International Conference on Co mputer, Co mmunication, Chemical, Materials and Electronic Eng ineering (IC4M E2), 11 -12

July, 2019

Performance Study of a Grid Connected Permanent

Magnet Synchronous Generator based Wind
Turbine System
Subir Datta1*, Akibul Islam1, Tridip Saikia1, Shuma Adhikari2
1
Department of Electrical Engineering, Mizoram University, Aizawl, M izoram, India
2
Department of Electrical Engineering, NIT Manipur, Imphal, M anipur, India
*[email protected]
Abstract— The performance of a grid connected permanent maximu m power at any wind velocity [8-11]. In this paper,
magnet synchronous generator (PMS G) based wind turbine Perturb and Observe (P & O) algorith m based MPPT is used
(WT) system under varying wind velocity is studied in this to design field oriented vector control scheme of MSC and
paper. To achieve variable speed and constant frequency grid voltage oriented vector control scheme is used to design
operation, two vector control schemes are adopted for GSC. The performance of PMSG based grid connected
controlling both the converters. Field oriented based vector WECS is studied under varying wind velocity at unity power
control scheme is used to control Machine Side Converter factor operation.
(MS C) and grid oriented based vector control scheme is used
to control Grid Side Converter (GS C). Maximum Power Point II. M ODELING OF WIND TURBINE
Tracker (MPPT) is also playing a very important role to
extract maximum power from wind an d hence, perturb an d The wind power is converted to mechanical power by the
observe (P & O) algorithm is used. Simulation of the complete mechanical part of the wind turbine which lies on the rotor
system is done in MATLAB/S imulink platform and time blades of the turbine which rotate at the area (A=πR2 ),
domain results are taken under varying wind velocity and where R= Blade length (m). The mechanical power is given
analyzed those results in details. as [12]:
1 3
Keywords—WECS, PMSG, MPPT algorithm, and Vector Control. P = ρ A C (λ,β ) V (1)
T 2 P w
I. INT RODUCTION
These days renewable energy has a growing energy Where p= Air density (kg/ m), Vw = Wind velocity (m/s) and
demand. Fossil fuels cannot be used as they caused pollution. CP = Po wer coefficient. The value of CP depends on the tip
The conventional energy resources are becoming finite and speed ratio (λ) and the turbine blade p itch angle (β). The
there will be a time when the conventional energy resources expression of power co-efficient (CP ) is given by [12]:
become extinct [1]. Then the renewable energies (like solar,
wind, tidal etc) will p lay a big role. The wind energy is the 2 (- 0.17 λ)
most suitable one because it is the most economically
C = 0.5 (λ - 0.22 β - 0.5) e (2)
p
suitable for producing electricity also it is pollution free. The The tip speed ratio expression is given as:
optimu m wind energy in Wind Energy Conversion system
(WECS) can be harness through wind turbine in case of R ωt
λ= (3)
variable speed [2]. Fro m the wind energy the maximu m Vω
power should be taken out and send it to the grid/load side to
get a desired output. Now-a-days, the variable speed WT-
III. GRID CONNECTED PMSG-WT SYST EM
system is preferred as its energy conversion efficiency is
high [3]. Different types of WT generators are used for The power electronic converter plays a significant role in
conversion of wind energy into electrical energy [3]. PMSG the WECSs. As the WT operates at variable speed according
is being very popular because it has certain merits over to availability of wind velocity, the voltage generated is of
others like maintenance cost is less due to absence of slip - variable magnitude and frequency. The study system consists
ring and brushes, achieving of higher power density, less of a back-to-back power converters decoupled by a dc-link
thermal stress on rotor circuit, absence of dc excitation capacitor. The PMSG is connected to grid through back-to-
system etc [4,5]. There are mainly two types of control back converter topology as shown in Fig.1. The active stator
schemes available: (i) scalar control scheme and (ii) vector power of PMSG is fed to grid but in exchange of reactive
control scheme. The vector control scheme is become very stator power flow does not take place due to dc-link
popular in WECS because it has the ability to achieve capacitor. The LC filter is used to make the dc link voltage
variable speed and constant frequency operation and ripple free. The WT converts the kinetic energy fro m wind to
decoupled control of active and reactive powers is possible mechanical energy. That mechanical energy is to be sent to
that means the speed, frequency, power can be controlled the rotor part of the generator through the gear box. The
independently [6,7]. To design vector control scheme and PMSG convert the mechanical energy to electrical energy.
capture maximu m power under varying wind speed, MPPT The electrical energy fro m PMSG is to be sending to grid via
is playing very vital role. Various MPPT methods are MSC and GSC. To get the maximu m power it should be
available to track the maximu m power fro m wind energy and operate at unity power factor for wh ich the reactive power
in the recent decades these methods are used for capture flow of the system should be maintain to zero. The MSC will
regulate the speed, active stator power and torque of the

