Sunteți pe pagina 1din 9

This article has been accepted for publication in a future issue of this journal, but has not been

fully edited. Content may change prior to final publication.

Direct Torque Control of Two-Phase Induction and


Synchronous Motors
Saleh Ziaeinejad, Younes Sangsefidi, Hamidreza Pairodin Nabi, Abbas Shoulaie

Abstract— In this paper, direct torque control (DTC) of or placing a capacitor in series with the auxiliary winding is
symmetrical two-phase induction and synchronous motors required for high-speed (e.g. rated speed) applications [3].
using different structures of two-phase inverters including two- It’s possible to overcome the stated problems by utilizing
, three- and four-leg inverters is studied. Essential equations three-leg inverter and using both motor windings. Difficulty
for controlling both electromagnetic torque and the stator flux of this method is that the auxiliary winding voltage should be
modulus are presented and switching tables are derived. higher than that of the main winding so using boost
Different inverter topologies are compared in terms of the converters or placing a capacitor in series with the auxiliary
control system operation in high speed applications and winding is required for high-speed (e.g. rated speed)
maximum speed of the rotating field deliverable by the drive applications [3]. Because of the cost attributed to these
system. Performance of DTC of two- phase induction and approaches, speed control of symmetrical two-phase
synchronous motors fed by different inverter topologies are induction motors is preferred. Another alternative is
evaluated and correctness of presented theories is validated by
utilization of two-phase synchronous drives which
experimental results.
introduces higher torque density and thus more compact size.
Keywords— two-phase induction motor, two-phase Common structures of two phase inverters shown in Fig.
synchronous motor, two-, three- and four-leg inverters, direct 1 have been studied in [4]-[9]. The operation of two-leg
torque control inverter shown in Fig.1.a is weak, but because of possessing
I. INTRODUCTION minimum number of switches, it is a low-cost approach. In
the applications in which the middle-point of DC voltage is
Because of some features such as low cost and simple accessible, using two-leg inverter is an acceptable approach.
and robust structure, single phase induction motors are When middle-point of DC voltage is inaccessible, utilizing
widely used in household products and low-power industrial this scheme needs not only two capacitors with high
equipments. Single phase induction motor is an capacitances but also a voltage equalizing circuit. Voltage
asymmetrical two-phase induction motor including the main equalization can be done by using resistors in parallel to
and the auxiliary windings. Three common types of single- capacitors. This increases volume of the drive system and
phase induction motors are split-phase, capacitor-start and power loss [10].
capacitor-run motors [1]. Single phase motors work well in
constant-speed applications, but show difficulties when Another approach is utilization of charge balancing
variable-speed performance is needed. In both-split-phase circuits [11] which makes the drive circuit complicated and
and capacitor-start single phase motors, operation in wide costly. A noticeable limitation of three-leg inverters in
speed range is practically inaccessible because in low speed controlling two-phase motors is that the RMS value of
operation, the centrifugal switch doesn’t disconnect the common leg current is higher than those of other two legs
auxiliary winding. This causes damages to the motor [5], [10] that necessitates utilization of switches with higher
structure since the auxiliary winding is not designed for current ratings at least in one leg. When integrated power
continuous operation. In capacitor-run motors, the modules (IPMs) are used, the entire module should be of
appropriate capacitance depends on the conditions of power higher current ratings. This makes the cost enhancement
supply and mechanical load. So, in variable speed more noticeable. This additional cost as well as better
application, torque ripple and over temperature are performance of four-leg inverter makes the use of four-leg
problematic [2], [3]. inverter justifiable in high performance applications [12].
It’s possible to overcome the stated problems by utilizing When good performance of motor drive is expected,
three-leg inverter and using both motor windings. Difficulty utilizing high performance control methods is inevitable.
of this method is that the auxiliary winding voltage should be Vector control of symmetrical two-phase induction motor is
higher than that of main winding, so using boost converters proposed in [13] as a solution for low-power level field; but
suggested scheme needs utilization of mechanical sensors
which makes the drive system costly. On the contrary, good
Saleh Ziaeinejad, Younes Sangsefidi, Hamidreza Peiro-Din-Nabi, and Abbas
Shoulaie are with the Department of Electrical Engineering, Iran University performance of DTC method without using speed or position
of Science and Technology, Tehran 16846-13114, Iran (e-mail: sensors makes it a viable choice in both high performance
s_ziaeinejad@elec.iust.ac.ir). and low-cost applications [14].

