Modeling and Simulation of an emulator of a Wind Turbine

Speed ​Using Vector Control of a Three-

phase Induction Motor

Andresa KR Sombra¹ Domenico Sgrò² Ruth PS Leão³, Raymond F. Sampaio 4, Philip C. Sampaio 5
Electrical Engineering Department
Federal University of Ceará
Fortaleza, Brazil
andresakelly@hotmail.com¹, domenico@dee.ufc.br², rleao@dee.ufc.br³, rfurtado@dee.ufc.br 4, felipecsamp@dee.ufc.br 5

Abstract - Among the renewable energy sources, wind power has They will depend on the wind characteristics at the installation site. In addition, the
Gained a prominent position worldwide due to the Increasing application system is not portable and is extremely expensive.
in the conversion into electric energy. The wind electricity generation has
A resource used for studies without the need for a wind turbine itself is the
Become the object of study of the impacts on the power grid of the
generation with variable output power. This paper AIMS to model and creation of a wind turbine emulator. The development of this emulator solves
simulate the wind turbine emulator using the three-phase induction motor problems of space, cost and portability, as well as allowing testing for different
(TIM) of squirrel cage rotor Which operates with variable speed. The direct wind characteristics may be used, including wind speed measurement data.
field-oriented control (DFOC) Has Been Applied to control the speed TIM.
The flux observer was used the current-voltage model. Simulations in the
MATLAB / Simulink simulator, using real wind speed data, made the full
system validation.
Aiming to create a test platform of an Aeolian-electric system, this paper
aims to model and simulate the speed control of a three-phase induction motor
(MIT) of 15 kW, which is capable of emulating a wind turbine. MIT will, at
another time, coupled to the shaft of a dual power 10 kW induction generator,
allowing testing under different operating conditions.
Keywords- wind turbine; emulator; three-phase induction motor control
DFO

I I. INTRODUCTION II. STATE OF ART

The emergence of policies for environmental sustainability and security of Several studies in the literature dealing with the control of electric motors in
energy supply has intensified the search for renewable energy, aiming to lower order to represent the wind speed. In [4], the author develops a wind turbine
environmental impacts and economic viability. In this sense, the generation of emulator using a synchronous permanent magnet motor (PMSM) and a
electricity from wind energy has global importance in the last 10 years. The synchronous generator of permanent magnet (PM), focusing on the dynamic
growth of the installed capacity of wind farms in the world, grew by 12.4% in compensation of the torque reference used in motor. However, according to an
2016 compared to 2015. In addition, the share of renewable sources in the in-depth study by [5], the use of a PMSM makes the proposed system even
global matrix increased from 10.3% in 2015 to 11.3 % in 2016 [1]. In terms of more expensive, which goes against the objective of the work, which is to create
electricity production in 10 years (2005 to 2015), the increase was over 700% a low-cost system.
[2].

In [3], the author develops an emulator using a MIT coupled to a PM,

producing even an analytical approximation of the curve which relates the power
Because of this descent, the wind generation has been the subject of coefficient ( )
several studies considering factors that can influence the power control, such the specific speed (). The control technique used to control the MIT was the
as: type of turbine technology, type of electrical generator and control strategies control climb volts-per-hertz / Which does not allow high dynamic performance in
of static converters associated [3]. motor control [6], unlike the vector control technique used in this work.

Creating a test for analyzing the behavior of a wind turbine environment and
its influence on the electric power production involves difficult task, because it In [7], the author controls a DC motor (direct current) coupled to a squirrel
requires a space for the installation of the turbine and the tests cage induction machine. However, the DC motor requires components such as
a commutator and brushes,
making it the most expensive motor and requiring frequent maintenance. In the term is given by (4).

The engine used for speed control in this work is a MIT, considered cheap
() = - (4)
compared to the other cited, reliable, robust and simple construction. ,

Furthermore, the control technique implemented is a vector control technique,

ensuring high dynamic performance. Fig. 1 shows the curve described by (3) for different values ​of pitch angle,
.

