Chapter-2

WIND TURBINE FUNCTIONING AND THEIR

AERODYNAMICS

2.1 Wind Power Technology

Wind turbines are the largest rotating machines on earth. The World’s
largest passenger airliner, the Airbus A-380 has a wing span of about 80 m,
where as Enercon E-126, 6 MW wind turbine has a rotor diameter of 126 m.
This HAWT is 138 m tall and has an overall height of 200 m from tower base
to blade tip. The HAWT are also the oldest devices for exploring the energy of
wind on land. In fact, the only older device for utilizing wind energy is the
sailing ship. For centuries, mankind was in need of mechanical energy, mainly
for agriculture, and increasingly for other purposes. In order to extract a useful
form of energy from wind that is free flowing everywhere, proper machines
were necessary and that is how windmills were born.

Figure 2.1 Principle of wind energy (Source: AWEA)

Chapter 2

The principal of wind energy is shown in Figure 2.1

The wind turbine blades are subjected to a force distribution that results
as mechanical torque at the rotor shaft. In modern HAWT the aerodynamic
driving force is mainly the lift force as in the aircraft flight, rather than drag
force used in ancient sailing ships. The shaft transfers the torque from the
blades to the generator, but these passages can be achieved through gear boxes.
Today after decades of industrial development, modern turbines transform the
wind energy to electric power (Knight and Carver, 2010). The knowledge about
the wind power technology has increased over the years. The development of
various windmill types from middle-ages to the industrial era can hardly be
considered as the result of systematic research and development. In the 17th and
18th centuries, when physical–mathematical thinking became more established,
Leonardo, Leibnitz, Bernoulli and Euler involved themselves in systematic
results on windmill technologies. Among the others, Danish professor Poul La
Cour carried out extensive scientific research in windmill blade aerodynamics
and windmill design during the second half of the 19th century, later he
experimented on how to generate electrical current with the help of wind power.
This marked the turning point from historical windmill building, to the modern
technology of power generating wind turbines. Lanchester and Betz were the
firsts to predict the maximum power output of an ideal wind turbine. The huge
and still increasing sizes of HAWT pose a real challenge to the design
engineers.

2.2 Functioning of HAWT

In this section the basics of the functioning of modern HAWT are briefly
summarized. The transformation of wind energy into electricity is not obtained
directly, but consists of many complicated steps. The latest technologies of

12 Division of Safety and Fire Engg., SOE, CUSAT

 Wind Turbine Functioning and Their Aerodynamics

aerodynamics, mechanics, control systems and electro technologies are

involved in such processes. The present research work deals with Horizontal
Axis Wind Turbines (HAWT).When compared to the Vertical Axis Wind
Turbines (VAWT), the HAWTs have

• Higher values of power co-efficient.

• They operate in the high atmospheric boundary layer instead of the

bottom shear layer, so they experience more uniform and intense wind
speeds.

• The HAWT have more stable mechanical behaviors and they can be
constructed in larger sizes.

The most commonly used wind turbine setups are of the following types:

1. A classical turbine with gearbox, i.e. the so-called Danish concept. A

gearbox with typically three steps connects the rotor shaft (the slow one)
to the generator shaft (the fast one), which is directly connected to the
electrical grid and therefore has a fixed rotational speed. The old type
configurations do not have an active aerodynamic control, but blades are
stall-regulated while wind changes.

2. A turbine without gearbox, with the axle directly connected to the

generator. It is a newer concept, compared to the previous one where a
large multi-pole generator allows the unique shaft to rotate with the low
speed of the rotor.

3. A hybrid concept, which combines the previous two is obtained by

adopting a smaller generator and a smaller (usually one-step) gearbox
for matching the required angular speed.

Division of Safety and Fire Engg., SOE, CUSAT 13

Chapter 2

Further, the following four concepts can be distinguished with regard to

the rotor- generator grid coupling (Reisenthel et al., 2003).

• Type A - fixed speed wind turbine concept corresponds to the “Danish

concept”. In this type of wind mill a squirrel cage induction generator
(SCIG) is directly connected to the grid through a transformer.

