Research Article Sensorless Direct Power Control of Induction Motor Drive Using Artificial Neural Network Abolfazl Halvaei Niasar and Hossein Rahimi Khoei Faculty of Electrical & Computer Engineering, University of Kashan, Kashan 87317-51167, Iran Correspondence should be addressed to Abolfazl Halvaei Niasar; [email protected] Received 27 September 2014; Revised 31 January 2015; Accepted 2 February 2015 Academic Editor: Matt Aitkenhead Copyright © 2015 A. Halvaei Niasar and H. Rahimi Khoei. is is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. is paper proposes the design of sensorless induction motor drive based on direct power control (DPC) technique. It is shown that DPC technique enjoys all advantages of pervious methods such as fast dynamic and ease of implementation, without having their problems. To reduce the cost of drive and enhance the reliability, an effective sensorless strategy based on artificial neural network (ANN) is developed to estimate rotor’s position and speed of induction motor. Developed sensorless scheme is a new model reference adaptive system (MRAS) speed observer for direct power control induction motor drives. e proposed MRAS speed observer uses the current model as an adaptive model. e neural network has been then designed and trained online by employing a back propagation network (BPN) algorithm. e estimator was designed and simulated in Simulink. Some simulations are carried out for the closed-loop speed control systems under various load conditions to verify the proposed methods. Simulation results confirm the performance of ANN based sensorless DPC induction motor drive in various conditions. 1. Introduction e electrical drive system is used to control the position, speed, and torque of the electric motors. Many works have been done on power converter topologies, control scheme of the electric drive systems, and the motor types in order to enhance and improve the performance of the electric motors so as to exactly perform and do what is required [1]. Induction motors (IMs) are widely used in industrial, commercial, and domestic applications as they are simple, rugged, and easy to maintain and of low cost. Since IMs demand well control performances, precise and quick torque and flux response, large torque at low speed, and wide speed range, the drive control system is necessary for IMs [2]. Control of induction motors can be done using vari- ous techniques. Most common techniques are (a) constant voltage/frequency control (V /F), (b) field orientation control (FOC), and (c) direct torque control (DTC). e first one is considered as scalar control since it adjusts only magnitude and frequency of the voltage or current with no concern about the instantaneous values of motor quantities. It does not require knowledge of parameters of the motor, and it is an open-loop control. us, it is a low cost simple solution for low performance applications such as fans and pumps. e other two methods are in the space vector control category because they utilize both magnitude and angular position of space vectors of motor variables, such as the voltage and flux. ey are employed in high performance applications, such as positioning drives or electric vehicles [3, 4]. Direct power control is a control method that directly selects output voltage vector states based on the power and flux errors using hysteresis controllers and without using current loops. In this respect, it is similar to the well- known direct torque control (DTC) method described in the literatures for various AC motors [5]. What is in common among these applications is that they all are power output devices needed to provide real power to the load. DPC technique has been basically applied to the generators, but it is tried to use it for control of electrical motors instead of DTC technique, due to the problems of torque estimation and dependency on the motor’s parameters. erefore, DPC technique enjoys all advantages of DTC such as fast dynamic and ease of implementation, without having the DTC’s problems. However, publications about direct power control Hindawi Publishing Corporation Advances in Artificial Neural Systems Volume 2015, Article ID 318589, 9 pages http://dx.doi.org/10.1155/2015/318589
Welcome message from author
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
Research ArticleSensorless Direct Power Control of Induction MotorDrive Using Artificial Neural Network
Abolfazl Halvaei Niasar and Hossein Rahimi Khoei
Faculty of Electrical amp Computer Engineering University of Kashan Kashan 87317-51167 Iran
Correspondence should be addressed to Abolfazl Halvaei Niasar halvaeikashanuacir
Received 27 September 2014 Revised 31 January 2015 Accepted 2 February 2015
Academic Editor Matt Aitkenhead
Copyright copy 2015 A Halvaei Niasar and H Rahimi Khoei This is an open access article distributed under the Creative CommonsAttribution License which permits unrestricted use distribution and reproduction in any medium provided the original work isproperly cited
This paper proposes the design of sensorless induction motor drive based on direct power control (DPC) technique It is shownthat DPC technique enjoys all advantages of pervious methods such as fast dynamic and ease of implementation without havingtheir problems To reduce the cost of drive and enhance the reliability an effective sensorless strategy based on artificial neuralnetwork (ANN) is developed to estimate rotorrsquos position and speed of induction motor Developed sensorless scheme is a newmodel reference adaptive system (MRAS) speed observer for direct power control induction motor drives The proposed MRASspeed observer uses the current model as an adaptive model The neural network has been then designed and trained online byemploying a back propagation network (BPN) algorithmThe estimator was designed and simulated in Simulink Some simulationsare carried out for the closed-loop speed control systems under various load conditions to verify the proposedmethods Simulationresults confirm the performance of ANN based sensorless DPC induction motor drive in various conditions
1 Introduction
The electrical drive system is used to control the positionspeed and torque of the electric motors Many works havebeen done on power converter topologies control scheme ofthe electric drive systems and the motor types in order toenhance and improve the performance of the electric motorsso as to exactly perform and dowhat is required [1] Inductionmotors (IMs) are widely used in industrial commercial anddomestic applications as they are simple rugged and easyto maintain and of low cost Since IMs demand well controlperformances precise and quick torque and flux responselarge torque at low speed and wide speed range the drivecontrol system is necessary for IMs [2]
Control of induction motors can be done using vari-ous techniques Most common techniques are (a) constantvoltagefrequency control (VF) (b) field orientation control(FOC) and (c) direct torque control (DTC) The first one isconsidered as scalar control since it adjusts only magnitudeand frequency of the voltage or current with no concernabout the instantaneous values of motor quantities It doesnot require knowledge of parameters of the motor and it is
an open-loop controlThus it is a low cost simple solution forlow performance applications such as fans and pumps Theother two methods are in the space vector control categorybecause they utilize both magnitude and angular position ofspace vectors of motor variables such as the voltage and fluxThey are employed in high performance applications such aspositioning drives or electric vehicles [3 4]
Direct power control is a control method that directlyselects output voltage vector states based on the power andflux errors using hysteresis controllers and without usingcurrent loops In this respect it is similar to the well-known direct torque control (DTC) method described in theliteratures for various AC motors [5] What is in commonamong these applications is that they all are power outputdevices needed to provide real power to the load DPCtechnique has been basically applied to the generators butit is tried to use it for control of electrical motors insteadof DTC technique due to the problems of torque estimationand dependency on the motorrsquos parameters Therefore DPCtechnique enjoys all advantages of DTC such as fast dynamicand ease of implementation without having the DTCrsquosproblems However publications about direct power control
Hindawi Publishing CorporationAdvances in Artificial Neural SystemsVolume 2015 Article ID 318589 9 pageshttpdxdoiorg1011552015318589
2 Advances in Artificial Neural Systems
Rs Lls Rr Llr
j120596r
rarr120582r
Lmrarr s
+
minus
Figure 1 The equivalent circuit of IM in the stationary referenceframe
are mainly aimed at either rectifiers [6] converters [7 8]dual-fed induction generators (DFIG) [9 10] or permanentmagnet synchronous generators (PMSG) [11 12] and there isno research on using the DPC technique for inductionmotor
Recently many researches have been carried on thedesign of speed-sensorless control schemes of inductionmotor drives The main reasons for the development of sen-sorless drives are reduction of complex hardware and hencecost increase in mechanical robustness and hence overallruggedness working under hostile environment higher reli-ability reduced maintenance and so forth Techniques rangefrom open-loop low performance strategy to closed-loophigh performance over the past decades [13ndash17]
Speed estimator employing artificial neural network(ANN) is an improvement over the classical mathematicalmodel based approaches [18ndash23] It is a major advantageof ANN based techniques that they do not require anymathematical model of the motor under consideration andthe drive development time can be substantially reduced Inthis study a speed estimator based on ANN based modelreference adaptive system (MRAS) has been studied andanalyzed For ANN the back propagation network (BPN)algorithm is used for online training to estimate the motorspeed The paper is organized as follows Section 2 givesthe dynamic model of induction motor Section 3 proposesthe original DPC strategy and employs it for inductionmotor Speed estimation technique using neural network ispresented in Section 4 Simulation results are presented inSection 5 and the conclusion is given in Section 6
2 Model of Induction Motor
Neglecting the motor core loss the saturation the slot effectand so forth the equivalent circuit of the IM in stationaryreference frame is shown in Figure 1
The mathematical model in stationary reference framecan be derived from the equivalent circuit as follows
119881119904= 119877119904
119894119904+
119889
120582119904
119889119905
0 = 119877
119903
119894119903+
119889
120582119903
119889119905
minus 119895120596119903
120582119903
(1)
where119871119904= 119871119897119904+ 119871119898
119871119903= 119871119897119903
+ 119871119898
120582119904= 119871119904
119894119904+ 119871119898
119894119903
120582119903= 119871119903
119894119903+ 119871119898
119894119904
(2)
The electromagnetic torque produced in the motor is
119879119890=
2
3
119901 Im (119894119904
120582119904
lowast
) (3)
where 119901 is number of motor pole pairs and lowast is conjugateoperator
The induction motor model in the 120572 minus 120573 fixed referenceframe can be described by the following equations
[
[
[
[
[
119906120572119904
119906120573119904
0
0
]
]
]
]
]
=
[
[
[
[
[
[
119877119904+ 119871119904119901 0 119871
119898119901 0
0 119877119904+ 119871119904119901 0 119871
119898119901
119871119898119901 120596
119903119871 119877
119903+ 119871119903119901 120596
119903119871
minus120596119903119871 119871
119898119901 minus120596
119903119871 119877
119903+ 119871119903119901
]
]
]
]
]
]
sdot
[
[
[
[
[
[
119894120572119904
119894120573119904
119894120572119903
119894120573119903
]
]
]
]
]
]
[
[
[
[
[
[
120595120572119904
120595120573119904
120595120572119903
120595120573119903
]
]
]
]
]
]
=
[
[
[
[
[
[
119871119904
0 119871119898
0
0 119871119904
0 119871119898
119871119898
0 119871119903
0
0 119871119898
0 119871119903
]
]
]
]
]
]
[
[
[
[
[
[
119894120572119904
119894120573119904
119894120572119903
119894120573119903
]
]
]
]
]
]
(4)
where the subscripts 119904 and 119903 stand for stator and rotorquantities 119906 and 119894 denote voltage and current respectively119877 denotes resistance and 120596
119903is the rotor speed 120595 denotes flux
linkage
3 Direct Power and Flux Control ofInduction Motors
The direct power control methods discussed in this paperbear certain similarity to the direct torque control (DTC)Therefore DPC is actually direct power and flux controlwith two parameters involved in the control strategy soit is also named as direct power and flux control (DPFC)in some publications [24] Direct power and flux control(DPFC) of IMs is a controlmethod that directly selects outputvoltage vector states based on the power and flux errors usinghysteresis controllers Figure 2 shows the block diagram of ageneral open-loop DPFC system
Advances in Artificial Neural Systems 3
Power control
Fluxcontrol
Stateselection
VS inverter IM
Load
Figure 2 General block diagram of direct power and flux controlsystem
31 Flux Control Principles Fixed linkage flux can provideenough electromagnetic torque and avoid magnetizing cur-rent saturation in the iron core of the IMTherefore in directtorque and flux control as well as in the proposed directpower and flux controls in the subsequent chapters the fluxis maintained constant [25] In the stator stationary referenceframe the frame rotation speed is zero and the rotor voltageis zero as well (for squirrel-cage IMs) Thus
119881119904= 119877119904
119894119904+
119889
120582119904
119889119905
(5)
0 = 119877
119903
119894119903+
119889
120582119903
119889119905
minus 119895120596119903
120582119903
(6)
With neglecting the small voltage drop across the statorresistance we have
119881119904cong
119889
120582119904
119889119905
(7)
Integrating (7) and writing it in a discrete form we obtain
120582119904(119905119899+1
) =
120582119904(119905119899) +
119881119904Δ119905 (8)
That is
Δ
120582119904cong
119881119904Δ119905 (9)
where Δ119905 = 119905119899+1
minus 119905119899equals the switching interval Therefore
within a switching interval Δ119905 the increase of stator flux isproportional to the stator voltage space vector This is theprinciple of direct flux control in DPC
32 Power Control Principles From the power flow charts ininduction motor it is evident that real output power is thepart that produces the torque and is what the user of thesystem is mostly interested in The output real power is givenby
119875out = 119879119890120596119898
= 119879119890
120596119903
119901
(10)
119879119890=
2
3
119901 Im (119894119904
120582119904
lowast
) =
2
3
119901 Im(
119871119903
120582119904minus 119871119898
120582119903
119871119904119871119903minus 119871
2
119898
120582119904
lowast
)
= minus
2
3
119901
119871119898
119871
2
120590
Im (
120582119903
120582119904
lowast
) =
2
3
119901
119871119898
119871
2
120590
Im (
120582119904
120582119903
lowast
)
=
2
3
119901
119871119898
119871
2
120590
120582119904120582119903sin (120579119904minus 120579119903)
(11)
120596m
Δ120596m+
minus
PlowastoutTlowast
e120596lowastm
Figure 3 Generating the output power reference obtained fromspeed loop
Substituting the torque in (10) with (11) the output powerbecomes
119875out =2
3
120596119903
119871119898
119871
2
120590
Im (
120582119904
120582119903
lowast
) =
2
3
119871119898
119871
2
120590
120596119903120582119904120582119903sin (120579119904minus 120579119903)
(12)
Since the magnitude of the stator flux is kept constantand the rotor flux does not change much due to its inertiathe rotor speed and angle can be considered constant tooThe formula above shows that the change of output powerdepends only on the change of stator flux angle The statorvoltage vector that can increase the stator angle needs tobe raised in order to increase the output power The realoutput power equation obtained above is only valid forexplanation of the principles of power control However itis not appropriate for the purpose of estimating the actualpower in simulations
33 Output Power Reference The output power reference isthe command value or set point for the power control In aclosed-loop speed control system the reference of the powercontroller is obtained from the output of the PI-type speedcontroller (see Figure 3) The speed error is defined as thedifference of the reference speed and the estimated actualspeed
Δ120596119898
= 120596
lowast
119898minus 120596119898 (13)
where 120596
lowast
119898is the reference speed (the asterisk denotes a
reference value) Then the reference torque can be obtainedthrough a conventional PI controller as
119879
lowast
119890= 119870119901(Δ120596119898) + 119870119894int (Δ120596
119898) 119889119905 (14)
The continuous standard form above can also be expressed ina discrete incremental PI control form which ismore suitablefor the digital implementation
119879
lowast
119890(119905119899+1
) = 119879
lowast
119890(119905119899) + 119870119901[Δ120596119898
(119905119899) minus Δ120596
119898(119905119899minus1
)]
+ 119870119894Δ119879Δ120596
119898(119905119899)
(15)
The subscript (119899) denotes the current sampling instant (119899minus1)
is the last instant and (119899+1) is the next oneThe proportionalgain is denoted by 119870
119901 119870119894is the integral gain which equals
119870119901divided by the integral time constant 119879
119894 and Δ119879 is the
sampling time interval between the 119899 and (119899 + 1) samplinginstants The output power reference according to (10) istherefore expressed as
119875
lowast
out = 119879
lowast
119890120596
lowast
119898 (16)
4 Advances in Artificial Neural Systems
Table 1 Combination of the power and flux controller outputs
119887119901= minus1 119887
119901= 0 119887
119901= 1
1 2 34 5 6
The process of obtaining the output power reference from thespeed reference is illustrated in Figure 3 For simulations theactual motor speed 120596
119898can be obtained as
120596119898
=
1
119869
int (119879119890minus 119879119897119889) 119889119905 (17)
In practice the speed is either measured directly or estimatedfrom the current and voltage signals The magnitude of thestator flux is kept constant in the simulation thus the fluxreference 120582
lowast
119904is a constant The error of the stator flux is
Δ120582119904= 120582
lowast
119904minus 120582119904 (18)
34 Power and Flux Hysteresis Controllers Both the outputpower and the stator flux controllers are of hysteresis typeDepending on the control error the output of the controlleris set to two or three discrete values The power controllerhas a three-level output [24 26] The values are 1 0 andminus1 representing an increase no change and a decrease ofthe controlled variable respectively The number of fluxcontroller output levels is two with 1 and 0 meaning increaseand decrease commands respectively Figure 4 illustratescharacteristics of these two controllers
35 Switching Table The task of the state selector in thedirect power control is to combine the outputs of the powercontroller and flux controller to select the values of theswitching variables a b and c These variables describe therequired voltage vectors of the inverter To make it easier toimplement the combination of the two controller outputs canbe expressed as follows
119887 = 3119887120582+ 119887119901+ 2 (19)
In the above equation variable 119887 = 1 2 3 4 5 6 while 119887120582
=
(0 1) and 119887119901= (minus1 0 1)
