MultiCraft

International Journal of Engineering, Science and Technology

Vol. 2, No. 10, 2010, pp. 67-78

INTERNATIONAL
JOURNAL OF
ENGINEERING,
SCIENCE AND
TECHNOLOGY


www.ijest-ng.com
2010 MultiCraft Limited. All rights reserved

Carrier-based pulse width modulation technique for a three-to-five phase
matrix converter for supplying five-phase two-motor drives

1
Mohammad Saleh,
2
Atif Iqbal,
3
SK. Moin Ahmed,
4
Akhtar Kalam and
5
Haitham Abu-Rub

1
Victoria University, Melbourne, AUSTRALIA
2
Qatar University, QATAR
3
Texas A&M University at Qatar, QATAR

4
Victoria University, Melbourne, AUSTRALIA
5
Texas A&M University at QATAR
E-mails: [email protected] (Mohammad Saleh), [email protected] (Atif Iqbal), [email protected] (SK. Moin Ahmed,),
[email protected] (Akhtar Kalam,), [email protected] (Haitham Abu-Rub)



Abstract

Multi-phase (more than three-phase) converters are required mainly for feeding variable speed multi-phase drive systems. This
paper discusses one such solution by using direct ac-ac converter that can be used to supply a five-phase two-motor drive system.
Matrix converter based two-motor drive system is presented for the first time in this paper. Two stator windings of two five-phase
machines when connected in seires/parallel while the rotors may be connected to different laods, are called series/parallel-
connected two-motor drive. Appropriate phase transposition is introuduced while connecting the series/parallel stator winding to
obtain decoupled control of the two machines. This paper proposes simple carrier-based Pulse width modulation (PWM) technique
of three-phase to five-phase matrix converter supplying series-connected five-phase AC machine drive system. The developed
modulation technique is based on the comparison of high frequency carrier signal with the input voltages, similar to the one used
in voltage source inverter. Although carrier-based scheme is widely employed for control of voltage source inverter, it is very
recently being used for a matrix converter. The similar concept is extended in this paper for controlling a three-phase to five-phase
matrix converter. The major aim of the PWM is to generate two independent fundamental frequency components that will
subsequently control two series/parallel connecetd five-phase motor. The viability of the proposed control techniques is proved
using analytically and simulation approach.

Keywords: Carrier-based PWM, Matrix converter, modulation, five-phase, AC machine, drive


1. Introduction

Controlled bidirectional power flow using direct ac-ac conversion using semiconductor switches arranged in the form of matrix
array, popularly known as Matrix converter, has recently attracted much attention worldwide. Matrix converter is becoming seri-
ous contender to its counterpart voltage source inverter due to some inherent attractive features such as operation at unity power
factor for any load, controlled bidirectional power flow, sinusoidal input and output currents etc. A comprehensive overview of the
development in the field of matrix converter research is presented in (Wheeler et al, 2002). It is to be noted here that the most
common configuration of the matrix converter discussed in the literature are three-phase to three-phase (Kazmierkowaski et al,
2002, Toliyat et al, 2004). Little attention has been paid on the development of matrix converter with output more than three.
The conventional structure for variable-speed drives consists of a three-phase motor supplied by a three-phase power electronic
converter. However, when the machine is connected to a modular power electronic converter, such as a voltage source inverter or a
matrix converter then the need for a specific number of phases, such as three, disappears. This is due to the fact that by simply add-
ing one leg increases the number of output phases. Nowadays, the development of modern power electronics, makes it possible to
consider the number of phases a degree of freedom, i.e., an additional design variable. Multi-phase (more than three) motor drives
have some inherent advantages over the traditional three-phase motor drives such as reducing the amplitude and increasing the fre-
quency of torque pulsations, reducing the rotor harmonic currents losses and lowering the dc link current harmonics. In addition,
Saleh et al. / International Journal of Engineering, Science and Technology, Vol. 2, No. 10, 2010, pp. 67-78