• Type B - variable speed wind turbine concept with variable rotor

resistance.

This type of wind mill uses optislip induction generator, where the rotor
winding of the generator is connected with an optically (without slip-rings)
controlled resistance. By changing the resistance the slip changes and thus the
power output can be varied in the range of 0-10 % of the synchronous speed.

• Type C - variable speed wind turbine concept with partial-scale power

converter.

This configuration denotes the variable speed controlled wind turbine

with doubly fed induction generator (DFIG) and pitch control. In this type the
stator is directly connected to the grid, while the rotor is connected through a
partial-scale power converter. The variable speed range is +/- 30 % around the
synchronous speed.

• Type D - variable speed wind turbine concept with full-scale power

converter.

This configuration of HAWT corresponds to the full variable speed,

pitch-controlled with the generator connected to the grid through a full-scale
power converter. The generator can be excited electrically (WRSG or WRIG)
or by permanent magnets (PMSG) or squirrel cage induction generator (SCIG).
Typically, a direct driven, multi-pole synchronous generator (no gearbox) is
used in this type. This concept has full control of the speed range from 0 to

14 Division of Safety and Fire Engg., SOE, CUSAT

 Wind Turbine Functioning and Their Aerodynamics

100% of the synchronous speed (Vermeer et al., 2003). However, it has a higher
power loss compared to Type C concept, as all the generated power has to pass
through the power converter.

Thus it is clear that the aerodynamic control and regulation systems

allow the turbine to follow the variable wind. All modern turbines have variable
pitch of 100. As wind intensity increases the pitch changes to follow the best
aerodynamic flow angles along the blades and angular speed increases for
absorbing the torque. The turbine must also yaw into the wind. For attaining
this, on the top of the nacelle an anemometer together with a wind vane which
determines intensity and direction of the incoming wind are kept. The output
signal commands the control electric motors of the HAWT.

The start-up wind velocity (cut-in) of a wind turbine is about 3 to 5 m/s.

While the wind speed increases the power production also grows along with it,
till the rated power is reached and then the exceeding output is limited. At a
fixed value of wind speed (cut-off), typically around 25 m/s the turbine stops
for safety reasons (standstill). For the correct functioning of a wind turbine,
control systems are needed till rated power is reached, these control systems
should make the turbine work with the maximum efficiency, whereas after the
rated or peak power is reached the target of the control system is to keep the
power generation as a constant (Sohn et al., 2005, Timmer and Rooij, 2003).
The various control systems are grouped into three families

• Stall control

This control system is a passive way of controlling the power output of a

wind turbine. The blades are aerodynamically shaped in order to work with
optimum angle of attack at the average wind speeds whereas, at higher speed
the flow angle increases, separation occurs on their sections and that is how lift

Division of Safety and Fire Engg., SOE, CUSAT 15

Chapter 2

force is limited. As wind keeps increasing, the lift force of the blades also
decreases because of increasing stall. The exceeding loads are in such a way to
completely dissipated and no possible control on the efficiency could be done in
the low-speed operating range. No rotational speed control is here available.
Small and economical turbine types adopt this regulation, as well as the older
models (Johnson, 1985).

• Pitch control

In this method of control system the blades can be pitched, i.e. turned
around their axes by means of hydro dynamic or electrical motors. In such a
way they follow the optimum angles till the rated power speed is reached and
then the loads are limited by loosing lift forces (Glauert, 1948).

• Active stall

In active stall control system stall is triggered intentionally. When the

wind mill is in stall only very small pitch excursions are required to give shorter
response times than those of pitch control. However, stall is an unstable
operating condition and thrust is allowed to remain large.