Alternatively (19) can also be represented by Table 1A whole stator flux cycle of 360∘ is divided equally into
6 sectors each one spanning 60∘ Combining with the sectornumbers from 1 through 6 produces lookup Table 2 for thestate selection [24]The concept of state selection is illustratedin Figure 5
Note that the stator flux angle 120579119904must be converted to a
sector number from 1 through 6 for the use of Table 2 for stateselection
36 Estimation of Stator Flux and Output Power The estima-tion of flux is implemented by integration of (5)
120582119904= int (V
119904minus
119894119904119877119904) 119889119905 (20)
Table 2 State selection lookup table
b = 1 b = 2 b = 3 b = 4 b = 5 b = 6Sector 1 1 0 2 5 7 6Sector 2 5 7 6 4 0 2Sector 3 4 0 1 6 7 3Sector 4 6 7 5 2 0 1Sector 5 2 0 4 3 7 5Sector 6 3 7 6 1 0 4
That output power can be calculated from (10) Figure 6illustrates the estimation of actual output power and flux fromstator voltage and current together with mechanical angularspeedThese estimated values are the feedback for the outputpower and stator flux controls
4 Speed Estimation Using Neural Network
In MRAS technique some state variables 119883119889and 119883
119902(eg
rotor flux linkage components 120595119889119903
and 120595119902119903 or back-EMF
components 119890119889 119890119902 etc) of the inductionmachine (which are
obtained by using measured quantities eg stator voltagesand currents) are estimated in a reference model and arethen compared with state variables
119883119889and
119883119902estimated by
using an adaptive model The difference between these statevariables is then formulated into a speed tuning signal (120576)which is then an input into an adaptation mechanism whichoutputs the estimated rotor speed ()
Speed estimator using ANN is a part of a model referenceadaptive system (MRAS) where ANN takes the role of theadaptive model ANN contains the adjustable and constantweights and the adjustable weights are proportional to therotor speed The adjustable weights are changed by usingthe error between the outputs of the reference and adaptivemodel Figure 7 shows the MRAS-based speed estimationscheme which contains an ANN with BPN adaptationtechnique [17]
The outputs of the reference model are the rotor fluxlinkage components in stationary reference frame given by
120595119889119903
=
119871119903
119871119898
[int (V119889119904
minus 119877119878119894119889119904) 119889119905 minus 119871
1015840
119904119894119889119904]
120595119902119903
=
119871119903
119871119898
[int (V119902119904
minus 119877119878119894119902119904) 119889119905 minus 119871
1015840
119904119894119902119904]
(21)
These two equations do not contain the rotor speed anddescribe the reference model The equations of adaptivemodel are given by
119889119903
=
1
119879
int (119871119898119894119889119904
minus 119889119903
minus 120596119903119879119903119902119903) 119889119905
119902119903
=
1
119879
int (119871119898119894119902119904
minus 119902119903
minus 120596119903119879119903119889119903) 119889119905
(22)
It is possible to implement (22) by a two-layer ANN contain-ing weights 119882
1(= 1 minus 119862) 119882
2(= 120596119903119879119903119862) and 119882
3(= 119862119897
119898)
that 119862 = 119879119879119903 119879 119879119903are sampling time and rotor time
Advances in Artificial Neural Systems 5
bp
0 ΔPout
+1
minus1
(a)
0
b120582
Δ120582s
1
(b)
Figure 4 Characteristics of the hysteresis controllers (a) power controller and (b) flux controller
b120582Δ120582s
ΔPout
Pout
minus
minus
+
+
120579s
a
bc
bp
120582s
Plowastout
120582lowasts
Figure 5 Block diagram of the inverter state selection
120582s120579s
rarr120582s
rarr s minus Rs
rarri s
s
rarr srarri s
Pout
120596m
Te 2
3p Im(rarri s
rarr120582lowast
s )
Figure 6 Estimation of actual output power and stator flux linkage
Reference model
Artificial neural network(ANN)
Back propagation weightadjustment
w2
sum
sum
120576d
120576q
1Z 1Z
1Z 1Z
iqs
ids
qs
ds
dr
qr
120595dr
120595qr
minus
Figure 7 MRAS-based rotor speed estimator using ANN
6 Advances in Artificial Neural Systems
w2
w1
w1
minusw2
w3
w3
ids(k minus 1)
iqs(k minus 1)
dr(k minus 1)
qr(k minus 1) dr(k)
qr(k)
Figure 8 ANN model for the estimation of rotor flux linkage
Power controller
Fluxcontroller
Stateselector
Voltage source inverter
IM
Load
Voltage sensor
Current sensor
Fluxestimator
Powerestimator
Speedcontroller
120596lowastm
Plowastout
bp
120582lowasts
120579s
120582s
Pout
rarr s
rarri s
120596m
Vdc
abcb120582
Figure 9 Block diagram of direct output power and flux control system for IM for closed-loop speed control
1400
1200
1000
800
600
400
200
0
minus200
Time (s)
Pow
er (W
)
0 05 1 15 2 25 3
Motor powerReference power
(a)
600
500
400
300
200
100
0
minus100
Spee
d (r
pm)
Time (s)0 05 1 15 2 25 3
Motor powerReference power
(b)
Figure 10 Simulation results DPC of induction motor (a) electromagnetic power and (b) rotor speed
Advances in Artificial Neural Systems 7
wr_est1
wr estimator
wr_est
Reference model
I_qs
I_ds
V_qs
V_ds
phi_dr
phi_qrLow pass filter
Zp
adjustment
ed
eqwr
Artificial neural network
I_qs
I_ds
wr
ANN-MRAS
I_qs
I_ds
wr
4
3
2
1
1z
1z+minus
+minus
iqs
ids
qs
ds
Lls
philowast_dr
philowast_qr
philowast_dr
philowast_qr
phi_dr_est
phi_qr_est
phi_dr
phi_qr
phi_dr_est
phi_qr_est
num(z)
den(z)
Back propagation
Figure 11 ANN based model of speed estimation
1500
1000
500
0
Spee
d (r
ads
)
MotorDifferentiated speedEstimation speed
Time (s)0 05 1 15
Figure 12 Response comparison of actual machine and ANN basedspeed estimator with error
constant The variable ANN weight1198822is proportional to the
rotor speed By using the backward difference method theequation of adaptive model is given below
119889119903
(119896) = 1198821119889119903
(119896 minus 1) minus 1198822119902119903
(119896 minus 1) + 1198823119894119889119904
(119896 minus 1)
119902119903
(119896) = 1198821119902119903
(119896 minus 1) minus 1198822119889119903
(119896 minus 1) + 1198823119894119902119904
(119896 minus 1)
(23)
which gives the value of rotor flux at the119870th sampling instantThese equations can be visualized by the very simple two-layer ANN shown in Figure 8
After taking learning factor 120578 andmomentum term 120572 intoaccount the estimated rotor speed is given below
119903(119896)
= 119903(119896 minus 1)
+
120578
119879
minus [120595119889119903
(119896) minus 119889119903
(119896)] 119902119903
(119896 minus 1)
+ [120595119902119903
(119896) minus 119902119903
(119896)] 119889119903
(119896 minus 1)
+
120572
119879
Δ1199082(119896 minus 1)
(24)
If the value of learning rate (120578) is chosen high it may lead tooscillations in the outputs of ANN and 120572 is chosen in rangebetween 01 and 08 The inclusion of momentum term intothe weight adjustment mechanism can significantly increasethe convergence which is extremely useful when the ANNshown in Figure 8 is used to estimate in real time [13]
8 Advances in Artificial Neural Systems
40
30
20
10
0
Pow
er (W
)
Reference powerElectromagnetic power
Time (s)0 05 1 15 2 25 3
minus10
minus20
minus30
(a)
2000
1800
1600
1400
1200
1000
800
600
400
200
0
Spee
d (r
pm)
Reference speedMotor speed
Time (s)0 05 1 15 2 25 3
(b)
Figure 13 Simulation results of DPC based on ANN of induction motor (a) electromagnetic power and (b) rotor speed
Table 3 Induction motor parameter
Parameter Value119875119899
2 hp120596rated 1500 [rpm]119881 300 [V]119868119899
21 [A]119869 0004 [kgsdotm2]119901 2119877119904
10 [Ω]119871119898
330 [mH]119871119897119904 119871119897119903
04 [mH]119861 0002 [Nsdotm(radsec)]
5 Simulation Results
51 Direct Power Control The simulation is carried out usingMATLAB The whole system setup is shown in Figure 9 Itcontains the power and flux controllers as inner loops andthe speed loop as outer loop
The sampling frequency is chosen as 50 kHz The hys-teresis tolerance for both power and flux controller is 1 ofrespective reference values For the PI speed controller (see(14)) the proportional gain is tuned to 119896
119901= 90 the overall
coefficient of the integral part is set to 01 which yieldedan integral gain 119896
119894= 5000 at the 50 kHz sampling rate
The reference speed profile is set in which the speed alwaysremains below the rated speed to avoid field weakening Theparameters of IM used in the simulations are listed in Table 3The simulation results are shown in Figure 10
52 Speed Estimator Based onANN Implementation ofANNbased speed estimator is carried out in MATLABSimulinkas shown in Figure 11
The response of ANN based speed estimator is comparedwith actual machine which is shown in Figure 12
53 Direct PowerControl withANN Theresult ofANNbasedspeed estimator with DPC controller is shown in Figure 13
6 Conclusion
This paper studies the possibility of direct power controlfor speed control of induction motor It has been shownthat DPC is basically derived from direct torque control(DTC) method and the simulation results show that thiscontrol scheme for IMsmotor has all the advantages of directtorque control method One of the major advantages of DPCmethod compared to DTC method is the easier calculationof actual power Moreover this paper proposes a new MRASspeed observer for DPC controlled induction motor drivesusing neural networks for speed estimation The simulationresults show that the proposed two-layers neural networkcan identify and track the motor speed accurately duringthe whole operating region Structure and algorithm are alsosimple Overall the dynamic response of this scheme of speedestimation shows a good performance
Conflict of Interests
The authors declare that there is no conflict of interestsregarding the publication of this paper
References
[1] P C Krause O Wasynozuk and S D Sudhoff Analysis ofElectric Machinery and Drive Systems IEEE Press 2nd edition2002
[2] I Boldea and S A Nasar Electric Drives Taylor amp Francis NewYork NY USA 2006
[3] P Vas The Control of AC Machines Oxford University PressNew York NY USA 1990
[4] A M Trzynadlowski Control ofInduction Motors Academicpress 2001
Advances in Artificial Neural Systems 9
[5] G S Buja and M P Kazmierkowski ldquoDirect torque control ofPWM inverter-fed ACmotorsmdasha surveyrdquo IEEE Transactions onIndustrial Electronics vol 51 no 4 pp 744ndash757 2004
[6] G Escobar A M Stankovic J M Carrasco E Galvan andR Ortega ldquoAnalysis and design of direct power control (DPC)for a three phase synchronous rectifier via output regulationsubspacesrdquo IEEE Transactions on Power Electronics vol 18 no3 pp 823ndash830 2003
[7] S Vazquez J A Sanchez J M Carrasco J I Leon and EGalvan ldquoA model-based direct power control for three-phasepower convertersrdquo IEEE Transactions on Industrial Electronicsvol 55 no 4 pp 1647ndash1657 2008
[8] Z Dawei X Lie and B W Williams ldquoImproved direct powercontrol of grid-connected DCAC convertersrdquo IEEE Transac-tions on Power Electronics vol 24 no 5 pp 1280ndash1292 2009
[9] L Xu and P Cartwright ldquoDirect active and reactive powercontrol of DFIG for wind energy generationrdquo IEEE Transactionson Energy Conversion vol 21 no 3 pp 750ndash758 2006
[10] D Zhi and L Xu ldquoDirect power control of DFIG with constantswitching frequency and improved transient performancerdquoIEEE Transactions on Energy Conversion vol 22 no 1 pp 110ndash118 2007
[11] Y Errami M Benchagra M Hilal M Maaroufi and M Ouas-said ldquoControl strategy for PMSG wind farm based on MPPTand direct power controlrdquo in Proceedings of the InternationalConference onMultimedia Computing and Systems (ICMCS rsquo12)pp 1125ndash1130 May 2012
[12] A Harrouz A Benatiallah and O Harrouz ldquoDirect powercontrol of a PMSG dedicated to standalone wind energysystemsrdquo in Proceedings of the 8th International Conference andExhibition on Ecological Vehicles and Renewable Energies (EVERrsquo13) pp 1ndash5 March 2013
[13] P Vas Sensorless Vector and Direct Torque Control OxfordUniversity Press 1998
[14] K Sedhuraman S Himavathi and A MuthuramalingamldquoPerformances comparison of neural architectures for on-linespeed estimation in sensorless IM drivesrdquo World Academy ofScience Engineering and Technology no 60 pp 1318ndash1325 2011
[15] S Meziane and R Benalla ldquoMRAS based speed control ofsensorlessinduction motor drivesrdquo ICGST-ACSE Journal vol 7no 1 pp 43ndash500 2007
[16] E Levi and M Wang ldquoImpact of parameter variations onspeed estimation in sensorless rotor flux oriented inductionmachinesrdquo in Proceedings of the 7th International Conferenceon Power Electronics and Variable Speed Drives pp 305ndash310London UK September 1998
[17] M E Elbulk L Tong and I Husain ldquoNeural-network-basedmodel reference adaptive systems for high-performance motordrives and motion controlsrdquo IEEE Transactions on IndustryApplications vol 38 no 3 pp 879ndash886 2002
[18] M T Hagan B Demuth andM BealeNeural Network DesignCengage Learning Pvt Ltd New Delhi India 2008
[19] S-H Kim T-S Park J-N Yoo and G-T Park ldquoSpeed-sensorless vector control of an induction motor using neuralnetwork speed estimationrdquo IEEE Transactions on IndustrialElectronics vol 48 no 3 pp 609ndash614 2001
[20] M Kuchar P Brandstetter and M Kaduch ldquoSensorless induc-tion motor drive with neural networkrdquo in Proceedings ofthe IEEE 35th Annual Power Electronics Specialists Conference(PESC rsquo04) pp 3301ndash3305 June 2004
[21] S Maiti V Verma C Chakraborty and Y Hori ldquoAn adaptivespeed sensorless induction motor drive with artificial neuralnetwork for stability enhancementrdquo IEEE Transactions onIndustrial Informatics vol 8 no 4 pp 757ndash766 2012
[22] P Girovsky and J Timko ldquoShaft sensor-less FOC control ofan induction motor using neural estimatorsrdquo Acta PolytechnicaHungarica vol 9 no 4 pp 31ndash45 2012
[23] D P Kothari and I J Nagrath Electric Machines chapter 9 TataMcGraw-Hill Education Private Limited 2004
[24] G Escobar A M Stankovic J M Carrasco E Galvan andR Ortega ldquoAnalysis and design of direct power control (DPC)for a three phase synchronous rectifier via output regulationsubspacesrdquo IEEE Transactions on Power Electronics vol 18 no3 pp 823ndash830 2003
[25] N R N Idris and A H M Yatim ldquoDirect torque controlof induction machines with constant switching frequency andreduced torque ripplerdquo IEEE Transactions on Industrial Elec-tronics vol 51 no 4 pp 758ndash767 2004
[26] S-H Huh K-B Lee D-W Kim I Choy and G-T ParkldquoSensorless speed control system using a neural networkrdquoInternational Journal of Control Automation and Systems vol3 no 4 pp 612ndash619 2005
Submit your manuscripts athttpwwwhindawicom
Computer Games Technology
International Journal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Distributed Sensor Networks
International Journal of
Advances in
FuzzySystems
Hindawi Publishing Corporationhttpwwwhindawicom
Volume 2014
International Journal of
ReconfigurableComputing
Hindawi Publishing Corporation httpwwwhindawicom Volume 2014
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Applied Computational Intelligence and Soft Computing
thinspAdvancesthinspinthinsp
Artificial Intelligence
HindawithinspPublishingthinspCorporationhttpwwwhindawicom Volumethinsp2014
Advances inSoftware EngineeringHindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Electrical and Computer Engineering
Journal of
Journal of
Computer Networks and Communications
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Hindawi Publishing Corporation
httpwwwhindawicom Volume 2014
Advances in
Multimedia
International Journal of
Biomedical Imaging
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
ArtificialNeural Systems
Advances in
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
RoboticsJournal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Computational Intelligence and Neuroscience
Industrial EngineeringJournal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Modelling amp Simulation in EngineeringHindawi Publishing Corporation httpwwwhindawicom Volume 2014
The Scientific World JournalHindawi Publishing Corporation httpwwwhindawicom Volume 2014
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Human-ComputerInteraction
Advances in
Computer EngineeringAdvances in
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
2 Advances in Artificial Neural Systems
Rs Lls Rr Llr
j120596r
rarr120582r
Lmrarr s
+
minus
Figure 1 The equivalent circuit of IM in the stationary referenceframe
are mainly aimed at either rectifiers [6] converters [7 8]dual-fed induction generators (DFIG) [9 10] or permanentmagnet synchronous generators (PMSG) [11 12] and there isno research on using the DPC technique for inductionmotor
Recently many researches have been carried on thedesign of speed-sensorless control schemes of inductionmotor drives The main reasons for the development of sen-sorless drives are reduction of complex hardware and hencecost increase in mechanical robustness and hence overallruggedness working under hostile environment higher reli-ability reduced maintenance and so forth Techniques rangefrom open-loop low performance strategy to closed-loophigh performance over the past decades [13ndash17]
Speed estimator employing artificial neural network(ANN) is an improvement over the classical mathematicalmodel based approaches [18ndash23] It is a major advantageof ANN based techniques that they do not require anymathematical model of the motor under consideration andthe drive development time can be substantially reduced Inthis study a speed estimator based on ANN based modelreference adaptive system (MRAS) has been studied andanalyzed For ANN the back propagation network (BPN)algorithm is used for online training to estimate the motorspeed The paper is organized as follows Section 2 givesthe dynamic model of induction motor Section 3 proposesthe original DPC strategy and employs it for inductionmotor Speed estimation technique using neural network ispresented in Section 4 Simulation results are presented inSection 5 and the conclusion is given in Section 6
2 Model of Induction Motor
Neglecting the motor core loss the saturation the slot effectand so forth the equivalent circuit of the IM in stationaryreference frame is shown in Figure 1
The mathematical model in stationary reference framecan be derived from the equivalent circuit as follows
119881119904= 119877119904
119894119904+
119889
120582119904
119889119905
0 = 119877
119903
119894119903+
119889
120582119903
119889119905
minus 119895120596119903
120582119903
(1)
where119871119904= 119871119897119904+ 119871119898
119871119903= 119871119897119903
+ 119871119898
120582119904= 119871119904
119894119904+ 119871119898
119894119903
120582119903= 119871119903
119894119903+ 119871119898
119894119904
(2)
The electromagnetic torque produced in the motor is
119879119890=
2
3
119901 Im (119894119904
120582119904
lowast
) (3)
where 119901 is number of motor pole pairs and lowast is conjugateoperator
The induction motor model in the 120572 minus 120573 fixed referenceframe can be described by the following equations
[
[
[
[
[
119906120572119904
119906120573119904
0
0
]
]
]
]
]
=
[
[
[
[
[
[
119877119904+ 119871119904119901 0 119871
119898119901 0
0 119877119904+ 119871119904119901 0 119871
119898119901
119871119898119901 120596
119903119871 119877
119903+ 119871119903119901 120596
119903119871
minus120596119903119871 119871
119898119901 minus120596
119903119871 119877
119903+ 119871119903119901
]
]
]
]
]
]
sdot
[
[
[
[
[
[
119894120572119904
119894120573119904
119894120572119903
119894120573119903
]
]
]
]
]
]
[
[
[
[
[
[
120595120572119904
120595120573119904
120595120572119903
120595120573119903
]
]
]
]
]
]
=
[
[
[
[
[
[
119871119904
0 119871119898
0
0 119871119904
0 119871119898
119871119898
0 119871119903
0
0 119871119898
