Tech Science Press

DOI: 10.32604/fdmp.2024.046158

ARTICLE

Optimized Design of Bio-Inspired Wind Turbine Blades

Yuanjun Dai1,4,*, Dong Wang1, Xiongfei Liu2 and Weimin Wu3
1
School of Mechanical Engineering, Shanghai DianJi University, Shanghai, 201306, China
2
School of Information Engineering, Yinchuan University of Science and Technology, Yinchuan, 750000, China
3
School of Mechanical and Power Engineering, Chongqing University of Science and Technology, Chongqing, 401331, China
4
School of Aeronautical Mechanical and Electrical Engineering, Chongqing Aerospace Polytechnic, Chongqing, 400021, China
*
Corresponding Author: Yuanjun Dai. Email: daiyj@sdju.edu.cn
Received: 20 September 2023 Accepted: 22 February 2024 Published: 23 July 2024

ED
ABSTRACT
To enhance the aerodynamic performance of wind turbine blades, this study proposes the adoption of a bionic
airfoil inspired by the aerodynamic shape of an eagle. Based on the blade element theory, a non-uniform extrac-
CT
tion method of blade elements is employed for the optimization design of the considered wind turbine blades.
Moreover, Computational Fluid Dynamics (CFD) is used to determine the aerodynamic performances of the eagle
airfoil and a NACA2412 airfoil, thereby demonstrating the superior aerodynamic performance of the former.
Finally, a mathematical model for optimizing the design of wind turbine blades is introduced and a comparative
analysis is conducted with respect to the aerodynamic performances of blades designed using a uniform extraction
A
approach. It is found that the blades designed using non-uniform extraction exhibit better aerodynamic
performance.
TR

KEYWORDS
Airfoil; wind turbines; blade design; CFD
RE

1 Introduction
Wind energy, as a green, pollution-free, and renewable energy source, has gradually become a focus of
attention [1,2]. Wind turbine blades are crucial components for converting wind energy into mechanical
energy, and their shape design significantly impacts the efficiency of the wind power generation system.
Therefore, efficient capture and utilization of wind energy to improve energy conversion efficiency are
the key points in wind turbine blade design [3–5]. The design of airfoil and blade design methods for
wind turbines are crucial for enhancing aerodynamic performance.
In the aspect of airfoil research, Rose et al. [6] introduced knowledge and technology from biomimetics
and conducted numerical studies on the tubercles inspired by the humpback whale (HW) using a straight
wing and a swept-back wing configuration. Metin et al. [7], inspired by biological systems in nature,
increased aerodynamic performance by adding small wings and annular blades at the wing tip. Yuan et al.
[8] proposed improving the aerodynamic performance of wind turbine blades by refining the trailing edge
of the airfoil. Wang et al. [9] proposed a multi-objective optimization model that combines maximizing
the power coefficient and minimizing the surface area of the blade for variable pitch wind turbines using

This work is licensed under a Creative Commons Attribution 4.0 International License, which
permits unrestricted use, distribution, and reproduction in any medium, provided the original
work is properly cited.
1648 FDMP, 2024, vol.20, no.7

a novel airfoil. Wang et al. [10] have established a wind turbine airfoil model based on the Generalized
Regression Neural Network (GRNN) and proposed an optimization design method to improve the
aerodynamic performance of the airfoil under multiple constraints. Yan et al. [11], focusing on sturgeons,
have conducted numerical simulation studies by employing 3D reverse engineering and B-spline curve
fitting techniques to establish physical models of both asymmetric and symmetric biomimetic airfoils.
Zhao et al. [12] have designed a novel serrated airfoil inspired by owl wings, enhancing the overall
performance of the airfoil. Hua et al. [13] and Lanzafame et al. [14] have also conducted research on
airfoil designs. In research on optimization design methods for wind turbine blades, Lee et al. [15]
examined various design parameters and utilized the second-order response surface method to establish
the correlation between the objective function and factors such as chord length and torsion angle in the
design calculations. They compared the performance of the optimized blade with the original design and
found significant improvements. Rodriguez et al. [16] proposed an integrated optimization methodology
for wind turbine blade design by combining computational fluid dynamics (CFD), blade element
momentum theory (BEM), and genetic algorithms (GA). Özkan et al. [17] have developed a novel

ED
optimization technique called the Artificial Bee Colony algorithm based on Blade Element Momentum
theory (ABC-BEM), which has been applied to designing small-scale wind turbine blades. A, Zhu et al.
[18] have proposed a multi-objective optimization design method for wind turbine systems based on a
tower-blade coupling model, aiming to explore the coupling effect between the tower and blades to
improve the performance of wind turbine systems. Zhu et al. [19] proposed a Matlab-based program
CT
combined with Ansys to achieve an optimal trade-off between maximum power generation and minimum
blade mass in wind turbine design. Zemamou et al. [20] investigated the optimization of traditional
Savonius blade design using Bézier curves and transient computational fluid dynamics (CFD)
simulations. Zhang et al. [21], Maki et al. [22], Tang et al. [23], among others, have also conducted
A
research on methods for wind turbine blade design.
In-depth research has been conducted on the optimization design of wind turbine blades. However, these
TR

studies predominantly focus on either airfoil or blade design methods, with limited research on integrating
both. This study applies a bionic airfoil design inspired by an eagle, implementing a non-uniform extraction
method of blade elements based on the blade element theory for wind turbine blade optimization.
2 Extraction and Aerodynamic Performance Analysis of the Bionic Airfoil
RE