978-1-7281-3060-6/19/$31.00 ©2019 IEEE

generator. The GSC is used in order to maintain fixed Which means the magnitude of Ψ q is equal to the magnitude
frequency or constant dc link voltage, so whatever the power of main flu x ψ. The act ive power equation in the dq-axis,
will be at constant frequency. So GSC will be the inverter 3
and MSC will be the rectifier. P= *(V I +V I ) (7)
2 dd qq
As Ψq = ψ then, Vq =0 and Vd =V; because the angle between
voltage and ψ is 90o . So putting Vd and Vq into (7), we get,
Gear
3 (8)
Box Control Scheme Control Scheme P= *(V I )
For MSC for GSC 2 dd
Here, Vd is constant; therefore, the active power P is directly
proportional to Id . So, it is possible to control the active
power by controlling Id .
The reactive power equation is given by:
PMSG MSC GSC Filter
~ 3
Q= *(V I -V I ) (9)
Wind 2 dq qd
GRID As Vq =0, so (9) beco mes:
Turbine
Fig.1.Grid connected PMSG-WT System 3 (10)
Q= *(V I )
2 dq
IV. MPPT A LGORIT HM As Vd is constant, therefore, the reactive power Q is directly
To extract the maximu m power fro m the wind energy, proportional to Iq . By controlling Id , the reactive power (Q)
MPPT algorithm is required, when the wind speed is on a can be control and hence, power factor can also be
large range of wind speed. The Perturb and Observe controlled to its desire value. The detail control scheme, fo r
flowchart, shown in Fig.2, is used to generate the reference MSC, is presented in Fig.4.
value of generator speed for different value of wind velocity
and its corresponding maximu m power is also generated start
fro m the flowchart. Here, the P m(k) and W m(k) are input
active power and the generator speed respectively of the Measure Pm(k) and Wm(k)
MPPT algorith m and these are sensed from the system Calculate
[Fig.1]. Thereafter, the small change of active power ΔPm(k)=Pm(k)-Pm(k-1)
(ΔP m(k)) and rotor speed (ΔW m(k)) are calcu lated by ΔWm(k)=Wm(k)-Wm(k-1)
subtracting delay value of active power (Pm(k-1)) and rotor
speed (W m(k-1)) fro m P m(k) and W m(k) respectively. Yes
After that, the reference value of generator rotor speed |ΔPm(k)<0|
(W mref) is calculated based on the different conditions
between ΔP m(k) and ΔP m(k-1) under varying wind speed, as Wm(ref)=Wm(k)-Wm(k-1) No
shown in Fig.2.
This method deals with two values (new and old values) No
Yes |ΔPm(k)>=ΔPm( k-1)|
i.e. the new reference power and the old reference power, Yes
w.r.t. the reference wind speed. The new value can be No ΔPm(k)>0
|ΔPm(k-1)| <0
carried out by taking an in itial value and subtract the initial Yes No
value fro m the first value the new value will be found and Yes ΔWm(k) >0
No
continue the process till the maximu m power is obtained. Kopt= ΔPm(k-1) / Wm(k-1)^3
Wm(ref)(k)= (Pm(k)/Kopt)^1/3 ΔWm(k)>0
V. CONT ROL SCHEME OF MSC No
ΔWm(ref)(k)=ΔWm(k-1)+(D*ΔPm(k)) Yes
The MSC is used to extract the maximu m power
corresponding to its reference generator speed under varying ΔWm(ref)(k)= ΔWm (k-1) -(D*ΔPm(k))
wind speed at unity power factor generation. In a variab le
speed WECS, the maximu m power depends on the power Wm(ref)(k)= |(1*Pm(k)/1)^1/3|
coefficient (Cp ) at different wind speeds. To achieve these
objectives, stator field oriented vector control scheme is Reference Speed
adopted for MSC and its phasor diagram is shown in
Fig.3(a). Here, d s - and q s - axis are synchronously rotating Fig.2. MPPT Flowchart
reference frame and α s - and β s -axis are stationary reference
frame. It is assumed that the stator flu x is aligned with q s - VI. CONT ROL SCHEME OF GSC
axis of synchronous rotating frame, to design the control The GSC controls and regulates the voltage across dc-
scheme for M SC. The stator flu x equation is given as: lin k capacitor and react ive power flow between grid and dc-
ψ=ψ +jψ (5) lin k capacitor at their predefined reference values. GSC is
d q also used to boost grid power quality. To achieve its
As the main flu x align in the d-axis so, ψd = 0. objectives, the grid voltage oriented vector control [13]
Then (5) becomes: technique is considered to design the control scheme fo r
ψ =ψ (6) GSC and it is designed on the synchronously rotating
q
reference frame i.e. d s -q s axis. The phasor diagram for grid
oriented vector control, used for GSC, is shown in Fig.3(b).
Here, the d co mponent of grid voltage is aligned in d-axis o f GSC FILTER
Vdc C
synchronously rotating reference frame and therefore,
Vgd =V and Vgq =0. The block d iagram of GSC is shown in
GRID
Fig.5 in details.
The dynamic model between GSC and grid, in PWM 3/2
synchronously rotating reference frame is given as:
Vag Vbg Vcg θg
d (11)
V =(V -RI -L I + ω LI ) 2/3 tanθg=Vgβ/Vgα
gd id gd dt gd g gq θg
d (12) ωg Vid Viq ωg
V =(V -RI -L I +ω L I ) DECOULING - - DECOUPLING
gq iq gq dt gq g g gd
TERMS +
TERMS
V=V +j0 (13) Igd + Igq
gd
PI PI