0 119871119903
]
]
]
]
]
]
[
[
[
[
[
[
119894120572119904
119894120573119904
119894120572119903
119894120573119903
]
]
]
]
]
]
(4)
where the subscripts 119904 and 119903 stand for stator and rotorquantities 119906 and 119894 denote voltage and current respectively119877 denotes resistance and 120596
119903is the rotor speed 120595 denotes flux
linkage
3 Direct Power and Flux Control ofInduction Motors
The direct power control methods discussed in this paperbear certain similarity to the direct torque control (DTC)Therefore DPC is actually direct power and flux controlwith two parameters involved in the control strategy soit is also named as direct power and flux control (DPFC)in some publications [24] Direct power and flux control(DPFC) of IMs is a controlmethod that directly selects outputvoltage vector states based on the power and flux errors usinghysteresis controllers Figure 2 shows the block diagram of ageneral open-loop DPFC system
Advances in Artificial Neural Systems 3
Power control
Fluxcontrol
Stateselection
VS inverter IM
Load
Figure 2 General block diagram of direct power and flux controlsystem
31 Flux Control Principles Fixed linkage flux can provideenough electromagnetic torque and avoid magnetizing cur-rent saturation in the iron core of the IMTherefore in directtorque and flux control as well as in the proposed directpower and flux controls in the subsequent chapters the fluxis maintained constant [25] In the stator stationary referenceframe the frame rotation speed is zero and the rotor voltageis zero as well (for squirrel-cage IMs) Thus
119881119904= 119877119904
119894119904+
119889
120582119904
119889119905
(5)
0 = 119877
119903
119894119903+
119889
120582119903
119889119905
minus 119895120596119903
120582119903
(6)
With neglecting the small voltage drop across the statorresistance we have
119881119904cong
119889
120582119904
119889119905
(7)
Integrating (7) and writing it in a discrete form we obtain
120582119904(119905119899+1
) =
120582119904(119905119899) +
119881119904Δ119905 (8)
That is
Δ
120582119904cong
119881119904Δ119905 (9)
where Δ119905 = 119905119899+1
minus 119905119899equals the switching interval Therefore
within a switching interval Δ119905 the increase of stator flux isproportional to the stator voltage space vector This is theprinciple of direct flux control in DPC
32 Power Control Principles From the power flow charts ininduction motor it is evident that real output power is thepart that produces the torque and is what the user of thesystem is mostly interested in The output real power is givenby
119875out = 119879119890120596119898
= 119879119890
120596119903
119901
(10)
119879119890=
2
3
119901 Im (119894119904
120582119904
lowast
) =
2
3
119901 Im(
119871119903
120582119904minus 119871119898
120582119903
119871119904119871119903minus 119871
2
119898
120582119904
lowast
)
= minus
2
3
119901
119871119898
119871
2
120590
Im (
120582119903
120582119904
lowast
) =
2
3
119901
119871119898
119871
2
120590
Im (
120582119904
120582119903
lowast
)
=
2
3
119901
119871119898
119871
2
120590
120582119904120582119903sin (120579119904minus 120579119903)
(11)
120596m
Δ120596m+
minus
PlowastoutTlowast
e120596lowastm
Figure 3 Generating the output power reference obtained fromspeed loop
Substituting the torque in (10) with (11) the output powerbecomes
119875out =2
3
120596119903
119871119898
119871
2
120590
Im (
120582119904
120582119903
lowast
) =
2
3
119871119898
119871
2
120590
120596119903120582119904120582119903sin (120579119904minus 120579119903)
(12)
Since the magnitude of the stator flux is kept constantand the rotor flux does not change much due to its inertiathe rotor speed and angle can be considered constant tooThe formula above shows that the change of output powerdepends only on the change of stator flux angle The statorvoltage vector that can increase the stator angle needs tobe raised in order to increase the output power The realoutput power equation obtained above is only valid forexplanation of the principles of power control However itis not appropriate for the purpose of estimating the actualpower in simulations
33 Output Power Reference The output power reference isthe command value or set point for the power control In aclosed-loop speed control system the reference of the powercontroller is obtained from the output of the PI-type speedcontroller (see Figure 3) The speed error is defined as thedifference of the reference speed and the estimated actualspeed
Δ120596119898
= 120596
lowast
119898minus 120596119898 (13)
where 120596
lowast
119898is the reference speed (the asterisk denotes a
reference value) Then the reference torque can be obtainedthrough a conventional PI controller as
119879
lowast
119890= 119870119901(Δ120596119898) + 119870119894int (Δ120596
119898) 119889119905 (14)
The continuous standard form above can also be expressed ina discrete incremental PI control form which ismore suitablefor the digital implementation
119879
lowast
119890(119905119899+1
) = 119879
lowast
119890(119905119899) + 119870119901[Δ120596119898
(119905119899) minus Δ120596
119898(119905119899minus1
)]
+ 119870119894Δ119879Δ120596
119898(119905119899)
(15)
The subscript (119899) denotes the current sampling instant (119899minus1)
is the last instant and (119899+1) is the next oneThe proportionalgain is denoted by 119870
119901 119870119894is the integral gain which equals
119870119901divided by the integral time constant 119879
119894 and Δ119879 is the
sampling time interval between the 119899 and (119899 + 1) samplinginstants The output power reference according to (10) istherefore expressed as
119875
lowast
out = 119879
lowast
119890120596
lowast
119898 (16)
4 Advances in Artificial Neural Systems
Table 1 Combination of the power and flux controller outputs
119887119901= minus1 119887
119901= 0 119887
119901= 1
1 2 34 5 6
The process of obtaining the output power reference from thespeed reference is illustrated in Figure 3 For simulations theactual motor speed 120596
119898can be obtained as
120596119898
=
1
119869
int (119879119890minus 119879119897119889) 119889119905 (17)
In practice the speed is either measured directly or estimatedfrom the current and voltage signals The magnitude of thestator flux is kept constant in the simulation thus the fluxreference 120582
lowast
119904is a constant The error of the stator flux is
Δ120582119904= 120582
lowast
119904minus 120582119904 (18)
34 Power and Flux Hysteresis Controllers Both the outputpower and the stator flux controllers are of hysteresis typeDepending on the control error the output of the controlleris set to two or three discrete values The power controllerhas a three-level output [24 26] The values are 1 0 andminus1 representing an increase no change and a decrease ofthe controlled variable respectively The number of fluxcontroller output levels is two with 1 and 0 meaning increaseand decrease commands respectively Figure 4 illustratescharacteristics of these two controllers
35 Switching Table The task of the state selector in thedirect power control is to combine the outputs of the powercontroller and flux controller to select the values of theswitching variables a b and c These variables describe therequired voltage vectors of the inverter To make it easier toimplement the combination of the two controller outputs canbe expressed as follows
119887 = 3119887120582+ 119887119901+ 2 (19)
In the above equation variable 119887 = 1 2 3 4 5 6 while 119887120582
=
(0 1) and 119887119901= (minus1 0 1)
Alternatively (19) can also be represented by Table 1A whole stator flux cycle of 360∘ is divided equally into
6 sectors each one spanning 60∘ Combining with the sectornumbers from 1 through 6 produces lookup Table 2 for thestate selection [24]The concept of state selection is illustratedin Figure 5
Note that the stator flux angle 120579119904must be converted to a
sector number from 1 through 6 for the use of Table 2 for stateselection
36 Estimation of Stator Flux and Output Power The estima-tion of flux is implemented by integration of (5)
120582119904= int (V
119904minus
119894119904119877119904) 119889119905 (20)
Table 2 State selection lookup table
b = 1 b = 2 b = 3 b = 4 b = 5 b = 6Sector 1 1 0 2 5 7 6Sector 2 5 7 6 4 0 2Sector 3 4 0 1 6 7 3Sector 4 6 7 5 2 0 1Sector 5 2 0 4 3 7 5Sector 6 3 7 6 1 0 4
That output power can be calculated from (10) Figure 6illustrates the estimation of actual output power and flux fromstator voltage and current together with mechanical angularspeedThese estimated values are the feedback for the outputpower and stator flux controls
4 Speed Estimation Using Neural Network
In MRAS technique some state variables 119883119889and 119883
119902(eg
rotor flux linkage components 120595119889119903
and 120595119902119903 or back-EMF
components 119890119889 119890119902 etc) of the inductionmachine (which are
obtained by using measured quantities eg stator voltagesand currents) are estimated in a reference model and arethen compared with state variables
119883119889and
119883119902estimated by
using an adaptive model The difference between these statevariables is then formulated into a speed tuning signal (120576)which is then an input into an adaptation mechanism whichoutputs the estimated rotor speed ()
Speed estimator using ANN is a part of a model referenceadaptive system (MRAS) where ANN takes the role of theadaptive model ANN contains the adjustable and constantweights and the adjustable weights are proportional to therotor speed The adjustable weights are changed by usingthe error between the outputs of the reference and adaptivemodel Figure 7 shows the MRAS-based speed estimationscheme which contains an ANN with BPN adaptationtechnique [17]
The outputs of the reference model are the rotor fluxlinkage components in stationary reference frame given by
120595119889119903
=
119871119903
119871119898
[int (V119889119904
minus 119877119878119894119889119904) 119889119905 minus 119871
1015840
119904119894119889119904]
120595119902119903
=
119871119903
119871119898
[int (V119902119904
minus 119877119878119894119902119904) 119889119905 minus 119871
1015840
119904119894119902119904]
(21)
These two equations do not contain the rotor speed anddescribe the reference model The equations of adaptivemodel are given by
119889119903
=
1
119879
int (119871119898119894119889119904
minus 119889119903
minus 120596119903119879119903119902119903) 119889119905
119902119903
=
1
119879
int (119871119898119894119902119904
minus 119902119903
minus 120596119903119879119903119889119903) 119889119905
(22)
It is possible to implement (22) by a two-layer ANN contain-ing weights 119882
1(= 1 minus 119862) 119882
2(= 120596119903119879119903119862) and 119882
3(= 119862119897
119898)
that 119862 = 119879119879119903 119879 119879119903are sampling time and rotor time
Advances in Artificial Neural Systems 5
bp
0 ΔPout
+1
minus1
(a)
0
b120582
Δ120582s
1
(b)
Figure 4 Characteristics of the hysteresis controllers (a) power controller and (b) flux controller
b120582Δ120582s
ΔPout
Pout
minus
minus
+
+
120579s
a
bc
bp
120582s
Plowastout
120582lowasts
Figure 5 Block diagram of the inverter state selection
120582s120579s
rarr120582s
rarr s minus Rs
rarri s
s
rarr srarri s
Pout
120596m
Te 2
3p Im(rarri s
rarr120582lowast
s )
Figure 6 Estimation of actual output power and stator flux linkage
Reference model
Artificial neural network(ANN)
Back propagation weightadjustment
w2
sum
sum
120576d
120576q
1Z 1Z
1Z 1Z
iqs
ids
qs
ds
dr
qr
120595dr
120595qr
minus
Figure 7 MRAS-based rotor speed estimator using ANN
6 Advances in Artificial Neural Systems
w2
w1
w1
minusw2
w3
w3
ids(k minus 1)
iqs(k minus 1)
dr(k minus 1)
qr(k minus 1) dr(k)
qr(k)
Figure 8 ANN model for the estimation of rotor flux linkage
Power controller
Fluxcontroller
Stateselector
Voltage source inverter
IM
Load
Voltage sensor
Current sensor
Fluxestimator
Powerestimator
Speedcontroller
120596lowastm
Plowastout
bp
120582lowasts
120579s
120582s
Pout
rarr s
rarri s
120596m
Vdc
abcb120582
Figure 9 Block diagram of direct output power and flux control system for IM for closed-loop speed control
1400
1200
1000
800
600
400
200
0
minus200
Time (s)
Pow
er (W
)
0 05 1 15 2 25 3
Motor powerReference power
(a)
600
500
400
300
200
100
0
minus100
Spee
d (r
pm)
Time (s)0 05 1 15 2 25 3
Motor powerReference power
(b)
Figure 10 Simulation results DPC of induction motor (a) electromagnetic power and (b) rotor speed
Advances in Artificial Neural Systems 7
wr_est1
wr estimator
wr_est
Reference model
I_qs
I_ds
V_qs
V_ds
phi_dr
phi_qrLow pass filter
Zp
adjustment
ed
eqwr
Artificial neural network
I_qs
I_ds
wr
ANN-MRAS
I_qs
I_ds
wr
4
3
2
1
1z
1z+minus
+minus
iqs
ids
qs
ds
Lls
philowast_dr
philowast_qr
philowast_dr
philowast_qr
phi_dr_est
phi_qr_est
phi_dr
phi_qr
phi_dr_est
phi_qr_est
num(z)
den(z)
Back propagation
Figure 11 ANN based model of speed estimation
1500
1000
500
0
Spee
d (r
ads
)
MotorDifferentiated speedEstimation speed
Time (s)0 05 1 15
Figure 12 Response comparison of actual machine and ANN basedspeed estimator with error
constant The variable ANN weight1198822is proportional to the
rotor speed By using the backward difference method theequation of adaptive model is given below
119889119903
(119896) = 1198821119889119903
(119896 minus 1) minus 1198822119902119903
(119896 minus 1) + 1198823119894119889119904
(119896 minus 1)
119902119903
(119896) = 1198821119902119903
(119896 minus 1) minus 1198822119889119903
(119896 minus 1) + 1198823119894119902119904
(119896 minus 1)
(23)
which gives the value of rotor flux at the119870th sampling instantThese equations can be visualized by the very simple two-layer ANN shown in Figure 8
After taking learning factor 120578 andmomentum term 120572 intoaccount the estimated rotor speed is given below
119903(119896)
= 119903(119896 minus 1)
+
120578
119879
minus [120595119889119903
(119896) minus 119889119903
(119896)] 119902119903
(119896 minus 1)
+ [120595119902119903
(119896) minus 119902119903
(119896)] 119889119903
(119896 minus 1)
+
120572
119879
Δ1199082(119896 minus 1)
(24)
If the value of learning rate (120578) is chosen high it may lead tooscillations in the outputs of ANN and 120572 is chosen in rangebetween 01 and 08 The inclusion of momentum term intothe weight adjustment mechanism can significantly increasethe convergence which is extremely useful when the ANNshown in Figure 8 is used to estimate in real time [13]
8 Advances in Artificial Neural Systems
40
30
20
10
0
Pow
er (W
)
Reference powerElectromagnetic power
Time (s)0 05 1 15 2 25 3
minus10
minus20
minus30
(a)
2000
1800
1600
1400
1200
1000
800
600
400
200
0
Spee
d (r
pm)
Reference speedMotor speed
Time (s)0 05 1 15 2 25 3
(b)
Figure 13 Simulation results of DPC based on ANN of induction motor (a) electromagnetic power and (b) rotor speed
Table 3 Induction motor parameter
Parameter Value119875119899
2 hp120596rated 1500 [rpm]119881 300 [V]119868119899
21 [A]119869 0004 [kgsdotm2]119901 2119877119904
10 [Ω]119871119898
330 [mH]119871119897119904 119871119897119903
04 [mH]119861 0002 [Nsdotm(radsec)]
5 Simulation Results
51 Direct Power Control The simulation is carried out usingMATLAB The whole system setup is shown in Figure 9 Itcontains the power and flux controllers as inner loops andthe speed loop as outer loop
The sampling frequency is chosen as 50 kHz The hys-teresis tolerance for both power and flux controller is 1 ofrespective reference values For the PI speed controller (see(14)) the proportional gain is tuned to 119896
119901= 90 the overall
coefficient of the integral part is set to 01 which yieldedan integral gain 119896
119894= 5000 at the 50 kHz sampling rate
The reference speed profile is set in which the speed alwaysremains below the rated speed to avoid field weakening Theparameters of IM used in the simulations are listed in Table 3The simulation results are shown in Figure 10
52 Speed Estimator Based onANN Implementation ofANNbased speed estimator is carried out in MATLABSimulinkas shown in Figure 11
The response of ANN based speed estimator is comparedwith actual machine which is shown in Figure 12
53 Direct PowerControl withANN Theresult ofANNbasedspeed estimator with DPC controller is shown in Figure 13
6 Conclusion
This paper studies the possibility of direct power controlfor speed control of induction motor It has been shownthat DPC is basically derived from direct torque control(DTC) method and the simulation results show that thiscontrol scheme for IMsmotor has all the advantages of directtorque control method One of the major advantages of DPCmethod compared to DTC method is the easier calculationof actual power Moreover this paper proposes a new MRASspeed observer for DPC controlled induction motor drivesusing neural networks for speed estimation The simulationresults show that the proposed two-layers neural networkcan identify and track the motor speed accurately duringthe whole operating region Structure and algorithm are alsosimple Overall the dynamic response of this scheme of speedestimation shows a good performance
Conflict of Interests
The authors declare that there is no conflict of interestsregarding the publication of this paper
References
[1] P C Krause O Wasynozuk and S D Sudhoff Analysis ofElectric Machinery and Drive Systems IEEE Press 2nd edition2002
[2] I Boldea and S A Nasar Electric Drives Taylor amp Francis NewYork NY USA 2006
[3] P Vas The Control of AC Machines Oxford University PressNew York NY USA 1990
[4] A M Trzynadlowski Control ofInduction Motors Academicpress 2001
Advances in Artificial Neural Systems 9
[5] G S Buja and M P Kazmierkowski ldquoDirect torque control ofPWM inverter-fed ACmotorsmdasha surveyrdquo IEEE Transactions onIndustrial Electronics vol 51 no 4 pp 744ndash757 2004
[6] G Escobar A M Stankovic J M Carrasco E Galvan andR Ortega ldquoAnalysis and design of direct power control (DPC)for a three phase synchronous rectifier via output regulationsubspacesrdquo IEEE Transactions on Power Electronics vol 18 no3 pp 823ndash830 2003
[7] S Vazquez J A Sanchez J M Carrasco J I Leon and EGalvan ldquoA model-based direct power control for three-phasepower convertersrdquo IEEE Transactions on Industrial Electronicsvol 55 no 4 pp 1647ndash1657 2008
[8] Z Dawei X Lie and B W Williams ldquoImproved direct powercontrol of grid-connected DCAC convertersrdquo IEEE Transac-tions on Power Electronics vol 24 no 5 pp 1280ndash1292 2009
[9] L Xu and P Cartwright ldquoDirect active and reactive powercontrol of DFIG for wind energy generationrdquo IEEE Transactionson Energy Conversion vol 21 no 3 pp 750ndash758 2006
[10] D Zhi and L Xu ldquoDirect power control of DFIG with constantswitching frequency and improved transient performancerdquoIEEE Transactions on Energy Conversion vol 22 no 1 pp 110ndash118 2007
[11] Y Errami M Benchagra M Hilal M Maaroufi and M Ouas-said ldquoControl strategy for PMSG wind farm based on MPPTand direct power controlrdquo in Proceedings of the InternationalConference onMultimedia Computing and Systems (ICMCS rsquo12)pp 1125ndash1130 May 2012
[12] A Harrouz A Benatiallah and O Harrouz ldquoDirect powercontrol of a PMSG dedicated to standalone wind energysystemsrdquo in Proceedings of the 8th International Conference andExhibition on Ecological Vehicles and Renewable Energies (EVERrsquo13) pp 1ndash5 March 2013
[13] P Vas Sensorless Vector and Direct Torque Control OxfordUniversity Press 1998
[14] K Sedhuraman S Himavathi and A MuthuramalingamldquoPerformances comparison of neural architectures for on-linespeed estimation in sensorless IM drivesrdquo World Academy ofScience Engineering and Technology no 60 pp 1318ndash1325 2011