68
owing to their redundant structure, multi-phase motor drives improve the system reliability Detail reviews on the development in
the area of multi-phase (more than three-phase) drive are presented in (Singh, 2002, Jones et al, 2002, Bojoi et al, 2006, Levi et al,
2007, Levi, 2008a, Levi, 2008b).
A novel concept for multi-motor drive systems, based on utilization of multi-phase machines and multi-phase inverters, have
been proposed recently (Levi et al, 2003a, Levi et al, 2004a, Levi et al, 2004b, Levi et al, 2003b, Iqbal et al, 2005, Yukosovic et
al, 2008, Jones et al, 2006, Jones, et al, 2009). Since vector control of any multi-phase machine requires only two stator current
components, the additional stator current components are used to control other machines. It has been shown that, by connecting
multi-phase stator windings in series (Levi et al, 2003a, Levi et al, 2004a, Levi et al, 2004b, Levi et al, 2003b, Iqbal et al, 2005)
and in parallel (Yukosovic et al, 2008, Jones, et al,2006, Jones , et al, 2009) with an appropriate phase transposition, it is possible
to control independently all the machines with supply coming from a single multi-phase voltage source inverter. One specific drive
system, covered by this general concept, is the five-phase series/parallel-connected two-motor drive, consisting of two five-phase
machines and supplied from a single five-phase voltage source inverter. Such topology has been analysed in a considerable depth
in (Levi et al, 2003b, Iqbal et al, 2005) for series connection and (Yukosovic et al, 2008, Jones et al, 2006) for parallel connection.
The multi-motor drive system discussed so far in the literature uses multi-phase voltage source inverter as their supply. In contrast
this paper proposes a multi-phase matrix converter to supply such drive topology.
This paper focuses on the development of novel topology of matrix converter which is a single stage power converter to produce
more than three phases. The performance of power electronic converters (ac to ac or ac-dc-ac) is highly dependent on their control
algorithms. Thus a number of modulation schemes are developed for voltage source inverters for three-phase output (Holtz, 1992,
Holmes, et al, 2003) and multi-phase output (Dujic, et al, 2009a, Dujic, et al, 2009b). Modulation methods of matrix converters
are complex and are generally classified in two different groups, called direct and indirect. The direct PWM method developed by
Alesina and Venturini (Alesina, et al, 1981) limits the output to half the input voltage. This limit was subsequently raised to 0.866
by taking advantage of third harmonic injection (Alesina, et al, 1989) and it was realized that this is maximum output that can be
obtained from a three-to-three phase matrix converter in the linear modulation region. Indirect method assumes a matrix converter
as a cascaded virtual three-phase rectifier and a virtual voltage source inverter with imaginary dc link. With this representation,
space vector PWM method of VSI is extended to a matrix converter (Casadei, et al, 1993, Huber, et al, 1995). Although the space
vector PWM method is suited to three-phase system but the complexity of implementation increases with the increase in the num-
ber of switches/phases. Motivated from the simple implementation, carrier-based PWM scheme is introduced recently for three to
three phase matrix converter (Wang, et al, 2006, Young, et al, 2006, Poh, et al, 2009).
In this paper, a carrier based PWM strategy is presented based on the comparison of the modulating signals (five-phase target
output voltages) with the high frequency triangular carrier wave for a three to five-phase matrix converter. The output voltage is
limited to 0.75 of the input voltage magnitude if common-mode voltage is not injected. Another scheme is suggested in the paper
utilising the injection of common mode voltage in the output five-phase target voltage. This results in enhanced output voltage
equal to 0.7886 of the input magnitude. Theoretically this is the maximum output magnitude that can be obtained in this matrix
converter configuration in the linear modulation region. Analytical approach is used to develop and analyse the proposed modula-
tion techniques and are further supported by simulation results. The major aim of the paper is to produce two fundamental fre-
quency output from the matrix converter that can be used to control two series/parallel connected five-phase machines.

2. Five-phase Two-motor Drive system

In five-phase system two set of orthogonal voltage/current components are produced namely d-q and x-y. In single-motor drive
system, only d-q components are utilised and the x-y components are free to flow creating losses. Thus concept of two-motor five-
phase drive system is developed where both these components are utilised, d-q by one machine and x-y by other machine. The ex-
tra set of current components (x-y) available in a five-phase system is effectively utilised in independently controlling an additional
five-phase machine when the stator windings of two five-phase machines are connected in series (Fig. 1a)/parallel (Fig. 1b) and
are supplied from a single five-phase VSI. In Fig. 1a, reference currents generated by two independent vector controllers, are
summed up as per the transposition rules and are supplied to the series-connected five-phase machines. As such the two five-phase
machines are supplied from one five-phase inverter but are controlled independently. More detail on this configuration of the drive
system is available in (Levi et al, 2003a, Levi et al, 2004a, Levi et al, 2004b, Levi et al, 2003b, Iqbal et al, 2005). The voltage and
current relationship of this drive topology is given as;

2 1
2 1
2 1
2 1
2 1
d e
s
E
b d
s
D
e c
s
C
c b
s
B
a a
s
A
v v v
v v v
v v v
v v v
v v v
+ =
+ =
+ =
+ =
+ =

2 1
2 1
2 1
2 1
2 1
d e
s
E
b d
s
D
e c
s
C
c b
s
B
a a
s
A
i i i
i i i
i i i
i i i
i i i
= =
= =
= =
= =
= =
(1)
Saleh et al. / International Journal of Engineering, Science and Technology, Vol. 2, No. 10, 2010, pp. 67-78

69
Here, s stand for source quantity, capital letters denote source output and small letters represents motor side quantities.
Another analogues stator winding connection of five-phase two-motor drive is parallel-connected drive (Fig.1b) which is directly
derived from the analogy of series and parallel circuits. The source (three to five-phase matrix converter) supplied both the ma-
chines and these machine are connected in parallel with appropriate phase transposition (Vukosovic et al, 2008, Jones, et al, 2006,
Jones , et al, 2009). The control decoupling is possible due to decoupling of the - and x-y components. The d-q components of
one machine become the x-y to the other and vice-versa. Here the vector controller produced voltage references in contrary to se-
ries-connected drive where current references are generated from the vector controller. Once again independent control is achieved
of the two five-phase motors. The voltage and current relations for this drive topology are given as;
2 1
2
1
2 1
2
1
2 1
d e
s
E
b
d
s
D
e c
s
C
c
b
s
B
a a
s
A
v v v
v v v
v v v
v v v
v v v
= =
= =
= =
= =
= =

