CHAPTER 6

Numerical Simulations of
Small Wind Turbines—
HAWT Style
Jimmy C. K. Tong

Here, the author introduces the important topic of product certification and
code compliance. Next, he focuses on cases that are considered for numerical
investigation of horizontal-axis wind turbine (HAWT) performance. Among
the cases are different design environments with varying wind gusts, wind
direction, and turbulence. Creation of the numerical mesh and application of
the flow boundary conditions are discussed in detail. So too are the various
turbulence models that modern computational simulators employ.
An emerging field in this area is the coupled interaction of the fluid
and the solid object. This analysis approach, termed Fluid Structural
Interaction, is finding use among research scientists as computer capa-
bilities increase. Finally, the author discusses the types of information
that are obtained from numerical simulation and how that information
guides the design process.

6.1 INTRODUCTION

Given the rapid development of wind energy in small communities and

residential buildings, the advancement of wind turbine design is crucial.
The adoption of green energy is undoubtedly a worldwide trend, as more
countries see the need to hand over a more sustainable earth to future gen-
erations. As the modern wind turbine has proven its effectiveness at least
for the last 30 years, transforming this technology from a bigger scale to
a smaller scale to overcome technical and cost challenges is commonly
faced by the small wind turbine designers and manufacturers. Compu-
tational Fluid Dynamic (CFD) software is becoming a standard tool in
132  • SMALL-SCALE WIND POWER

the development of small wind turbines. This numerical simulation tool

can assist designers and engineers to design and analyze turbines that are
subjected to different wind conditions, allowing them to gain insight into
turbine behaviors under various scenarios and to streamline design iter-
ations to reach an optimal design with a shorter cycle. As a result, new
wind turbine models can be developed for the intended market with higher
performance and lower production cost in a shorter time.
This chapter is focused on the use of numerical simulation in the
design of HAWTs. The two major categories of objectives for using
numerical simulation are as follows:

1. Product certification and code compliance

2. Power generation and structural safety performance

In order to introduce a new wind turbine model into a market, most coun-
tries require or encourage product certification that is in compliance with
local or international standards. To qualify for the certification, the wind
turbine will need to pass the safety and standard performance testing and
design verifications specified by the designated authority. This rigorous
process is employed to ensure that the wind turbine will meet a certain
quality standard and is safe to use, and the advertised performance is com-
parable to other products available in the market.
Design verification is part of the certification process, and numeri-
cal simulation can be helpful to demonstrate that the structural and safety
requirements are met under the operating conditions. In general, numerical
simulation uses the code-specified incoming wind condition to determine
the wind loading acting on the blades and the turbine structure. Hence, the
wind turbine can be designed to meet required material safety factors. This
will be discussed in more detail later in this chapter.
Beyond design verification, the numerical simulation is also bene-
ficial for designers to improve the performance of the wind turbine. In
particular, a few computer codes are developed for blade design in order
to capture kinetic energy from moving air. As the entire wind turbine com-
prises more components, the general CFD software is applicable for the
complete system analyses. Details about the approach of numerical simu-
lation, the analysis process with inputs and outputs, and the design itera-
tion are discussed later in this chapter.

6.2 PRODUCT CERTIFICATION AND CODE

COMPLIANCE

Product certification is a necessary process before a product can be

introduced to customers. For most electronic products, the Underwriters
 NUMERICAL SIMULATIONS OF SMALL WIND TURBINES • 133

Laboratories (UL) and Conformité Européenne (CE) certifications are

widely used around the world. Most countries adopt well-recognized
international standards, while local codes still vary from place to place
where the local authorities have jurisdiction over the local practice. For
small wind turbine industries, the International Electrotechnical Commis-
sion (IEC) has published the following list of standards that covers the
design and testing for wind turbine performance and safety:

IEC 61400-2: 2006 Ed 2.0 Design requirements for small wind tur-
bines [1].
IEC 61400-11: 2012 Ed 3.0 Acoustic noise measurement techniques [2].
IEC 61400-12-1: 2005 Ed 1.0 Power performance measurements of
electricity-producing wind turbines [3].
IEC 61400-14: 2005 Ed 1.0 Declaration of apparent sound level and
tonality values [4].
IEC 61400-21: 2008 Ed 2.0 Measurement and assessment of power
quality characteristics of grid-connected wind turbines [5].
IEC 61400-22 2010 Ed 1.0 Conformity testing and certification [6].
IEC 61400-23: 2001 Ed 1.0 Full scale structural testing of rotor
blades [7].