[15] S Meziane and R Benalla ldquoMRAS based speed control ofsensorlessinduction motor drivesrdquo ICGST-ACSE Journal vol 7no 1 pp 43ndash500 2007
[16] E Levi and M Wang ldquoImpact of parameter variations onspeed estimation in sensorless rotor flux oriented inductionmachinesrdquo in Proceedings of the 7th International Conferenceon Power Electronics and Variable Speed Drives pp 305ndash310London UK September 1998
[17] M E Elbulk L Tong and I Husain ldquoNeural-network-basedmodel reference adaptive systems for high-performance motordrives and motion controlsrdquo IEEE Transactions on IndustryApplications vol 38 no 3 pp 879ndash886 2002
[18] M T Hagan B Demuth andM BealeNeural Network DesignCengage Learning Pvt Ltd New Delhi India 2008
[19] S-H Kim T-S Park J-N Yoo and G-T Park ldquoSpeed-sensorless vector control of an induction motor using neuralnetwork speed estimationrdquo IEEE Transactions on IndustrialElectronics vol 48 no 3 pp 609ndash614 2001
[20] M Kuchar P Brandstetter and M Kaduch ldquoSensorless induc-tion motor drive with neural networkrdquo in Proceedings ofthe IEEE 35th Annual Power Electronics Specialists Conference(PESC rsquo04) pp 3301ndash3305 June 2004
[21] S Maiti V Verma C Chakraborty and Y Hori ldquoAn adaptivespeed sensorless induction motor drive with artificial neuralnetwork for stability enhancementrdquo IEEE Transactions onIndustrial Informatics vol 8 no 4 pp 757ndash766 2012
[22] P Girovsky and J Timko ldquoShaft sensor-less FOC control ofan induction motor using neural estimatorsrdquo Acta PolytechnicaHungarica vol 9 no 4 pp 31ndash45 2012
[23] D P Kothari and I J Nagrath Electric Machines chapter 9 TataMcGraw-Hill Education Private Limited 2004
[24] G Escobar A M Stankovic J M Carrasco E Galvan andR Ortega ldquoAnalysis and design of direct power control (DPC)for a three phase synchronous rectifier via output regulationsubspacesrdquo IEEE Transactions on Power Electronics vol 18 no3 pp 823ndash830 2003
[25] N R N Idris and A H M Yatim ldquoDirect torque controlof induction machines with constant switching frequency andreduced torque ripplerdquo IEEE Transactions on Industrial Elec-tronics vol 51 no 4 pp 758ndash767 2004
[26] S-H Huh K-B Lee D-W Kim I Choy and G-T ParkldquoSensorless speed control system using a neural networkrdquoInternational Journal of Control Automation and Systems vol3 no 4 pp 612ndash619 2005
Submit your manuscripts athttpwwwhindawicom
Computer Games Technology
International Journal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Distributed Sensor Networks
International Journal of
Advances in
FuzzySystems
Hindawi Publishing Corporationhttpwwwhindawicom
Volume 2014
International Journal of
ReconfigurableComputing
Hindawi Publishing Corporation httpwwwhindawicom Volume 2014
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Applied Computational Intelligence and Soft Computing
thinspAdvancesthinspinthinsp
Artificial Intelligence
HindawithinspPublishingthinspCorporationhttpwwwhindawicom Volumethinsp2014
Advances inSoftware EngineeringHindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Electrical and Computer Engineering
Journal of
Journal of
Computer Networks and Communications
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Hindawi Publishing Corporation
httpwwwhindawicom Volume 2014
Advances in
Multimedia
International Journal of
Biomedical Imaging
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
ArtificialNeural Systems
Advances in
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
RoboticsJournal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Computational Intelligence and Neuroscience
Industrial EngineeringJournal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Modelling amp Simulation in EngineeringHindawi Publishing Corporation httpwwwhindawicom Volume 2014
The Scientific World JournalHindawi Publishing Corporation httpwwwhindawicom Volume 2014
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Human-ComputerInteraction
Advances in
Computer EngineeringAdvances in
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Advances in Artificial Neural Systems 3
Power control
Fluxcontrol
Stateselection
VS inverter IM
Load
Figure 2 General block diagram of direct power and flux controlsystem
31 Flux Control Principles Fixed linkage flux can provideenough electromagnetic torque and avoid magnetizing cur-rent saturation in the iron core of the IMTherefore in directtorque and flux control as well as in the proposed directpower and flux controls in the subsequent chapters the fluxis maintained constant [25] In the stator stationary referenceframe the frame rotation speed is zero and the rotor voltageis zero as well (for squirrel-cage IMs) Thus
119881119904= 119877119904
119894119904+
119889
120582119904
119889119905
(5)
0 = 119877
119903
119894119903+
119889
120582119903
119889119905
minus 119895120596119903
120582119903
(6)
With neglecting the small voltage drop across the statorresistance we have
119881119904cong
119889
120582119904
119889119905
(7)
Integrating (7) and writing it in a discrete form we obtain
120582119904(119905119899+1
) =
120582119904(119905119899) +
119881119904Δ119905 (8)
That is
Δ
120582119904cong
119881119904Δ119905 (9)
where Δ119905 = 119905119899+1
minus 119905119899equals the switching interval Therefore
within a switching interval Δ119905 the increase of stator flux isproportional to the stator voltage space vector This is theprinciple of direct flux control in DPC
32 Power Control Principles From the power flow charts ininduction motor it is evident that real output power is thepart that produces the torque and is what the user of thesystem is mostly interested in The output real power is givenby
119875out = 119879119890120596119898
= 119879119890
120596119903
119901
(10)
119879119890=
2
3
119901 Im (119894119904
120582119904
lowast
) =
2
3
119901 Im(
119871119903
120582119904minus 119871119898
120582119903
119871119904119871119903minus 119871
2
119898
120582119904
lowast
)
= minus
2
3
119901
119871119898
119871
2
120590
Im (
120582119903
120582119904
lowast
) =
2
3
119901
119871119898
119871
2
120590
Im (
120582119904
120582119903
lowast
)
=
2
3
119901
119871119898
119871
2
120590
120582119904120582119903sin (120579119904minus 120579119903)
(11)
120596m
Δ120596m+
minus
PlowastoutTlowast
e120596lowastm
Figure 3 Generating the output power reference obtained fromspeed loop
Substituting the torque in (10) with (11) the output powerbecomes
119875out =2
3
120596119903
119871119898
119871
2
120590
Im (
120582119904
120582119903
lowast
) =
2
3
119871119898
119871
2
120590
120596119903120582119904120582119903sin (120579119904minus 120579119903)
(12)
Since the magnitude of the stator flux is kept constantand the rotor flux does not change much due to its inertiathe rotor speed and angle can be considered constant tooThe formula above shows that the change of output powerdepends only on the change of stator flux angle The statorvoltage vector that can increase the stator angle needs tobe raised in order to increase the output power The realoutput power equation obtained above is only valid forexplanation of the principles of power control However itis not appropriate for the purpose of estimating the actualpower in simulations
33 Output Power Reference The output power reference isthe command value or set point for the power control In aclosed-loop speed control system the reference of the powercontroller is obtained from the output of the PI-type speedcontroller (see Figure 3) The speed error is defined as thedifference of the reference speed and the estimated actualspeed
Δ120596119898
= 120596
lowast
119898minus 120596119898 (13)
where 120596
lowast
119898is the reference speed (the asterisk denotes a
reference value) Then the reference torque can be obtainedthrough a conventional PI controller as
119879
lowast
119890= 119870119901(Δ120596119898) + 119870119894int (Δ120596
119898) 119889119905 (14)
The continuous standard form above can also be expressed ina discrete incremental PI control form which ismore suitablefor the digital implementation
119879
lowast
119890(119905119899+1
) = 119879
lowast
119890(119905119899) + 119870119901[Δ120596119898
(119905119899) minus Δ120596
119898(119905119899minus1
)]
+ 119870119894Δ119879Δ120596
119898(119905119899)
(15)
The subscript (119899) denotes the current sampling instant (119899minus1)
is the last instant and (119899+1) is the next oneThe proportionalgain is denoted by 119870
119901 119870119894is the integral gain which equals
119870119901divided by the integral time constant 119879
119894 and Δ119879 is the
sampling time interval between the 119899 and (119899 + 1) samplinginstants The output power reference according to (10) istherefore expressed as
119875
lowast
out = 119879
lowast
119890120596
lowast
119898 (16)
4 Advances in Artificial Neural Systems
Table 1 Combination of the power and flux controller outputs
119887119901= minus1 119887
119901= 0 119887
119901= 1
1 2 34 5 6
The process of obtaining the output power reference from thespeed reference is illustrated in Figure 3 For simulations theactual motor speed 120596
119898can be obtained as
120596119898
=
1
119869
int (119879119890minus 119879119897119889) 119889119905 (17)
In practice the speed is either measured directly or estimatedfrom the current and voltage signals The magnitude of thestator flux is kept constant in the simulation thus the fluxreference 120582
lowast
119904is a constant The error of the stator flux is
Δ120582119904= 120582
lowast
119904minus 120582119904 (18)
34 Power and Flux Hysteresis Controllers Both the outputpower and the stator flux controllers are of hysteresis typeDepending on the control error the output of the controlleris set to two or three discrete values The power controllerhas a three-level output [24 26] The values are 1 0 andminus1 representing an increase no change and a decrease ofthe controlled variable respectively The number of fluxcontroller output levels is two with 1 and 0 meaning increaseand decrease commands respectively Figure 4 illustratescharacteristics of these two controllers
35 Switching Table The task of the state selector in thedirect power control is to combine the outputs of the powercontroller and flux controller to select the values of theswitching variables a b and c These variables describe therequired voltage vectors of the inverter To make it easier toimplement the combination of the two controller outputs canbe expressed as follows
119887 = 3119887120582+ 119887119901+ 2 (19)
In the above equation variable 119887 = 1 2 3 4 5 6 while 119887120582
=
(0 1) and 119887119901= (minus1 0 1)
Alternatively (19) can also be represented by Table 1A whole stator flux cycle of 360∘ is divided equally into
6 sectors each one spanning 60∘ Combining with the sectornumbers from 1 through 6 produces lookup Table 2 for thestate selection [24]The concept of state selection is illustratedin Figure 5
Note that the stator flux angle 120579119904must be converted to a
sector number from 1 through 6 for the use of Table 2 for stateselection
36 Estimation of Stator Flux and Output Power The estima-tion of flux is implemented by integration of (5)
120582119904= int (V
119904minus
119894119904119877119904) 119889119905 (20)
Table 2 State selection lookup table
b = 1 b = 2 b = 3 b = 4 b = 5 b = 6Sector 1 1 0 2 5 7 6Sector 2 5 7 6 4 0 2Sector 3 4 0 1 6 7 3Sector 4 6 7 5 2 0 1Sector 5 2 0 4 3 7 5Sector 6 3 7 6 1 0 4
That output power can be calculated from (10) Figure 6illustrates the estimation of actual output power and flux fromstator voltage and current together with mechanical angularspeedThese estimated values are the feedback for the outputpower and stator flux controls
4 Speed Estimation Using Neural Network
In MRAS technique some state variables 119883119889and 119883
119902(eg
rotor flux linkage components 120595119889119903
and 120595119902119903 or back-EMF
components 119890119889 119890119902 etc) of the inductionmachine (which are
obtained by using measured quantities eg stator voltagesand currents) are estimated in a reference model and arethen compared with state variables
119883119889and
119883119902estimated by
using an adaptive model The difference between these statevariables is then formulated into a speed tuning signal (120576)which is then an input into an adaptation mechanism whichoutputs the estimated rotor speed ()
Speed estimator using ANN is a part of a model referenceadaptive system (MRAS) where ANN takes the role of theadaptive model ANN contains the adjustable and constantweights and the adjustable weights are proportional to therotor speed The adjustable weights are changed by usingthe error between the outputs of the reference and adaptivemodel Figure 7 shows the MRAS-based speed estimationscheme which contains an ANN with BPN adaptationtechnique [17]
The outputs of the reference model are the rotor fluxlinkage components in stationary reference frame given by
120595119889119903
=
119871119903
119871119898
[int (V119889119904
minus 119877119878119894119889119904) 119889119905 minus 119871
1015840
119904119894119889119904]
120595119902119903
=
119871119903
119871119898
[int (V119902119904
minus 119877119878119894119902119904) 119889119905 minus 119871
1015840
119904119894119902119904]
(21)
These two equations do not contain the rotor speed anddescribe the reference model The equations of adaptivemodel are given by
119889119903
=
1
119879
int (119871119898119894119889119904
minus 119889119903
minus 120596119903119879119903119902119903) 119889119905
119902119903
=
1
119879
int (119871119898119894119902119904
minus 119902119903
minus 120596119903119879119903119889119903) 119889119905
(22)
It is possible to implement (22) by a two-layer ANN contain-ing weights 119882
1(= 1 minus 119862) 119882
2(= 120596119903119879119903119862) and 119882
3(= 119862119897
119898)
that 119862 = 119879119879119903 119879 119879119903are sampling time and rotor time
Advances in Artificial Neural Systems 5
bp
0 ΔPout
+1
minus1
(a)
0
b120582
Δ120582s
1
(b)
Figure 4 Characteristics of the hysteresis controllers (a) power controller and (b) flux controller
b120582Δ120582s
ΔPout
Pout
minus
minus
+
+
120579s
a
bc
bp
120582s
Plowastout
120582lowasts
Figure 5 Block diagram of the inverter state selection
120582s120579s
rarr120582s
rarr s minus Rs
rarri s
s
rarr srarri s
Pout
120596m
Te 2
3p Im(rarri s
rarr120582lowast
s )
Figure 6 Estimation of actual output power and stator flux linkage
Reference model
Artificial neural network(ANN)
Back propagation weightadjustment
w2
sum
sum
120576d
120576q
1Z 1Z
1Z 1Z
iqs
ids
qs
ds
dr
qr
120595dr
120595qr
minus
Figure 7 MRAS-based rotor speed estimator using ANN
6 Advances in Artificial Neural Systems
w2
w1
w1
minusw2
w3
w3
ids(k minus 1)
iqs(k minus 1)
dr(k minus 1)
qr(k minus 1) dr(k)
qr(k)
Figure 8 ANN model for the estimation of rotor flux linkage
Power controller
Fluxcontroller
Stateselector
Voltage source inverter
IM
Load
Voltage sensor
Current sensor
Fluxestimator
Powerestimator
Speedcontroller
120596lowastm
Plowastout
bp
120582lowasts
120579s
120582s
Pout
rarr s
rarri s
120596m
Vdc
abcb120582
Figure 9 Block diagram of direct output power and flux control system for IM for closed-loop speed control
1400
1200
1000
800
600
400
200
0
minus200
Time (s)
Pow
er (W
)
0 05 1 15 2 25 3
Motor powerReference power
(a)
600
500
400
300
200
100
0
minus100
Spee
d (r
pm)
Time (s)0 05 1 15 2 25 3
Motor powerReference power
(b)
Figure 10 Simulation results DPC of induction motor (a) electromagnetic power and (b) rotor speed
Advances in Artificial Neural Systems 7
wr_est1
wr estimator
wr_est
Reference model
I_qs
I_ds
V_qs
V_ds
phi_dr
phi_qrLow pass filter
Zp
adjustment
ed
eqwr
Artificial neural network
I_qs
I_ds
wr
ANN-MRAS
I_qs
I_ds
wr
4
3
2
1
1z
1z+minus
+minus
iqs
ids
qs
ds
Lls
philowast_dr
philowast_qr
philowast_dr
philowast_qr
phi_dr_est
phi_qr_est
phi_dr
phi_qr
phi_dr_est
phi_qr_est
num(z)
den(z)
Back propagation
Figure 11 ANN based model of speed estimation
1500
1000
500
0
Spee
d (r
ads
)
MotorDifferentiated speedEstimation speed
Time (s)0 05 1 15
Figure 12 Response comparison of actual machine and ANN basedspeed estimator with error
constant The variable ANN weight1198822is proportional to the
rotor speed By using the backward difference method theequation of adaptive model is given below
119889119903
(119896) = 1198821119889119903
(119896 minus 1) minus 1198822119902119903
(119896 minus 1) + 1198823119894119889119904
(119896 minus 1)
119902119903
(119896) = 1198821119902119903
(119896 minus 1) minus 1198822119889119903
(119896 minus 1) + 1198823119894119902119904
(119896 minus 1)
(23)
which gives the value of rotor flux at the119870th sampling instantThese equations can be visualized by the very simple two-layer ANN shown in Figure 8
After taking learning factor 120578 andmomentum term 120572 intoaccount the estimated rotor speed is given below
119903(119896)
= 119903(119896 minus 1)
+
120578
119879
minus [120595119889119903
(119896) minus 119889119903
(119896)] 119902119903
(119896 minus 1)
+ [120595119902119903
(119896) minus 119902119903
(119896)] 119889119903
(119896 minus 1)
+
120572
119879
Δ1199082(119896 minus 1)
(24)
If the value of learning rate (120578) is chosen high it may lead tooscillations in the outputs of ANN and 120572 is chosen in rangebetween 01 and 08 The inclusion of momentum term intothe weight adjustment mechanism can significantly increasethe convergence which is extremely useful when the ANNshown in Figure 8 is used to estimate in real time [13]
8 Advances in Artificial Neural Systems
40
30
20
10
0
Pow
er (W
)
Reference powerElectromagnetic power
Time (s)0 05 1 15 2 25 3
minus10
minus20
minus30
(a)
2000
1800
1600
1400
1200
1000
800
600
400
200
0
Spee
d (r
pm)
Reference speedMotor speed
Time (s)0 05 1 15 2 25 3
(b)
Figure 13 Simulation results of DPC based on ANN of induction motor (a) electromagnetic power and (b) rotor speed
Table 3 Induction motor parameter
Parameter Value119875119899
2 hp120596rated 1500 [rpm]119881 300 [V]119868119899
21 [A]119869 0004 [kgsdotm2]119901 2119877119904
10 [Ω]119871119898
330 [mH]119871119897119904 119871119897119903
04 [mH]119861 0002 [Nsdotm(radsec)]
5 Simulation Results
51 Direct Power Control The simulation is carried out usingMATLAB The whole system setup is shown in Figure 9 Itcontains the power and flux controllers as inner loops andthe speed loop as outer loop
The sampling frequency is chosen as 50 kHz The hys-teresis tolerance for both power and flux controller is 1 ofrespective reference values For the PI speed controller (see(14)) the proportional gain is tuned to 119896
119901= 90 the overall
coefficient of the integral part is set to 01 which yieldedan integral gain 119896
119894= 5000 at the 50 kHz sampling rate
The reference speed profile is set in which the speed alwaysremains below the rated speed to avoid field weakening Theparameters of IM used in the simulations are listed in Table 3The simulation results are shown in Figure 10
52 Speed Estimator Based onANN Implementation ofANNbased speed estimator is carried out in MATLABSimulinkas shown in Figure 11
The response of ANN based speed estimator is comparedwith actual machine which is shown in Figure 12
53 Direct PowerControl withANN Theresult ofANNbasedspeed estimator with DPC controller is shown in Figure 13
6 Conclusion
