Sparse Matrix Converter: Modeling and PWM

Control
L. Rmili, S. Rahmani K. Al-Haddad, Fellow, IEEE
Laboratory of Biophysics and Medical Technology (BMT), Canada Research Chair in Energy Conversion and Power
ISTMT of the University of Tunis El-Manar, Tunisia Electronics CRC-ECPE
Av. Dr. Zouhaier Essafi, 1006 Tunisia École de Technologie Supérieure, 1100 Notre-Dame,
[email protected], [email protected] Montréal, Québec H3C 1K3, Canada
Phone: (514) 396 8874, Fax: (514) 396 8684
[email protected]

Abstract—The matrix converter (MC) is an alternative ac-ac necessary to improve the displacement factor, the quality of the input
power converter by connecting directly input to output phases current signal, the output voltage signal and the ratio voltage [15]-
through bidirectional switches and without using any dc-link or [20]. A state model of the conversion system is established in fixed
energy storing element, therefore is called an all-silicon reference frame and in a rotating d-q reference frame in order to ease
converter. Two topologies of matrix converter are established and limited control lows to direct and quadrature component. The
such as direct topology and the indirect topology. This paper is study of the matrix converter developed in section IV is mainly
devoting to presents the topology of the indirect sparse matrix focused on modulation scheme of the pulse width modulation
converter (ISMC). So, the three phases to three phase sparse technique [21]-[27]. Finally, simulation results are shows in figs. 6-18
matrix converter (SMC) topology fed an RL load is modulated presenting the current input and output signals and the output
using the Pulse Width Modulation technique (PWM). Simulation voltages of the converter with spectra identifying its THD, this is
results are shown to prove the usefulness of the control scheme. before and after the integration of the designated filters and with
different load frequencies such as 25 Hz, 50 Hz and 100 Hz.
Keywords—Sparse matrix converter; modelling; pulse width
modulation;, total harmonic distortion. II. INDIRECT SPARSE MATRIX CONVERTER TOPOLOGY
The indirect matrix converter is an alternative of conventional
I. INTRODUCTION converters such as cyclo-converter, dimmer and conventional
The matrix converter is a bidirectional ac-ac power converter converter (rectifier-capacitor-inverter). IMC is used for directly
without dc-link capacitor. The main advantages of MC are the power conversion ac/ac, instead of the ac/dc/ac conversion. Four
sinusoidal input and output waveforms, variable output voltages three phase ac/ac converters are an equivalent structure combining a
amplitude and frequency, operation with unity power factor, four current source rectifier and a voltage source inverter connected by
quadrant operations and the absence of the energy storing element. virtual dc-link as shown in Fig.1. The rectifier stage is based on six
These advantages over other types of conventional converters bidirectional switches S1-S6. The bidirectional switches of rectifier
promotes the integration of this MC topology in several areas of stage consisting of a pair of devices with turn-off capability are
industrial applications such as aerospace industries who have a great usually insulated gate bipolar transistors (IGBTs) in either a common
interest in this type of converter, the marine propulsion industry [1], collector or a common emitter back-to-back arrangement and each
the electrical drive machines with variable speed embedded systems IGBT has an anti-parallel diode. The inverter stage is a three phase
[2]-[4], field of renewable energy based on wind [5], and fuel cells. voltage source topology based on six switches S7-S12. This
Two types of topologies for the matrix converter have been configuration of IMC generates less switching and conduction loss
established by researchers as direct matrix converter topology and compared to other configurations, but it has a complex control for the
indirect matrix converter topology [6]-[10]. The direct topology is a number of switches to handle. All this leads to the development of
one compact stage composed from nine bidirectional switches, but other configurations with the aim of reducing the number of
the indirect topology is composed from two stages, the rectifier and transistors required which facilitates the control of the matrix
the inverter. The first indirect matrix converter topology (IMC) converter.
established is shown in Fig.1. This structure presents a large number
of transistors mainly in the rectifier stage. This topology has a
complex control for the number of switches to handle, which leads to
the development of other topologies in order to reduce the number of
transistors required, which facilitates the control of the matrix
converter. More topologies are developed such as sparse matrix
converter shown in Fig.3. This structure is the subject of the study in
this paper. It presents a reduced number of switches in the rectifier
such as each arm of the rectifier is constituted by three transistors
instead of four as shown in Fig.2. In section II, this topology is
Fig. 1. Indirect matrix converter
developed [11]-[14].
The dimensioning of the input and output filters link to the sparse In order to reduce the number of transistor, one IGBT from each
matrix converter fed through a balanced three-phase system and arm of the rectifier is eliminated as in shown in Fig.2. So three
feeding an R-L load is established in section III. These two filters are