The standards of three countries are described next as examples:

In the United States, the American Wind Energy Association (AWEA)
has the “Small Wind Turbine Performance and Safety Standard 9.1—
2009,” and a list of turbines that have received the Small Wind Certifica-
tion Council approval can be found on its website [8].
Similarly, the British Wind Energy Association (BWEA) in the
United Kingdom has its “Small Wind Turbine Performance and Safety
Standard—2008,” and those turbines that have received accreditation are
listed on the Microgeneration Certificate Scheme (MCS) website [9].
In Denmark, the Danish Energy Authority (DEA) requires wind tur-
bines to be certified under the Danish Certification Scheme, and those
turbines that have received accreditation are listed on the Danish Certifi-
cation website [10].

6.2.1 DESIGN LOAD CASES AND EXTERNAL CONDITIONS

While there are local code variations, most of the codes are very similar
to the IEC 61400-2 [1]. Taking IEC 61400-2 as the basis for wind turbine
design is commonly accepted by developers who intend their products for
the global market. The standard specifies the design methodology, which
includes simplified load equations, aeroelastic modeling, and mechanical
134  • SMALL-SCALE WIND POWER

load testing. Out of these methodologies, aeroelastic modeling is the best

option for balancing accuracy and development time. Numerical simula-
tion is a good tool to assist in this process, and compliance with the stan-
dard will drive how the numerical simulation should be constructed. The
standard lists a set of design load cases (DLC; see Table 6.1), which in turn
yields design loadings that generate the limit states. Then, the aeroelastic
modeling of the wind turbine design is used to verify the safety operations
under the specified cases.
In Table 6.1, seven basic design situations are covered and one or
more DLC are specified for each situation. For each DLC, a wind condi-
tion and the type of analysis required (F for fatigue loading and U for ulti-
mate loading) are prescribed. The several wind conditions used in these
DLC are as follows and the details of each wind profile can be found in
Chapter 6 of the standard IEC 61400-2 [1]:

ECD: Extreme coherent gust with direction change

ECG: Extreme coherent gust
EDC50: Extreme direction change with recurrence period of 50 year
EOG1: Extreme operating gust with recurrence period of 1 year
EOG50: Extreme operating gust with recurrence period of 50 year
EWM: Extreme wind speed model
NTM: Normal turbulence model
NWP: Normal wind profile model

Table 6.1. Set of Design Load Cases for aeroelastic models from IEC
61400-2.
Design DLC Wind condition Other Type of
­situation ­conditions analysis
1) Power 1.1 NTM Vin < Vhub < F, U
­production Vout or 3Vave
1.2 ECD Vhub < Vdesign U

1.3 EOG50 Vin < Vhub U

<Vout
or 3Vave
1.4 EDC50 Vin < Vhub U
<Vout or 3Vave
1.5 ECG Vhub = Vdesign U

(Continued)
 NUMERICAL SIMULATIONS OF SMALL WIND TURBINES • 135

Design DLC Wind condition Other Type of

­situation ­conditions analysis

2) Power pro- 2.1 NWP Vhub = Vdesign Control U

duction plus
or Vout or system
occurrence 2,5 Vave fault
of fault 2.2 NTM Vin < Vhub < Control or F, U

Vout Ve1 protection
system
fault
2.3 EOG1 Vin < Vout or Loss of U

2,5 Vave electrical
connection
3) Normal shut 3.1 NTM Vin < Vhub < F
down
Vout
3.2 EOG1 Vin = Vout or U

Vmax, shutdown
4) Emergency 4.1 NTM To be stated U
or manual by the
shut down manufacturer
5) Parked 5.1 EWM Vhub = Ve50 Possible U
(standing loss of
still or electrical
idling) power
network
5.2 NTM Vhub < 0,7 Vref F

6) Parked and 6.1 EWM Vhub = Ve1 U

fault condi-
tion
7) Transport, 7.1 To be stated by the U
assembly, manufacturer
maintenance
and repair

The author thanks the International Electrotechnical Commission (IEC) for

permission to reproduce information from its International Standard ICE
61400-2 ed 2.0 (2006). All such extracts are copyright of IEC, Geneva,
Switzerland. All rights reserved. Further information on the IEC is avail-
able from www.iec.ch. IEC has no responsibility for the placement and
context in which the extracts and contents are reproduced by the author,
nor is IEC in any way responsible for the content or accuracy therein.
136  • SMALL-SCALE WIND POWER