Copyright (c) 2011 IEEE. Personal use is permitted. For any other purposes, permission must be obtained from the IEEE by emailing pubs-permissions@ieee.org.
This article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication.

flux (xy), and stationary (DQ) reference frames. According


to [17] and [18], by using voltage-current equations, stator
fluxes of D and Q axes in stationary reference frame and
electromagnetic torque of two-phase induction motor can be
defined as:
λ =∫ −r i dt (1)
(a) (b)
λ = ∫( − r i ) dt (2)

= λ +λ (3)

=P λ i −λ i =P × =
(c)
P sin (θ − θ ) (4)
Figure 1. Different inverter topologies used in two-phase motor drives (a)
two-leg inverter (b) three-leg inverter (c) four-leg inverter where r is resistance of the stator winding; ,
respectively are the stator flux vector and the rotor flux
vector; , , i , i , λ and λ respectively are
stator voltage components, stator current components and
stator flux components in Q and D axes. Also, L , L , L
and P respectively are stator inductance, rotor inductance
referred to stator side, magnetizing inductance and number
of motor pole pairs. In addition, θ , θ and δ are stator flux
angle, rotor flux angle and load angle, respectively and

Figure 2. stator and rotor flux vectors in deferent reference frames


= (1 − ) (5)

Direct torque control of two-phase induction motors A similar analysis is performed to study the DTC method
using two- and three-leg inverters has been studied in the of synchronous motor. Regarding [19] and [20], using
literature [15], [16].In both inverter topologies, in some voltage-current equations of the motor, stator flux
regions of stator flux, the torque control fails. In high speeds, components in stationary reference frame and
the trouble gets worse. Changing the switching table for electromagnetic torque of two-phase synchronous motor can
improving the torque control is proposed in [15], [16]. Using be calculated as:
this approach not only makes the control system λ =∫ −r i dt + λ | (6)
complicated, but also discards flux control in some areas of
stator flux. In high speed operation, utilizing modified λ = ∫( − r i )dt + λ | (7)
switching table leads to losing flux control in a wide area and
inefficient operation of control system [16]. | |= + (8)
In this paper, because of prominent features of two-phase
| |
motors in variable speed applications, utilizing these motors =P λ i −λ i = 2λ L sin(θ − θ ) −
is suggested instead of single phase induction motors. Main
application of two-phase motor drives would be in low- |λ | L − L sin(2(θ − θ )) (9)
power industrial applications as well as in household where is rotor linkage flux and and are d-axis
products such as refrigerators, vacuum cleaners, fans, air and q-axis inductances, respectively. Initial values of D- and
conditioners, water pumps etc. Direct torque control as a low Q- axis stator flux components are:
cost, high reliable and high performance control method for
two-phase induction and synchronous motors is proposed λ | = sin (10)
and switching tables for two-, three- and four- leg inverter
topologies are presented. Different inverter topologies are λ | = cos (11)
compared in terms of drive system ability to control where is initial angle between rotor flux vector and D
electromagnetic torque and maximum speed of rotating field axis. Block diagram of direct torque control of two-phase
provided by the drive system. Performance of DTC of two- induction and synchronous motor is shown in Fig.3 in which
phase induction and synchronous motors fed by different at each sampling period, the stator flux modulus and
inverter topologies are evaluated and correctness of electromagnetic torque are found and a voltage vector which
presented formulas is validated by experimental results. reduces the errors of these parameters is chosen. Electrical
time constant of the rotor of induction motors is large [18]
II. DIRECT TORQUE CONTROL OF TWO-PHASE INDUCTION and so the rotor flux doesn’t considerably change in a short
AND SYNCHRONOUS MOTORS interval. Thus, a quick torque response can be achieved by
A. Basic Principles rotating the voltage vector in a way that θ would increase
rapidly. Like induction motor, step change in
For analyzing DTC method, first, equations of two-phase
electromagnetic torque of synchronous motor can be resulted
induction motor are studied. It’s assumed that air gap length
by accelerating stator flux vector leading to change the angle
is constant and magnetic circuit is linear. Fig.2 shows the
between stator and rotor linkage flux vectors [20].
stator and rotor linkage flux vectors in rotor flux (dq), stator

Copyright (c) 2011 IEEE. Personal use is permitted. For any other purposes, permission must be obtained from the IEEE by emailing pubs-permissions@ieee.org.
This article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication.

curve in stable operating region is always positive, so


increase of the load angle leads to increased electromagnetic
torque. Therefore, DTC method works properly. When a
salient-pole synchronous motor with < is being
controlled using DTC method, slope of “load angle-torque”
curve should be positive to have a perfect torque control.
Regarding (9), a constraint should be met:
|λ |∗ < λ (12)
If (12) is not met, a delay will occur during motor
starting period [19].
Figure 3. Block diagram of direct torque control of two-phase induction
and synchronous motor 2. Switching table of DTC method is designed to control
synchronous motor in stable region of “load angle-torque”
curve. If the command torque exceeds the maximum
deliverable torque of motor, loss of synchronism occurs [20],
[21]. So, the command torque should be less than maximum
deliverable torque. The maximum deliverable torque of
synchronous motor can be derived as:
| |
= 2λ L sin( ) − |λ | L −
L ) sin(2 ) (13)
Figure 4. Four stator flux sectors generated by voltage vectors of two-leg where is the maximum load angle that can be
inverter expressed as:

= cos | |
− | | | |
+2 (14)

where

ε= (15)

3. In the case of synchronous motor control, regarding


(6)-(11), precise calculation of stator flux components
requires diagnosis of initial position. As position encoder is
Figure 5. Six stator flux sectors generated by voltage vectors of three-leg not utilized in DTC method, the estimation of initial rotor
inverter position is formidable. When the initial position estimation is
TABLE I. DTC SWITCHING TABLE FOR TWO-PHASE MOTORS FED BY incorrect, the motor initially rotates in wrong direction that
TWO -LEG INVERTER restricts usage of synchronous motor DTC drive in some
applications [22].
B. DTC method utilizing two-leg inverter
When the middle point of DC-link power supply is
accessible, DTC of a two-phase motor by a two-leg inverter
is possible. Fig.4 shows available voltage vectors generated
by two-leg inverter and the way of distinguishing four stator
TABLE II. DTC SWITCHING TABLE FOR TWO-PHASE MOTORS FED BY
THREE -LEG INVERTER flux sectors using a method similar to presented ones in
[19],[23] for three-phase motors. Appropriate switching table
for DTC of two-phase induction and synchronous motor fed
by two-leg inverter is presented in Table I. When the stator
flux vector is near the border of a sector, the component of
voltage vector tangent to circular route of the stator flux
becomes so small that is not able to reduce torque error and
so electromagnetic control method is unable to keep the
electromagnetic torque error in allowed band.
Using DTC method, switching table of two-phase C. DTC method utilizing three-leg inverter
synchronous motor derive resembles that of induction motor Using a three-leg inverter for controlling two-phase
drive, but synchronous motor control method has some motor obviates the need of accessing middle point of DC
constraints: link voltage. For controlling both flux and electromagnetic
torque, six distinct sectors can be specified for the stator flux.
1. In DTC method, when increasing the electromagnetic The way of doing this and available voltage vectors are
torque is desired, the switching table increases the load shown in Fig. 5. Table II presents the appropriate switching
angle. In “load angle-torque” curve of cylindrical and table of two-phase induction and synchronous motor fed by
salient-pole synchronous motors with > , slope of the three-leg inverter.

Copyright (c) 2011 IEEE. Personal use is permitted. For any other purposes, permission must be obtained from the IEEE by emailing pubs-permissions@ieee.org.
This article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication.

III. MAXIMUM ROTATING FIELD SPEED OF TWO-PHASE


MOTOR IN DTC METHOD
In DTC method, in order to increase the electromagnetic
torque, angle of the stator flux vector should increase.
Keeping the modulus of the stator flux constant, maximum
speed of the rotating field is limited. For obtaining maximum
possible speed, it’s assumed that mechanical load is light and
cannot be equal to command torque even in high speed. In
this case, DTC system continually applies voltage vectors
Figure 6. Eight stator flux sectors generated by voltage vectors of four-leg which rotate the stator flux in forward direction;
inverter using two approaches to generate zero voltage
electromagnetic torque is not controlled and the rotating field
rotates in highest speed possible. For the sake of simplicity,
TABLE III. DTC SWITCHING TABLE FOR TWO-PHASE MOTORS FED BY
FOUR-LEG INVERTER the stator ohmic voltage drop is neglected in the calculations.
A. Maximum speed of rotating field using two-leg inverter
In a two-phase DTC drive using two-leg inverter there
are four identical flux sectors, so analysis of the maximum
speed in one sector is adequate. It’s assumed that stator flux
vector is in sector 1, so according to Fig. 4, in order to
achieve the maximum rotating speed, one of or vectors
are chosen to accelerate flux vector based on the output of
flux hysteresis controller. The stator flux vector is being
rotated by a component of voltage vector tangential to its
circular route. The tangential components of voltage vectors
generated by two-leg inverter are presented in table IV.
Maximum linear speed of stator flux vector is equal to
weighted average of the tangential component of or
V = ( ) + ( ) (16)