This paper studies the possibility of direct power controlfor speed control of induction motor It has been shownthat DPC is basically derived from direct torque control(DTC) method and the simulation results show that thiscontrol scheme for IMsmotor has all the advantages of directtorque control method One of the major advantages of DPCmethod compared to DTC method is the easier calculationof actual power Moreover this paper proposes a new MRASspeed observer for DPC controlled induction motor drivesusing neural networks for speed estimation The simulationresults show that the proposed two-layers neural networkcan identify and track the motor speed accurately duringthe whole operating region Structure and algorithm are alsosimple Overall the dynamic response of this scheme of speedestimation shows a good performance
Conflict of Interests
The authors declare that there is no conflict of interestsregarding the publication of this paper
References
[1] P C Krause O Wasynozuk and S D Sudhoff Analysis ofElectric Machinery and Drive Systems IEEE Press 2nd edition2002
[2] I Boldea and S A Nasar Electric Drives Taylor amp Francis NewYork NY USA 2006
[3] P Vas The Control of AC Machines Oxford University PressNew York NY USA 1990
[4] A M Trzynadlowski Control ofInduction Motors Academicpress 2001
Advances in Artificial Neural Systems 9
[5] G S Buja and M P Kazmierkowski ldquoDirect torque control ofPWM inverter-fed ACmotorsmdasha surveyrdquo IEEE Transactions onIndustrial Electronics vol 51 no 4 pp 744ndash757 2004
[6] G Escobar A M Stankovic J M Carrasco E Galvan andR Ortega ldquoAnalysis and design of direct power control (DPC)for a three phase synchronous rectifier via output regulationsubspacesrdquo IEEE Transactions on Power Electronics vol 18 no3 pp 823ndash830 2003
[7] S Vazquez J A Sanchez J M Carrasco J I Leon and EGalvan ldquoA model-based direct power control for three-phasepower convertersrdquo IEEE Transactions on Industrial Electronicsvol 55 no 4 pp 1647ndash1657 2008
[8] Z Dawei X Lie and B W Williams ldquoImproved direct powercontrol of grid-connected DCAC convertersrdquo IEEE Transac-tions on Power Electronics vol 24 no 5 pp 1280ndash1292 2009
[9] L Xu and P Cartwright ldquoDirect active and reactive powercontrol of DFIG for wind energy generationrdquo IEEE Transactionson Energy Conversion vol 21 no 3 pp 750ndash758 2006
[10] D Zhi and L Xu ldquoDirect power control of DFIG with constantswitching frequency and improved transient performancerdquoIEEE Transactions on Energy Conversion vol 22 no 1 pp 110ndash118 2007
[11] Y Errami M Benchagra M Hilal M Maaroufi and M Ouas-said ldquoControl strategy for PMSG wind farm based on MPPTand direct power controlrdquo in Proceedings of the InternationalConference onMultimedia Computing and Systems (ICMCS rsquo12)pp 1125ndash1130 May 2012
[12] A Harrouz A Benatiallah and O Harrouz ldquoDirect powercontrol of a PMSG dedicated to standalone wind energysystemsrdquo in Proceedings of the 8th International Conference andExhibition on Ecological Vehicles and Renewable Energies (EVERrsquo13) pp 1ndash5 March 2013
[13] P Vas Sensorless Vector and Direct Torque Control OxfordUniversity Press 1998
[14] K Sedhuraman S Himavathi and A MuthuramalingamldquoPerformances comparison of neural architectures for on-linespeed estimation in sensorless IM drivesrdquo World Academy ofScience Engineering and Technology no 60 pp 1318ndash1325 2011
[15] S Meziane and R Benalla ldquoMRAS based speed control ofsensorlessinduction motor drivesrdquo ICGST-ACSE Journal vol 7no 1 pp 43ndash500 2007
[16] E Levi and M Wang ldquoImpact of parameter variations onspeed estimation in sensorless rotor flux oriented inductionmachinesrdquo in Proceedings of the 7th International Conferenceon Power Electronics and Variable Speed Drives pp 305ndash310London UK September 1998
[17] M E Elbulk L Tong and I Husain ldquoNeural-network-basedmodel reference adaptive systems for high-performance motordrives and motion controlsrdquo IEEE Transactions on IndustryApplications vol 38 no 3 pp 879ndash886 2002
[18] M T Hagan B Demuth andM BealeNeural Network DesignCengage Learning Pvt Ltd New Delhi India 2008
[19] S-H Kim T-S Park J-N Yoo and G-T Park ldquoSpeed-sensorless vector control of an induction motor using neuralnetwork speed estimationrdquo IEEE Transactions on IndustrialElectronics vol 48 no 3 pp 609ndash614 2001
[20] M Kuchar P Brandstetter and M Kaduch ldquoSensorless induc-tion motor drive with neural networkrdquo in Proceedings ofthe IEEE 35th Annual Power Electronics Specialists Conference(PESC rsquo04) pp 3301ndash3305 June 2004
[21] S Maiti V Verma C Chakraborty and Y Hori ldquoAn adaptivespeed sensorless induction motor drive with artificial neuralnetwork for stability enhancementrdquo IEEE Transactions onIndustrial Informatics vol 8 no 4 pp 757ndash766 2012
[22] P Girovsky and J Timko ldquoShaft sensor-less FOC control ofan induction motor using neural estimatorsrdquo Acta PolytechnicaHungarica vol 9 no 4 pp 31ndash45 2012
[23] D P Kothari and I J Nagrath Electric Machines chapter 9 TataMcGraw-Hill Education Private Limited 2004
[24] G Escobar A M Stankovic J M Carrasco E Galvan andR Ortega ldquoAnalysis and design of direct power control (DPC)for a three phase synchronous rectifier via output regulationsubspacesrdquo IEEE Transactions on Power Electronics vol 18 no3 pp 823ndash830 2003
[25] N R N Idris and A H M Yatim ldquoDirect torque controlof induction machines with constant switching frequency andreduced torque ripplerdquo IEEE Transactions on Industrial Elec-tronics vol 51 no 4 pp 758ndash767 2004
[26] S-H Huh K-B Lee D-W Kim I Choy and G-T ParkldquoSensorless speed control system using a neural networkrdquoInternational Journal of Control Automation and Systems vol3 no 4 pp 612ndash619 2005
Submit your manuscripts athttpwwwhindawicom
Computer Games Technology
International Journal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Distributed Sensor Networks
International Journal of
Advances in
FuzzySystems
Hindawi Publishing Corporationhttpwwwhindawicom
Volume 2014
International Journal of
ReconfigurableComputing
Hindawi Publishing Corporation httpwwwhindawicom Volume 2014
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Applied Computational Intelligence and Soft Computing
thinspAdvancesthinspinthinsp
Artificial Intelligence
HindawithinspPublishingthinspCorporationhttpwwwhindawicom Volumethinsp2014
Advances inSoftware EngineeringHindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Electrical and Computer Engineering
Journal of
Journal of
Computer Networks and Communications
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Hindawi Publishing Corporation
httpwwwhindawicom Volume 2014
Advances in
Multimedia
International Journal of
Biomedical Imaging
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
ArtificialNeural Systems
Advances in
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
RoboticsJournal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Computational Intelligence and Neuroscience
Industrial EngineeringJournal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Modelling amp Simulation in EngineeringHindawi Publishing Corporation httpwwwhindawicom Volume 2014
The Scientific World JournalHindawi Publishing Corporation httpwwwhindawicom Volume 2014
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Human-ComputerInteraction
Advances in
Computer EngineeringAdvances in
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
4 Advances in Artificial Neural Systems
Table 1 Combination of the power and flux controller outputs
119887119901= minus1 119887
119901= 0 119887
119901= 1
1 2 34 5 6
The process of obtaining the output power reference from thespeed reference is illustrated in Figure 3 For simulations theactual motor speed 120596
119898can be obtained as
120596119898
=
1
119869
int (119879119890minus 119879119897119889) 119889119905 (17)
In practice the speed is either measured directly or estimatedfrom the current and voltage signals The magnitude of thestator flux is kept constant in the simulation thus the fluxreference 120582
lowast
119904is a constant The error of the stator flux is
Δ120582119904= 120582
lowast
119904minus 120582119904 (18)
34 Power and Flux Hysteresis Controllers Both the outputpower and the stator flux controllers are of hysteresis typeDepending on the control error the output of the controlleris set to two or three discrete values The power controllerhas a three-level output [24 26] The values are 1 0 andminus1 representing an increase no change and a decrease ofthe controlled variable respectively The number of fluxcontroller output levels is two with 1 and 0 meaning increaseand decrease commands respectively Figure 4 illustratescharacteristics of these two controllers
35 Switching Table The task of the state selector in thedirect power control is to combine the outputs of the powercontroller and flux controller to select the values of theswitching variables a b and c These variables describe therequired voltage vectors of the inverter To make it easier toimplement the combination of the two controller outputs canbe expressed as follows
119887 = 3119887120582+ 119887119901+ 2 (19)
In the above equation variable 119887 = 1 2 3 4 5 6 while 119887120582
=
(0 1) and 119887119901= (minus1 0 1)
Alternatively (19) can also be represented by Table 1A whole stator flux cycle of 360∘ is divided equally into
6 sectors each one spanning 60∘ Combining with the sectornumbers from 1 through 6 produces lookup Table 2 for thestate selection [24]The concept of state selection is illustratedin Figure 5
Note that the stator flux angle 120579119904must be converted to a
sector number from 1 through 6 for the use of Table 2 for stateselection
36 Estimation of Stator Flux and Output Power The estima-tion of flux is implemented by integration of (5)
120582119904= int (V
119904minus
119894119904119877119904) 119889119905 (20)
Table 2 State selection lookup table
b = 1 b = 2 b = 3 b = 4 b = 5 b = 6Sector 1 1 0 2 5 7 6Sector 2 5 7 6 4 0 2Sector 3 4 0 1 6 7 3Sector 4 6 7 5 2 0 1Sector 5 2 0 4 3 7 5Sector 6 3 7 6 1 0 4
That output power can be calculated from (10) Figure 6illustrates the estimation of actual output power and flux fromstator voltage and current together with mechanical angularspeedThese estimated values are the feedback for the outputpower and stator flux controls
4 Speed Estimation Using Neural Network
In MRAS technique some state variables 119883119889and 119883
119902(eg
rotor flux linkage components 120595119889119903
and 120595119902119903 or back-EMF
components 119890119889 119890119902 etc) of the inductionmachine (which are
obtained by using measured quantities eg stator voltagesand currents) are estimated in a reference model and arethen compared with state variables
119883119889and
119883119902estimated by
using an adaptive model The difference between these statevariables is then formulated into a speed tuning signal (120576)which is then an input into an adaptation mechanism whichoutputs the estimated rotor speed ()
Speed estimator using ANN is a part of a model referenceadaptive system (MRAS) where ANN takes the role of theadaptive model ANN contains the adjustable and constantweights and the adjustable weights are proportional to therotor speed The adjustable weights are changed by usingthe error between the outputs of the reference and adaptivemodel Figure 7 shows the MRAS-based speed estimationscheme which contains an ANN with BPN adaptationtechnique [17]
The outputs of the reference model are the rotor fluxlinkage components in stationary reference frame given by
120595119889119903
=
119871119903
119871119898
[int (V119889119904
minus 119877119878119894119889119904) 119889119905 minus 119871
1015840
119904119894119889119904]
120595119902119903
=
119871119903
119871119898
[int (V119902119904
minus 119877119878119894119902119904) 119889119905 minus 119871
1015840
119904119894119902119904]
(21)
These two equations do not contain the rotor speed anddescribe the reference model The equations of adaptivemodel are given by
119889119903
=
1
119879
int (119871119898119894119889119904
minus 119889119903
minus 120596119903119879119903119902119903) 119889119905
119902119903
=
1
119879
int (119871119898119894119902119904
minus 119902119903
minus 120596119903119879119903119889119903) 119889119905
(22)
It is possible to implement (22) by a two-layer ANN contain-ing weights 119882
1(= 1 minus 119862) 119882
2(= 120596119903119879119903119862) and 119882
3(= 119862119897
119898)
that 119862 = 119879119879119903 119879 119879119903are sampling time and rotor time
Advances in Artificial Neural Systems 5
bp
0 ΔPout
+1
minus1
(a)
0
b120582
Δ120582s
1
(b)
Figure 4 Characteristics of the hysteresis controllers (a) power controller and (b) flux controller
b120582Δ120582s
ΔPout
Pout
minus
minus
+
+
120579s
a
bc
bp
120582s
Plowastout
120582lowasts
Figure 5 Block diagram of the inverter state selection
120582s120579s
rarr120582s
rarr s minus Rs
rarri s
s
rarr srarri s
Pout
120596m
Te 2
3p Im(rarri s
rarr120582lowast
s )
Figure 6 Estimation of actual output power and stator flux linkage
Reference model
Artificial neural network(ANN)
Back propagation weightadjustment
w2
sum
sum
120576d
120576q
1Z 1Z
1Z 1Z
iqs
ids
qs
ds
dr
qr
120595dr
120595qr
minus
Figure 7 MRAS-based rotor speed estimator using ANN
6 Advances in Artificial Neural Systems
w2
w1
w1
minusw2
w3
w3
ids(k minus 1)
iqs(k minus 1)
dr(k minus 1)
qr(k minus 1) dr(k)
qr(k)
Figure 8 ANN model for the estimation of rotor flux linkage
Power controller
Fluxcontroller
Stateselector
Voltage source inverter
IM
Load
Voltage sensor
Current sensor
Fluxestimator
Powerestimator
Speedcontroller
120596lowastm
Plowastout
bp
120582lowasts
120579s
120582s
Pout
rarr s
rarri s
120596m
Vdc
abcb120582
Figure 9 Block diagram of direct output power and flux control system for IM for closed-loop speed control
1400
1200
1000
800
600
400
200
0
minus200
Time (s)
Pow
er (W
)
0 05 1 15 2 25 3
Motor powerReference power
(a)
600
500
400
300
200
100
0
minus100
Spee
d (r
pm)
Time (s)0 05 1 15 2 25 3
Motor powerReference power
(b)
Figure 10 Simulation results DPC of induction motor (a) electromagnetic power and (b) rotor speed
Advances in Artificial Neural Systems 7
wr_est1
wr estimator
wr_est
Reference model
I_qs
I_ds
V_qs
V_ds
phi_dr
phi_qrLow pass filter
Zp
adjustment
ed
eqwr
Artificial neural network
I_qs
I_ds
wr
ANN-MRAS
I_qs
I_ds
wr
4
3
2
1
1z
1z+minus
+minus
iqs
ids
qs
ds
Lls
philowast_dr
philowast_qr
philowast_dr
philowast_qr
phi_dr_est
phi_qr_est
phi_dr
phi_qr
phi_dr_est
phi_qr_est
num(z)
den(z)
Back propagation
Figure 11 ANN based model of speed estimation
1500
1000
500
0
Spee
d (r
ads
)
MotorDifferentiated speedEstimation speed
Time (s)0 05 1 15
Figure 12 Response comparison of actual machine and ANN basedspeed estimator with error
constant The variable ANN weight1198822is proportional to the
rotor speed By using the backward difference method theequation of adaptive model is given below
119889119903
(119896) = 1198821119889119903
(119896 minus 1) minus 1198822119902119903
(119896 minus 1) + 1198823119894119889119904
(119896 minus 1)
119902119903
(119896) = 1198821119902119903
(119896 minus 1) minus 1198822119889119903
(119896 minus 1) + 1198823119894119902119904
(119896 minus 1)
(23)
which gives the value of rotor flux at the119870th sampling instantThese equations can be visualized by the very simple two-layer ANN shown in Figure 8
After taking learning factor 120578 andmomentum term 120572 intoaccount the estimated rotor speed is given below
119903(119896)
= 119903(119896 minus 1)
+
120578
119879
minus [120595119889119903
(119896) minus 119889119903
(119896)] 119902119903
(119896 minus 1)
+ [120595119902119903
(119896) minus 119902119903
(119896)] 119889119903
(119896 minus 1)
+
120572
119879
Δ1199082(119896 minus 1)
(24)
If the value of learning rate (120578) is chosen high it may lead tooscillations in the outputs of ANN and 120572 is chosen in rangebetween 01 and 08 The inclusion of momentum term intothe weight adjustment mechanism can significantly increasethe convergence which is extremely useful when the ANNshown in Figure 8 is used to estimate in real time [13]
8 Advances in Artificial Neural Systems
40
30
20
10
0
Pow
er (W
)
Reference powerElectromagnetic power
Time (s)0 05 1 15 2 25 3
minus10
minus20
minus30
(a)
2000
1800
1600
1400
1200
1000
800
600
400
200
0
Spee
d (r
pm)
Reference speedMotor speed
Time (s)0 05 1 15 2 25 3
(b)
Figure 13 Simulation results of DPC based on ANN of induction motor (a) electromagnetic power and (b) rotor speed
Table 3 Induction motor parameter
Parameter Value119875119899
2 hp120596rated 1500 [rpm]119881 300 [V]119868119899
21 [A]119869 0004 [kgsdotm2]119901 2119877119904
10 [Ω]119871119898
330 [mH]119871119897119904 119871119897119903
04 [mH]119861 0002 [Nsdotm(radsec)]
5 Simulation Results
51 Direct Power Control The simulation is carried out usingMATLAB The whole system setup is shown in Figure 9 Itcontains the power and flux controllers as inner loops andthe speed loop as outer loop
The sampling frequency is chosen as 50 kHz The hys-teresis tolerance for both power and flux controller is 1 ofrespective reference values For the PI speed controller (see(14)) the proportional gain is tuned to 119896
119901= 90 the overall
coefficient of the integral part is set to 01 which yieldedan integral gain 119896
119894= 5000 at the 50 kHz sampling rate
The reference speed profile is set in which the speed alwaysremains below the rated speed to avoid field weakening Theparameters of IM used in the simulations are listed in Table 3The simulation results are shown in Figure 10
52 Speed Estimator Based onANN Implementation ofANNbased speed estimator is carried out in MATLABSimulinkas shown in Figure 11
The response of ANN based speed estimator is comparedwith actual machine which is shown in Figure 12
53 Direct PowerControl withANN Theresult ofANNbasedspeed estimator with DPC controller is shown in Figure 13
6 Conclusion
This paper studies the possibility of direct power controlfor speed control of induction motor It has been shownthat DPC is basically derived from direct torque control(DTC) method and the simulation results show that thiscontrol scheme for IMsmotor has all the advantages of directtorque control method One of the major advantages of DPCmethod compared to DTC method is the easier calculationof actual power Moreover this paper proposes a new MRASspeed observer for DPC controlled induction motor drivesusing neural networks for speed estimation The simulationresults show that the proposed two-layers neural networkcan identify and track the motor speed accurately duringthe whole operating region Structure and algorithm are alsosimple Overall the dynamic response of this scheme of speedestimation shows a good performance
Conflict of Interests
The authors declare that there is no conflict of interestsregarding the publication of this paper
References
[1] P C Krause O Wasynozuk and S D Sudhoff Analysis ofElectric Machinery and Drive Systems IEEE Press 2nd edition2002
[2] I Boldea and S A Nasar Electric Drives Taylor amp Francis NewYork NY USA 2006
[3] P Vas The Control of AC Machines Oxford University PressNew York NY USA 1990
[4] A M Trzynadlowski Control ofInduction Motors Academicpress 2001
Advances in Artificial Neural Systems 9
[5] G S Buja and M P Kazmierkowski ldquoDirect torque control ofPWM inverter-fed ACmotorsmdasha surveyrdquo IEEE Transactions onIndustrial Electronics vol 51 no 4 pp 744ndash757 2004
[6] G Escobar A M Stankovic J M Carrasco E Galvan andR Ortega ldquoAnalysis and design of direct power control (DPC)for a three phase synchronous rectifier via output regulationsubspacesrdquo IEEE Transactions on Power Electronics vol 18 no3 pp 823ndash830 2003
[7] S Vazquez J A Sanchez J M Carrasco J I Leon and EGalvan ldquoA model-based direct power control for three-phasepower convertersrdquo IEEE Transactions on Industrial Electronicsvol 55 no 4 pp 1647ndash1657 2008