The aforementioned eight different types of wind conditions describe spe-

cific wind scenarios. In different combinations with the DLC, the resulting
loading for the wind turbine can be determined. These wind conditions
describe, in general, the magnitude, turbulence, direction, wind shear,
and time variation of these parameter changes. Furthermore, the behav-
iors of these wind conditions are different when the recurrence period is
longer. For instance, a higher magnitude of extreme wind can occur when
the recurrence period is 50 year as compared to extreme winds that recur
yearly.
Together with the required DLC and the corresponding specified wind
conditions, these form the inputs to the numerical simulations. The next
section of this chapter further discusses how these inputs are incorporated
into the analyses.

6.3 APPROACH TO NUMERICAL SIMULATION OF

HORIZONTAL-AXIS WIND TURBINE

With the code and standard, which cover the certification process and
define the safety and performance requirements of a wind turbine,
numerical simulation is a common tool for the developers to properly
assess the design and to effectively make design improvement through
the iterative product development cycle. Out of the various important
components of a wind turbine, like the blades, generator, brakes, con-
trols of torque, pitch and yaw, and the tower, the numerical simula-
tion discussed in this chapter only covers the mechanical aspect of the
design. Other aspects, such as electrical and electromagnetic aspects, are
not discussed.
As the blades and the rest of the turbine system (rotor, nacelle, and
tower) are components that directly interact with the wind, numerical sim-
ulation for mechanical/structural loading and power generation (through
transforming the kinetic energy from moving air into rotating mechanical
energy for the generator to turn into electricity) can be determined through
aeroelastic modeling and CFD modeling. For aeroelastic modeling, the
Blade Element Momentum (BEM) theory has been commonly employed
for blade design and airfoil selection. BEM allows simple calculations of
steady loads, thrust, and power for different settings of wind speed, rota-
tional speed, and pitch angle. Beyond steady calculations, and with the
advancement of computation power, CFD is becoming the standard tool
applied to wind turbines subjected to complex flow situations. Designers
and engineers are still using both tools, independently or in combination,
 NUMERICAL SIMULATIONS OF SMALL WIND TURBINES • 137

Figure 6.1. Streamline patterns from a small horizontal-axis wind turbine by

CFD simulation.

Steady wind profile

Instantaneous wind field

Figure 6.2. Schematic of wind profile.

for the turbine design. Figure 6.1 shows streamline patterns from a typical
CFD simulation for a conventional three-bladed HAWT.

6.3.1 INPUTS TO NUMERICAL SIMULATIONS

As mentioned in the last section, the requirements specified by the standards

provide most of the inputs to the numerical simulations. For instance, a visual
of the wind profile acting on the wind turbine that describes a case for sim-
ulation is illustrated in Figure 6.2. The inputs can be categorized as follows:
138  • SMALL-SCALE WIND POWER

External wind conditions: wind speed, direction, shear, turbulence,

temperature, and density
Turbine geometrical specifications: blade design, nacelle dimension
and weight, hub height, material properties, and tower design
Turbine operational specifications: Revolutions per minute (RPM),
pitch angle, and brake torque

6.3.2 PROPER SELECTION OF SOLUTION DOMAIN

In formulating the numerical simulation, it is recognized that not only are

there geometrical complexities to be considered, but also relevant fluid
flow processes that take place in the surrounding environment of the wind
turbine. Furthermore, the flow is inherently three-dimensional. Consider-
able care is required to construct a mesh that would properly resolve these
issues. Numerical simulation literature often states that once the condi-
tions are specified on the boundaries of the chosen domain, the flow pat-
terns are immediately affected by the features inside the solution domain.
However, this phenomenon is a strong sign that the specified boundary
conditions might not be appropriate. A recommendation for eliminating
this type of error is to extend the solution domain so that the prescribed
flow pattern remains for a while before being altered by the modeled fea-
tures. Therefore, using an extended solution domain in order to obtain
results of high accuracy is deemed necessary.
The observations set forth in the preceding paragraphs merely illus-
trate the importance of the proper selection of the boundaries of the solu-
tion domain. Moreover, sufficient upstream and downstream spaces are
needed to allow proper specification of the boundary conditions. The solu-
tion domain can be considered large enough when the boundary condi-
tions specified are able to maintain the original characteristics for a while
before being affected by the presence of the wind turbine. In other words,
the arbitrarily chosen virtual boundaries should not have any impact on
the flow results of interest. This idea follows the same principle of design-
ing a wind tunnel in such a way that the edge effects are minimized.