TABLE IV. TANGENTIAL COMPONENTS OF VOLTAGE VECTORS Where is tangential component of and ( ) is the
GENERATED BY TWO-LEG INVERTER probability of applying voltage vector . Probability of
Tangential component V V applying each voltage vector is proportional to the maximum
time interval in which applying that vector maintains the
Tangential component √2 π √2 π modulus of stator flux in allowed range. Since the amplitude
V cos (θ + ) cos (θ − )
2 4 2 4 of four voltage vectors generated by two-leg inverter are all
Tangential component V V
the same, the time interval in which a certain voltage vector
is applied is proportional to the length of the path of stator
Tangential component √2 3π √2 5π flux while the modulus of stator flux is in allowed range.
V cos (θ − ) cos (θ − )
2 4 2 4
First, it’s supposed that the angle between the stator flux
vector and D-axis is and the modulus of stator flux has
In sectors 1 and 4, when the stator flux vector is near the reached to its minimum allowed value. As shown in Fig. 6,
border of the sector, the component of voltage vector tangent for increasing both torque and flux, is selected. In Fig. 7,
to circular route of the stator flux becomes so small that is | |∗ is desired modulus of the stator flux and x is the length
not able to reduce torque error and so electromagnetic of flux-developing vector. In “OAB” triangle, equation (17)
control method is unable to keep the electromagnetic torque is established:
error in allowed band. However, because of possessing
∗ ∗
larger voltage vectors, the ability of three-leg inverter to + −
−2 − ×
control electromagnetic torque is more than that of two-leg 3 ∗
cos 4+ = | | + (17)
inverter and its uncontrollable torque region is smaller [16].
where is the width of hysteresis flux controller.
D. DTC method utilizing four-leg inverter
Assuming = | |∗ and = | |∗ gives
Like three-leg inverters, four-leg inverters also do not
need access to middle-point of DC link voltage. For
controlling both flux and electromagnetic torque, eight = 4 + (1 + ) sin 4+ −
distinct sectors can be specified for the stator flux. The way
of doing this and available voltage vectors are shown in Fig. (1 + ) sin
4+ (18)
6. Table III presents appropriate switching table for DTC “k” has a small value; so regarding Maclaurin
method. Four-leg inverter is able to control electromagnetic
torque in a wide range of speed and does not have series √ + + 1 → ≅ 1 + , (18) results:
uncontrollable torque region which is problematic in DTC
using aforementioned inverters. As there’s always a vector = (19)
suitable for controlling both flux and torque, these two
parameters can be properly controlled by four-leg inverter. where is normalized length of flux-developing vector.

Copyright (c) 2011 IEEE. Personal use is permitted. For any other purposes, permission must be obtained from the IEEE by emailing pubs-permissions@ieee.org.
This article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication.

So, the maximum rotational speed of the stator flux


vector throughout a sector is

= ∫ ≅ 0.56 | (24)
( ) | |∗ |∗

Regarding (24), the maximum rotating field speed of


Figure 7. Selecting vector in the sector1 in order to develop stator flux two-phase motor depends on DC link voltage and the
modulus of the stator flux.
B. Maximum speed of rotating field using three-leg inverter
Six sectors of the stator flux can be divided into two
identical groups, each including three consecutive sectors.
So, sectors 1, 2 and 3 are chosen for calculations. According
to Fig. 5, tangential components of voltage vectors generated
Figure 8. Selecting vector in the sector1 in order to weaken stator flux by three-leg inverter are presented in Table V.
TABLE V. TANGENTIAL COMPONENT OF VOLTAGE VECTORS Like those in two-leg inverter, essential equations for
GENERATED BY THREE-LEG INVERTER calculating the lengths of flux-developing and flux-
Tangential component v v v
weakening vectors generated by three-leg inverter and the
probability of applying these vectors are derived as functions
Tangential Component π π
√2cos (θ + ) cos (θ) cos (θ − ) of the stator flux angle. Table VI presents probability of
V 4 2
applying each voltage vector in sectors 1, 2 and 3 when the
Tangential component v v v maximum speed of rotating field is expected. ( ) presents
Tangential Component 3π 3π the probability of using in sector j.
√2cos (θ − ) cos (θ − π) cos (θ − )
V 4 2
According to probability functions and tangential
TABLE VI. PROBABILITY OF USING DIFFERENNT VOLTAGE VECTORS
components of voltage vectors, maximum angular speed of
OF THREE-LEG INVERTER IN SECTORS 1, 2, 3 the stator flux in each sector can be calculated as:
π