2.1 Extraction of Biomimetic Airfoil

In nature, birds are the most adept at flying, with their wing shapes evolved through natural processes to
achieve outstanding flight capabilities. The wings of the sparrowhawk are formed by a lateral arrangement of
airfoil shapes. Through fluttering and clever utilization of aerodynamic principles, they generate upward lift
and thrust to propel the body forward, resembling the operating conditions of wind turbines. This study
employs a portable 3D scanner to scan the sparrowhawk airfoils and acquire a point cloud image. The
resulting scan data is processed using image processing software called ImageWare, as illustrated in
Fig. 1. To capture the optimal cross-sectional airfoil, multiple interceptions and selections of airfoils were
performed. Firstly, cross-section 1 was selected as the reference, and relevant parameters were set. The
sparrowhawk airfoil model was divided equally at specified intervals, resulting in a total of 60 intercepted
airfoils. Through a comparative analysis, the airfoil located 72 mm from cross-section 1 was identified as
the most suitable. Subsequently, cross-section 2 was created for further optimization, with corresponding
parameters set and the reference position placed at −72.0 mm. Twenty airfoils were intercepted in this
process, and after comparison, the airfoil positioned −73.25 mm from cross-section 1 was found to better
meet the conditions. Moving forward, cross-section 3 was created to refine the selection. With predefined
parameters, the cross-section position was set at −73.25 mm, and 20 airfoils with a spacing of 0.1 mm
were intercepted. After comparison, the airfoil at −73.05 mm from cross-section 1 was determined as the
FDMP, 2024, vol.20, no.7 1649

optimal choice, as shown in Fig. 2. The coordinates of the sparrowhawk airfoil were exported from
Geomagic Studio to Profili software for the establishment of the biomimetic airfoil model, as illustrated
in Fig. 3.

ED
Figure 1: Sparrowhawk airfoil scanning process

CT
A
TR

Figure 2: Extraction process of sparrowhawk airfoil type

RE

Figure 3: Sparrowhawk airfoil type QY-7305

Lastly, the final selected sparrowhawk airfoil section was derived, and the collection and fitting of
coordinate points were carried out. Using Matlab software, the upper and lower airfoil profiles were fitted
to obtain the coordinate equation of the airfoil. The resulting equation is presented below:
Upper wing type:
Y1 ¼ 0:0006x3  0:0367x2 þ 1:1094x þ 1:1985 (1)
Lower wing type:
Y2 ¼ 0:011x3 þ 0:045x2  0:3793x  1:3623 (2)
The airfoil is named QY-7305, where “QY” represents the initial capitalization of the Chinese name of
the Sparrowhawk and “7305” indicates the cross section at −73.05 mm with the characteristic plane as the
reference.
1650 FDMP, 2024, vol.20, no.7

2.2 Numerical Computation and Analysis of Airfoil

2.2.1 Modeling and Mesh Generation
Mesh generation of the airfoil was conducted using the ICEM-CFD software, employing a structured
grid. The computational domain was defined as a circular region with the leading edge of the airfoil as
the center and a radius equal to 10 times the airfoil chord length. The first layer height of the mesh on the
airfoil surface was set to 2.2 × 10−5, with a boundary layer consisting of 10 layers. The wall mesh growth
rate was set at 1.1. These conditions were maintained for mesh independent verification, ensuring that
variations in cell size did not impact the accuracy of the computational results. Tables 1 and 2 present the
influence of different mesh cell numbers on the lift coefficient for the QY-7305 and NACA2412 airfoils
at an angle of attack of 6°. It is noted that all cases listed in the tables satisfied the condition of y+ being
less than 1.

Table 1: Mesh-independent verification (QY-7305)

ED
No. Number of grids Lift coefficient Correlation error %
1 63942 1.5133 1.836
2 78635 1.5369 0.305
3 87985 1.5389 0.175
4 95960
CT 1.5408 0.05
5 11658 1.5416 0
A
Table 2: Mesh-independent verification (NACA2412)
TR

No. Number of grids Lift coefficient Correlation error %

1 59689 0.7058 2.16
2 73659 0.7125 1.23
3 85460 0.7189 0.346
RE

4 93567 0.7201 0.18

5 10563 0.7214 0

The total number of grids for the selected QY-7305 and NACA2412 airfoils in this study were
95960 and 93567, respectively. The mesh distribution of the QY-7305 airfoil is shown in Fig. 4, and the
mesh distribution of the NACA2412 airfoil is shown in Fig. 5.