III. M ODELAGEM OF T URBINA And ÓLICA AND MIT

THE. Wind turbine

The extracted power of wind is given by equation (1).

= (,) (1)

On what:

power extracted by the wind turbine; sweeping

area of ​the rotor blades (m²);

λ specific speed ( tip speed ratio) ( PU); blade pitch angle

(°); Fig. 1. Variation of the power coefficient for different pitch angle values.

(,) turbine power coefficient.

The maximum power coefficient, i.e. yield of the turbine model used is
The specific speed is calculated by the ratio between the linear speed at the
41.1%, and occurs when the pitch angle is equal to 0 ° and the specific speed is
7.95.
blade tip and the incident wind speed, according to (2). =

The extracted power curve of the wind turbine in angular velocity function is
(2) plotted, from (1) and (3), for different values ​of wind speed, as shown in Fig. 2,
taken from [9].
On what:

angular velocity of the blade tip (rad / sec); radius of

the wind turbine (m); Incident wind speed (m / s);

In (1) it is seen that the power extracted from the wind depends on the
turbine power coefficient , which describes
the wind turbine efficiency. It is nonlinear and varies with the
Specific speed and the pitch angle of the blades . The power coefficient of a wind
turbine may be obtained by different alternatives as a method of paddle
elements, lookup table ( lookup table) analytical approach and [8]. In the
literature different analytical approaches are found to

, differing, often only

the coefficients for the function. These variations are used to represent faithfully Fig. 2. power curve of the turbine used for different wind speed values.
a specific turbine [9].

It is highlighted, even in Fig. 2, the line of maximum values ​of power

The approach used in this work, obtained in [9] is described by (3). ( , ) = coefficients for the different wind speed values.

B. Dynamic modeling of induction motor

- 0.4 to 5 ⋅ (3)
For the dynamic model MIT, some considerations are made [10] such that:
the gap is constant; the
rotor windings and stator are equal; the magnetic saturation is disregarded; the MIT mechanical power; MIT
zero-sequence component is zero, since the motor is connected to
mechanical speed; torque developed
non-grounded star; the electrical angle between the windings are equal; the
magnetic losses are disregarded. by MIT.

The relationship between the synchronous speed and MIT is given by (11).
The equations representing the voltages and currents in a squirrel cage MIT
the dq reference are described in (5) and (6) [11] .The reference speed used for
dq axes of their own angular velocity of the rotor flux.
= (11)

On what:
=. + + . (5)
MIT number of pole pairs. Equating (9) and (10), one obtains the
relation given by (12) to the torque of the machine.
= 0. + + (-). (6)

On what:
=. × (12)
stator voltage vector on the dq rotating reference; vector of stator

currents in the rotating dq reference; Vector rotor currents in the On what:

rotating dq reference; equivalent resistance of the stator; equivalent relationship of the mutual inductance divided by
rotor self inductance .
resistance of the rotor; angular speed of the rotor flux; synchronous
The technique used for the speed control is the vector control technique of
engine speed; the type field-oriented control (FOC), which is based on the rotor flux setting the
d-axis. Thus, the relationship to the torque given by (12) can be written as (13).

Vector stator flows in rotating dq reference; Vector stator flows in

= . . . (13)
rotating dq reference. The relationship between the rotor and flows MIT

and stator currents are given by (7) and (8) [11]. On what:

d flow in the rotor axis; stator current

in the q axis.
=. + . (7)

IV. C ONTROL OF C AMPO The RIENTADO D IRETO (DFO) The use of

FOC and enables smooth operation of the engine under wide speed range,
=. + . (8) generate torque at zero speed, and have high dynamic performance, makes it
possible to reduce the dynamic model of the machine three-phase induction in
On what: a model of a dc motor.

idutância own stator; rotor self

In the dc motor (MCC), the switch action makes the control simple,
inductance;
considering that becomes mechanically fixed and orthogonal to the angle
mutual inductance between the rotor and the stator. The mechanical between the magnetomotive force and the machine stream [10]. Thus, the MCC
power of the machine can be written in two different ways, according to (9) and course causes the decoupling between the control torque and flux, which is the
goal of creating the FOC control [12] cited in [10].
(10).