ω = ∫ ( ) + ( ) dθ ≅ 0.794 (25)
π | |∗ | |∗
π

ω = ∫π ( ) v + ( ) dθ ≅ 1.122 (26)
π | |∗ | |∗

π
ω = ∫π ( ) + ( ) dθ ≅ 1.122 (27)
π | |∗ | |∗

As the stator flux transverses half of the path with


average speed ω and remaining path with average speed
ω = ω maximum angular speed of stator flux is
Now, it’s assumed that the modulus of stator flux has
reached to its maximum allowed value. In this condition, as 0.93 ∗ .
| |
shown in Fig. 8, in order to increase electromagnetic torque
and reduce the modulus of the stator flux, vector is C. Maximum speed of rotating field using four-leg inverter
selected. In Fig.8, y is the length of flux weakening vector. Like two- and three-leg inverter topologies, for
Pursuing calculations in “ ” triangle analogous to those determining maximum rotating field speed of two-phase
pursued in “ ” triangle results: motors fed by four-leg inverter, it’s assumed that voltage
vectors which rotate the stator flux vector with highest
= (20) possible speed and also maintain the modulus of the stator
flux in allowed range are used. The values of tangential
is normalized length of flux-weakening vector and is components of voltage vectors generated by the inverter are
equal to ∗.So, the probabilities of applying each of these required for calculation of maximum speed of rotating field.
| |
These values are shown in Table VII based on the voltage
two flux-developing and flux-weakening vectors are as: vectors illustrated in Fig.6. Eight sectors of the stator flux
can be divided into four identical groups each containing 90°
= = (21)
( ) of the stator flux angle, so performing calculations in 90° is
and adequate. Areas 1, 2 that respectively represent ≤θ<0
, 0 ≤ θ < are selected for calculations.
( ) = = (22)
Like aforementioned two topologies, the lengths of flux-
Regarding (16), (21) and (22), maximum linear speed of developing and flux-weakening vectors are calculated and
stator flux vector can be expressed as: then the probability distribution functions are derived as a
function of stator flux angle as shown in Table VIII, where
V = ( ) + ( ) = (23)
( ) ( ) is the probability of using in area j.

Copyright (c) 2011 IEEE. Personal use is permitted. For any other purposes, permission must be obtained from the IEEE by emailing pubs-permissions@ieee.org.
This article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication.

TABLE VII. TANGENTIAL COMPONENT OF VOLTAGE VECTORS TABLE X. SYNCHRONOUS MOTOR PARAMETERS
GENERATED BY FOUR -LEG INVERTER
Stator resistance 4.5 Ω
Tangential Tangential Component Tangential Tangential Component
component component
d- axis inductance 323 mH
V V
q- axis inductance 110 mH
π 3π
v √2cos (θ + ) v √2cos (θ − ) Number of poles 4
4 4
Rotor flux 1
v cos (θ) v cos (θ − π)
π 5π
v √2cos (θ − ) v √2cos (θ − )
4 4
π 6π
v cos (θ − ) v cos (θ − )
2 4

TABLE VIII. PROBABILITY OF USING DIFFERENNT VOLTAGE VECTORS


OF FOUR -LEG INVERTER IN 90°

Figure 10. Test setup configuration

Configuration of the experimental setup is shown in


Fig.10. A four-leg IGBT-based voltage source inverter is
connected to DC-link bus to drive the motor. Regarding
According to probability functions and tangential desired inverter configuration, 2, 3 or 4 legs of the inverter
components of voltage vectors, maximum angular speed of are used. A DC generator feeding a resistive load is used as
stator flux of direct-torque-controlled-two-phase motor fed the mechanical load of the motor. A TMS320F2812 Digital
by four-leg inverter can be calculated as: Signal Processor (DSP) is connected to computer for
programming. Time interval for sampling and computations
= ∫ ≅ 1.12 | (28) is 40 µs. In order to display and save considered waveforms
| | ( ) |
in the computer, Advantech USB-4711A Data Acquisition
Thus, keeping the stator flux and DC link voltage Module is utilized.
constant, four-leg inverter provides the possibility of
operation at higher speeds. A. DTC of two-phase motors
Results of direct torque control of two-phase induction
IV. EXPERIMENTAL RESULTS motor using two-leg inverter is shown in Fig. 11. Command
In order to validate theoretical analysis, two-phase torque changes ±3 each 1.5 sec. First change is applied
induction and synchronous motor drive systems has been at = 0.25 .
implemented in the laboratory. The used induction motor is a According to Fig. 11-a, electromagnetic torque quickly
2 KW n-phase motor (n=2,3,4,6 and 12) which was set in responds to changes of command torque. Rotating field
two-phase configuration. The used synchronous motor is a 2 speed shown in Fig. 11-b is obtained by processing
KW two-phase motor made for the tests. Fig. 9 shows the derivative of stator flux vector angle in accordance with
used induction and synchronous motors with parameters time. Due to Fig. 11-a and 11-b, as the speed of rotating field
listed in Table IX and Table X, respectively. increases, ability of drive system to control the
electromagnetic torque decreases. In addition, fig. 11-d
shows that the stator flux vector traverses a circular path and
its modulus is controlled.
Results of DTC of two-phase induction motor using a
three-leg inverter is shown in Fig. 12. Like two-leg inverter,
ability of the drive system to control electromagnetic torque
decreases by increasing rotating field speed; But three-leg
inverter control the electromagnetic torque in a better way
Figure 9. Two-phase induction motor and DC generator (left), two-phase
synchronous motor and DC generator (right)
since voltage vectors generated by this topology are larger
than those of two-leg inverter. Regarding Fig. 12-d, the stator
TABLE IX. INDUCTION MOTOR PARAMETERS flux modulus is properly controlled.
Stator resistance 2.6 Ω As shown in Fig. 13-a, the drive system with four-leg
Rotor resistance 1.0 Ω
inverter properly controls the electromagnetic torque.
Stator leakage inductance 7.3 mH
Rotor leakage inductance 7.3 mH
Regarding Fig. 13-d, the drive system is also able to control
Magnetizing inductance 238.5 mH the stator flux modulus properly. Results of DTC of two-
Number of poles 4 phase synchronous motor using two-, three- and four-leg
inverters are demonstrated in figures 14, 15 and 16,
respectively.