The significant traits of a wind turbine are summarized below (Glauert,

1948).

o Number of blades: 1, 2, 3, 4

o Rotational speed: 20-50 RPM

o Tip-speed (ωR): less than 70 m/s for noise limitation

o Rated power: from 1 kW (Micro-turbine) to 10 MW (Multi–MW WT)

o Control system: stall-regulation, active-pitch, variable speed (rotational)

o Tower-height/Rotor-diameter ratio: H/D

16 Division of Safety and Fire Engg., SOE, CUSAT

 Wind Turbine Functioning and Their Aerodynamics

The increasing penetration of large wind turbines into electrical power

systems is inspiring continuously the designers to develop both custom
generators and power electronics, and to implement modern control system
strategies. The wind turbine design objectives have thus changed over the years
from being conventional-driven to being optimized-driven, within the operating
regime and market environment. In addition to becoming larger, wind turbine
design concepts have been progressing from fixed speed, stall-controlled and
drive trains with gearbox to variable speed, pitch-controlled with or without
gearbox (Patel, 1999).

2.3 A Brief Review of the Aerodynamic Models

The size of commercial wind turbines has increased dramatically in the
last 25 years, from a rated power of approximately 50 kW and a rotor diameter
of 10 to 15 m up to today’s commercially available 5-10 MW machines with a
rotor diameter of more than 120 m (Liu and Jameson, 1992, Reisenthel et al.,
2003). This development has forced the design tools to change, from simple
static calculations assuming a constant wind to dynamic simulation software
that uses unsteady aerodynamic loads to model the aero elastic response of the
entire wind turbine construction (Johnson, 1985).

Here, the basic theories for different classical aerodynamic models are
treated. The following models will be reviewed in the present section.

o Blade Element Momentum method (BEM).

o Lifting line, panel and vortex methods.

o Actuator disc method.

o Navier-Stokes solvers.

Division of Safety and Fire Engg., SOE, CUSAT 17

Chapter 2

2.3.1 Basic Definitions

• Tip speed ratio

The tip speed ratio is a very important parameter for wind turbine
design. It expresses the ratio between the tip speed and the undisturbed wind
speed.

λ = ωR/U∞ ………………….2.1
Where, ω is the angular velocity, R the radius of the blades and U∞ the
undisturbed wind speed. The tip speed ratio dictates the operating condition of a
wind turbine and it affects a number of flow parameters that will be discussed
later. Typical values for modern turbines are 6 to 8. For tip speed ratio at radius
r, it is common to use r instead of R in equation 2.1.

• Solidity

The turbine solidity is defined as the ratio between the blade areas to the
area of the disk.

σ= ……………….2.2

And like the local tip speed ratio the local turbine solidity is also
defined, while referring to certain radius r, as follows

σ= ………………….2.3.

Here B is the number of blades and c the chord length of the local blade
section (Patel, 1999).

Power coefficient

The kinetic energy of an air mass m moving at a free stream velocity U

can be expressed as

18 Division of Safety and Fire Engg., SOE, CUSAT

 Wind Turbine Functioning and Their Aerodynamics

E = mU2 ………………….2.4

And for unit mass

e = U2 ………………….2.5

Taking a certain cross–section area A, by which the air passes at the

velocity U, the volume Q during a certain time unit is

Q = UA ………………….2.6
And the mass flow with the air density ρ is

m=ρQ ρUA ………………….2.7

With the product of the kinetic energy of the moving air and the mass
flow, one yields the amount of energy passing through a cross–section A per
second. It is physically the power P (Johnson, 1985, CWET 2010).

P0 = e = ρU3A ……………….2.8

This is the available power in the wind. However, if we had taken the
whole energy, the flow would be arrested. Thus there should be a power
coefficient CP that determines the portion of the flow energy captured by the
wind energy converter (Liu, 2010, Johnson, 1985).

P = CPP0 CP = ……………….2.9

According to Betz, CP can never exceed a value of CP max = 16/27

(Overgaard & Lund, 2010).

Division of Safety and Fire Engg., SOE, CUSAT 19

Chapter 2

• Thrust coefficient

In a similar way the thrust coefficient could be defined for the integral
thrust force acting on the rotor T (Velte et al., 2008, Selig and Tangler, 1999).

CT = ………………….2.10

Induction factors

The incoming wind flow is affected by the turbine blade as it approaches

the rotor. The effective apparent velocity seen by the blade vortex system is
modified as illustrated in Figure 2.1.