2 1
2
1
2 1
2
1
2 1
d e
s
E
b
d
s
D
e c
s
C
c
b
s
B
a a
s
A
i i i
i i i
i i i
i i i
i i i
+ =
+ =
+ =
+ =
+ =
(2)

Fig.1a Five-phase series-connected two-motor drive structure

Fig. 1b. Five-phase parallel-connected two-motor drive structure
If current control in the rotating reference frame is to be utilized, appropriate PWM scheme for five-phase VSI needs to be devel-
oped to generate voltage references instead of current references for both series and parallel connected drives. The PWM technique
developed and reported in the paper thus produces two-frequency voltage reference output.

3. Three-to-Five Matrix Converter
The power circuit topology of a three-phase to five-phase matrix converter is illustrated in Fig. 2. There are five legs with each leg
having three bidirectional power switches connected in series. Each power switch is bidirectional in nature with anti-parallel con-
nected IGBTs and diodes. The input is similar to a three-phase to three-phase matrix converter having LC filters and the output is
five-phases with 72 phase displacement between each phases.
Saleh et al. / International Journal of Engineering, Science and Technology, Vol. 2, No. 10, 2010, pp. 67-78

70
sa
i
sb
i
sc
i
a
i
b
i
c
i
a
v
b
v
c
v
11
S
21
S
31
S
12
S
22
S
32
S
13
S
23
S
33
S
14
S
24
S
34
S
15
S
25
S
35
S
A
i
B
i
C
i
D
i
E
i

Fig. 2. Power Circuit topology of Three-phase to five-phase matrix converter

The load to the matrix converter is assumed as star-connected five-phase ac machine. Switching function is defined as
S
jk
= {1 for closed switch, 0 for open switch}, j = {a,b,c} (input), k= {A,B,C,D,E} (output). The switching constraint is S
ak
+ S
bk
+
S
ck
= 1.
The modulation technique is developed by assuming input side as three-phase controlled rectifier and the output is a five-phase
voltage source inverter with a fictitious dc link.

4. Carrier-Based Pulse Width Modulation Technique for Two-motor Drives

Carrier-based PWM scheme developed in this section follows the similar concept presented in (Jones , et al, 2009). Since the in-
put side is three-phase, the analytical treatment remains the same as that of (Jones , et al, 2009). However, the output is now in-
creased to five and hence the analysis will be modified to suit the requisite output phase number. A balanced three-phase system is
assumed at the input and unbalance study will be reported separately.
) 3 / 4 cos(
) 3 / 2 cos(
) cos(

=
=
=
t V v
t V v
t V v
b
b
a

(4)
Since the matrix converter outputs voltages with frequency decoupled from the input voltages, the duty ratios of the switches are
to be calculated accordingly. The five-phase output voltage duty ratios should be calculated in such a way that output voltages re-
mains independent of input frequency. In other words, the five-phase output voltages can be considered in synchronous reference
frame and the three-phase input voltages can be considered to be in stationary reference frame, so that the input frequency term
will be absent in output voltages. Considering the above, duty ratios of output phase j is chosen as

) 3 / 4 cos(
), 3 / 2 cos(
), cos(



=
=
=
t k
t k
t k
j cj
j bj
j aj

(5)
Where is the phase shift at the input side. The input and output voltages are related as:

c
b
a
cE bE aE
cD bD aD
cC bC aC
cB bB aB
cA bA aA
E
D
C
B
A
V
V
V
V
V
V
V
V





(6A)

Therefore the phase A output voltage can be obtained by using the above duty ratios as

)] 3 / 4 cos( ) 3 / 4 cos( ) 3 / 2 cos( ) 3 / 2 cos( ) cos( ) [cos( + + = t t t t t t V k V
A A

(6B)
Saleh et al. / International Journal of Engineering, Science and Technology, Vol. 2, No. 10, 2010, pp. 67-78

71
In general equation (6) can be written as
) cos( V k V
A A

2
3
=

(7)
In eq (7), ) cos( term indicates that the output voltage is affected by . The term k
A
is defined in equation (10). Thus, the out-
put voltage
A
V is independent of the input frequency and only depends on the amplitude V of the input voltage and
A
k is a ref-
erence output voltage time-varying modulating signal for the output phase A with the desired output frequency
2 1 o o
+ ,
1 o
is
the operating frequency of machine-1 or the first fundamental output frequency and
2 o
is the operating frequency of machine-2
or the second fundamental output frequency. The fundamental output voltage magnitude corresponding to
1 o
is given as m
1
and
corresponding to
2 o
is given as m
2
.