6.3.3 MESHING AND NODE DEPLOYMENT OF SOLUTION

DOMAIN

The key to attaining numerical accuracy in simulation studies is to con-

struct an appropriate mesh. Most commercial CFD software is based
on the finite-volume method of discretizing the solution domain. The
 NUMERICAL SIMULATIONS OF SMALL WIND TURBINES • 139

accepted approach for seeking mesh-independent solutions is to system-

atically refine the mesh. This approach, although widely accepted, pos-
sibly has invalid indications. For example, if the elements that compose
the mesh are all diminished by the same multiplicative factor without
accounting for regions of high gradients, mesh independence may appear
to exist whereas, in fact, proper attention to high-gradient zones may indi-
cate a different conclusion. Another approach, although less systematic,
is to concentrate on regions of high gradients. Here, a combination of the
two approaches is recommended to assess mesh independence.
In view of the previous paragraph, studies of mesh independence can
be performed by comparing the simulation results of coarser and finer
meshes based on the aforementioned approach. Along the same lines, the
results of interest and of high-gradient zones can be set as figures of merit
for the attainment of mesh-independent solutions.
Attention is now turned to two types of CFD analysis—(i) blade and
rotor level and (ii) entire turbine system level. The first analysis may have
an advantage of focusing on the blade performance, whereas the second
one can investigate the system response for the entire turbine. For the first
level of CFD analysis, which is conducted on the blade only or on the
blades with a rotor, a sample mesh with the blade and rotor is shown in
Figure 6.3. The solution domain contains the surfaces of a single blade
and a portion of the rotor. As this is a three-bladed HAWT application, the
fluid domain is chosen to be one-third of the entire circle. With sufficient
upstream and downstream fluid domain included, the entire mesh forms
like a pie as shown in Figure 6.3. Mesh refinement is performed near the

Figure 6.3. Meshing for wind turbine blade and rotor CFD simulation.
Source: Reproduced from Bazilevs et al. [11].
140  • SMALL-SCALE WIND POWER

blade and rotor surfaces to resolve the boundary layer of the fluid on the
leading and trailing edges and tip of the blade. With only one-third of the
domain modeled, it is possible to obtain a solution using a numerical tech-
nique of periodicity. The periodic boundary conditions can be used when
the flows going out through one boundary reappear as the flows going in
through the opposite boundary. If the flow pattern is too complex for the
periodicity to be appropriate, the same mesh can be repeated two more
times every 60°. When the complete face of the blade and rotor is modeled,
then the usual CFD analysis without using periodicity can be conducted.
The other type of analysis covers the entire wind turbine system,
which includes all the blades, rotor, nacelle, and tower structure. This
analysis can be useful to simulate a real site condition or a wind tunnel
test situation. A sample mesh for this simulation is shown in Figure 6.4.
As seen in Figure 6.4, the solution domain is chosen for a wind tunnel.
In both the wind tunnel and real site location, the recommended distances to
form the solution domain are 2–5 times the rotor diameter as the upstream
distance, 10–20 times as the downstream distance, and 1–3 times from the
edge as the distance from both sides. Mesh refinement is done near all
surfaces to allow finer resolution of the flow pattern to be developed within
the boundary layer. In particular, special attention on mesh refinement is
required near the leading and trailing edges and the tip of the blades.
Since wind turbines are a type of rotating machinery, a numerical
simulation technique using a rotating frame of reference is appropriate

Interface zone 1
Interface zone 2

Cylinder part Wind tunnel part

Rectangular part

Figure 6.4. Meshing for entire wind turbine system CFD simulation.
Source: Reprinted from Moa et al. [12].
 NUMERICAL SIMULATIONS OF SMALL WIND TURBINES • 141

to simulate the moving blades and rotor. The fluid inside this rotating
domain is considered to be moving with the frame, and this simplifies the
computation and allows faster completion of the calculation. As shown in
Figure 6.4, the surfaces of blades and rotor are meshed in the rectangle
part and the rectangle part is inside the cylinder part, which causes the full
rotation of the blades and rotor. Both the rectangle and cylinder parts are
considered moving domains, whereas the wind tunnel part is treated as a
stationary domain. The sliding mesh technique is employed and the fluid
interfaces are defined between the aforementioned parts. Hence, interface
zone 1 is located between the rectangle and cylinder parts, and interface
zone 2 is located between the cylinder and wind tunnel parts. Although
this technique may be computationally demanding, it is the known method
for achieving high accuracy unsteady solutions for moving and stationary
domains that comprise multiple frames of reference.