Figure 4: QY-7305 airfoil overall grid and local grid diagram

FDMP, 2024, vol.20, no.7 1651

Figure 5: NACA2412 airfoil overall grid and local grid diagram

2.2.2 Numerical Method

The study of the aerodynamic characteristics of wind turbine airfoils is essentially an investigation into
the impact of turbulent flow on the aerodynamic properties of wind turbine airfoils. Turbulence is a complex
and highly nonlinear flow, requiring a comprehensive understanding of numerical methods for turbulence
computation to better simulate and predict various performance aspects of wind turbine airfoils.

ED
Considering the balance between computational accuracy and resource requirements, this paper adopts a
method based on the Reynolds-Averaged Navier-Stokes approach. The flow velocity studied in this paper
is lower than the Mach number of 0.3. Considering the influence of viscous forces, it is simulated as a
steady-state incompressible flow.
Continuity equation:
CT
@Ui
¼0 (3)
@xi
Momentum equation:
A
@Ui Uj 1 @P @Ui @U 0 i U 0 j
¼ þm  (4)
TR

@xi q @xi @xi xj @xj

Reynolds stress tensor equation:


0 0 @Ui @Uj 2
U i U j ¼ mt þ  kdij (5)
RE

@xj @xi 3
1
k ¼ U 0iU 0j (6)
2
In the equation, Ui Uj respectively represent the velocity vectors in the i and j directions, P is the fluid
pressure, q is the air density, and U 0 i U 0 j is the Reynolds stress term generated by the motion of the fluid unit
mass. mt is the vortex viscosity coefficient. dij is Kronecker delta.
x ¼ e=k (7)
The SST k  x model is a variant of the Shear-Stress Transport turbulence model used for simulating
turbulent flow fields. This model combines two different types of turbulence models, namely the k  e
model and the k  x model, to achieve better performance in different flow regions. The SST k − ω
model introduces a correction term near the wall to enhance the accuracy of capturing turbulence
structures in the vicinity of the wall. The equations for this model are as follows:


@ @ @ @k
ðqkÞ þ ðqkli Þ ¼ k þ Gk  Yk þ Sk
@t @xi @xj @xj

 (8)
@ @ @ @x
ðqxÞ þ ðqxli Þ ¼ x þ Gx  Yx þ Sx
@t @xi @xj @xj
1652 FDMP, 2024, vol.20, no.7

@Uj
Gk ¼ qU 0 i U 0 j (9)
@xi
xa 0 0 @Uj
Gx ¼ q U iU j (10)
k @xi
Yk ¼ qb f b  kx (11)

Yx ¼ qbi f b  x2 (12)

bi ¼ F1 bi;1 þ ð1  F1 Þbi;2 (13)

lt
k ¼lþ (14)
rk

ED
lt
x ¼lþ (15)
rx

1
CT
rk ¼ (16)
F1 =rk;1 þ ð1  F1 Þ=rk;2
A
1
rx ¼ (17)
F1 =rx;1 þ ð1  F1 Þ=rx;2
TR

The turbulent viscosity is calculated as follows:

a k
lt ¼ q (18)
x
RE

In the equations, Gk and Gx represent the generation terms of turbulence kinetic energy due to the mean
velocity gradients, k and x are the effective diffusivity coefficients for k and x, Yk and Yx denote the
dissipation terms due to turbulence, Sk and Sx is a user-defined source term. The Constant coefficients in
the SST k  x model are shown in Table 3.
Table 3: Constant coefficients in the SST k − ω model
Model constants Value
α* 1
α 0.52
β* 0.09
βi,1 0.075
βi,2 0.0828
σk,1 1.176
σk,2 1
σω,1 2
σω,2 1.168
FDMP, 2024, vol.20, no.7 1653

The airfoil surface is set as a no-slip wall boundary, and the boundary conditions for the domain are
chosen using a pressure far-field approach. The incoming flow velocity is adjusted by changing the Mach
number setting. The angle of attack is altered by modifying the X and Y components of the incoming
flow direction. The present study employs a pressure-based algorithm and a second-order upwind
differencing scheme for discretization, using the SIMPLE pressure-velocity coupling method with double-
precision solving. During the solving process, in addition to monitoring the residuals of the variables, lift
coefficient (CL), and drag coefficient (CD) must also be tracked to ensure true convergence of the
solution. The convergence criteria for all calculations are residuals less than 1 × 10−5 [24,25].
2.2.3 Analysis of Computational Results
The analysis focuses on varying the incoming angle of attack, specifically examining the range of
0–12 degrees in this paper. As shown in Fig. 6, the pressure distributions on the QY-7305 airfoil are
presented for angles of attack of 0°, 6°, and 12°. Similarly, Fig. 7 illustrates the pressure distributions on
the NACA2412 airfoil at angles of attack of 0°, 6°, and 12°. Taking the 0° angle of attack as an example,
the pressure on the upper surface of the QY-7305 airfoil is approximately −106 Pa, while the pressure