ωr

Figure 2.2 Relative velocity concepts

By decomposing the velocity vector into its axial or stream wise and
tangential component and introducing the induction factors, we get these
relationships.

20 Division of Safety and Fire Engg., SOE, CUSAT

Chapter 2

Γ = ∫ uds ………………….2.14

Lift and drag coefficients define the characteristic of an airfoil (Johnson,

1985).

Figure 2.3 BEM annular control volumes.

L
Cl = ………………….2.15
0.5ρU 2 C

D
Cd = ……………….......2.16
0.5ρU 2 C

Where L and D forces per unit span, the flow speed and C the chord
length. These coefficients are function of the angle of attack, Reynolds number
and Mach number (Patel, 1999).

• Pressure coefficient

Dealing with airfoils, pressure coefficient is often defined as follows

CP = −
P - P∞
………………….2.17
0.5ρ U2
This is an inverse to the general definition, in order to have the positive
suction, which is the main contribution to lift generation (Overgaard and Lund,
2010; Nelson et al., 2008; Glauert, 1948).

22 Division of Safety and Fire Engg., SOE, CUSAT

 Wind Turbine Functioning and Their Aerodynamics

2.3.2 The BEM Method

Loads and performance calculations of wind turbines are today routinely

performed by the Blade-Element Momentum method. The method is indeed
computationally cheap and thus very fast, even with providing very satisfactory
results. The BEM method consists of dividing the flow in annular control
volumes and applying momentum balance and energy conservation in each
control volume. The annuli are bounded by stream surfaces that enclose the
rotor and extend from far upstream to far downstream. Basic assumptions of
this method are that, the induced velocity in the rotor plane is equal to one half
of the induced velocity in the ultimate wake, and that the flow can be analyzed
by dividing the blade into a number of independent elements. Moreover the
loads for each blade are uniformly distributed azimuth-wise, which means the
rotor would have an infinite number of blades. For each blade element, the
aerodynamic forces are obtained using tabulated airfoil data, which stem from
wind tunnel measurements and corrected for three-dimensional effects. The
BEM could be a design as well as verification method and when using BEM as
a design method, the following inputs and outputs are defined:

• Input:

Rated power P, power coefficient CP, mean wind speed V0, number of
blades B

• Output:

Rotor diameter D, chord Ci(r) and twist θi(r) radial distribution.

This is in principle the BEM method, but in order to get good results it is
necessary to apply, at least, the following two corrections.

Division of Safety and Fire Engg., SOE, CUSAT 23

Chapter 2

• The first is called Prandtl’s tip loss factor, which corrects the
assumption of an infinite number of blades.
• The second correction is called Glauert’s correction and is an empirical
relation between the thrust coefficient CT and the axial interference
factor for values greater than approximately 0.3, where the relation
derived from the one-dimensional momentum theory is no longer valid.

Since the BEM model is simple and very fast to run on a computer, it
has gained an enormous popularity and is today the common design code in use
by industry. However, owing to limitations of representing all the various flow
situations encountered in practice, it has become necessary to introduce
different empirical corrections. Such situations include phenomena related to
dynamic inflow, yaw misalignment, tip loss and heavily loaded rotors. Thus
there exists an increasing need to establish experiments and to develop more
advanced models to evaluate the basic assumptions underlying the BEM model.
Although today there seems to be no realistic alternative that may replace the
BEM model as an industrial design tool, a number of more advanced in viscid
models have been developed to overcome the limitations of the momentum
approach (Liu, 2010; Madsen et al., 1998; Johnson, 1985).