Copyright (c) 2011 IEEE. Personal use is permitted. For any other purposes, permission must be obtained from the IEEE by emailing pubs-permissions@ieee.org.
This article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication.

Accordingly, using any of these inverter topologies, the


stator flux modulus is properly controlled. Unlike DTC using
four-leg inverter which controls electromagnetic torque
properly, when either two- or three-leg inverter is used,
ability of drive system to control electromagnetic torque
decreases.
(a)
B. Maximum speed of rotating field in two-phase DTC
drives
Theoretical analysis for calculating maximum rotating
field speed in DTC drive system using different inverter
topologies is also validated by experimental tests. In these
tests, the motor load is removed so the motor rotates in
(b) highest speed limited by DC link voltage and stator flux
modulus. Rotating field speed of no-load two-phase
induction motor controlled by DTC method using different
inverter topologies is shown in Fig.17 when reference of the
stator flux modulus is 0.8 wb and DC link voltage is 100 V.
Comparison between theoretical and experimental results is
presented in Table XI. Difference between theoretical and
(c)
experimental results is due to neglecting ohmic voltage drop
of the stator windings in calculations.
Rotating field speed of no-load two-phase synchronous
motor controlled by DTC method using different inverter
topologies is shown in Fig. 18 when the stator flux modulus
reference is 1 wb and DC link voltage is 100 V. Table XII
shows theoretical and experimental results. The difference
(d)
between results is due to neglecting ohmic voltage drop of
Figure 11. DTC of induction motor using two-leg inverter (a)
electromagnetic torque (b) rotating field speed (c) D- and Q- axes
the stator windings in calculations.
components of the stator flux (d) stator flux trajectory

(a)

(a)

(b)

(b)

(c)

(c)

(d)
(d) Figure 13. DTC of induction motor using four-leg inverter (a)
electromagnetic torque (b) rotating field speed (c) D- and Q- axes
Figure 12. DTC of induction motor using three-leg inverter (a) components of the stator flux (d) stator flux trajectory
electromagnetic torque (b) rotating field speed (c) D- and Q- axes
components of the stator flux (d) stator flux trajectory

Copyright (c) 2011 IEEE. Personal use is permitted. For any other purposes, permission must be obtained from the IEEE by emailing pubs-permissions@ieee.org.
This article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication.

(a) (a)

(b) (b)

(c) (c)

(d)
(d)
Figure 14. DTC of synchronous motor using two-leg inverter (a) Figure 16. DTC of synchronous motor using four-leg inverter (a)
electromagnetic torque (b) rotating field speed (c) D- and Q- axes electromagnetic torque (b) rotating field speed (c) D- and Q- axes
components of the stator flux (d) stator flux trajectory components of the stator flux (d) stator flux trajectory

TABLE XI. THEORETICAL AND EXPERIMENTAL RESULT OF MAXIMUM


ROTATING FIELD SPEED OF TWO- PHASE INDUCTION MOTOR

Inverter topology Two- leg Three- leg Four- leg


Maximum speed of rotating
668 rpm 1110 rpm 1337 rpm
field (calculations)
Maximum speed of rotating
637 rpm 1011 rpm 1243 rpm
field (experimental)
(a)

(b)
Figure 17. speed of rotating field of no-load two-phase induction motor
controlled by DTC method using different inverter topologies

TABLE XII. THEORETICAL AND EXPERIMENTAL RESULT OF MAXIMUM


ROTATING FIELD OF TWO - PHASE SYNCHRONOUS MOTOR

Inverter topology Two- leg Three- leg Four- leg


Maximum speed of rotating field
(c) 535 rpm 888 rpm 1070 pm
(calculations)
Maximum speed of rotating field
472 rpm 783 rpm 950 rpm
(experimental)

(d)
Figure 15. DTC of synchronous motor using three-leg inverter (a)
electromagnetic torque (b) rotating field speed (c) D- and Q- axes Figure 18. speed of rotating field of no-load two-phase synchronous
components of the stator flux (d) stator flux trajectory motor controlled by DTC method using different inverter topologies

Copyright (c) 2011 IEEE. Personal use is permitted. For any other purposes, permission must be obtained from the IEEE by emailing pubs-permissions@ieee.org.
This article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication.