ED
on the lower surface is around 24 Pa, resulting in a pressure difference of 130 Pa. On the other hand, the
NACA2412 airfoil exhibits a pressure of about −45 Pa on the upper surface and approximately 2 Pa on
the lower surface. Consequently, the pressure difference amounts to approximately 47 Pa for the
NACA2412 airfoil. Notably, at a 0° angle of attack, the QY-7305 airfoil displays a larger pressure
CT
difference, indicating higher lift.
A
TR

Figure 6: Pressure clouds of QY-7305 airfoil at angles of attack of 0°, 6° and 12°
RE

Figure 7: Pressure clouds of NACA2412 airfoil at angles of attack of 0°, 6° and 12°

The lift-to-drag ratio is a key metric for assessing the aerodynamic performance of an airfoil. As shown
in Fig. 8, it illustrates the characteristic curve of the lift-to-drag ratio for the QY-7305 and NACA2412 airfoils
across angles of attack from 0 to 12 degrees. Analyzing the lift-to-drag ratio characteristic curve provides
insights into the aerodynamic performance of the airfoils. In the graph, it is evident that the lift-to-drag
ratio of the QY-7305 airfoil is higher than that of the NACA2412 airfoil at the optimal angle of attack.
Additionally, the QY-7305 airfoil maintains a superior lift-to-drag ratio over a broader range of angles of
attack compared to the NACA2412 airfoil, indicating better aerodynamic performance for the QY-
7305 airfoil.
1654 FDMP, 2024, vol.20, no.7

ED
Figure 8: Comparative analysis of lift-to-drag ratio of airfoil

3 Aerodynamic Profile Design of Wind Turbine Blades

CT
The key design parameters for wind turbine blades include the selection and design of the blade airfoil,
the number of blades (B), the blade diameter (D), the design wind speed (V), the chord length (C), and the
twist angle (h). These parameters collectively determine the performance characteristics of the blades.

3.1 Determination of the Overall Parameters of the Wind Turbine

A
The design parameters of the wind turbine in this paper are shown in Table 4. The wind turbine’s rated
power is custom set to 44 kW, with a transmission efficiency and wind energy utilization coefficient of
TR

0.85 and 0.40 respectively. The incoming wind speed is 12 m/s, and the airfoil is chosen as the QY-7305.
For small wind turbines, the typical tip speed ratio is between 3–7. In this study, a tip speed ratio of 6 is
selected. For tip speed ratios within the range of 5–8, the number of blades for the wind turbine is chosen
between 2–4. In this study, 3 blades are selected.
RE

Table 4: Design parameters of wind turbine blades

Rated Wind turbine drive Wind energy Air Design Number Blade tip Airfoil
power efficiency η1 × η2 utilization factor density wind speed of blades speed ratio type
(kW) CP (kg/m³) (m/s) B λ0
44 0.85 0.40 1.225 12 3 6 QY-
7305

The diameter of the wind turbine can be determined using the following equation for power of the airfoil:

2
1 D
P ¼ pqgCp V3 (19)
2 4
obtained by transforming Eq. (19):
sffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi rffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
8P 8  44000
D¼ ¼ ¼ 12:5 ðmÞ
pqgCpV 3
p  1:225  0:85  0:4  123
FDMP, 2024, vol.20, no.7 1655

It is calculated that the diameter of wind wheel D = 12.5 m, the hub of wind turbine is taken as 0.5 m, and
the radius of wind wheel is taken as R = 6 m.

3.2 Establish the Mathematical Model of Blade Optimization Design Method

The Wilson design method, which is based on the blade element theory, divides the blade into equal
sections and uses the maximum power coefficient of each section as the objective function for
optimization and solution. Increasing the number of segmented elements in the Wilson design method
leads to more element cross sections being aligned with the objective function, resulting in a wind turbine
blade design that closely approaches an optimal shape and exhibits improved aerodynamic performance.
Excessive elements complicate the calculation and modeling process, while insufficient elements
compromise accuracy and lead to errors in fitting chord length and twist angle, thereby affecting the
optimization of wind turbine blade shape design. This paper proposes a method for non-uniformly
acquiring wind turbine blade elements, enabling the adjustment of element numbers within an optimal
range to achieve high calculation accuracy. This approach facilitates the creation of 3D models and

ED
enhances wind turbine energy utilization. Additionally, relevant correction functions are employed to
rectify the calculated element twist angle and element chord length parameters, thereby providing greater
convenience for wind turbine blade design based on the easy 3D model generation.
The basic mathematical model for pneumatic shape calculation is as follows:
CT
To maximize the wind energy utilization factor of the entire wind turbine, the wind energy utilization
factor of each blade section must be maximized, where the objective function is:
8
dCp ¼ ð1  aÞbFk3 dk (20)
k20
A
The constraints for this objective function is:
TR

að1  aFÞ ¼ bð1 þ bÞk2 (21)

where:
2
F ¼ arccosðeðf Þ Þ (22)
RE

p
B Rr
f ¼ (23)
2 R sin u
After obtaining the induced factors a and b, the chord length C and the installation angle can be obtained
by the follo airfoil equation h:
ð1  aÞ 1
tan u ¼ (24)
ð1  bÞ k
h¼ua (25)