6.3.4 BOUNDARY CONDITIONS

The boundary conditions are specified by the standard, and the inputs
described in the previous section are to be translated as the boundary
­conditions. Beyond the conditions from product certification and code
compliance, conditions from the actual site location can also be used in
the simulation.

6.3.5 NUMERICAL SOLVER

Focus is now turned back to CFD simulation to further determine the next
step for completing the analysis. The wind blows in open air; therefore,
most flow situations are turbulent flows. Due to the complexity of the
different flow patterns, numerous models are developed to balance the
accuracy and speed of the computation. In addition to flow modeling, sim-
ulation with multi-physics is becoming more common for achieving bet-
ter designs. One combined model worth mentioning is the fluid–structure
interaction (FSI) model, which is discussed later.

6.3.6 TURBULENCE FLOW MODELS

As the development of computational power continues to increase and

more complex flow patterns are able to be formulated, more turbulent flow
models can be created, ranging from a few specialized generic types of
142  • SMALL-SCALE WIND POWER

flow patterns to more generalized codes that cover wide applications. The
guiding principles of formulating the CFD code include the law of con-
servation of mass and Newton’s second law of motion. In the specialized
literature for fluid mechanics, Newton’s second law for flowing fluid is
sometimes called momentum conservation or, alternatively, the Navier–
Stokes equations. Conservation of mass states that mass cannot be created
or destroyed, while Newton’s second law balances forces with changes
of momentum. The following list of turbulent flow models is available
through most commonly known software:

1. Reynolds-Averaged Navier-Stokes Simulation (RANS)

Standard k−ε model
RNG k−ε model
Realizable k−ε model
Standard k-ω model
Shear–stress transport (SST) k-ω model
Reynolds stress model (RSM)
2. Detached Eddy Simulation (DES)
Spalart–Allmaras model
3. Large Eddy Simulation (LES)
Smagorinsky model
Algebraic Dynamic model
Localized Dynamic model
4. Direct Numerical Simulation (DNS)

The aforementioned list covers four major groups of models, presented

in the order of their complexity and their demand for computational
resources. The first group of models is known as RANS, where the
numerical formulation of the term “turbulent viscosity” is introduced in
the Navier–Stokes equations. It is adjusted to suit different flow patterns
and account for the effects of turbulence. As each of the specific models
in this group uses different methodologies to solve equations in different
ensemble-averaged forms, different equations of turbulent viscosity are
used in each of these models.
The second group is DES, which uses a hybrid method combining the
RANS and LES approaches to treat near-wall regions and the bulk flow
respectively.
The third group is LES, where the momentum and energy transfer of
large energy-carrying turbulent structures are computed exactly by using
the governing equations, while the effect of the sub-grid scales (SGS)
of turbulence is modeled or approximated. LES applies low-pass filters
 NUMERICAL SIMULATIONS OF SMALL WIND TURBINES • 143

according to turbulence theory and available computational resources

to separate the “small” length and time scales in the velocity field. It is
known to be more accurate than RANS but less demanding than DNS.
Lastly, the DNS approach resolves all scales of turbulence by solving
the Navier–Stokes equations directly without any turbulence modeling.
DNS requires a very fine mesh; hence, it has a very high demand for com-
putational resources in order to achieve high accuracy.

6.3.7 FLUID–STRUCTURE INTERACTION

In the context of wind turbine applications, FSI is the interaction of the

deformable structure, like the blade and tower, with the surrounding fluid
flow. For instance, the blade deforms from wind loading and the deformed
shape changes the flow, which in turn affects the performance. This inter-
action mostly affects the blade structure design and aerodynamic perfor-
mance. To carry out the analysis of FSI, it is recommended to perform a
two-way iterative loop with a mapping of calculated results between a CFD
model and another Finite Element (FE) model. For instance, the pressure
results on the blade are first obtained in the CFD model, and then mapped
on the FE model with the original shape. After the FE calculation, the
deformed shape is passed back to the CFD model to form a new geometry
for flow calculation. The process continues until convergence is found.