2.3.3 Lifting Line, Panel and Vortex Methods

Three dimensional inviscid aerodynamic models have been developed in

an attempt to obtain a more detailed description of the 3D flow that develops
around a wind turbine. The fact that viscous effects are neglected is certainly
restrictive as regards the usage of such models on wind turbines. However, they
should be given the credit of contributing to a better understanding of dynamic
inflow effects as well as providing a better insight into the overall flow
development. There have been attempts to introduce viscous effects using

24 Division of Safety and Fire Engg., SOE, CUSAT

 Wind Turbine Functioning and Their Aerodynamics

viscous–inviscid interaction techniques. But they are not used in wind turbine
functioning and aerodynamics, as they are not yet reached the required maturity
so as to become engineering tools, although they are full 3D models that can be
used in aero elastic analyses. In vortex models, the rotor blades, trailing and
shed vortices in the wake are represented by lifting lines or surfaces. The vortex
strength on the blades is determined from the bound circulation that stems from
the amount of lift created locally by the flow past the blades. The trailing wake
is generated by the span wise variation of the bound circulation, while the shed
wake is generated by a temporal variation, and ensures that the total circulation
over each section along the blade remains constant in time. Knowing the
strength and position of the vortices, the induced velocity w(x) can be found at
any point using the Biot-Savart induction law (Liu and Jameson, 1992, Patel,
1999).

w(x)= ……………….2.18

Where w is the vortices, the point x where the potential is computed and
x′ is the point of integration, which is taken over the region where the vortices
is non-zero. In the lifting lines models, the bound circulation is found from
airfoil data table, just as in the BEM method. The inflow is determined as the
sum of the induced velocity, the blade velocity, and the undisturbed wind
velocity (Sohn et al., 2005, Knight and Carver, 2010).

The relationship between the bound circulation and the lift is the well-
known Kutta-Joukowski theorem, and using this together with the definition of
the lift coefficient, a simple relationship between the bound circulation and the
lift coefficient can be derived.

Division of Safety and Fire Engg., SOE, CUSAT 25

Chapter 2

L= ρUrel = = …….2.19

In its simplest form the wake from one blade is prescribed as a hub
vortex plus a spiraling tip vortex, or as a series of ring vortices. Actually, one of
the most recurrent applications of vortex methods has been wake dynamics. The
inviscid and incompressible flow past the blades can be found by applying a
surface distribution of sources and dipoles. This is the approach followed by
panel methods (Liu and Jameson, 1990, Velte, 2008). The background is
green’s theorem which allows obtaining an integral representation of any
potential flow field in terms of singularity distribution. An external potential
flow field is given, possible varying in time and space, to which is added a
perturbation scalar potential. The active boundary of the flow includes both the
solid boundaries as well as the wake surfaces. Dipoles are added so as to
develop circulation into the flow to simulate lift.

2.3.4 Actuator Disk Method

The actuator disk model is probably the oldest analytical tool for
evaluating rotor performance. In this model, the rotor is represented by a
permeable disk that allows the flow to pass through the rotor, at the same time it
is subjected to the influence of the surface forces. The classical actuator disk
model is based on conservation of mass, momentum and energy, and constitutes
the main ingredient in the D momentum theory, as originally formulated by
Rankine and Froude. Combining it with a blade-element analysis, we end up
with the classical Blade-Element Momentum Technique by Glauert. In its
general form, however, the actuator disk might as well be combined with the
Euler or NS equations. Thus, as will be told in the following, no physical
restrictions have to be imposed on the kinematics of the flow. In a numerical
actuator disk model, the NS (or Euler) equations are typically solved by a

26 Division of Safety and Fire Engg., SOE, CUSAT

 Wind Turbine Functioning and Their Aerodynamics

second order accurate finite difference/volume scheme, as in a usual CFD

computation. However, the geometry of the blades and the viscous flow around
the blades are not resolved. Instead, the swept surface of the rotor is replaced by
surface forces that act upon the incoming flow. The main limitation of the axi
symmetric assumption is that the forces are distributed evenly along the
actuator disk; hence the influence of the blades is taken as an integrated
quantity in the azimuth direction. To overcome this limitation, an extended 3D
actuator disk model has recently been developed. The model combines a 3D NS
solver with a technique in which body forces are distributed radialy along each
of the rotor blades. Thus, the kinematics of the wake is determined by a full 3D
NS simulation, whereas the influence of the rotating blades on the flow field is
included using tabulated airfoil data to represent the loading on each blade. As
in the axis symmetric model, airfoil data and subsequent loading are determined
iteratively by computing local angles of attack from the movement of the blades
and the local flow field. The concept enables one to study in detail the dynamics
of the wake and the tip vortices and their influence on the induced velocities in
the rotor plane (Dahl et al., 1998; Timmer and Van Rooij, 2003; Patel, 1999).