BCCL 8paFsin2 u ð1  aFÞ

¼ (26)
r cos u ð1  aÞ2

V ð1  aÞC
Re ¼ (27)
l sin l
In the above formulas, CP is the wind energy utilization coefficient, a is the axial induction factor, b is
the tangential induction factor, F is the tip loss factor, B is the number of blades, r is the distance from the
1656 FDMP, 2024, vol.20, no.7

blade element cross-section to the rotor center, φ is the inflow angle, a is the angle of attack, h is the twist
angle, V is the incoming wind speed, C is the chord length, CL is the lift coefficient, μ is the dynamic
viscosity, and the dynamic viscosity is set to 1.385  10-5 m2/s.
Firstly, according to the blade element theory, the wind turbine blade is divided into three parts along the
spreading direction, which are blade root, blade middle and blade tip. The spacing between the blade
elements at the root is 1.2 m, divided into two equal parts; the spacing between the blade elements at the
middle of the blade is 0.3 m, divided into nine equal parts; the spacing between the blade elements at the
tip is 0.9 divided into one equal part; each blade element is a separate airfoil that needs to be calculated.
For each blade element, the optimization problem is solved using Eq. (20) as the objective function and
(21)–(23) as the constraints to determine the induced factors “a” and “b” of the blade element cross section,
as well as the slight blade loss factor “F”.
A fitting function is employed to fit the best lift coefficients for different Reynolds numbers to a curve.
Using the initial Reynolds number, the initial lift coefficient and chord length are calculated based on

ED
Eq. (26).
The fitting function is called to fit the best angle of attack corresponding to different Reynolds numbers
to a curve, and according to Eq. (27), the Reynolds number is calculated to obtain the best angle of attack by
fitting the fitted curve, and the twist angle is obtained by Eq. (25).
CT
The obtained chord lengths and twist angles are ideal, so it is necessary to make linear corrections to the
obtained chord lengths and twist angles in order to make the blades meet the requirements of machining,
structure, etc.
Obtain the 3D coordinates of the airfoil sections of each blade element and import the 3D coordinates
A
into Solidworks for modeling.
4 Programming and Computation
TR

The program is developed using the Wilson optimization design method, and the optimization flowchart
is presented in Fig. 9.
In the programming, it is important to divide it into three parts: blade root, blade middle, and blade tip.
Within the entire loop, if statements are inserted. After calculating the blade root, the various parameters are
RE

iterated to calculate the blade middle to prevent inaccurate results due to re-iteration. Similarly, after
calculating the blade middle, the parameters are iterated to calculate the blade tip. This process continues
until the calculation for the entire blade element is completed.
The ideal wind energy utilization coefficient for each cross-section of the blade element is presented in
Fig. 10. It can be observed that the wind energy utilization coefficient at each cross-section falls within the
Baez limit of 0.593. The position where the blade primarily generates power is towards the middle.
Moreover, the CP (coefficient of power) at the middle position of the blade exceeds 0.5, and CPmax
(maximum coefficient of power) is almost above 0.55, which is very close to the Baez limit. Meanwhile,
the program output, which characterizes whether the Fmincon function finds the optimal value under the
given constraint, indicates successful program execution after computing the optimal value.
The iterative calculations result in optimized chord lengths and twist angles for each blade element,
maximizing the utilization of wind energy. However, in this ideal state, the chord length and twist angle
exhibit a nonlinear relationship along the length of the blade, posing challenges in production and
processing. To meet the design requirements for production and processing, it is necessary to utilize the
curve-fitting function Polyfit in Matlab to perform fitting operations on the optimized chord lengths and
twist angles. If the fitting iterations are too low, it may result in a significant deviation from the optimized
FDMP, 2024, vol.20, no.7 1657

parameters, reducing wind energy utilization. On the other hand, higher fitting iterations may approach
theoretical data, but they might not achieve the desired correction effect, making installation and
manufacturing more inconvenient. After comprehensive consideration and practical analysis, a third-
degree polynomial fitting is applied to correct both the chord length and twist angle in this project. Due
to the lower wind energy utilization at the blade root and tip, the chord length and twist angle of each
blade element section within the range of 0.4 R–0.9 R are subjected to fitting correction. Moreover, at
0.4 R–0.9 R, the blade elements are positioned in the middle of the blade with small spacing, leading to a
better fitting effect. The curves in Fig. 11 depict the chord length and twist angle after fitting correction.
Through the analysis of the curves, it can be observed that the fitting results ensure a smooth transition in
the chord length at the blade tip, and excessive twist angles at the blade root are corrected.