The five-phase reference output voltages can then be represented as
) 5 / 8 cos(
) 5 / 6 cos(
) 5 / 4 cos(
) 5 / 2 cos(
), cos(
1 1 1
1 1 1
1 1 1
1 1 1
1 1 1
=
=
=
=
=
t m k
t m k
t m k
t m k
t m k
o E
o D
o C
o B
o A

(8)


) 5 / 8 cos(
) 5 / 6 cos(
) 5 / 4 cos(
) 5 / 2 cos(
) cos(
2 2 2
2 2 2
2 2 2
2 2 2
2 2 2
=
=
=
=
=
t m k
t m k
t m k
t m k
t m k
o E
o D
o C
o B
o A
(9)


2 1
2 1
2 1
2 1
2 1
D E E
B D D
E C C
C B B
A A A
k k k
k k k
k k k
k k k
k k k
+ =
+ =
+ =
+ =
+ =
(10)
Therefore, from (7), the output voltages are obtained as;
)
5
6 cos( ) cos(
2
3
)
5
8 cos( ) cos(
2
3
)
5
2 cos( ) cos(
2
3
)
5
6 cos( ) cos(
2
3
)
5
8 cos( ) cos(
2
3
)
5
4 cos( ) cos(
2
3
)
5
4 cos( ) cos(
2
3
)
5
2 cos( ) cos(
2
3
) cos( ) cos(
2
3
) cos( ) cos(
2
3
2 2 1 1
2 2 1 1
2 2 1 1
2 2 1 1
2 2 1 1

=
t V k t V k V
t V k t V k V
t V k t V k V
t V k t V k V
t V k t V k V
o D o E E
o D o D D
o E o C C
o C o B B
o A o A A

(11)
IV A Application of Offset Duty Ratio

In the above discussion, duty-ratios become negative (see eq. (5)) which are not practically realizable. For the switches con-
nected to output phase-A, at any instant, the condition 1 d , d , d 0
cA bA aA
should be valid. Therefore, offset duty ratios should to
be added to the existing duty-ratios, so that the net resultant duty-ratios of individual switches are always positive. Furthermore,
Saleh et al. / International Journal of Engineering, Science and Technology, Vol. 2, No. 10, 2010, pp. 67-78

72
the offset duty-ratios should be added equally to all the output phases to ensure that the effect of resultant output voltage vector
produced by the offset duty-ratios is null in the load. That is, the offset duty-ratios can only add the common-mode voltages in the
output. Considering the case of output phase-A;

0 ) 3 / 4 cos( ) 3 / 2 cos( ) cos( = + + = + + t k t k t k
A A A cA bA aA

(12)
Absolute values of the duty-ratios are added to cancel the negative components from individual duty ratios. Thus the minimum
individual offset duty ratios should be

) / t cos( k ) t ( D
and
) / t cos( k ) t ( D
) t cos( k ) t ( D
A cA c
A bA b
A aA a



= =
= =
= =
3 4
3 2

(13)

The effective duty ratios are
) t ( D
), t ( D
), t ( D
c cA
'
cA
b bA
'
bA
a aA
'
aA
+ =
+ =
+ =



(14)

Other output phases can be written in the similar fashion. The net duty ratio 1 0
'
ak
should be within the range of 0 to 1.
For the worst case
(15)

The maximum value of
A
k is equal to 0.5. In any switching cycle the output phase should not be open circuited. Thus the sum of
the duty ratios in (13) must equal unity. But the summation ) t ( D ) t ( D ) t ( D
c b a
+ + is less than or equal to unity. Hence another
offset duty-ratio { } [ ] 3 1 / ) t ( D ) t ( D ) t ( D
c b a
+ + is added to ) ( ) ( ), ( t D and t D t D
c b a
in (13). The addition of this offset duty-ratio
in all switches will maintain the output voltages and input currents unaffected. Similarly, the duty-ratios are calculated for the other
five output phases.
If
E D C B A
k k k k k , , , , are chosen to be 5-phase sinusoidal references as given in equation (9), the input voltage capability is not
fully utilized for output voltage generation and the output magnitude remains only 50% of the input magnitude. To overcome this,
an additional common mode term equal to ] 2 / )} , , , , min( ) , , , , {max(
E D C B A E D C B A
k k k k k k k k k k + is added as in the carrier-based
PWM principle as implemented in two-level inverters. Thus the amplitude of ) , , , , (
E D C B A
k k k k k can be enhanced from 0.5 to
0.5257.

IV B Without Common-mode voltage addition

In the above section, two offsets are added to the original duty ratios to form the following effective duty ratio for output phase
p;
) 3 / 4 cos( 3 / )}) ( ) ( ) ( { 1 ( ) (
) 3 / 2 cos( 3 / )}) ( ) ( ) ( { 1 ( ) (
) cos( 3 / )}) ( ) ( ) ( { 1 ( ) (
+ + + + =
+ + + + =
+ + + + =
t k t D t D t D t D
t k t D t D t D t D
t k t D t D t D t D
p c b a c cp
p c b a b bp
p c b a a ap

(16)

Where F , E , D , C , B , A p
These duty ratios (equation (16)) can be compared to the triangular carrier wave to generate the gating signals for the bidirec-
tional power switches. The output phase voltage magnitude will reach 75% of the input voltage magnitude with this method. To
further enhance the output voltage magnitude, common mode voltage of the output reference signals are added to formulate the
new duty ratios as discussed in the next section.