βs Igd
qs ds -
Igq +
ψ
+ - I*gd
Igq* PI
θs αs Vdc - + Vdc*
(a) Fig.5.Control Blocks of GSC

qs βs
ds VII. DISCUSSION AND A NALYSIS OF SIMULATION RESULT S
Vd
Simu link model of Fig.1 is developed in
MATLAB/Simulink p latform for an extensive study of the
θg system performances. The wind velocity pattern of the study
αs system is shown in Fig.6(a). The time do main simulation
results are presented in Figs.6, 7 and 8.
(b) During 0-1 Sec: The wind speed is considered constant
Fig.3.Phasor diagram for (a) MSC and (b) GSC i.e.12m/s (rated speed). The electromagnetic torque (Te),
produces inside the PMSG, is tracking the mechanical torque
(Tm), as shown in Fig.7(b) and for generator operation both
the torques are showing negative. It is noticed, fro m Fig.6
PMSG MSC C (d), that the generator actual rotor speed increases and track
ωm its reference value (generated fro m waveform fro m MPPT
algorithm) by controlling the RSC. Therefore, the stator
active power, tip speed ratio and power coefficient are also
reached to its optimum values as shown in Fig.7(a), Fig.6(c)
PWM
and Fig.6 (b) respectively.
Vas Vbs Vcs It is also observed, from Fig.7(d), that the dc-link voltage is
maintained constant and it tracks its reference value by
θs 2/3 ωe controlling the dc-link voltage based PI-controller of GSC.
During 1-2 Sec: Wind speed is decreased in ramp fro m
ωe Vqs Vds 12m/s to 9m/s, as shown in Fig.6(a). It is observed that the
DECOUPLING + DECOUPLING reference generator rotor speed which is obtained from
Ids TERMS + TERMS
- - MPPT is also decreased and the actual rotor speed of the
generator is exactly chased its reference speed as illustrated
PI PI in Fig.6(d). It is seen, from Fig.7(a, b), that the active power
and the electromagnetic torque are also chased their
- Iqs - predefined reference values by controlling-d-axis current
(Fig.8(c)).
Iqs* + Ids During 2-3 Sec: Wind velocity is considered 9 m/s and in
+
PI this interval all the time domain responses of the study
Ids*=0 system are tracked their reference values like earlier time
Iqs WIND interval, and the details are shown in the Figs.6, 7 and 8.
ωm - SPEED x The dc-lin k voltage (Fig.7(d )) is maintained fixed
+ irrespective to the variation o f wind velocity. This
indicates the quality of performance of the controller for
ωm* MPPT
GSC.
Fig.4.Control Block for MSC
 x It is noticed, fro m Fig.7(c), that the actual reactive power VIII. CONCLUSION
is tracked its reference value i.e. reactive power flo w Simu lation model of a grid connected PMSG-WT system
between PMSG and grid is maintained zero throughout is developed in MATLAB/Simulink (MATLA B-2013b)
operation and achieved unity power factor operation. environment to study the performance of the system under
This is obtained by controlling zero q-axis current varying wind velocity. Vector control approaches are used to
(Fig.8(d)) irrespective of variab le wind velocity. control both the converters. Flu x oriented and grid voltage
x The transient oscillations are observed in few Figs oriented vector control schemes are adopted for controlling
during init ial period of time. It is occurred that due to the MSC and GSC respectively to obtain variable speed constant
mis match in electro magnetic and mechanical torques. frequency operation at unity power factor generation. To
extract maximu m power from a particular wind speed, P and
O based MPPT algorith m is incorporated with the vector
control scheme of the MSC. An extensive simulation study is
done and simulink results are analyzed in details. Results
show that the system works according to control schemes
designed at and can offer good dynamic response under any
wind velocity.