[8] Z Dawei X Lie and B W Williams ldquoImproved direct powercontrol of grid-connected DCAC convertersrdquo IEEE Transac-tions on Power Electronics vol 24 no 5 pp 1280ndash1292 2009
[9] L Xu and P Cartwright ldquoDirect active and reactive powercontrol of DFIG for wind energy generationrdquo IEEE Transactionson Energy Conversion vol 21 no 3 pp 750ndash758 2006
[10] D Zhi and L Xu ldquoDirect power control of DFIG with constantswitching frequency and improved transient performancerdquoIEEE Transactions on Energy Conversion vol 22 no 1 pp 110ndash118 2007
[11] Y Errami M Benchagra M Hilal M Maaroufi and M Ouas-said ldquoControl strategy for PMSG wind farm based on MPPTand direct power controlrdquo in Proceedings of the InternationalConference onMultimedia Computing and Systems (ICMCS rsquo12)pp 1125ndash1130 May 2012
[12] A Harrouz A Benatiallah and O Harrouz ldquoDirect powercontrol of a PMSG dedicated to standalone wind energysystemsrdquo in Proceedings of the 8th International Conference andExhibition on Ecological Vehicles and Renewable Energies (EVERrsquo13) pp 1ndash5 March 2013
[13] P Vas Sensorless Vector and Direct Torque Control OxfordUniversity Press 1998
[14] K Sedhuraman S Himavathi and A MuthuramalingamldquoPerformances comparison of neural architectures for on-linespeed estimation in sensorless IM drivesrdquo World Academy ofScience Engineering and Technology no 60 pp 1318ndash1325 2011
[15] S Meziane and R Benalla ldquoMRAS based speed control ofsensorlessinduction motor drivesrdquo ICGST-ACSE Journal vol 7no 1 pp 43ndash500 2007
[16] E Levi and M Wang ldquoImpact of parameter variations onspeed estimation in sensorless rotor flux oriented inductionmachinesrdquo in Proceedings of the 7th International Conferenceon Power Electronics and Variable Speed Drives pp 305ndash310London UK September 1998
[17] M E Elbulk L Tong and I Husain ldquoNeural-network-basedmodel reference adaptive systems for high-performance motordrives and motion controlsrdquo IEEE Transactions on IndustryApplications vol 38 no 3 pp 879ndash886 2002
[18] M T Hagan B Demuth andM BealeNeural Network DesignCengage Learning Pvt Ltd New Delhi India 2008
[19] S-H Kim T-S Park J-N Yoo and G-T Park ldquoSpeed-sensorless vector control of an induction motor using neuralnetwork speed estimationrdquo IEEE Transactions on IndustrialElectronics vol 48 no 3 pp 609ndash614 2001
[20] M Kuchar P Brandstetter and M Kaduch ldquoSensorless induc-tion motor drive with neural networkrdquo in Proceedings ofthe IEEE 35th Annual Power Electronics Specialists Conference(PESC rsquo04) pp 3301ndash3305 June 2004
[21] S Maiti V Verma C Chakraborty and Y Hori ldquoAn adaptivespeed sensorless induction motor drive with artificial neuralnetwork for stability enhancementrdquo IEEE Transactions onIndustrial Informatics vol 8 no 4 pp 757ndash766 2012
[22] P Girovsky and J Timko ldquoShaft sensor-less FOC control ofan induction motor using neural estimatorsrdquo Acta PolytechnicaHungarica vol 9 no 4 pp 31ndash45 2012
[23] D P Kothari and I J Nagrath Electric Machines chapter 9 TataMcGraw-Hill Education Private Limited 2004
[24] G Escobar A M Stankovic J M Carrasco E Galvan andR Ortega ldquoAnalysis and design of direct power control (DPC)for a three phase synchronous rectifier via output regulationsubspacesrdquo IEEE Transactions on Power Electronics vol 18 no3 pp 823ndash830 2003
[25] N R N Idris and A H M Yatim ldquoDirect torque controlof induction machines with constant switching frequency andreduced torque ripplerdquo IEEE Transactions on Industrial Elec-tronics vol 51 no 4 pp 758ndash767 2004
[26] S-H Huh K-B Lee D-W Kim I Choy and G-T ParkldquoSensorless speed control system using a neural networkrdquoInternational Journal of Control Automation and Systems vol3 no 4 pp 612ndash619 2005
Submit your manuscripts athttpwwwhindawicom
Computer Games Technology
International Journal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Distributed Sensor Networks
International Journal of
Advances in
FuzzySystems
Hindawi Publishing Corporationhttpwwwhindawicom
Volume 2014
International Journal of
ReconfigurableComputing
Hindawi Publishing Corporation httpwwwhindawicom Volume 2014
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Applied Computational Intelligence and Soft Computing
thinspAdvancesthinspinthinsp
Artificial Intelligence
HindawithinspPublishingthinspCorporationhttpwwwhindawicom Volumethinsp2014
Advances inSoftware EngineeringHindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Electrical and Computer Engineering
Journal of
Journal of
Computer Networks and Communications
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Hindawi Publishing Corporation
httpwwwhindawicom Volume 2014
Advances in
Multimedia
International Journal of
Biomedical Imaging
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
ArtificialNeural Systems
Advances in
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
RoboticsJournal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Computational Intelligence and Neuroscience
Industrial EngineeringJournal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Modelling amp Simulation in EngineeringHindawi Publishing Corporation httpwwwhindawicom Volume 2014
The Scientific World JournalHindawi Publishing Corporation httpwwwhindawicom Volume 2014
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Human-ComputerInteraction
Advances in
Computer EngineeringAdvances in
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Advances in Artificial Neural Systems 5
bp
0 ΔPout
+1
minus1
(a)
0
b120582
Δ120582s
1
(b)
Figure 4 Characteristics of the hysteresis controllers (a) power controller and (b) flux controller
b120582Δ120582s
ΔPout
Pout
minus
minus
+
+
120579s
a
bc
bp
120582s
Plowastout
120582lowasts
Figure 5 Block diagram of the inverter state selection
120582s120579s
rarr120582s
rarr s minus Rs
rarri s
s
rarr srarri s
Pout
120596m
Te 2
3p Im(rarri s
rarr120582lowast
s )
Figure 6 Estimation of actual output power and stator flux linkage
Reference model
Artificial neural network(ANN)
Back propagation weightadjustment
w2
sum
sum
120576d
120576q
1Z 1Z
1Z 1Z
iqs
ids
qs
ds
dr
qr
120595dr
120595qr
minus
Figure 7 MRAS-based rotor speed estimator using ANN
6 Advances in Artificial Neural Systems
w2
w1
w1
minusw2
w3
w3
ids(k minus 1)
iqs(k minus 1)
dr(k minus 1)
qr(k minus 1) dr(k)
qr(k)
Figure 8 ANN model for the estimation of rotor flux linkage
Power controller
Fluxcontroller
Stateselector
Voltage source inverter
IM
Load
Voltage sensor
Current sensor
Fluxestimator
Powerestimator
Speedcontroller
120596lowastm
Plowastout
bp
120582lowasts
120579s
120582s
Pout
rarr s
rarri s
120596m
Vdc
abcb120582
Figure 9 Block diagram of direct output power and flux control system for IM for closed-loop speed control
1400
1200
1000
800
600
400
200
0
minus200
Time (s)
Pow
er (W
)
0 05 1 15 2 25 3
Motor powerReference power
(a)
600
500
400
300
200
100
0
minus100
Spee
d (r
pm)
Time (s)0 05 1 15 2 25 3
Motor powerReference power
(b)
Figure 10 Simulation results DPC of induction motor (a) electromagnetic power and (b) rotor speed
Advances in Artificial Neural Systems 7
wr_est1
wr estimator
wr_est
Reference model
I_qs
I_ds
V_qs
V_ds
phi_dr
phi_qrLow pass filter
Zp
adjustment
ed
eqwr
Artificial neural network
I_qs
I_ds
wr
ANN-MRAS
I_qs
I_ds
wr
4
3
2
1
1z
1z+minus
+minus
iqs
ids
qs
ds
Lls
philowast_dr
philowast_qr
philowast_dr
philowast_qr
phi_dr_est
phi_qr_est
phi_dr
phi_qr
phi_dr_est
phi_qr_est
num(z)
den(z)
Back propagation
Figure 11 ANN based model of speed estimation
1500
1000
500
0
Spee
d (r
ads
)
MotorDifferentiated speedEstimation speed
Time (s)0 05 1 15
Figure 12 Response comparison of actual machine and ANN basedspeed estimator with error
constant The variable ANN weight1198822is proportional to the
rotor speed By using the backward difference method theequation of adaptive model is given below
119889119903
(119896) = 1198821119889119903
(119896 minus 1) minus 1198822119902119903
(119896 minus 1) + 1198823119894119889119904
(119896 minus 1)
119902119903
(119896) = 1198821119902119903
(119896 minus 1) minus 1198822119889119903
(119896 minus 1) + 1198823119894119902119904
(119896 minus 1)
(23)
which gives the value of rotor flux at the119870th sampling instantThese equations can be visualized by the very simple two-layer ANN shown in Figure 8
After taking learning factor 120578 andmomentum term 120572 intoaccount the estimated rotor speed is given below
119903(119896)
= 119903(119896 minus 1)
+
120578
119879
minus [120595119889119903
(119896) minus 119889119903
(119896)] 119902119903
(119896 minus 1)
+ [120595119902119903
(119896) minus 119902119903
(119896)] 119889119903
(119896 minus 1)
+
120572
119879
Δ1199082(119896 minus 1)
(24)
If the value of learning rate (120578) is chosen high it may lead tooscillations in the outputs of ANN and 120572 is chosen in rangebetween 01 and 08 The inclusion of momentum term intothe weight adjustment mechanism can significantly increasethe convergence which is extremely useful when the ANNshown in Figure 8 is used to estimate in real time [13]
8 Advances in Artificial Neural Systems
40
30
20
10
0
Pow
er (W
)
Reference powerElectromagnetic power
Time (s)0 05 1 15 2 25 3
minus10
minus20
minus30
(a)
2000
1800
1600
1400
1200
1000
800
600
400
200
0
Spee
d (r
pm)
Reference speedMotor speed
Time (s)0 05 1 15 2 25 3
(b)
Figure 13 Simulation results of DPC based on ANN of induction motor (a) electromagnetic power and (b) rotor speed
Table 3 Induction motor parameter
Parameter Value119875119899
2 hp120596rated 1500 [rpm]119881 300 [V]119868119899
21 [A]119869 0004 [kgsdotm2]119901 2119877119904
10 [Ω]119871119898
330 [mH]119871119897119904 119871119897119903
04 [mH]119861 0002 [Nsdotm(radsec)]
5 Simulation Results
51 Direct Power Control The simulation is carried out usingMATLAB The whole system setup is shown in Figure 9 Itcontains the power and flux controllers as inner loops andthe speed loop as outer loop
The sampling frequency is chosen as 50 kHz The hys-teresis tolerance for both power and flux controller is 1 ofrespective reference values For the PI speed controller (see(14)) the proportional gain is tuned to 119896
119901= 90 the overall
coefficient of the integral part is set to 01 which yieldedan integral gain 119896
119894= 5000 at the 50 kHz sampling rate
The reference speed profile is set in which the speed alwaysremains below the rated speed to avoid field weakening Theparameters of IM used in the simulations are listed in Table 3The simulation results are shown in Figure 10
52 Speed Estimator Based onANN Implementation ofANNbased speed estimator is carried out in MATLABSimulinkas shown in Figure 11
The response of ANN based speed estimator is comparedwith actual machine which is shown in Figure 12
53 Direct PowerControl withANN Theresult ofANNbasedspeed estimator with DPC controller is shown in Figure 13
6 Conclusion
This paper studies the possibility of direct power controlfor speed control of induction motor It has been shownthat DPC is basically derived from direct torque control(DTC) method and the simulation results show that thiscontrol scheme for IMsmotor has all the advantages of directtorque control method One of the major advantages of DPCmethod compared to DTC method is the easier calculationof actual power Moreover this paper proposes a new MRASspeed observer for DPC controlled induction motor drivesusing neural networks for speed estimation The simulationresults show that the proposed two-layers neural networkcan identify and track the motor speed accurately duringthe whole operating region Structure and algorithm are alsosimple Overall the dynamic response of this scheme of speedestimation shows a good performance
Conflict of Interests
The authors declare that there is no conflict of interestsregarding the publication of this paper
References
[1] P C Krause O Wasynozuk and S D Sudhoff Analysis ofElectric Machinery and Drive Systems IEEE Press 2nd edition2002
[2] I Boldea and S A Nasar Electric Drives Taylor amp Francis NewYork NY USA 2006
[3] P Vas The Control of AC Machines Oxford University PressNew York NY USA 1990
[4] A M Trzynadlowski Control ofInduction Motors Academicpress 2001
Advances in Artificial Neural Systems 9
[5] G S Buja and M P Kazmierkowski ldquoDirect torque control ofPWM inverter-fed ACmotorsmdasha surveyrdquo IEEE Transactions onIndustrial Electronics vol 51 no 4 pp 744ndash757 2004
[6] G Escobar A M Stankovic J M Carrasco E Galvan andR Ortega ldquoAnalysis and design of direct power control (DPC)for a three phase synchronous rectifier via output regulationsubspacesrdquo IEEE Transactions on Power Electronics vol 18 no3 pp 823ndash830 2003
[7] S Vazquez J A Sanchez J M Carrasco J I Leon and EGalvan ldquoA model-based direct power control for three-phasepower convertersrdquo IEEE Transactions on Industrial Electronicsvol 55 no 4 pp 1647ndash1657 2008
[8] Z Dawei X Lie and B W Williams ldquoImproved direct powercontrol of grid-connected DCAC convertersrdquo IEEE Transac-tions on Power Electronics vol 24 no 5 pp 1280ndash1292 2009
[9] L Xu and P Cartwright ldquoDirect active and reactive powercontrol of DFIG for wind energy generationrdquo IEEE Transactionson Energy Conversion vol 21 no 3 pp 750ndash758 2006
[10] D Zhi and L Xu ldquoDirect power control of DFIG with constantswitching frequency and improved transient performancerdquoIEEE Transactions on Energy Conversion vol 22 no 1 pp 110ndash118 2007
[11] Y Errami M Benchagra M Hilal M Maaroufi and M Ouas-said ldquoControl strategy for PMSG wind farm based on MPPTand direct power controlrdquo in Proceedings of the InternationalConference onMultimedia Computing and Systems (ICMCS rsquo12)pp 1125ndash1130 May 2012
[12] A Harrouz A Benatiallah and O Harrouz ldquoDirect powercontrol of a PMSG dedicated to standalone wind energysystemsrdquo in Proceedings of the 8th International Conference andExhibition on Ecological Vehicles and Renewable Energies (EVERrsquo13) pp 1ndash5 March 2013
[13] P Vas Sensorless Vector and Direct Torque Control OxfordUniversity Press 1998
[14] K Sedhuraman S Himavathi and A MuthuramalingamldquoPerformances comparison of neural architectures for on-linespeed estimation in sensorless IM drivesrdquo World Academy ofScience Engineering and Technology no 60 pp 1318ndash1325 2011
[15] S Meziane and R Benalla ldquoMRAS based speed control ofsensorlessinduction motor drivesrdquo ICGST-ACSE Journal vol 7no 1 pp 43ndash500 2007
[16] E Levi and M Wang ldquoImpact of parameter variations onspeed estimation in sensorless rotor flux oriented inductionmachinesrdquo in Proceedings of the 7th International Conferenceon Power Electronics and Variable Speed Drives pp 305ndash310London UK September 1998
[17] M E Elbulk L Tong and I Husain ldquoNeural-network-basedmodel reference adaptive systems for high-performance motordrives and motion controlsrdquo IEEE Transactions on IndustryApplications vol 38 no 3 pp 879ndash886 2002
[18] M T Hagan B Demuth andM BealeNeural Network DesignCengage Learning Pvt Ltd New Delhi India 2008
[19] S-H Kim T-S Park J-N Yoo and G-T Park ldquoSpeed-sensorless vector control of an induction motor using neuralnetwork speed estimationrdquo IEEE Transactions on IndustrialElectronics vol 48 no 3 pp 609ndash614 2001
[20] M Kuchar P Brandstetter and M Kaduch ldquoSensorless induc-tion motor drive with neural networkrdquo in Proceedings ofthe IEEE 35th Annual Power Electronics Specialists Conference(PESC rsquo04) pp 3301ndash3305 June 2004
[21] S Maiti V Verma C Chakraborty and Y Hori ldquoAn adaptivespeed sensorless induction motor drive with artificial neuralnetwork for stability enhancementrdquo IEEE Transactions onIndustrial Informatics vol 8 no 4 pp 757ndash766 2012
[22] P Girovsky and J Timko ldquoShaft sensor-less FOC control ofan induction motor using neural estimatorsrdquo Acta PolytechnicaHungarica vol 9 no 4 pp 31ndash45 2012
[23] D P Kothari and I J Nagrath Electric Machines chapter 9 TataMcGraw-Hill Education Private Limited 2004
[24] G Escobar A M Stankovic J M Carrasco E Galvan andR Ortega ldquoAnalysis and design of direct power control (DPC)for a three phase synchronous rectifier via output regulationsubspacesrdquo IEEE Transactions on Power Electronics vol 18 no3 pp 823ndash830 2003
[25] N R N Idris and A H M Yatim ldquoDirect torque controlof induction machines with constant switching frequency andreduced torque ripplerdquo IEEE Transactions on Industrial Elec-tronics vol 51 no 4 pp 758ndash767 2004
[26] S-H Huh K-B Lee D-W Kim I Choy and G-T ParkldquoSensorless speed control system using a neural networkrdquoInternational Journal of Control Automation and Systems vol3 no 4 pp 612ndash619 2005
Submit your manuscripts athttpwwwhindawicom
Computer Games Technology
International Journal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Distributed Sensor Networks
International Journal of
Advances in
FuzzySystems
Hindawi Publishing Corporationhttpwwwhindawicom
Volume 2014
International Journal of
ReconfigurableComputing
Hindawi Publishing Corporation httpwwwhindawicom Volume 2014
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Applied Computational Intelligence and Soft Computing
thinspAdvancesthinspinthinsp
Artificial Intelligence
HindawithinspPublishingthinspCorporationhttpwwwhindawicom Volumethinsp2014
Advances inSoftware EngineeringHindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Electrical and Computer Engineering
Journal of
Journal of
Computer Networks and Communications
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Hindawi Publishing Corporation
httpwwwhindawicom Volume 2014
Advances in
Multimedia
International Journal of
Biomedical Imaging
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
ArtificialNeural Systems
Advances in
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
RoboticsJournal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Computational Intelligence and Neuroscience
Industrial EngineeringJournal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Modelling amp Simulation in EngineeringHindawi Publishing Corporation httpwwwhindawicom Volume 2014
The Scientific World JournalHindawi Publishing Corporation httpwwwhindawicom Volume 2014
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Human-ComputerInteraction
Advances in
Computer EngineeringAdvances in
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
6 Advances in Artificial Neural Systems
w2
w1
w1
minusw2
w3
w3
ids(k minus 1)
iqs(k minus 1)
dr(k minus 1)
qr(k minus 1) dr(k)
qr(k)
Figure 8 ANN model for the estimation of rotor flux linkage
Power controller
Fluxcontroller
Stateselector
Voltage source inverter
IM
Load
Voltage sensor
Current sensor
Fluxestimator
Powerestimator
Speedcontroller
120596lowastm
Plowastout
bp
120582lowasts
120579s
120582s
Pout
rarr s
rarri s
120596m
Vdc
abcb120582
Figure 9 Block diagram of direct output power and flux control system for IM for closed-loop speed control
1400
1200
1000
800
600
400
200
0
minus200
Time (s)
Pow
er (W
)
0 05 1 15 2 25 3
Motor powerReference power
(a)
600
500
400
300
200
100
0
minus100
Spee
d (r
pm)
Time (s)0 05 1 15 2 25 3
Motor powerReference power
(b)
Figure 10 Simulation results DPC of induction motor (a) electromagnetic power and (b) rotor speed
Advances in Artificial Neural Systems 7
wr_est1
wr estimator
wr_est
Reference model
I_qs
I_ds
V_qs
V_ds
phi_dr
phi_qrLow pass filter
Zp
adjustment
ed
eqwr
Artificial neural network
I_qs
I_ds
wr
ANN-MRAS
I_qs
I_ds
wr
4
3
2
1
1z
1z+minus
+minus
iqs
ids
qs
ds
Lls
philowast_dr
philowast_qr
philowast_dr
philowast_qr
phi_dr_est
phi_qr_est
phi_dr
phi_qr
phi_dr_est
phi_qr_est
num(z)
den(z)
Back propagation