6.3.8 SOFTWARE

Numerical simulation using CFD is a widely adopted tool in both research

and industry. Many software packages are available through commercial
licenses and open sources. Many of these codes are developed for generic
applications and some of these are tailor-made for wind turbine design. To
name a few brands as reference, ANSYS/Fluent/CFX, Star-CD, Pointwise,
ACUSIM, SimPack, and FAST are among the most common ones. Besides
CFD, a specially designed code, Bladed, which uses BEM, is also a reliable
tool and is widely employed by wind turbine developers in the industry.

6.4 POWER GENERATION AND STRUCTURAL

SAFETY PERFORMANCE

To achieve higher performance of the wind turbine design, numerical sim-

ulation is an effective tool to provide an accurate and quick assessment
144  • SMALL-SCALE WIND POWER

on the design under the specified environmental conditions. Using the

inputs as prescribed in the codes and standards and determined by oper-
ational design parameters, the turbine design is modeled in the properly
chosen solution domain using the inputs as boundary conditions. With a
fine enough meshing determined by a mesh-independence study to ensure
the simulation quality and the appropriate flow solver, the results can be
extracted.
Among the vast amount of data in a complete set of simulation runs,
the key results that drive the turbine design are pressure and thrust coef-
ficients of the blade, the power curve and power coefficient of the wind
turbine, the loading experienced at the blade, the drive train, the tower of
the turbine, and the wake losses created by the wind turbine. Other result-
ing factors associated with each design iteration are the material used and
cost.
From the numerical simulation that focuses on the details of the blade
design (i.e., the selection of airfoil, blade length, chord length, twist angle,
relative thickness, and pitch axis), the calculated results of the pressure
and thrust coefficients are helpful to designers. The coefficients can be
used to give designers the insights to balance the power that can be cap-
tured by the blade and the structural loading acting on the blade. Based on
trade-off analysis, an optimal design is reached for the specified condition.
Figure 6.5 shows an example of calculated pressure coefficients along a
blade from numerical simulation and compares the values from a wind
tunnel test.
For the overall wind turbine performance, the same approach is used
to evaluate the “effectiveness” of the design; it balances between the power
generation and the structural integrity for safe operation. An example of
a typical power curve is shown in Figure 6.6. In the figure, the result-
ing power curves for models #1–#3 illustrate different design approaches,
some with higher power generation while others more focused on reduc-
ing loading on the turbine structure. As for the structural aspect, the code
and standard have suggested a preset format for the present load data of
the blade, drive train, and the tower of the turbine. The format helps to
organize a large amount of data and allows attention to be focused on the
limiting cases for design verification and improvement. An example of the
format is shown in Table 6.2.
Finally, by analyzing the performance of each design iteration,
improvements are made possible by changing the design parameters. To
speed up and shorten the design iteration cycle, many optimization algo-
rithms are available to run in conjunction with the numerical simulations.
Many successful products and projects are examples of this process.
 NUMERICAL SIMULATIONS OF SMALL WIND TURBINES • 145

0° YAW 7m/s EXP 0°

2.5 80% R EXP 180°
Error
2 WMB 0°
WMB 180°
1.5
-Cp

0.5

-0.5

-1
0 0.2 0.4 0.6 0.8 1
Y/c

Figure 6.5. Pressure coefficient (Cp) on the blade surface along the entire blade
length.
Source: Reproduced from Gómez-Iradi, Barakos, and Munduate [13].

12

10
Power generation (kW)

8
Model #1
6
Model #2

4 Model #3

0
0 5 10 15 20 25 30
Wind speed at hub height (m/s)

Figure 6.6. Power curves for a typical 10 kW horizontal-axis wind turbine.