ED
CT
A
TR
RE

Figure 9: Program flow chart

Fitting polynomial for the chord length:

CðrÞ ¼ 0:003518r3 þ 0:05453r2  0:325r þ 0:9038 (28)
1658 FDMP, 2024, vol.20, no.7

ED
Figure 10: Variation of wind energy utilization coefficient along the radius
CT
A
TR
RE

Figure 11: Chord length and twist angle fitting curve graph

Fitting polynomial for the twist angle:

hðrÞ ¼ 0:1673r3 þ 2:498r2  14:25r þ 29:89 (29)
The optimized values of each prime surface parameter for the blades are presented in Table 5.

5 3D Modeling of the Blade

The wind turbine blade is a vital component of the wind turbine, and its complex shape poses challenges
in achieving accurate modeling. The accuracy of the modeling directly impacts the efficiency of wind energy
utilization.
FDMP, 2024, vol.20, no.7 1659

Table 5: Parameters of each blade element cross-section of the blade

Radius/m a b Twist Fitting twist Chord Fitting chord F Re
angle/° angle/° length/m length/m
1.2 0.3351 0.1422 18.88 16.10 0.5990 0.5862 0.9593 790613
2.4 0.3393 0.0387 7.84 7.77 0.3924 0.3893 0.9810 876708
2.7 0.3398 0.0308 6.34 6.33 0.3547 0.3546 0.9822 878994
3 0.3405 0.0251 5.10 5.10 0.3246 0.3246 0.9822 884600
3.3 0.3414 0.0209 4.06 4.06 0.2988 0.2987 0.9811 888609
3.6 0.3427 0.0177 3.17 3.16 0.2765 0.2764 0.9787 891745
3.9 0.3446 0.0152 2.40 2.39 0.2572 0.2570 0.9744 894175
4.2 0.3473 0.0133 1.72 1.71 0.2402 0.2401 0.9673 895912

ED
4.5 0.3513 0.0118 1.11 1.10 0.2250 0.2250 0.9554 896697
4.8 0.3574 0.0107 0.55 0.54 0.2113 0.2111 0.9351 895710
5.1 0.3669 0.0099 0.01 0.00 0.1981 0.1980 0.8989 890634
6 0.1419 0.0039 1.11 −1.82
CT 0.0000 0.1570 0.0000 0

By using the method of coordinate transformation, we first export the two-dimensional coordinates of
each blade element section in Profili software, and then use the formula for spatial coordinate
transformation to calculate the three-dimensional spatial coordinates of each blade element section.
A
Finally, we import the three-dimensional coordinates of each blade element section into the modeling
software to complete the modeling process. The basic process of coordinate transformation and the
specific steps for solving three-dimensional coordinates are as follows:
TR

(1) Obtain the raw airfoil coordinates (x0, y0) using Profili software.
(2) The acquired cross-section will be translated to the aerodynamic center, enabling the determination
of two-dimensional coordinates (x1, y1). The coordinates of the aerodynamic center are assumed to be
(0.25C, 0). (x1, y1) = (x0, y0) − (0.25C, 0).
RE

(3) Through the implementation of a three-dimensional coordinate transformation, the resulting

coordinates are (x, y, z).
qffiffiffiffiffiffiffiffiffiffiffiffiffiffi

y1
x ¼ C  x1 þ y1  cos arctg þ h
2 2
x1
qffiffiffiffiffiffiffiffiffiffiffiffiffiffi

y1 (30)
y ¼ C  x1 þ y1  sin arctg þ h
2 2
x1
z¼r

The calculated coordinates of each airfoil section are saved, and then imported into SolidWorks for the
purpose of modeling. The resulting 3D model of the wind turbine is presented in Fig. 12.
6 Analysis and Discussion of Aerodynamic Performance Calculation Results
6.1 Aerodynamic Parameter Calculation Results
Aerodynamic performance calculation plays a crucial role in blade design and verification processes.
Verifying its aerodynamic performance becomes an essential step to evaluate the design outcomes.
Conversely, the results obtained from aerodynamic performance calculations offer valuable feedback for
refining the aerodynamic shape of the blade.
1660 FDMP, 2024, vol.20, no.7

Figure 12: 3D model of the wind turbine

In this study, the aerodynamic performance parameters are computed using the fitting formulas for chord

ED
length and twist angle obtained through the non-uniformly sampled blade element method. Additionally, a
comparison is performed using the uniformly sampled blade element method for analysis. Both methods
utilize twelve cross-sectional blade elements to calculate each parameter, with chord length and twist
angle corrections made at 0.4 R–0.9 R. The chord length and twist angle functions obtained through
fitting using two methods are used to calculate the aerodynamic performance parameters, including the
CT
wind energy utilization coefficient (CP), torque coefficient (CM), and thrust coefficient (CT). The curve
depicting the variation of CP with tip speed ratio is shown in Fig. 13, the curve depicting the variation of
CM with tip speed ratio is shown in Fig. 14, and the curve depicting the variation of CT with tip speed
ratio is shown in Fig. 15.
A
TR
RE

