
In the text

	Fig. 32 Total deformation of Kevlar-49.
In the text

	Fig. 33 Strain energy of Kevlar-49.
In the text

	Fig. 34 Maximum shear stress of Kevlar-49.
In the text

	Fig. 35 Equivalent (von-Mises) Stress of Kevlar-49.
In the text

	Fig. 36 Total deformation of epoxy carbon UD (395 GPa).
In the text

	Fig. 37 Strain energy of epoxy carbon UD (395 GPa).
In the text

	Fig. 38 Maximum shear stress of epoxy carbon UD (395 GPa).
In the text

	Fig. 39 Equivalent (von-Mises) stress of epoxy carbon UD (395 GPa).
In the text

	Fig. 40 Total deformation of carbon fiber (395 GPa).
In the text

	Fig. 41 Strain energy of carbon fiber (395 GPa).
In the text

	Fig. 42 Maximum shear stress of carbon fiber (395 GPa).
In the text

	Fig. 43 Equivalent (von-Mises) stress of carbon fiber (395 GPa).
In the text

	Fig. 44 Total deformation of carbon nanotube.
In the text

	Fig. 45 Strain energy of carbon nanotube.
In the text

	Fig. 46 Maximum shear stress of carbon nanotube.
In the text

	Fig. 47 Equivalent (von-Mises) stress of carbon nanotube.
In the text

	Fig. 48 Graphical representations of ANSYS results for several reinforcements and a composite: (a) maximum total deformation, (b) maximum equivalent stress, (c) maximum shear stress, (d) maximum strain energy.
In the text

	Fig. 48 Continued.
In the text