1 2 0
A
k .
Saleh et al. / International Journal of Engineering, Science and Technology, Vol. 2, No. 10, 2010, pp. 67-78

73
IV C With Common mode voltage Addition

The duty ratios can further be modified by injection common mode voltage of the output voltage references to improve the output
voltage magnitude. The output voltage magnitude increases and reaches its limiting value of 78.8% of the input magnitude. The
common mode voltage that is added to obtain new duty ratios are;
2
Min Max
cm
V V
V

= (17)
Where
{ }
{ }
F E D C B A min
F E D C B A max
k , k , k , k , k , k min V
k , k , k , k , k , k max V
=
=
(18)
The duty ratio for output phase p can be written as;
) 3 / 4 cos( ] [ 3 / )}) ( ) ( ) ( { 1 ( ) (
) 3 / 2 cos( ] [ 3 / )}) ( ) ( ) ( { 1 ( ) (
) cos( ] [ 3 / )}) ( ) ( ) ( { 1 ( ) (
+ + + + + =
+ + + + + =
+ + + + + =
t V k t D t D t D t D d
t V k t D t D t D t D d
t V k t D t D t D t D d
cm p c b a c cp
cm p c b a b bp
cm p c b a a ap

(19)
Where F , E , D , C , B , A p
The five-phase output voltages can be written as;
) 5 / 2 cos( ) 5 / 8 cos(
) 5 / 6 cos( ) 5 / 6 cos(
) 5 / 8 cos( ) 5 / 4 cos(
) 5 / 4 cos( ) 5 / 2 cos(
) cos( ) cos(
2 2 1 1
2 2 1 1
2 2 1 1
2 2 1 1
2 2 1 1





+ =
+ =
+ =
+ =
+ =
t m t m k
t m t m k
t m t m k
t m t m k
t m t m k
o o E
o o D
o o C
o o B
o o A
(20)
Where is input frequency and
1 o
and
2 o
are the output frequencies of machine-1 and machine-2 respectively, m
1
and m
2

are the modulation indices for machine-1 and machine 2, respectively. For unity power factor has to chosen zero. The modulat-
ing signals are shown in Fig. 4, after adding the output common mode voltages. These duty ratios (equation (19)) are then com-
pared with the high frequency triangular carrier signals to generate the gating signals as illustrated in Fig. 3, for phase A. Similarly
fifteen more duty ratios will be compared with the triangular carrier to generate overall gating signals.
A complete block schematic of the PWM signal generation is presented in Fig. 4. The reference voltages for two machines with
the desired speeds and appropriate voltage magnitudes are generated. These references are then summed according to the phase
transposition rule. The overall modulating signal thus generated is given to the PWM block. This PWM block then generate ap-
propriate gate signals for the matrix converter. The matrix converter then produces appropriate voltages which drive the two se-
ries/parallel connected machines.

0.0178 0.018 0.0182 0.0184
0
1
0
1
0
1
0
1
Time (s)
d
aA
d
bA
d
cA
S
11
S
21
S
31
Carrier Signal

Fig. 3. Gate signal generation for output phase A.
Saleh et al. / International Journal of Engineering, Science and Technology, Vol. 2, No. 10, 2010, pp. 67-78

74

Fig. 4. Block diagram of Carrier-based PWM for two frequency output
5. Simulation Results

Matalab/Simulink model is developed for the proposed matrix converter control. The input voltage is fixed at 100 V to show the
exact gain at the output side. The switching frequency of the devices is kept at 6 kHz. The purpose here is to show two fundamen-
tal components of current produced by the matrix converter. These voltage components are independent from each other and thus
can independently control the two machines. The results shown here is only limited to the production of the appropriate voltage
components. The motor behaviour is not discussed in this paper and will be reported separately. It is assumed that one voltage
component has frequency of 30 Hz and second voltage component has 60 Hz. To respect the v/f=constant control the voltage mag-
nitude of the lower frequency component is half compared to the higher frequency component. For the simulation purpose a R-L
load is connected with R = 10 and L = 10 mH. Simulation results are shown for the modulation with common mode voltage ad-
dition in the output target voltage. Thus the maximum output of the matrix converter is limited to 78.8 V as the input is 100 V. The
results without common mode voltage addition will remain the same except with the lower output magnitude.
The resulting waveforms are presented in Fig. 5a,b to 7. The input source side and converter side waveforms are presented in
Fig. 5a,b. The results clearly shows unity power factor at the input side. The converter side current shows PWM signal and the
spectrum is clearly sinusoidal with no lower order harmonics.