= × (9) The FOC is basically a set of conditions that secure the stator current vector
and the rotor flux vector, or flux of the machine [10]. For this reason, it is called
control for guidance flow.

= . (10)
The field-oriented control has two classifications: Direct FOC (DFOC) or
indirect FOC (IFOC). this classification
On what:
It depends on the information used to reference the control to the field On what:
coordinate system [12] cited in [10]. At first, the flow of the machine module and
stator voltage vector in the αβ reference; vector of stator
in angle should be measured directly on the machine by using sensors, or can
be estimated by means of mathematical models that relate quantities other
currents in the αβ reference; Vector rotor currents in the
machine with the flow measurement. In the second, the information is obtained
by calculating the slip between the magnetic field of the stator and the rotor [10]. αβ reference; Vector stator flows in αβ reference; Vector

rotor flows in the αβ reference; constant rotor time;

The flow in the case of DFOC control, need to be replenished, either by

measurements or estimations. Despite this feedback, usually performed by
means of estimators flows, require a greater complexity and computational
effort, it allows you to control the machine in closed loop flow. Given that the = 1 -. .
control technique used in this work is the DFOC type and the cost of using flow
sensors is high, it is necessary to use a flux estimator, also called flow observer. In this work, it was decided to combine estimators, forming the
voltage-current model in order to obtain a high dynamic performance,
considering that the voltage model works best for high speed and the current
model, at low speeds [ 14].

Fig. 3 shows the diagram of the estimator used. The controller used was a
THE. Modeling of flow observer
proportional integral PI controller that provides alternating current voltage model
For DFOC control, are required module information flow, used for control of
for the model around 10 Hz.
this closed loop, and the angle between the vector flow and stationary axis α,
and is used to coordinate changes made in the control scheme.

Knowing that the rotor flux corresponds to the resultant vector of the flow in
the stationary axes α and β (transformed Clarke [13]), the module and the
estimated angle of the flux vector can be calculated according to (14) and (15)
respectively.

= + (14)

= tan (15)

On what:

estimated rotor flux modulus; estimated rotor flux in the Fig. 3. Model combined flow estimator.

stationary axis α; estimated rotor flux in the stationary shaft

B. Specification of the controllers
β; angle estimated rotor flux.
Considering that the control DFOC provides a decoupling between the
control flux and torque, each of the controllers, flow and speed, outputs the
stator current in one of the dq axes. Thus, the current control should be much
The estimation of the rotor flux can be made using the equation model for faster than the flow controls and velocity, so as to avoid interference between
the current estimator according to (16) or by the voltage model equation the controls.
according to (17).

The current control is designed for a bandwidth around 1000 Hz. Given the
0= -. + + (16)
current controller of speed, flow control is designed to be 5 times slower,
ensuring the tracking reference almost instantaneous current control .

= - -. . . (17)
 The speed control is designed for a bandwidth of 6 Hz, considering the current control, flow speed and is shown in Fig. 4.
mechanical limitations of a real machine. The complete diagram of control,
containing the

Fig. 4. Diagram complete control of vector control speed.

V. R UTCOME OF S IMULAÇÃO

For the simulation, the Simulink / MATLAB was used, using the three-phase
induction motor model with the parameters described in Table 1 and wind
turbine in Table 2.

They simulated the speed control response to data measured winds in the
city of Camocim, CE - Brazil, at a height of 40.2 m [15]. For the simulation, we
used wind data month of August of 2005.

Feature Value

power 15 kW

Line Voltage 380 Vrms

Chain 30.5 Arms

Torque 8.23 kgf.m Fig. 5. wind speed data.

Yield 92.4%
Fig. 5 shows the wind data used for simulation and Fig. 6, the reference
Power factor 0.81 signal corresponding to MIT.A response of the speed controller, as the reference
Table 1. rated characteristics of the MIT signal, is shown in Fig. 7.

Feature Value Analyzing the speed control response, it is observed that the engine could
power 15 kW get high desired dynamic performance, and proven by the start of the simulation,
Wind speed 9m/s
which is imposed sudden change of speed, rated at 100 rad / s.