REFERENCES
[1] T. Ackermann, “Wind Power in Power systems,” John Wiley & Sons,
Ltd, England, 2005.
[2] Z. Chen, J. M. Guerro, and F. Blaabjerg, “ A review of the state of the
art of power electronics for wind turbines,” IEEE Trans. Energy
Convers., vol.24, no.8, 2009, pp.1859 - 1875
[3] H. S. Kim and D. D. Chuanlu, “Wind energy conversion system from
electrical perspective-A survey,” Smart grid & renewable energy,
Fig.6. (a) Variation of Wind Speed (m/s), (b) Power Co-efficient, (c) Tip vol.1, no.3, 2010, pp. 119-131.
Speed Ratio (λ), and (d) Variation of Generator Rotor Speed (rad/s) [4] A. H. Kasem, A. Boudy, A. Daoud, S. S. Desouky, A. A. Salem,
“ Converters controls and flicker study of PMSG-based grid connected
wind turbines,” Ain Shams Engineering Journal, vol.4, no.1, 2013,
pp.75-91.
[5] S. Vijayalakshmi, S. kumar, S. Saravanam, R.V. Sandip, and V.
Sridhar, “ Modelling and control of a wind turbine using PMSG,”
International Journal of Engineering Science and Technology, vol.3,
no.3, 2011.
[6] S. Li, T . A. Haskew, R. P. Swatlokis, and W. Gathings, “ Optimal and
Direct Current Vector Control of Direct Driven PMSG wind turbine,”
IEEE Trans. on Power Electronics, vol.27, no.5, 2012, pp.2325-2337.
[7] Y. Errami, M. Ouassaid, and M. Marroufi, “ A performance
comparison of a nonlinear and a linear control for grid connected
PMSG wind energy conversion system,” International Journal of
Electrical Power and Energy System, vol.68, 2015, pp.180-194.
[8] J. S. Thongam, P. Bouchard, H. Ezzaidi, and M. Ouhrouche, “ Wind
Speed Sensorless maximum power point tracking conrol of variable
speed wind energy conversion systems,” IEEE International Electric
Machines and Drives Conference, 3-6 May, 2009, pp.1832-1837.
[9] M. Chinchilla, S. Arnaltes, and J. C. Burgos, “ Control of Permanent -
Fig.7. (a) Active Power (Watts), (b) Mechanical (T m) and Electromagnetic Magnet Generators Applied to Variable-Speed Wind-Energy Systems
(T e) Torques (N-m), (c) Reactive Power (VAr) and (d) DC-link Voltage Connected to the Grid’” IEEE Trans. On Energy Conv., vol.21, no.1,
(Volts) 2006, pp.130-135.
[10] A. K. Abdelsalam, A. M. Massoud, S. Ahmed and P. Enjeti, “High -
performance adaptive perturb and observe MPPT technique for
photovoltaic-based microgrids,” IEEE Trans. Power Electronics,
vol.26, no.4, 2011, pp.1010-1021.
[11] L. Piegari and R. Rizzo, “ Adaptive perturb and observe algorithm for
photovoltaic maximum power point tracking,” IET Renew. Power
Generation, vol.4, no.4, 2010, pp.317-328.
[12] S. Datta, J. P. Mishra, and A. K. Roy, “ Active and reactive power
control of a grid connected speed sensor-less DFIG based wind
energy conversion system,” IEEE International conference on Energy,
Power and Environment: Towards Sustainable Growth, 12-13 June,
2015, pp.1-6.
[13] S. Sharma, J. P. Mishra, and S. Datta, “ Sliding Mode Power Control
of a DFIG based Variable Speed Wind Energy Conversion System,”
Annual IEEE India Conference, 17-20 Dec., 2015, pp.1-6.

Fig.8. (a) d-axis Grid Voltage (Volts), (b) q-axis Grid Voltage (Volts), (c)
d-axis Grid Current (Amps), and (d) q-axis Grid Current (Amps)