The main motivation for developing such types of model is to analysis

and verifies the validity of the basic assumptions that are employed in the
simpler and more practical engineering models.

2.3.5 Navier-Stokes Solvers

The first applications of CFD to wings and rotor configurations were

studied way back in the way late seventies and early eighties, in connection
with aircraft wings and helicopter rotors, using potential flow solvers.

A shift towards unsteady Euler solvers was seen through the 80’s. Full-
Reynolds Averaged Navier Stokes (RANS) equations were solved first for

Division of Safety and Fire Engg., SOE, CUSAT 27

Chapter 2

helicopter rotor computations, including viscous effects, in the late eighties and
early nineties, but the first full NS computations of rotor aerodynamics were
reported in the literature in the late nineties (Knight and Carver, 2010).

Recently, Detached Eddy Simulations (DES) and Large Eddy

Simulation (LES) techniques, which are becoming increasingly important in
several engineering fields, have been directed at solving turbine’s flow even
though high computational efforts still remain a drawback for a practical usage.

Nowadays, an extensive literature on N–S CFD studies on wind turbine

exists. Since this method has been chosen for the study of WT aerodynamics, a
more comprehensive description of the CFD methodologies will be discussed in
the forth coming chapters.

Figure 2.4 Wind power distributions

(Source www.WINDPOWER.org 1998)

28 Division of Safety and Fire Engg., SOE, CUSAT

 Wind Turbine Functioning and Their Aerodynamics

In the Figure 2.4 the colors indicate the wind power distribution as
stated below.

Gray: Total Potential Wind Power, Blue, Usable Power (Betz’ Law) and
Red: Turbine Power Output

The present work aims at maximizing the turbine power output to match
with the Betz limit.

2.4 Literature Review on Aerodynamics

Liu and Jameson (1990) carried out an experimental study over force
constants, speed/power relationship and velocity contours using the panel
method. Liu and Jameson (1992) predicted the end wall effect boundary layer
condition of flow analysis. Castino et al., (1976) developed a new model which
treated both the wind speed and wind direction dynamics. Their analyses were
made over stochastic properties of wind speed with the help of two models (i)
Markov chain model and (ii) discrete auto regressive model of order 2
(DAR(2)). In their analyses they concluded that, it was impossible to describe
the time series of the wind velocity.

The solution space was parameterized by Mach number, pitch angle,

grid resolution, and solution algorithm settings were addressed in the paper by
Patrick et al., (2003). The static load carrying capacity, buckling stability, blade
tip deflection and natural frequencies at various rotational speeds and fatigue
strength were evaluated by Sohn et al., (2005) to satisfy the strength
requirements in accordance with the IEC 61400-1 and GL Regulations.

The aim of the research carried out by Velte (2008) was to investigate
the effect of vortex generators in a turbulent, separating, low Reynold’s number
(Re 20000) boundary layer over geometry which was similar to flow past a
wind turbine blade. A qualitative comparison with LES simulations of the flow

Division of Safety and Fire Engg., SOE, CUSAT 29

Chapter 2

behind vortex generator was made which showed that LES was capable of
reproducing flow structure.

In their report Roberl et al., (2008) proposed a methodology called

PACE (Plasma Aerodynamic Control Effectors). This paper severs as an
experimental report on significant aerodynamics characteristics.

Overgaard and Lund (2010) have carried out a Finite Element Analysis
of about the geometric nonlinear and inter laminar progressive failure of a
generic wind turbine blade which was undergoing a static flap wise load.
Results were compared with the experimented findings and the aero dynamical
loading and thrust were found numerically.

Knight and Carver (2010) showed that the energy capture of STAR
rotor was 12 % for long run and 36 % for short run with the measured results of
3322 and 2524 data points respectively, when compared to other rotors run
simultaneously.

30 Division of Safety and Fire Engg., SOE, CUSAT