Figure 13: CP with tip speed ratio

6.2 Comparative Analysis of Aerodynamic Parameter Results

Figs. 13–15 illustrate that the calculated wind energy utilization coefficient (CP) reaches maximum
values of 0.4614 and 0.4516 when designing wind turbine blades using the non-uniform and uniform
blade element methods, respectively, at a tip speed ratio of 6. Similarly, the torque coefficient (CM)
reaches values of 0.0769 and 0.0752, while the thrust coefficient (CT) reaches values of 0.801 and 0.755.
FDMP, 2024, vol.20, no.7 1661

The wind energy utilization coefficient (CP) holds utmost significance for assessing the aerodynamic
performance of wind turbine blades, reflecting their efficiency in capturing wind energy. Within the tip
speed ratio range of 5 to 10, the wind energy utilization coefficient calculated using the non-uniform
blade element method consistently surpasses that of the uniform blade element method. Additionally, it
aligns with the tip speed ratio at the design point, attaining its peak value when the tip speed ratio is 6.
Consequently, designing blades using the non-uniform blade element method leads to enhanced
aerodynamic performance.

ED
CT
A
TR

Figure 14: CM with tip speed ratio

RE

Figure 15: CT with tip speed ratio

1662 FDMP, 2024, vol.20, no.7

This study utilized two distinct methods for conducting calculations and analysis in wind turbine design.
Specifically, 10, 15, 20, and 30 blade elements were employed, and the resulting outcomes are presented in
Table 6. The findings demonstrate that the non-uniform blade element method surpasses the uniform blade
element method in terms of aerodynamic performance calculations when a smaller number of blade elements
is used. However, when a larger number of blade elements is employed, the calculated aerodynamic
performance parameters are essentially the same for both methods. This is primarily due to the fact that
with a larger number of blade elements, both methods exhibit greater accuracy in fitting the chord length
and twist angle within the 0.4 R–0.9 R range. In contrast, when fewer blade elements are utilized, the
non-uniform blade element method achieves a smaller spacing between blade elements within the 0.4 R–
0.9 R range, thereby enabling higher precision in fitting the chord length and twist angle and resulting in
more accurate aerodynamic performance calculations. Consequently, employing the non-uniform blade
element method not only yields superior aerodynamic performance for the geometric shape of wind
turbine blades, but also circumvents the issue of excessive blade elements, which can lead to
cumbersome calculations and modeling.

ED
Table 6: Aerodynamic calculation results of different blade elements
Number of blade elements CT 10 12 15 20 30
Wind energy utilization coefficient obtained through 0.4588 0.4614 0.4621 0.4622 0.4623
non-uniform design method
Wind energy utilization coefficient obtained through 0.4472 0.4516 0.4596 0.4617 0.4621
uniform design method
A
TR

7 Conclusion
In this study, we extracted an airfoil inspired by the eagle and employed it in conjunction with the non-
uniform blade element method to enhance the design of wind turbine blades. The resulting conclusions are
outlined below:
RE

(1) Using biomimicry, the airfoil of an eagle was extracted and compared with the aerodynamic
characteristics of the NACA2412 airfoil. Through the analysis, it was determined that the eagle airfoil
demonstrated superior aerodynamic performance.
(2) The blade was designed using the non-uniform blade element method and then compared with the
same method itself. The analysis showed that the blades designed using this approach demonstrated superior
aerodynamic performance, while effectively mitigating the challenge of increased complexity in calculations
and modeling arising from a higher number of blade elements.
(3) When a smaller number of blade elements is used, the non-uniform blade element method
outperforms the uniform blade element method.
Acknowledgement: None.

Funding Statement: This study was supported by the National Natural Science Foundation Projects (Grant
Number 51966018), the Chongqing Natural Science Foundation of China (Grant Number cstc2020jcyj-
msxmX0314), the Key Research & Development Program of Xinjiang (Grant Number 2022B01003),
Ningxia Key Research and Development Program of Foreign Science and Technology Cooperation
Projects (202204), and the Key Scientific Research Project in Higher Education Institution from the
Ningxia Education Department (2022115).
FDMP, 2024, vol.20, no.7 1663

Author Contributions: The authors confirm contribution to the paper as follows: supervision, project
administration, funding acquisition and writing-review & editing: Yuanjun Dai, Dong Wang, Xiongfei
Liu, Weimin Wu; study conception and design: Yuanjun Dai, Dong Wang; data collection: Dong Wang;
analysis and interpretation of results: Dong Wang; draft manuscript preparation: Dong Wang. All authors
reviewed the results and approved the final version of the manuscript.
Availability of Data and Materials: The authors confirm that the data supporting the findings of this study
are available within the article.
Conflicts of Interest: The authors declare that they have no conflicts of interest to report regarding the
present study.