V. CONCLUSION Induction Motor as a Replacement," IEEE Transactions on Power


Electronics, vol.28, no.1, pp. 526-536, January 2013.
In this paper, at first, problems associated with speed [14] D. Casadei, F. Profumo, G. Serra, A. Tani, "FOC and DTC: Two
control of single phase motors are described and substitution Viable Schemes for Induction Motors Torque Control," IEEE
of single phase motors with two-phase motors in variable- Transactions on Power Electronics, vol.17, no.5, pp. 779- 787,
speed applications is proposed. Then, direct torque control of September 2002.
two-phase induction and synchronous motors using two- [15] F.A.S. Neves, E.B.S. Filho, J.M.S. Cruz, R.P. Landim, Z.D. Lins,
three- and four-leg inverters is analyzed. It’s shown that A.G.H. Accioly, "Single-Phase Induction Motor Drives with Direct
Torque Control," 28th Annual Conference of the Industrial
when a two-phase motor controlled by DTC method is fed Electronics Society, vol.1, pp. 241- 246 5-8 November 2002.
by four-leg inverter, unwanted torque ripple doesn’t exist, [16] Y. Sangsefidi, S. Ziaeinejad, A. Shoulaie, “Direct Torque Control of
while torque ripple is problematic when the motor is fed by Two-Phase Induction Motors Fed by Two- and Three- leg Inverters”,
two- or three-leg inverter. In addition, it’s shown that having 20th Iranian Conference on Electrical Engineering (ICEE2012),
certain DC link voltage and desired stator flux, maximum pp.444-449, 15-17 May 2012.
possible speed of the motor controlled by DTC method [17] Chee Mun Ong, "Dynamic Simulation of Electric Machinery Using
increases when the motor is fed by an inverter possessing MATLAB SIMULINK," Prentice Hall, New Jersey 1998.
higher number of legs. Inverter having the least number of [18] P. Vas, "Sensor-less Vector and Direct Torque Control," Oxford
University Press, U.K. 1998.
legs which can provide the required speed is the most
inexpensive option. Experimental results are included to [19] L. Zhong, M.F. Rahman, W.Y. Hu, K.W. Lim, "Analysis of Direct
Torque Control in Permanent Magnet Synchronous Motor Drives,"
evaluate performance of direct torque controlled two-phase IEEE Transactions on Power Electronics, vol.12, no.3, pp.528-536,
motor and validate theoretical calculations of maximum May 1997.
speed of rotating field in two-phase DTC drive system. [20] O. Pyrhönen, “Analysis and Control of Excitation, Field Weakening
and Stability in Direct Torque Controlled Electrically Excited
REFERENCES Synchronous Motor Drives,” Dept. Elect. Eng., Lappeenranta
University of Technology, Lappeenranta, Finland, 1998.
[1] D. Jang, "PWM Methods for Two-Phase Inverters," IEEE Industry
Applications Magazine, vol.13, no.2, pp.50-61, March-April 2007. [21] J. Luukko, O. Pyrhonen, M. Niemela, J. Pyrhonen, "Limitation of the
Load Angle in a Direct-Torque-Controlled Synchronous Machine
[2] E.R. Collins, Jr, R.E. Ashley III; , "Operating Characteristics of
Drive," IEEE Transactions on Industrial Electronics , vol.51, no.4,
Single-Phase Capacitor Motors Driven from Variable Frequency
pp. 793- 798, August 2004.
Supplies," Conference Record of the 1991 IEEE Industry
Applications Society Annual Meeting, vol. 1, pp.52-57 September- [22] M.F. Rahman, Md.E. Haque, L. Tang, L. Zhong, "Problems
October 1991 Associated with the Direct Torque Control of an Interior Permanent-
Magnet Synchronous Motor Drive and their Remedies," IEEE
[3] F. Blaabjerg, F. Lungeanu, K. Skaug, M. Tonnes, "Two-Phase
Transactions on Industrial Electronics, vol.51, no.4, pp. 799- 809,
Induction Motor Drives," IEEE Industry Applications Magazine, ,
August 2004.
vol.10, no.4, pp.24- 32, July-August 2004.
[23] I. Takahashi, T. Noguchi, "A New Quick-Response and High-
[4] S.S. Wekhande, B.N. Chaudhari, S.V. Dhopte, R.K. Sharma, "A Low
Efficiency Control Strategy of an Induction Motor," IEEE