Figure 11 ANN based model of speed estimation
1500
1000
500
0
Spee
d (r
ads
)
MotorDifferentiated speedEstimation speed
Time (s)0 05 1 15
Figure 12 Response comparison of actual machine and ANN basedspeed estimator with error
constant The variable ANN weight1198822is proportional to the
rotor speed By using the backward difference method theequation of adaptive model is given below
119889119903
(119896) = 1198821119889119903
(119896 minus 1) minus 1198822119902119903
(119896 minus 1) + 1198823119894119889119904
(119896 minus 1)
119902119903
(119896) = 1198821119902119903
(119896 minus 1) minus 1198822119889119903
(119896 minus 1) + 1198823119894119902119904
(119896 minus 1)
(23)
which gives the value of rotor flux at the119870th sampling instantThese equations can be visualized by the very simple two-layer ANN shown in Figure 8
After taking learning factor 120578 andmomentum term 120572 intoaccount the estimated rotor speed is given below
119903(119896)
= 119903(119896 minus 1)
+
120578
119879
minus [120595119889119903
(119896) minus 119889119903
(119896)] 119902119903
(119896 minus 1)
+ [120595119902119903
(119896) minus 119902119903
(119896)] 119889119903
(119896 minus 1)
+
120572
119879
Δ1199082(119896 minus 1)
(24)
If the value of learning rate (120578) is chosen high it may lead tooscillations in the outputs of ANN and 120572 is chosen in rangebetween 01 and 08 The inclusion of momentum term intothe weight adjustment mechanism can significantly increasethe convergence which is extremely useful when the ANNshown in Figure 8 is used to estimate in real time [13]
8 Advances in Artificial Neural Systems
40
30
20
10
0
Pow
er (W
)
Reference powerElectromagnetic power
Time (s)0 05 1 15 2 25 3
minus10
minus20
minus30
(a)
2000
1800
1600
1400
1200
1000
800
600
400
200
0
Spee
d (r
pm)
Reference speedMotor speed
Time (s)0 05 1 15 2 25 3
(b)
Figure 13 Simulation results of DPC based on ANN of induction motor (a) electromagnetic power and (b) rotor speed
Table 3 Induction motor parameter
Parameter Value119875119899
2 hp120596rated 1500 [rpm]119881 300 [V]119868119899
21 [A]119869 0004 [kgsdotm2]119901 2119877119904
10 [Ω]119871119898
330 [mH]119871119897119904 119871119897119903
04 [mH]119861 0002 [Nsdotm(radsec)]
5 Simulation Results
51 Direct Power Control The simulation is carried out usingMATLAB The whole system setup is shown in Figure 9 Itcontains the power and flux controllers as inner loops andthe speed loop as outer loop
The sampling frequency is chosen as 50 kHz The hys-teresis tolerance for both power and flux controller is 1 ofrespective reference values For the PI speed controller (see(14)) the proportional gain is tuned to 119896
119901= 90 the overall
coefficient of the integral part is set to 01 which yieldedan integral gain 119896
119894= 5000 at the 50 kHz sampling rate
The reference speed profile is set in which the speed alwaysremains below the rated speed to avoid field weakening Theparameters of IM used in the simulations are listed in Table 3The simulation results are shown in Figure 10
52 Speed Estimator Based onANN Implementation ofANNbased speed estimator is carried out in MATLABSimulinkas shown in Figure 11
The response of ANN based speed estimator is comparedwith actual machine which is shown in Figure 12
53 Direct PowerControl withANN Theresult ofANNbasedspeed estimator with DPC controller is shown in Figure 13
6 Conclusion
This paper studies the possibility of direct power controlfor speed control of induction motor It has been shownthat DPC is basically derived from direct torque control(DTC) method and the simulation results show that thiscontrol scheme for IMsmotor has all the advantages of directtorque control method One of the major advantages of DPCmethod compared to DTC method is the easier calculationof actual power Moreover this paper proposes a new MRASspeed observer for DPC controlled induction motor drivesusing neural networks for speed estimation The simulationresults show that the proposed two-layers neural networkcan identify and track the motor speed accurately duringthe whole operating region Structure and algorithm are alsosimple Overall the dynamic response of this scheme of speedestimation shows a good performance
Conflict of Interests
The authors declare that there is no conflict of interestsregarding the publication of this paper
References
[1] P C Krause O Wasynozuk and S D Sudhoff Analysis ofElectric Machinery and Drive Systems IEEE Press 2nd edition2002
[2] I Boldea and S A Nasar Electric Drives Taylor amp Francis NewYork NY USA 2006
[3] P Vas The Control of AC Machines Oxford University PressNew York NY USA 1990
[4] A M Trzynadlowski Control ofInduction Motors Academicpress 2001
Advances in Artificial Neural Systems 9
[5] G S Buja and M P Kazmierkowski ldquoDirect torque control ofPWM inverter-fed ACmotorsmdasha surveyrdquo IEEE Transactions onIndustrial Electronics vol 51 no 4 pp 744ndash757 2004
[6] G Escobar A M Stankovic J M Carrasco E Galvan andR Ortega ldquoAnalysis and design of direct power control (DPC)for a three phase synchronous rectifier via output regulationsubspacesrdquo IEEE Transactions on Power Electronics vol 18 no3 pp 823ndash830 2003
[7] S Vazquez J A Sanchez J M Carrasco J I Leon and EGalvan ldquoA model-based direct power control for three-phasepower convertersrdquo IEEE Transactions on Industrial Electronicsvol 55 no 4 pp 1647ndash1657 2008
[8] Z Dawei X Lie and B W Williams ldquoImproved direct powercontrol of grid-connected DCAC convertersrdquo IEEE Transac-tions on Power Electronics vol 24 no 5 pp 1280ndash1292 2009
[9] L Xu and P Cartwright ldquoDirect active and reactive powercontrol of DFIG for wind energy generationrdquo IEEE Transactionson Energy Conversion vol 21 no 3 pp 750ndash758 2006
[10] D Zhi and L Xu ldquoDirect power control of DFIG with constantswitching frequency and improved transient performancerdquoIEEE Transactions on Energy Conversion vol 22 no 1 pp 110ndash118 2007
[11] Y Errami M Benchagra M Hilal M Maaroufi and M Ouas-said ldquoControl strategy for PMSG wind farm based on MPPTand direct power controlrdquo in Proceedings of the InternationalConference onMultimedia Computing and Systems (ICMCS rsquo12)pp 1125ndash1130 May 2012
[12] A Harrouz A Benatiallah and O Harrouz ldquoDirect powercontrol of a PMSG dedicated to standalone wind energysystemsrdquo in Proceedings of the 8th International Conference andExhibition on Ecological Vehicles and Renewable Energies (EVERrsquo13) pp 1ndash5 March 2013
[13] P Vas Sensorless Vector and Direct Torque Control OxfordUniversity Press 1998
[14] K Sedhuraman S Himavathi and A MuthuramalingamldquoPerformances comparison of neural architectures for on-linespeed estimation in sensorless IM drivesrdquo World Academy ofScience Engineering and Technology no 60 pp 1318ndash1325 2011
[15] S Meziane and R Benalla ldquoMRAS based speed control ofsensorlessinduction motor drivesrdquo ICGST-ACSE Journal vol 7no 1 pp 43ndash500 2007
[16] E Levi and M Wang ldquoImpact of parameter variations onspeed estimation in sensorless rotor flux oriented inductionmachinesrdquo in Proceedings of the 7th International Conferenceon Power Electronics and Variable Speed Drives pp 305ndash310London UK September 1998
[17] M E Elbulk L Tong and I Husain ldquoNeural-network-basedmodel reference adaptive systems for high-performance motordrives and motion controlsrdquo IEEE Transactions on IndustryApplications vol 38 no 3 pp 879ndash886 2002
[18] M T Hagan B Demuth andM BealeNeural Network DesignCengage Learning Pvt Ltd New Delhi India 2008
[19] S-H Kim T-S Park J-N Yoo and G-T Park ldquoSpeed-sensorless vector control of an induction motor using neuralnetwork speed estimationrdquo IEEE Transactions on IndustrialElectronics vol 48 no 3 pp 609ndash614 2001
[20] M Kuchar P Brandstetter and M Kaduch ldquoSensorless induc-tion motor drive with neural networkrdquo in Proceedings ofthe IEEE 35th Annual Power Electronics Specialists Conference(PESC rsquo04) pp 3301ndash3305 June 2004
[21] S Maiti V Verma C Chakraborty and Y Hori ldquoAn adaptivespeed sensorless induction motor drive with artificial neuralnetwork for stability enhancementrdquo IEEE Transactions onIndustrial Informatics vol 8 no 4 pp 757ndash766 2012
[22] P Girovsky and J Timko ldquoShaft sensor-less FOC control ofan induction motor using neural estimatorsrdquo Acta PolytechnicaHungarica vol 9 no 4 pp 31ndash45 2012
[23] D P Kothari and I J Nagrath Electric Machines chapter 9 TataMcGraw-Hill Education Private Limited 2004
[24] G Escobar A M Stankovic J M Carrasco E Galvan andR Ortega ldquoAnalysis and design of direct power control (DPC)for a three phase synchronous rectifier via output regulationsubspacesrdquo IEEE Transactions on Power Electronics vol 18 no3 pp 823ndash830 2003
[25] N R N Idris and A H M Yatim ldquoDirect torque controlof induction machines with constant switching frequency andreduced torque ripplerdquo IEEE Transactions on Industrial Elec-tronics vol 51 no 4 pp 758ndash767 2004
[26] S-H Huh K-B Lee D-W Kim I Choy and G-T ParkldquoSensorless speed control system using a neural networkrdquoInternational Journal of Control Automation and Systems vol3 no 4 pp 612ndash619 2005
Submit your manuscripts athttpwwwhindawicom
Computer Games Technology
International Journal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Distributed Sensor Networks
International Journal of
Advances in
FuzzySystems
Hindawi Publishing Corporationhttpwwwhindawicom
Volume 2014
International Journal of
ReconfigurableComputing
Hindawi Publishing Corporation httpwwwhindawicom Volume 2014
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Applied Computational Intelligence and Soft Computing
thinspAdvancesthinspinthinsp
Artificial Intelligence
HindawithinspPublishingthinspCorporationhttpwwwhindawicom Volumethinsp2014
Advances inSoftware EngineeringHindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Electrical and Computer Engineering
Journal of
Journal of
Computer Networks and Communications
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Hindawi Publishing Corporation
httpwwwhindawicom Volume 2014
Advances in
Multimedia
International Journal of
Biomedical Imaging
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
ArtificialNeural Systems
Advances in
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
RoboticsJournal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Computational Intelligence and Neuroscience
Industrial EngineeringJournal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Modelling amp Simulation in EngineeringHindawi Publishing Corporation httpwwwhindawicom Volume 2014
The Scientific World JournalHindawi Publishing Corporation httpwwwhindawicom Volume 2014
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Human-ComputerInteraction
Advances in
Computer EngineeringAdvances in
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Advances in Artificial Neural Systems 7
wr_est1
wr estimator
wr_est
Reference model
I_qs
I_ds
V_qs
V_ds
phi_dr
phi_qrLow pass filter
Zp
adjustment
ed
eqwr
Artificial neural network
I_qs
I_ds
wr
ANN-MRAS
I_qs
I_ds
wr
4
3
2
1
1z
1z+minus
+minus
iqs
ids
qs
ds
Lls
philowast_dr
philowast_qr
philowast_dr
philowast_qr
phi_dr_est
phi_qr_est
phi_dr
phi_qr
phi_dr_est
phi_qr_est
num(z)
den(z)
Back propagation
Figure 11 ANN based model of speed estimation
1500
1000
500
0
Spee
d (r
ads
)
MotorDifferentiated speedEstimation speed
Time (s)0 05 1 15
Figure 12 Response comparison of actual machine and ANN basedspeed estimator with error
constant The variable ANN weight1198822is proportional to the
rotor speed By using the backward difference method theequation of adaptive model is given below
119889119903
(119896) = 1198821119889119903
(119896 minus 1) minus 1198822119902119903
(119896 minus 1) + 1198823119894119889119904
(119896 minus 1)
119902119903
(119896) = 1198821119902119903
(119896 minus 1) minus 1198822119889119903
(119896 minus 1) + 1198823119894119902119904
(119896 minus 1)
(23)
which gives the value of rotor flux at the119870th sampling instantThese equations can be visualized by the very simple two-layer ANN shown in Figure 8
After taking learning factor 120578 andmomentum term 120572 intoaccount the estimated rotor speed is given below
119903(119896)
= 119903(119896 minus 1)
+
120578
119879
minus [120595119889119903
(119896) minus 119889119903
(119896)] 119902119903
(119896 minus 1)
+ [120595119902119903
(119896) minus 119902119903
(119896)] 119889119903
(119896 minus 1)
+
120572
119879
Δ1199082(119896 minus 1)
(24)
If the value of learning rate (120578) is chosen high it may lead tooscillations in the outputs of ANN and 120572 is chosen in rangebetween 01 and 08 The inclusion of momentum term intothe weight adjustment mechanism can significantly increasethe convergence which is extremely useful when the ANNshown in Figure 8 is used to estimate in real time [13]
8 Advances in Artificial Neural Systems
40
30
20
10
0
Pow
er (W
)
Reference powerElectromagnetic power
Time (s)0 05 1 15 2 25 3
minus10
minus20
minus30
(a)
2000
1800
1600
1400
1200
1000
800
600
400
200
0
Spee
d (r
pm)
Reference speedMotor speed
Time (s)0 05 1 15 2 25 3
(b)
Figure 13 Simulation results of DPC based on ANN of induction motor (a) electromagnetic power and (b) rotor speed
Table 3 Induction motor parameter
Parameter Value119875119899
2 hp120596rated 1500 [rpm]119881 300 [V]119868119899
21 [A]119869 0004 [kgsdotm2]119901 2119877119904
10 [Ω]119871119898
330 [mH]119871119897119904 119871119897119903
04 [mH]119861 0002 [Nsdotm(radsec)]
5 Simulation Results
51 Direct Power Control The simulation is carried out usingMATLAB The whole system setup is shown in Figure 9 Itcontains the power and flux controllers as inner loops andthe speed loop as outer loop
The sampling frequency is chosen as 50 kHz The hys-teresis tolerance for both power and flux controller is 1 ofrespective reference values For the PI speed controller (see(14)) the proportional gain is tuned to 119896
119901= 90 the overall
coefficient of the integral part is set to 01 which yieldedan integral gain 119896
119894= 5000 at the 50 kHz sampling rate
The reference speed profile is set in which the speed alwaysremains below the rated speed to avoid field weakening Theparameters of IM used in the simulations are listed in Table 3The simulation results are shown in Figure 10
52 Speed Estimator Based onANN Implementation ofANNbased speed estimator is carried out in MATLABSimulinkas shown in Figure 11
The response of ANN based speed estimator is comparedwith actual machine which is shown in Figure 12
53 Direct PowerControl withANN Theresult ofANNbasedspeed estimator with DPC controller is shown in Figure 13
6 Conclusion
This paper studies the possibility of direct power controlfor speed control of induction motor It has been shownthat DPC is basically derived from direct torque control(DTC) method and the simulation results show that thiscontrol scheme for IMsmotor has all the advantages of directtorque control method One of the major advantages of DPCmethod compared to DTC method is the easier calculationof actual power Moreover this paper proposes a new MRASspeed observer for DPC controlled induction motor drivesusing neural networks for speed estimation The simulationresults show that the proposed two-layers neural networkcan identify and track the motor speed accurately duringthe whole operating region Structure and algorithm are alsosimple Overall the dynamic response of this scheme of speedestimation shows a good performance
Conflict of Interests
The authors declare that there is no conflict of interestsregarding the publication of this paper
References
[1] P C Krause O Wasynozuk and S D Sudhoff Analysis ofElectric Machinery and Drive Systems IEEE Press 2nd edition2002
[2] I Boldea and S A Nasar Electric Drives Taylor amp Francis NewYork NY USA 2006
[3] P Vas The Control of AC Machines Oxford University PressNew York NY USA 1990
[4] A M Trzynadlowski Control ofInduction Motors Academicpress 2001
Advances in Artificial Neural Systems 9
[5] G S Buja and M P Kazmierkowski ldquoDirect torque control ofPWM inverter-fed ACmotorsmdasha surveyrdquo IEEE Transactions onIndustrial Electronics vol 51 no 4 pp 744ndash757 2004
[6] G Escobar A M Stankovic J M Carrasco E Galvan andR Ortega ldquoAnalysis and design of direct power control (DPC)for a three phase synchronous rectifier via output regulationsubspacesrdquo IEEE Transactions on Power Electronics vol 18 no3 pp 823ndash830 2003
[7] S Vazquez J A Sanchez J M Carrasco J I Leon and EGalvan ldquoA model-based direct power control for three-phasepower convertersrdquo IEEE Transactions on Industrial Electronicsvol 55 no 4 pp 1647ndash1657 2008
[8] Z Dawei X Lie and B W Williams ldquoImproved direct powercontrol of grid-connected DCAC convertersrdquo IEEE Transac-tions on Power Electronics vol 24 no 5 pp 1280ndash1292 2009
[9] L Xu and P Cartwright ldquoDirect active and reactive powercontrol of DFIG for wind energy generationrdquo IEEE Transactionson Energy Conversion vol 21 no 3 pp 750ndash758 2006
[10] D Zhi and L Xu ldquoDirect power control of DFIG with constantswitching frequency and improved transient performancerdquoIEEE Transactions on Energy Conversion vol 22 no 1 pp 110ndash118 2007
[11] Y Errami M Benchagra M Hilal M Maaroufi and M Ouas-said ldquoControl strategy for PMSG wind farm based on MPPTand direct power controlrdquo in Proceedings of the InternationalConference onMultimedia Computing and Systems (ICMCS rsquo12)pp 1125ndash1130 May 2012
[12] A Harrouz A Benatiallah and O Harrouz ldquoDirect powercontrol of a PMSG dedicated to standalone wind energysystemsrdquo in Proceedings of the 8th International Conference andExhibition on Ecological Vehicles and Renewable Energies (EVERrsquo13) pp 1ndash5 March 2013
[13] P Vas Sensorless Vector and Direct Torque Control OxfordUniversity Press 1998
[14] K Sedhuraman S Himavathi and A MuthuramalingamldquoPerformances comparison of neural architectures for on-linespeed estimation in sensorless IM drivesrdquo World Academy ofScience Engineering and Technology no 60 pp 1318ndash1325 2011
[15] S Meziane and R Benalla ldquoMRAS based speed control ofsensorlessinduction motor drivesrdquo ICGST-ACSE Journal vol 7no 1 pp 43ndash500 2007
[16] E Levi and M Wang ldquoImpact of parameter variations onspeed estimation in sensorless rotor flux oriented inductionmachinesrdquo in Proceedings of the 7th International Conferenceon Power Electronics and Variable Speed Drives pp 305ndash310London UK September 1998
[17] M E Elbulk L Tong and I Husain ldquoNeural-network-basedmodel reference adaptive systems for high-performance motordrives and motion controlsrdquo IEEE Transactions on IndustryApplications vol 38 no 3 pp 879ndash886 2002
[18] M T Hagan B Demuth andM BealeNeural Network DesignCengage Learning Pvt Ltd New Delhi India 2008
[19] S-H Kim T-S Park J-N Yoo and G-T Park ldquoSpeed-sensorless vector control of an induction motor using neuralnetwork speed estimationrdquo IEEE Transactions on IndustrialElectronics vol 48 no 3 pp 609ndash614 2001
[20] M Kuchar P Brandstetter and M Kaduch ldquoSensorless induc-tion motor drive with neural networkrdquo in Proceedings ofthe IEEE 35th Annual Power Electronics Specialists Conference(PESC rsquo04) pp 3301ndash3305 June 2004
[21] S Maiti V Verma C Chakraborty and Y Hori ldquoAn adaptivespeed sensorless induction motor drive with artificial neuralnetwork for stability enhancementrdquo IEEE Transactions onIndustrial Informatics vol 8 no 4 pp 757ndash766 2012
[22] P Girovsky and J Timko ldquoShaft sensor-less FOC control ofan induction motor using neural estimatorsrdquo Acta PolytechnicaHungarica vol 9 no 4 pp 31ndash45 2012