0 0.01 0.02 0.03 0.04 0.05 0.06 0.07 0.08
-100
-80
-60
-40
-20
0
20
40
60
80
100
Time [s]
S
o
u
r
c
e

s
i
d
e

v
o
l
t
a
g
e

a
n
d

c
u
r
r
e
n
t

[
V
,
A
]

Fig. 5a. Input side waveforms of 3 to 5-phase Matrix Converter: upper trace
Saleh et al. / International Journal of Engineering, Science and Technology, Vol. 2, No. 10, 2010, pp. 67-78

75
0 0.01 0.02 0.03 0.04 0.05 0.06 0.07 0.08
-10
0
10
Selected signal: 4 cycles. FFT window (in red): 2 cycles
Time [s]
C
o
n
v
e
r
t
e
r

s
id
e

c
u
r
r
e
n
t

[
A
]
0 100 200 300 400 500
0
20
40
60
80
100
Frequency [Hz]
Fundamental (50Hz) = 5.782 , THD= 81.68%
C
o
n
v
e
r
t
e
r

s
i
d
e

c
u
r
r
e
n
t

[
%

o
f

F
u
n
d
.
]

Fig. 5b. Input voltage and current, bottom trace, Spectrum Input current.

The output side filtered voltages are given in Fig. 6 and the spectrum of the output PWM signal voltages are presented in Fig.
7a,b,c. The output filtered phase voltages shows superimposed fundamental and second harmonic components. The spectrum of
phase A voltage and the transformed voltage are given in Fig. 7a,b,c. It is clearly evident that the phase A voltage contains two
fundamental components at 30 Hz and 60 Hz. These voltages are then decoupled and appear in - plane (60 Hz) and x-y plane
(30 Hz). Thus is aim of the control is achieved. Also the magnitude of the two voltages follows v/f=constant rule.
0.01 0.02 0.03 0.04 0.05 0.06 0.07 0.08
-80
-60
-40
-20
0
20
40
60
80
Time [s]
O
u
t
p
u
t

f
i
l
t
e
r
e
d

p
h
a
s
e

v
o
l
t
a
g
e
s

[
V
,
A
]

Fig. 6. Output filtered five-phase voltages
0 0.02 0.04 0.06 0.08
-100
0
100
Selected signal: 2.4 cycles. FFT window (in red): 2 cycles
Time [s]
O
u
t
p
u
t

p
h
a
s
e

v
o
l
t
a
g
e

[
V
]
0 100 200 300 400 500
0
50
100
150
200
Frequency [Hz]
Fundamental (30Hz) = 25.87
O
u
t
p
u
t

p
h
a
s
e

v
o
l
t
a
g
e


[
%

o
f

F
u
n
d
.
]

Fig. 7a. Spectrum of output voltages; phase A
0 0.02 0.04 0.06 0.08
-100
0
100
Selected signal: 4.8 cycles. FFT window (in red): 2 cycles
Time [s]
V

[
V
]
0 100 200 300 400 500
0
20
40
60
80
100
Frequency [Hz]
Fundamental (60Hz) = 51.33
V

[
%

o
f

F
u
n
d
.
]

Fig. 7b. Spectrum of output voltages; -axis voltage
Saleh et al. / International Journal of Engineering, Science and Technology, Vol. 2, No. 10, 2010, pp. 67-78

76
0 0.02 0.04 0.06 0.08
-100
0
100
Selected signal: 2.4 cycles. FFT window (in red): 2 cycles
Time [s]
V
x

[
V
]
0 100 200 300 400 500
0
20
40
60
80
100
Frequency [Hz]
Fundamental (30Hz) = 25.86
V
x

[
%

o
f

F
u
n
d
.
]

c.
Fig. 7c. Spectrum of output voltages; x-axis voltage.

6. Conclusion

The paper presents a novel solution for supplying five-phase series/parallel connected two-motor drive system using a three-
phase input to five-phase output matrix converter The input to the matrix converter is three-phase ac supply and the output is five-
phase variable voltage and variable frequency. It possesses all the advantages offered by a conventional matrix converter, however,
the output voltage magnitude is found to be limited to 78.86% of the input voltage magnitude in the linear modulation region. The
proposed PWM strategy is derived from the analogy of the modulation of a voltage source inverter. The analytical findings are
confirmed using simulation approach. The decoupled voltage components are realized. The proposed structure will be further stud-
ied for feeding a five-phase two-motor drive and the drive behaviour will be investigated.