978-1-4673-7554-2/15/$31.00 ©2015 IEEE 411

components will be eliminated in all compared to the previous IV. MODELING SYSTEM CONVERSION
configuration shown in Fig.1 and a Sparse Matrix Converter (SMC)
shown in Fig.3. The main advantage of the minimizing number of A. Indirect matrix converter model
switches is to facilitate the development of the control of the The model of two-stage matrix converter shown in Fig.4 is given
converter. by the matrix expressions (1)-(2), resulting from the product of the
The four-quadrant switch current-source type rectifier is able to instantaneous states of the rectifier stage and the inverter stage.
operate with positive and negative dc-link voltage, but the dc-link
voltage of the IMC shows a fixed polarity. That’s why it is important . (1)
to reduce the number of unipolar turn-off power semi-conductors of
the rectifier stage and the techniques of reduction appear in Fig.2 for . (2)
one arm of the IMC rectifier.
′
. . ′ . (3)
′ ′ ′
′

(4)
′ ′
′
′ (5)
′
where is the connection matrix of the IMC, is the
(a) (b) (c) connection matrix of the rectifier stage and is the connection
Fig. 2. Modification of the arm structure of the rectifier stage of the IMC (a), matrix of the inverter stage. If represents the conduction time of the
into the SMC arm topology (c) Indirect matrix converter corresponding switch during the switching period T, then the
modulation coefficients relative to the conduction time is expressed
by equation (6), , , for the rectifier, , , for the
inverter.
(6)
The equations described above in “connection function” are
transformed in “modulation function” and the expression of output
voltage and input current are obtained in equations (7) and (8).
. (7)

Fig. 3. Sparse matrix converter . (8)

In this topology there are only 9 IGBTs in the rectifier stage m′

instead of 12 at the previous topology, keeping the same number of . m ′ . ′ ′ ′ (9)
diode. m ′

III. PASSIVE FILTER STRUCTURE

′ (10)
Passive filters as classic methods for power quality improvement ′ ′
in ac- ac system consist of series LC tuned for removing a specific
′
harmonic or blocking a bandwidth of severe harmonics of nonlinear
load currents. These filters have low impedances for the tuned ′ (11)
frequencies such as 3rd, 5th and 7th. Low cost is a great benefit of ′
these filters but because of their LC constant parameter, they cannot
be efficient power quality improvement facilities for dynamic where is the conversion matrix of the IMC, is the
nonlinear loads. Another problem of installation of passive filters is conversion matrix of the rectifier stage and is the conversion
probable resonances between the impedance of passive filter and the matrix of the inverter stage.
system resulting in increasing the harmonics and lower power quality B. Input filter model in fixed (a,b,c) frame
of the alternative system.
The input filter is located between the mains and the indirect
Due to the discontinuous input currents, the matrix converter matrix converter (Fig.4) to compensate harmonics in the system. It
behaves as a source of current harmonics, which are injected back can be modeled in a three-phase fixed base by equations (12) and (13)
into the alternative mains. Since these current harmonics results in below representing the relationship between instantaneous electrical
voltage distortions that affect the overall operation of the ac-ac quantities.
system, they have to be reduced. Therefore, it is necessary to use
effective power quality improvement facilities for improving the . . (12)
power quality of the matrix converter. Usage of LC input passive
filter can be a successful procedure in limitation or removal of input
current harmonics. An output passive filter can improve the quality of
the output voltage.