Radius of spades 1.23 m

Table 2. Nominal characteristics of the wind turbine

 adequate performance control, for both low
speeds, or transient, and for high speeds.

The control system was tested by using actual wind data, obtaining a
satisfactory result, validating therefore the work proposition.

In the future, we intend to analyze other effects as well as sudden changes

of reference, such as shadow effects of the tower and the turbine mechanical
oscillations.

R EFERENCES

[1] Abeeólica, "Annual report," 2016. [Online]. available:

http://www.abeeolica.org.br/wp- content / uploads / 2017/06 /
424_Relatorio_Anual_2016_final.pdf. [Accessed: 10-Jul-2017]. [2] GWEC,

"Global statistics, "2016. [Online]. available:

http://gwec.net/global-figures/wind-energy-global-status/. [Accessed:
Fig. 6. speed reference signal to the MIT.
10-Nov-2017].

[3] JCF Soltoski PTP dos Santos, CHI and Font, "Development
small scale of the wind turbine emulator work bench, "2016. [4] J. Yan, Y. Feng,

Dong J. and" Study on Dynamic characteristic of

wind turbine emulator based on PMSM, " Renew. Energy, vol. 97, pp. 731-736, 2016.

[5] Santos OWP, "Comparison of the induction motor and motor

permanent magnets comparison between the induction motor and permanent
magnet motor. "UFRGS. Porto Alegre, RS, 2015. [6] R. Reginatto, "Control for
induction motor driven field
with adaptation parameters via MRAC, "no. February. Federal University of
Santa Catarina. Florianópolis, 1993. [7] R. Himani and Dahiya, "Condition
Monitoring of the wind turbine
generator using the standalone wind turbine emulator " Front. Energy,
vol. 10, no. 3, pp. 1-12, 2016.

[8] A. Perdana, "Dynamic models of wind turbines the contribution

Fig. 7. MIT speed control response towards the establishment of standardized models of wind turbines for power
system stability studies, "Chalmers University of Technology, Goteborg, Sweden.,
The following reference is to a reasonable time for operating the real
2008. [9] JB Almada," Modeling, control and management of operating
machine, with a view to the mechanical limitations of the engine.
microrredes with renewable sources, "Federal University of Ceará,
2013.

SAW. W CONCLUSION [10] RA Benvenuti, "Platform for simulation and implementation

control techniques of three-phase induction machines and pentafásicas.
The proposed wind turbine emulation system consists of a low-cost system
"UFRGS. Porto Alegre, RS, 2014. [11] PC Krause, O. Wasynczuk, and SD
that aims to solve the problems of a test platform using a real wind turbine, such
Sudhoff, Analysis of Electric
as: cost, portability, space and the limitation of tests to site system installation.
Machinery and Drive Systems, 2nd ed. Wiley-IEEE Press, 2002. [12] R.

Reginatto and J. Rico, "Modeling of induction motor.".

Federal University of Santa Catarina. Florianópolis, p. 37,
The use of a wind turbine emulator lets you test the performance of an 1993.
Aeolian-electric system under a wide variety of wind characteristics.
[13] E. Clarke, "Circuit Analysis of AC Power Systems: Symmetrical and
related components, "in General electric series, v.1, J. Wiley & Sons,
1950, p. 540.
The used vector control strategy allows a high dynamic performance, so
that the wind speed is represented faithfully. [14] C. Lascu, Boldea I., and F. Blaabjerg, "A modified direct torque
control for induction motor sensorless drive, " IEEE Trans. Ind. Appl.,
vol. 36, no. 1, pp. 122-130, 2000. [15] Secretariat of the State of Ceará
This work presents the modeling of a wind turbine and a three-phase Insfraestrutura, "Data
induction motor with short-circuited rotor. Also presented the modeling of flow anenométricos. " [Online]. available:
estimator using the voltage-current model, which ensures a http://www.seinfra.ce.gov.br/index.php/downloads/category/6- energy.
[Accessed: 10-Jul-2017].