Cost Inverter Drive for 2-Phase Induction Motor," Proceedings of the
Transactions on Industry Applications , vol.IA-22, no.5, pp.820-827,
IEEE 1999 International Conference on Power Electronics and Drive
September 1986.
Systems, vol.1, no., pp.428-431 vol.1, 1999.
[5] M.B.R. Correa, C.B. Jacobina, A.M.N. Lima, E.R.C. da Silva,
"Single-Phase Induction Motor Drives Systems," Fourteenth Annual Saleh Ziaeinejad received the B.Sc. and M.Sc.
Applied Power Electronics Conference and Exposition, vol.1, pp.403- degrees with honors in electrical engineering from Iran
409, March 1999. University of Science and Technology (IUST), Tehran,
[6] L.M.C. Mhango, G.K. Creighton, "Novel Two-Phase Inverter-Fed Iran, in 2009 and 2012, respectively. His fields of
interest include power electronic converters, electrical
Induction-Motor Drive," Proceedings of Institute of Electrical eng. ,
vol.131, no.3, pp.99-104, May 1984. machines and drives, photovoltaic systems and battery
management systems.
[7] D. Jang, D.Y. Yoon, "Space-Vector PWM Technique for Two-Phase
Inverter-Fed Two-Phase Induction Motors," IEEE Transactions on Younes Sangsefidi received the B.Sc. and M.Sc.
Industry Applications, vol.39, no.2, pp. 542- 549, March-April 2003. degrees with honors in electrical engineering from Iran
[8] M.T. Bartholet, T. Nussbaumer, D. Krahenbuhl, F. Zurcher, J.W. University of Science and Technology (IUST), Tehran,
Kolar, "Modulation Concepts for the Control of a Two-Phase Iran, in 2009 and 2012, respectively. His fields of
Bearingless Slice Motor Utilizing Three-Phase Power Modules," interest include power electronic converter design,
IEEE Transactions on Industry Applications, vol.46, no.2, pp.831- motor drives, photovoltaic systems and battery
840, March-April 2010. systems.
[9] V. Kinnares, C. Charumit, "Modulating Functions of Space Vector
PWM for Three-Leg VSI-Fed Unbalanced Two-Phase Induction Hamidreza Pairodin Nabi was born in 1986 in Tehran,
Motors ," IEEE Transactions on Power Electronics, vol.24, no.4, pp. Iran. He received his M.Sc degree in electrical
1135-1139, April 2009. engineering from Iran University of Science &
[10] M.T. Bartholet, T. Nussbaumer, J.W. Kolar, "Comparison of Voltage- Technology (IUST), Iran, in 2011. He is currently
Source Inverter Topologies for Two-Phase Bearingless Slice Motors," pursuing the Ph.D. degree at the same university. His
IEEE Transactions on Industrial Electronics, vol.58, no.5, pp.1921- research interests include modeling, design and control
1925, May 2011. of power electronic converters, motor drive systems and
multiphase variable speed drives.
[11] H. Ertl, T. Wiesinger, J.W. Kolar, F.C. Zach, “A Simple Active
Method to Avoid the Balancing Losses of DC Link Capacitors”, 24 th
International Conference on Power Electronics (PCIM), Nürnberg, Abbas Shoulaie received Ph.D. degree in electrical
Germany, pp.1- 6, May 2003. engineering from the Universit´e des Sciences et
[12] Y. Kumsuwan, W. Srirattanawichaikul, S. Premrudeepreechacharn, Technologies de Lille 1 (U.S.T.L.), Montpellier, France,
K. Higuchi, H.A. Toliyat, "A Carrier-Based Unbalanced PWM in 1984. He is currently a Professor of electrical
Method for Four-Leg Voltage Source Inverter Fed Asymmetrical engineering at IUST. His current research interests
Two-Phase Induction Motor," 2010 International Power Electronics
include power electronics, magnetic systems and linear
Conference (IPEC), pp.2469-2476, June 2010.
motors, flexible alternating current transmission
[13] D. Jang, "Problems Incurred in Vector Controlled Single-phase
systems (FACTS) devices, and HVDC.
Induction Motor, and a Proposal for a Vector Controlled Two-phase

Copyright (c) 2011 IEEE. Personal use is permitted. For any other purposes, permission must be obtained from the IEEE by emailing pubs-permissions@ieee.org.

S-ar putea să vă placă și