146  • SMALL-SCALE WIND POWER

Table 6.2. Suggested format to present load results for extreme (top) and
fatigue (bottom) loads.
Results of the extreme load evaluation at a component
Design Safety Fx Fy Fz Fres Mx My Mz Mres
Load factor [kN] [kN] [kN] [kN] [kNm] [kNm] [kNm] [kNm]
case
DLC with
Fx Min

DLC with
Fx Max

DLC with
Fy Min

DLC with
Fy Max

DLC with
Fz Min

DLC with
Fz Max

DLC with
Fres
Min
DLC with
Fres
Max
DLC with
Mx Min

DLC with
Mx Max

DLC with
My Min

DLC with
My Max

DLC with
Mz Min

DLC with
Mz Max

DLC with
Mres
Min
DLC with
Mres
Max
 NUMERICAL SIMULATIONS OF SMALL WIND TURBINES • 147

Results of the fatigue load evaluation at a component

Inverse S-N Fx Fy Fz Mx My Mz
slopes [kN] [kN] [kN] [kNm] [kNm] [kNm]

6.5 CONCLUDING REMARKS

To deliver a new HAWT to the market for small-power applications with a

shortened design cycle, numerical simulation is a necessary tool for designers
and engineers. Numerical simulation allows developers to gain the insights
of the turbine design and to evaluate the effects of varying the design param-
eters in a much shorter time than using traditional techniques. By running the
code-required load cases under the specified wind conditions, the simulated
results can be used to provide evidence that a wind turbine is in compliance
with the code and standard; therefore, the first objective is achieved. Going
beyond the code and standard, the other objective of employing numerical
simulation is to achieve an optimized design. Designing wind turbines is a
complex process and many parameters affect both power generation and
structural performance. Numerical simulation allows turbine developers to
balance multidimensional targets with multifaceted constraints.
While there are attractive and obvious reasons to use numerical sim-
ulation to assist the design process, special care is required to formulate
the approach to achieve accurate and meaningful results. Many steps have
been explained in this chapter, starting with the proper selection of the
solution domain, the mesh-independence study for mesh quality, the cor-
rect translation of the design inputs as boundary conditions, and ending
with the choice of an appropriate numerical solver to obtain the results.
Thoughtful evaluation of the selected outputs is necessary to identify the
degree of impact by each of the design parameters. Through the design
iteration cycle, the determination of a possible range of the design parame-
ters affects the final attainment of the optimal design and the time required
148  • SMALL-SCALE WIND POWER

for this process. Therefore, careful planning for each of the aforemen-
tioned steps is recommended.

6.6 ACKNOWLEDGMENTS

The author would like to express appreciation to Janet Pau and Wai Heng
Ken Cheng for their support for this chapter. In addition, the author would
like to thank professor Ephraim Sparrow for his friendship and inspiration
for professional excellence.

REFERENCES

[1]. International Electrotechnical Commission; IEC 61400-2: 2006 Ed 2.0

Design requirements for small wind turbines.
[2]. International Electrotechnical Commission; IEC 61400-11: 2012 Ed 3.0
Acoustic noise measurement techniques.
[3]. International Electrotechnical Commission; IEC 61400-12-1: 2005 Ed 1.0
Power performance measurements of electricity producing wind turbines.
[4]. International Electrotechnical Commission;IEC 61400-14: 2005 Ed 1.0
Declaration of apparent sound level and tonality values.
[5]. International Electrotechnical Commission; IEC 61400-21: 2008 Ed 2.0
Measurement and assessment of power quality characteristics of grid con-
nected wind turbines.
[6]. International Electrotechnical Commission; IEC 61400-22: 2010 Ed 1.0
Conformity testing and certification.
[7]. International Electrotechnical Commission;IEC 61400-23: 2001 Ed 1.0 Full
scale structural testing of rotor blades.
[8]. http://www.smallwindcertification.org
[9]. http://www.microgenerationcertification.org/mcs-consumer/product-search
.php
[10]. http://www.dawt.dk/DK/Godkendte_small_WT.htm
[11]. Bazilevs, Y., Hsu, M.C., Akkerman, I., Wright, S., Takizawa, K., Henicke,
B., Spielman, T., and Tezduyar, T.E. (2011). 3D simulation of wind turbine
rotors at full scale. Part I: Geometry modeling and aerodynamics. Interna-
tional Journal for Numerical Methods in Fluids, 65, 207–235.
[12]. Moa, J.O., Choudhrya, A., Arjomandia, M., and Lee, Y.H. (2013). Large
eddy simulation of the wind turbine wake characteristics in the numerical
wind tunnel model. Journal of Wind Engineering and Industrial Aerody-
namics, 112, 11–24.
[13]. Gómez-Iradi, S., Barakos, G.N., and Munduate, X. (2010). A CFD investi-
gation of the near-blade 3D flow for a complete wind turbine configuration.
2010 European Wind Energy Conference & Exhibition, EWEC, Warsaw.