Nomenclature:
1 1 k k
i , v Stator voltages and currents, respectively for machine 1
2 2 k k
i , v Stator voltages and currents, respectively for machine 2
aj
Duty ratio of input phase a and output phase j
bj
Duty ratio of input phase b and output phase j
cj
Duty ratio of input phase c and output phase j
max
V Maximum of output/input voltages
min
V Minimum of output/input voltages
cm
V Common mode voltage
1
m Modulation index for machine 1
2
m Modulation index for machine 2
j
k Output voltage references


References

Alesina, A. and Venturini, M, 1981. Solid state power conversion: A Fourier analysis approach to generalised transformer synthe-
sis. IEEE Trans. Circuit System, Vol. 28, No. CS-4, pp. 319-330, April.
Alesina, A. and Venturini, M, 1989. Analysis and design of optimum amplitude nine-switch direct ac-ac converters. IEEE Trans.
Power Elect. Vol. PE-4, No. 1, pp. 101-112.
Bojoi, R., Farina, F., Profumo, F., and Tenconi, 2006. Dual three induction machine drives control-A survey. IEEJ Tran. On Ind.
Appl.,Vol. 126, No. 4, pp. 420-429.
Casadei, D, Grandi, G., Serra, G. and Tani, A, 2009. Space vector control of matrix converters with unity power factor and sinu-
soidal input/output waveforms. Proc. EPE Conf., Vol. 7, pp. 170-175, 1993.
Saleh et al. / International Journal of Engineering, Science and Technology, Vol. 2, No. 10, 2010, pp. 67-78

77
Dujic, D., Jones, M., Levi, E., 2009a. Analysis of output current ripple rms in multi-phase drives using space vector approach.
IEEE Trans. On Power Elect. Vol. 24, No. 8, pp. 1926-1938, Aug.
Dujic, D., Jones, M., and Levi, E., 2009b, Generalised space vector PWM for sinusoidal output voltage generation with multi-
phase voltage source inverts, Int. Journal of Ind. Elect. and Drives, Vol. 1, No.1, pp. 1-13.
Holmes, D.G., and Lipo, T.A., 2003, Pulse width modulation for power converters: Principle and Practice, Piscataway, NJ: IEEE
Press.
Holtz, J., 1992. Pulsewidth modulation-A survey, IEEE Trans. On Ind. Elect., Vol. 39, No. 5, pp. 410-420, Dec..
Huber, L., and Borojevic, D, 1995, Space vector modulated three-phase to three-phase matrix converter with input power factor
correction, IEEE Trans. Ind. Appl. Vol. 31, No. 6, pp. 1234-1246, Nov./Dec..
Iqbal, A., Vukosavic, S.N, Levi, E., Jones, M., Toliyat, H.A. 2005. Dynamics of a series-connected two-motor five-phase drive
system with a single-inverter supply, IEEE Ind. Appl. Society Annual Meeting IAS, Hong Kong, pp. 1081-1088.
Jones, M., and Levi, E, 2002, A literature survey of state-of-the-art in multiphase ac drives, Proc. 37
th
Int. Universities Power
Eng. Conf. UPEC, Stafford, UK, pp. 505-510.
Jones, M., Levi, E., and Vukosavic, S.N., 2006. Independent control of two five-phase induction machines connected in parallel to
a single inverter supply. Proc. IEEE IECON, pp. 1257-1262.
Jones, M., Vukosavic S.N., and Levi, E., 2009. Parallel-connected multiphase multi-drive systems with single inverter supply.
IEEE Trans. On Ind. Elect., Vol. 56, No. 6, pp. 2047-2057.
Kazmierkowaski, M.P., Krishnan, R., and Blaabjerg, F, 2002. Control in Power Electronics-Selected problems. Academic Press,
USA.
Levi, E., Bojoi,R., Profumo, F., Toliyat, H.A, and Williamson, S. 2007. Multi-phase induction motor drives-A technology status
review. IET Elect. Power Appl. Vol. 1, No. 4, pp. 489-516, July.
Levi, E. 2008a Guest editorial, IEEE Trans. Ind. Electronics, Vol.55, No. 5, May 2008, pp. 1891-1892.
Levi, E. 2008b. Multi-phase Machines for variable speed applications. IEEE Trans. On Ind. Elect. Vol. 55, No. 5, May, pp. 1893-
1909.
Levi, E., Iqbal, A., Vukosavic, S.N., and Toliyat, H.A., 2003. Modelling and control of a five-phase series-connected two-motor
drive. Proc. IEEE Ind. Elec. Society, Annual Meeting IECON, Roanoke, Virginia, pp. 208-213.
Levi, E., Jones, M., and Vukosavic, S.N. 2003. Even-phase multi-motor vector controlled drive with single inverter supply and se-
ries connection of stator windings. IEE Proc. Electric Power Applications, Vol. 150, No. 5, pp. 580-590.
Levi, E., Jones, M., Vukosavic, S.N., Toliyat, H.A., 2004a. A novel concept of a multi-phase, multi-motor vector controlled drive
system supplied from a single voltage source inverter. IEEE Trans. on Power Electronics, Vol. 19, No. 2, pp. 320-335.
Levi, E., Jones, M., Vukosavic, S.N., Toliyat, H.A., 2004b. A five-phase two-machine vector controlled induction motor drive
supplied from a single inverter. EPE Journal, Vol. 14, No. 3, pp. 38-48.
Poh C., Loh, R. Rong, F. Blaabjerg and P. Wang, 2009. Digital carrier modulation and sampling issues of matrix converter. IEEE
Trans. On Power Elect., Vol. 24, No. 7, July.
Toliyat, H.A., and Campbell, S, 2004. : DSP based Electromechanical Motion Control, CRC Press, USA.
Vukosavic, S.N., Jones, M., Levi, E., and Dujic, D., 2008. Experimental performance evaluation of a five-phase parallel-connected
two-motor drive. Prof. Int. conf. on Power Electronics, machines and drives, PEMD-2008, pp. 686-690.
Wheeler, P.W., Rodriguez, J., Clare, J,C., Empringham, L., and Weinstein,A., 2002, Matrix converters: a technology review.
IEEE Trans. On Ind. Elect. Vol. 49, No. 2, April, pp. 276-288.
Wang, B., and Venkataramanan,G. 2006. A carrier-based PWM algorithm for indirect matrix converters. in Proc. IEEE-PESC
2006, pp. 27802787.
Young-Doo Yoon and Seung-Ki Sul. 2006. Carrier-based modulation technique for matrix converter. IEEE Trans. Power Elect.,
Vol. 21, No. 6, pp. 16911703, November.