References
1. Mitchell S, Ogbonna I, Volkov K. Improvement of self-starting capabilities of vertical axis wind turbines with new
design of turbine blades. Sustain. 2021;13(7):3854–24.

ED
2. Gen MS, Zkan R. Optimum layer sequence analysis for composite blade using ACP-FSI model. Int J Sust Aviat.
2021;7(4):354–17.
3. Zhang J, Yang SU, Yong LI, Cong X. Numerical study on the flow control of vortex generators on large wind
turbine blades. Int J Intell Energy. 2021;43(12):66–5.
4. Muhsen H, Al-Kouz W, Khan W. Small wind turbine blade design and optimization. Symmetry. 2019;12(1):18–4.
CT
5. Tarfaoui M, Shah OR, Nachtane M. Design and optimization of composite offshore wind turbine blades. J Energy
Resour Technol. 2019;141(5):051204–9.
6. Rose JBR, Natarajan SG, Gopinathan VT. Biomimetic flow control techniques for aerospace applications: a
comprehensive review. Rev Environ Sci Biotechnol. 2021;20(3):645–32.
A
7. Metin U, Özdemir M, Yildirim ÇV, Çoban S. A novel biomimetic wing design and optimizing aerodynamic
performance. J Aviat. 2022;6(1):12–3.
TR

8. Yuan SK, Li RN. Analysis influence of trailing edge modification on aerodynamic performance of airfoils for wind
turbine. Appli Mechanic Materi. 2012;220–223(2):900–4.
9. Wang Q, Wang J, Chen J, Lou S, Sun J. Aerodynamic shape optimized design for wind turbine blade using new
airfoil series. J Mech Sci Technol. 2015;29(7):2871–11.
10. Wang X, Ju H, Lu J. Wind turbine airfoils optimization design by generalized regression neural network under
RE

small sample. J Mech Sci Technol. 2023;37(1):217–11.

11. Yan H, Su X, Zhang H, Hang J, Zhou L, Liu Z, et al. Design approach and hydrodynamic characteristics of a novel
bionic airfoil. Ocean Eng. 2020;216(8):108076–10.
12. Zhao M, Cao H, Zhang M, Liao C, Zhou T. Optimal design of aeroacoustic airfoils with owl-inspired trailing-edge
serrations. Bioinspir Biomim. 2021;16(5):056004–14.
13. Hua X, Zhang C, Wei J, Hu X, Wei H. Wind turbine bionic blade design and performance analysis. J Vis Commun
Image R. 2019;60(3):258–7.
14. Lanzafame R, Messina M. Design and performance of a double-pitch wind turbine with non-twisted blades. Renew
Energy. 2009;34(5):1413–7.
15. Lee SL, Shin SJ. Wind turbine blade optimal design considering multi-parameters and response surface method.
Energies. 2020;13(7):1639–23.
16. Rodriguez CV, Celis C. Design optimization methodology of small horizontal axis wind turbine blades using a
hybrid CFD/BEM/GA approach. J Braz Soc Mech Sci Eng. 2022;44(6):254–25.
17. Özkan R, Genç MS. Aerodynamic design and optimization of a small-scale wind turbine blade using a novel
artificial bee colony algorithm based on blade element momentum (ABC-BEM) theory. Energy Convers
Manag. 2023;283(4):116937–21.
18. Zhu J, Zhou Z, Cai X. Multi-objective aerodynamic and structural integrated optimization design of wind turbines
at the system level through a coupled blade-tower model. Renew Energy. 2020;150(1):523–15.
1664 FDMP, 2024, vol.20, no.7

19. Zhu J, Cai X, Gu R. Aerodynamic and structural integrated optimization design of horizontal-axis wind turbine
blades. Energies. 2016;9(2):66.
20. Zemamou M, Toumi A, Mrigua K, Lahlou Y, Aggour M. A novel blade design for Savonius wind turbine based on
polynomial bezier curves for aerodynamic performance enhancement. Int J Green Energy. 2020;17(11):652–14.
21. Zhang B, Song B, Mao Z, Tian W, Li B, Li B. A novel parametric modeling method and optimal design for
Savonius wind turbines. Energies. 2017;10(3):301–20.
22. Maki K, Sbragio R, Vlahopoulos N. System design of a wind turbine using a multi-level optimization approach.
Renew Energy. 2012;43(1):101–10.
23. Tang X, Huang X, Peng R, Liu X. A direct approach of design optimization for small horizontal axis wind turbine
blades. Procedia CIRP. 2015;36(1):12–5.
24. Belamadi R, Djemili A, Ilinca A, Mdouki R. Aerodynamic performance analysis of slotted airfoils for application
to wind turbine blades. J Wind Eng Ind Aerodyn. 2016;151(5):79–99.
25. Kepekci H. Comparative numerical aerodynamics performance analysis of NACA0015 and NACA4415 airfoils.
Int J Eng Sci. 2022;2(1):144–8.

ED
CT
A
TR
RE