[23] D P Kothari and I J Nagrath Electric Machines chapter 9 TataMcGraw-Hill Education Private Limited 2004
[24] G Escobar A M Stankovic J M Carrasco E Galvan andR Ortega ldquoAnalysis and design of direct power control (DPC)for a three phase synchronous rectifier via output regulationsubspacesrdquo IEEE Transactions on Power Electronics vol 18 no3 pp 823ndash830 2003
[25] N R N Idris and A H M Yatim ldquoDirect torque controlof induction machines with constant switching frequency andreduced torque ripplerdquo IEEE Transactions on Industrial Elec-tronics vol 51 no 4 pp 758ndash767 2004
[26] S-H Huh K-B Lee D-W Kim I Choy and G-T ParkldquoSensorless speed control system using a neural networkrdquoInternational Journal of Control Automation and Systems vol3 no 4 pp 612ndash619 2005
Submit your manuscripts athttpwwwhindawicom
Computer Games Technology
International Journal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Distributed Sensor Networks
International Journal of
Advances in
FuzzySystems
Hindawi Publishing Corporationhttpwwwhindawicom
Volume 2014
International Journal of
ReconfigurableComputing
Hindawi Publishing Corporation httpwwwhindawicom Volume 2014
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Applied Computational Intelligence and Soft Computing
thinspAdvancesthinspinthinsp
Artificial Intelligence
HindawithinspPublishingthinspCorporationhttpwwwhindawicom Volumethinsp2014
Advances inSoftware EngineeringHindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Electrical and Computer Engineering
Journal of
Journal of
Computer Networks and Communications
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Hindawi Publishing Corporation
httpwwwhindawicom Volume 2014
Advances in
Multimedia
International Journal of
Biomedical Imaging
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
ArtificialNeural Systems
Advances in
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
RoboticsJournal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Computational Intelligence and Neuroscience
Industrial EngineeringJournal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Modelling amp Simulation in EngineeringHindawi Publishing Corporation httpwwwhindawicom Volume 2014
The Scientific World JournalHindawi Publishing Corporation httpwwwhindawicom Volume 2014
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Human-ComputerInteraction
Advances in
Computer EngineeringAdvances in
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
8 Advances in Artificial Neural Systems
40
30
20
10
0
Pow
er (W
)
Reference powerElectromagnetic power
Time (s)0 05 1 15 2 25 3
minus10
minus20
minus30
(a)
2000
1800
1600
1400
1200
1000
800
600
400
200
0
Spee
d (r
pm)
Reference speedMotor speed
Time (s)0 05 1 15 2 25 3
(b)
Figure 13 Simulation results of DPC based on ANN of induction motor (a) electromagnetic power and (b) rotor speed
Table 3 Induction motor parameter
Parameter Value119875119899
2 hp120596rated 1500 [rpm]119881 300 [V]119868119899
21 [A]119869 0004 [kgsdotm2]119901 2119877119904
10 [Ω]119871119898
330 [mH]119871119897119904 119871119897119903
04 [mH]119861 0002 [Nsdotm(radsec)]
5 Simulation Results
51 Direct Power Control The simulation is carried out usingMATLAB The whole system setup is shown in Figure 9 Itcontains the power and flux controllers as inner loops andthe speed loop as outer loop
The sampling frequency is chosen as 50 kHz The hys-teresis tolerance for both power and flux controller is 1 ofrespective reference values For the PI speed controller (see(14)) the proportional gain is tuned to 119896
119901= 90 the overall
coefficient of the integral part is set to 01 which yieldedan integral gain 119896
119894= 5000 at the 50 kHz sampling rate
The reference speed profile is set in which the speed alwaysremains below the rated speed to avoid field weakening Theparameters of IM used in the simulations are listed in Table 3The simulation results are shown in Figure 10
52 Speed Estimator Based onANN Implementation ofANNbased speed estimator is carried out in MATLABSimulinkas shown in Figure 11
The response of ANN based speed estimator is comparedwith actual machine which is shown in Figure 12
53 Direct PowerControl withANN Theresult ofANNbasedspeed estimator with DPC controller is shown in Figure 13
6 Conclusion
This paper studies the possibility of direct power controlfor speed control of induction motor It has been shownthat DPC is basically derived from direct torque control(DTC) method and the simulation results show that thiscontrol scheme for IMsmotor has all the advantages of directtorque control method One of the major advantages of DPCmethod compared to DTC method is the easier calculationof actual power Moreover this paper proposes a new MRASspeed observer for DPC controlled induction motor drivesusing neural networks for speed estimation The simulationresults show that the proposed two-layers neural networkcan identify and track the motor speed accurately duringthe whole operating region Structure and algorithm are alsosimple Overall the dynamic response of this scheme of speedestimation shows a good performance
Conflict of Interests
The authors declare that there is no conflict of interestsregarding the publication of this paper
References
[1] P C Krause O Wasynozuk and S D Sudhoff Analysis ofElectric Machinery and Drive Systems IEEE Press 2nd edition2002
[2] I Boldea and S A Nasar Electric Drives Taylor amp Francis NewYork NY USA 2006
[3] P Vas The Control of AC Machines Oxford University PressNew York NY USA 1990
[4] A M Trzynadlowski Control ofInduction Motors Academicpress 2001
Advances in Artificial Neural Systems 9
[5] G S Buja and M P Kazmierkowski ldquoDirect torque control ofPWM inverter-fed ACmotorsmdasha surveyrdquo IEEE Transactions onIndustrial Electronics vol 51 no 4 pp 744ndash757 2004
[6] G Escobar A M Stankovic J M Carrasco E Galvan andR Ortega ldquoAnalysis and design of direct power control (DPC)for a three phase synchronous rectifier via output regulationsubspacesrdquo IEEE Transactions on Power Electronics vol 18 no3 pp 823ndash830 2003
[7] S Vazquez J A Sanchez J M Carrasco J I Leon and EGalvan ldquoA model-based direct power control for three-phasepower convertersrdquo IEEE Transactions on Industrial Electronicsvol 55 no 4 pp 1647ndash1657 2008
[8] Z Dawei X Lie and B W Williams ldquoImproved direct powercontrol of grid-connected DCAC convertersrdquo IEEE Transac-tions on Power Electronics vol 24 no 5 pp 1280ndash1292 2009
[9] L Xu and P Cartwright ldquoDirect active and reactive powercontrol of DFIG for wind energy generationrdquo IEEE Transactionson Energy Conversion vol 21 no 3 pp 750ndash758 2006
[10] D Zhi and L Xu ldquoDirect power control of DFIG with constantswitching frequency and improved transient performancerdquoIEEE Transactions on Energy Conversion vol 22 no 1 pp 110ndash118 2007
[11] Y Errami M Benchagra M Hilal M Maaroufi and M Ouas-said ldquoControl strategy for PMSG wind farm based on MPPTand direct power controlrdquo in Proceedings of the InternationalConference onMultimedia Computing and Systems (ICMCS rsquo12)pp 1125ndash1130 May 2012
[12] A Harrouz A Benatiallah and O Harrouz ldquoDirect powercontrol of a PMSG dedicated to standalone wind energysystemsrdquo in Proceedings of the 8th International Conference andExhibition on Ecological Vehicles and Renewable Energies (EVERrsquo13) pp 1ndash5 March 2013
[13] P Vas Sensorless Vector and Direct Torque Control OxfordUniversity Press 1998
[14] K Sedhuraman S Himavathi and A MuthuramalingamldquoPerformances comparison of neural architectures for on-linespeed estimation in sensorless IM drivesrdquo World Academy ofScience Engineering and Technology no 60 pp 1318ndash1325 2011
[15] S Meziane and R Benalla ldquoMRAS based speed control ofsensorlessinduction motor drivesrdquo ICGST-ACSE Journal vol 7no 1 pp 43ndash500 2007
[16] E Levi and M Wang ldquoImpact of parameter variations onspeed estimation in sensorless rotor flux oriented inductionmachinesrdquo in Proceedings of the 7th International Conferenceon Power Electronics and Variable Speed Drives pp 305ndash310London UK September 1998
[17] M E Elbulk L Tong and I Husain ldquoNeural-network-basedmodel reference adaptive systems for high-performance motordrives and motion controlsrdquo IEEE Transactions on IndustryApplications vol 38 no 3 pp 879ndash886 2002
[18] M T Hagan B Demuth andM BealeNeural Network DesignCengage Learning Pvt Ltd New Delhi India 2008
[19] S-H Kim T-S Park J-N Yoo and G-T Park ldquoSpeed-sensorless vector control of an induction motor using neuralnetwork speed estimationrdquo IEEE Transactions on IndustrialElectronics vol 48 no 3 pp 609ndash614 2001
[20] M Kuchar P Brandstetter and M Kaduch ldquoSensorless induc-tion motor drive with neural networkrdquo in Proceedings ofthe IEEE 35th Annual Power Electronics Specialists Conference(PESC rsquo04) pp 3301ndash3305 June 2004
[21] S Maiti V Verma C Chakraborty and Y Hori ldquoAn adaptivespeed sensorless induction motor drive with artificial neuralnetwork for stability enhancementrdquo IEEE Transactions onIndustrial Informatics vol 8 no 4 pp 757ndash766 2012
[22] P Girovsky and J Timko ldquoShaft sensor-less FOC control ofan induction motor using neural estimatorsrdquo Acta PolytechnicaHungarica vol 9 no 4 pp 31ndash45 2012
[23] D P Kothari and I J Nagrath Electric Machines chapter 9 TataMcGraw-Hill Education Private Limited 2004
[24] G Escobar A M Stankovic J M Carrasco E Galvan andR Ortega ldquoAnalysis and design of direct power control (DPC)for a three phase synchronous rectifier via output regulationsubspacesrdquo IEEE Transactions on Power Electronics vol 18 no3 pp 823ndash830 2003
[25] N R N Idris and A H M Yatim ldquoDirect torque controlof induction machines with constant switching frequency andreduced torque ripplerdquo IEEE Transactions on Industrial Elec-tronics vol 51 no 4 pp 758ndash767 2004
[26] S-H Huh K-B Lee D-W Kim I Choy and G-T ParkldquoSensorless speed control system using a neural networkrdquoInternational Journal of Control Automation and Systems vol3 no 4 pp 612ndash619 2005
Submit your manuscripts athttpwwwhindawicom
Computer Games Technology
International Journal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Distributed Sensor Networks
International Journal of
Advances in
FuzzySystems
Hindawi Publishing Corporationhttpwwwhindawicom
Volume 2014
International Journal of
ReconfigurableComputing
Hindawi Publishing Corporation httpwwwhindawicom Volume 2014
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Applied Computational Intelligence and Soft Computing
thinspAdvancesthinspinthinsp
Artificial Intelligence
HindawithinspPublishingthinspCorporationhttpwwwhindawicom Volumethinsp2014
Advances inSoftware EngineeringHindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Electrical and Computer Engineering
Journal of
Journal of
Computer Networks and Communications
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Hindawi Publishing Corporation
httpwwwhindawicom Volume 2014
Advances in
Multimedia
International Journal of
Biomedical Imaging
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
ArtificialNeural Systems
Advances in
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
RoboticsJournal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Computational Intelligence and Neuroscience
Industrial EngineeringJournal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Modelling amp Simulation in EngineeringHindawi Publishing Corporation httpwwwhindawicom Volume 2014
The Scientific World JournalHindawi Publishing Corporation httpwwwhindawicom Volume 2014
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Human-ComputerInteraction
Advances in
Computer EngineeringAdvances in
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Advances in Artificial Neural Systems 9
[5] G S Buja and M P Kazmierkowski ldquoDirect torque control ofPWM inverter-fed ACmotorsmdasha surveyrdquo IEEE Transactions onIndustrial Electronics vol 51 no 4 pp 744ndash757 2004
[6] G Escobar A M Stankovic J M Carrasco E Galvan andR Ortega ldquoAnalysis and design of direct power control (DPC)for a three phase synchronous rectifier via output regulationsubspacesrdquo IEEE Transactions on Power Electronics vol 18 no3 pp 823ndash830 2003
[7] S Vazquez J A Sanchez J M Carrasco J I Leon and EGalvan ldquoA model-based direct power control for three-phasepower convertersrdquo IEEE Transactions on Industrial Electronicsvol 55 no 4 pp 1647ndash1657 2008
[8] Z Dawei X Lie and B W Williams ldquoImproved direct powercontrol of grid-connected DCAC convertersrdquo IEEE Transac-tions on Power Electronics vol 24 no 5 pp 1280ndash1292 2009
[9] L Xu and P Cartwright ldquoDirect active and reactive powercontrol of DFIG for wind energy generationrdquo IEEE Transactionson Energy Conversion vol 21 no 3 pp 750ndash758 2006
[10] D Zhi and L Xu ldquoDirect power control of DFIG with constantswitching frequency and improved transient performancerdquoIEEE Transactions on Energy Conversion vol 22 no 1 pp 110ndash118 2007
[11] Y Errami M Benchagra M Hilal M Maaroufi and M Ouas-said ldquoControl strategy for PMSG wind farm based on MPPTand direct power controlrdquo in Proceedings of the InternationalConference onMultimedia Computing and Systems (ICMCS rsquo12)pp 1125ndash1130 May 2012
[12] A Harrouz A Benatiallah and O Harrouz ldquoDirect powercontrol of a PMSG dedicated to standalone wind energysystemsrdquo in Proceedings of the 8th International Conference andExhibition on Ecological Vehicles and Renewable Energies (EVERrsquo13) pp 1ndash5 March 2013
[13] P Vas Sensorless Vector and Direct Torque Control OxfordUniversity Press 1998
[14] K Sedhuraman S Himavathi and A MuthuramalingamldquoPerformances comparison of neural architectures for on-linespeed estimation in sensorless IM drivesrdquo World Academy ofScience Engineering and Technology no 60 pp 1318ndash1325 2011
[15] S Meziane and R Benalla ldquoMRAS based speed control ofsensorlessinduction motor drivesrdquo ICGST-ACSE Journal vol 7no 1 pp 43ndash500 2007
[16] E Levi and M Wang ldquoImpact of parameter variations onspeed estimation in sensorless rotor flux oriented inductionmachinesrdquo in Proceedings of the 7th International Conferenceon Power Electronics and Variable Speed Drives pp 305ndash310London UK September 1998
[17] M E Elbulk L Tong and I Husain ldquoNeural-network-basedmodel reference adaptive systems for high-performance motordrives and motion controlsrdquo IEEE Transactions on IndustryApplications vol 38 no 3 pp 879ndash886 2002
[18] M T Hagan B Demuth andM BealeNeural Network DesignCengage Learning Pvt Ltd New Delhi India 2008
[19] S-H Kim T-S Park J-N Yoo and G-T Park ldquoSpeed-sensorless vector control of an induction motor using neuralnetwork speed estimationrdquo IEEE Transactions on IndustrialElectronics vol 48 no 3 pp 609ndash614 2001
[20] M Kuchar P Brandstetter and M Kaduch ldquoSensorless induc-tion motor drive with neural networkrdquo in Proceedings ofthe IEEE 35th Annual Power Electronics Specialists Conference(PESC rsquo04) pp 3301ndash3305 June 2004
[21] S Maiti V Verma C Chakraborty and Y Hori ldquoAn adaptivespeed sensorless induction motor drive with artificial neuralnetwork for stability enhancementrdquo IEEE Transactions onIndustrial Informatics vol 8 no 4 pp 757ndash766 2012
[22] P Girovsky and J Timko ldquoShaft sensor-less FOC control ofan induction motor using neural estimatorsrdquo Acta PolytechnicaHungarica vol 9 no 4 pp 31ndash45 2012
[23] D P Kothari and I J Nagrath Electric Machines chapter 9 TataMcGraw-Hill Education Private Limited 2004
[24] G Escobar A M Stankovic J M Carrasco E Galvan andR Ortega ldquoAnalysis and design of direct power control (DPC)for a three phase synchronous rectifier via output regulationsubspacesrdquo IEEE Transactions on Power Electronics vol 18 no3 pp 823ndash830 2003
[25] N R N Idris and A H M Yatim ldquoDirect torque controlof induction machines with constant switching frequency andreduced torque ripplerdquo IEEE Transactions on Industrial Elec-tronics vol 51 no 4 pp 758ndash767 2004
[26] S-H Huh K-B Lee D-W Kim I Choy and G-T ParkldquoSensorless speed control system using a neural networkrdquoInternational Journal of Control Automation and Systems vol3 no 4 pp 612ndash619 2005
Submit your manuscripts athttpwwwhindawicom
Computer Games Technology
International Journal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Distributed Sensor Networks
International Journal of
Advances in
FuzzySystems
Hindawi Publishing Corporationhttpwwwhindawicom
Volume 2014
International Journal of
ReconfigurableComputing
Hindawi Publishing Corporation httpwwwhindawicom Volume 2014
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Applied Computational Intelligence and Soft Computing
thinspAdvancesthinspinthinsp
Artificial Intelligence
HindawithinspPublishingthinspCorporationhttpwwwhindawicom Volumethinsp2014
Advances inSoftware EngineeringHindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Electrical and Computer Engineering
Journal of
Journal of
Computer Networks and Communications
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Hindawi Publishing Corporation
httpwwwhindawicom Volume 2014
Advances in
Multimedia
International Journal of
Biomedical Imaging
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
ArtificialNeural Systems
Advances in
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
RoboticsJournal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Computational Intelligence and Neuroscience
Industrial EngineeringJournal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Modelling amp Simulation in EngineeringHindawi Publishing Corporation httpwwwhindawicom Volume 2014
The Scientific World JournalHindawi Publishing Corporation httpwwwhindawicom Volume 2014
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Human-ComputerInteraction
Advances in
Computer EngineeringAdvances in
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Submit your manuscripts athttpwwwhindawicom
Computer Games Technology
International Journal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Distributed Sensor Networks
International Journal of
Advances in
FuzzySystems
Hindawi Publishing Corporationhttpwwwhindawicom
Volume 2014
International Journal of
ReconfigurableComputing
Hindawi Publishing Corporation httpwwwhindawicom Volume 2014
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Applied Computational Intelligence and Soft Computing
thinspAdvancesthinspinthinsp
Artificial Intelligence
HindawithinspPublishingthinspCorporationhttpwwwhindawicom Volumethinsp2014
Advances inSoftware EngineeringHindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Electrical and Computer Engineering
Journal of
Journal of
Computer Networks and Communications
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Hindawi Publishing Corporation
httpwwwhindawicom Volume 2014
Advances in
Multimedia
International Journal of
Biomedical Imaging
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
ArtificialNeural Systems
Advances in
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
RoboticsJournal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Computational Intelligence and Neuroscience
Industrial EngineeringJournal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Modelling amp Simulation in EngineeringHindawi Publishing Corporation httpwwwhindawicom Volume 2014
The Scientific World JournalHindawi Publishing Corporation httpwwwhindawicom Volume 2014
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Human-ComputerInteraction
Advances in
Computer EngineeringAdvances in
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Related Documents