Biographical notes

Mohammad Saleh is Master in Telecommunication and Post Graduate diploma in Education and Training from Victoria University. He has completed his Bache-
lor of Electrical Engineering from University of North Carolina at Charlotte, USA in 1989. He is presently employed at Qatar Petroleum, Qatar and also registered
at Victoria University, Australia for his PhD degree.

Atif Iqbal Atif Iqbal received his B.Sc. and M.Sc. Engineering (Electrical) degrees in 1991 and 1996, respectively, from the Aligarh Muslim University, Aligarh,
India and PhD in 2006 from Liverpool John Moores University, UK. He has been employed as Lecturer in the Department of Electrical Engineering, Aligarh Mus-
lim University, Aligarh since 1991 and is working as Associate Professor in the same university. He is presently with Texas A&M University at Qatar on research
assignment. He is recipient of Maulana Tufail Ahmad Gold Medal for standing first at B.Sc. Engg. Exams in 1991 at AMU, and EPSRC, Govt. Of UK, fellowship
from 2002-2005 for pursuing PhD studies. His principal research interest is Modelling and Control of Power Electronics Converters & Drives.

SK. Moin Ahmed born in 1983 at Hoogly, West Bengal, India. He received his B.Tech (Electrical) and M.Tech. (Power System & drives) in 2006 & 2008, respec-
tively, from Department of Electrical Engineering, Aligarh Muslim University, Aligarh. He is gold medalist for securing top position in M.Tech. He is a recipient
of Torento fellowship funded by AMU. Currently he is pursuing his PhD at AMU and employed as Senior research fellow in a CSIR, New Delhi, funded project.

Saleh et al. / International Journal of Engineering, Science and Technology, Vol. 2, No. 10, 2010, pp. 67-78

78
Akhtar Kalam - Professor Akhtar Kalam has been at Victoria University, Melbourne since 1985 and a former Deputy Dean of the Faculty of Health, Engineering
and Science for 7 years. He has wide experience in educational institutions and industry across four continents. He received his B.Sc. and B.Sc. Engineering from
Calcutta University and Aligarh Muslim University, India in 1969 and 1973 respectively. He completed his MS and Ph.D. at the University of Oklahoma, USA and
the University of Bath, UK in 1975 and 1981 respectively. He has held teaching appointments at the University of Technology, Baghdad, Iraq and Capricornia In-
stitute of Advanced Education, Rockhampton, Queensland. He is regularly invited to deliver lectures, work on industrial projects and examine external thesis over-
seas. His major areas of interests are power system analysis, communication, control, protection and cogeneration systems. He has been actively engaged in the
teaching of Energy Systems to undergraduates, postgraduates and providing professional courses to the industry both in Australia and overseas. He regularly offers
professional development courses on Power System Protection, Renewable Energy and Cogeneration & Gas Turbine Operation to the Energy Supply Association
of Australia (ESAA) and Australian Power Institute (API). He also runs postgraduate distance education programme on Power System Protection for the ESAA. He
has conducted research, provided industrial consultancy and published over three hundred and fifty seven publications on his area of expertise and written over 29
books in the area. Professor Kalam is a Fellow of EA, IET, AIE and a member of IEEE.

Haitham Abu-Rub has done his PhD from Gdansk University, Poland in 1995 and MSc in Electrical Engg From Gdynia Maritime University. He is presently
serving as Visiting Senior Associate Professor in the Deptt. Of Electrical & Computer Engg., Texas A&M University at Qatar. His area of interest are Field ori-
ented control of electrical machines, nonlinear control of electrical drive, sensorless control, parameters estimation, power electronics, and electrical machines. Has
earned many international prestigious awards like the American Fulbright Scholarship (Texas A&M University), the German Alexander von Humboldt Fellowship
(Wuppertal university), the German DAAD Scholarship (Bochum University), and the British Royal Society Scholarship (Southampton University). Also received
the "Best Presentation Award" at the 27th Annual Conference of the IEEE Industrial Electronics Society in Denver (IECON'01).


Received September 2010
Accepted November 2010
Final acceptance in revised form December 2010