412
 2 1 1 0 1 0 0
0 0 0
3 3 3
1 2 1 .
0 0 0 0 0 1
3 3 3 0
0 0 1 1 2
0 0 0 0 0 0
3 3 3 0 0 0
2 1 1
. + .
3 3 3 0 0 0 0 0 0
1 2 1 0 0 0
3 3 3 0 0
1 1 2 0
+ . (19)
3 3 3 0 0
0
(13)
, (20)
1 1
0 0 0 0 This model can be noted in the following form (21)
1 1
0 0 0 0 ) (21)
1 1
0 0 0 0
V. PWM CONTROL OF SPARSE MATRIX CONVERTER
where represents the transformation matrix of the line to the
TOPOLOGY
voltage.
2 1 1 The Pulse Width Modulation (PWM) control technique is used to
1 2 1 (14) calculate the duty cycle of the switches. This technique is inspired
1 1 2 from comparison of a reference sine wave with a triangular wave in
order to generate pulses with varied duty cycle according to the
C. Input filter model in d-q frame frequency of the reference sine wave. The PWM signals switch the
To obtain a continuous model of the input filter in rotating d-q power switches and control the power energy supply to the load
reference, the Parck transformation matrix given by expression (15) is feeding by matrix converter [21]-[27]. This technique of modulation
applied to equations (12) and (13) to give the d-q model of input filter is applied to the system shown in Fig.5 formed by sparse matrix
represented in state space by the equation (18). converter feeding an R-L load from balanced three phases AC
voltage.

(15) VI. SIMULATION RESULTS

The simulation results of the sparse matrix converter feeding an
RL load before filtering are presented in Figs.6 to 8. Table I gives the
̅ list of specification elements of the circuit. Figs.9 to 17 presents the
. (16)
simulation results of the system after filtering and Table II present a
summary of the simulation results for different output frequencies.
(17) The output voltage Vdc of the rectifier is shown in Fig.18.

0 Figs.6 (a)-(b) show the phase 1 input current (ie) and its harmonic
0 0 spectrum, respectively before filtering. The input current has a THD
0 0 0 of 75.45% and its harmonic spectrum show that has high frequency
. + . harmonics. The output current has a THD of 2.74%. It is shown in
0 0 1 0
0 1 Figs.7. (a)-(b). The output voltage (vs) of phase 1 and its harmonic
0 0 spectrum are shown in Figs.8. (a)-(b), respectively. It has a THD of
1 0 92.02% and show the presence of high frequency harmonics.
0 1
. (18) An input and output LC filters are connected to mitigate the high-
0 0
0 0 frequency ripple from the input currents and output voltages with
output frequency equal 50Hz. Figs.9 (a)-(b) show the phase 1 input
current and its harmonic spectrum after compensation. The measured
D. State model indirect matrix converter THD of the input current in phase 1 is reduced to 1.8% from 75.45%.
Such as control laws and level model (18) are not It is important to notice that the input current is kept free of
available, and to obtain a quadrature component decoupled , harmonics.
and with a power factor near unity. It is interesting to
The THDs of the output voltage and output current in phase 1 are
transfer this state model to a controllable canonical form (19) taking
reduced to 0.3% from 92.02% and to 0.16% from 2.74%,
into account the transformation (21) follows.
respectively. The performance of the input and output filter are
observed to be excellent.

413
 Fig. 4. Topologie of indirect matrix converter

Fundamental (50Hz) = 25.11 , THD= 1.80%

120

M a g ( % o f F u n d a m e n ta l)
Fundamental (50Hz) = 13.52 , THD= 75.45%
120 20 100
M a g (% o f F u n d a m e n ta l)

ie ( A )
20 100 80
0
60
80
ie ( A )

0 60 -20 40

20
40
-20 0.1 0.11 0.12 0.13 0.14 0.15 0
20 0 5 10 15 20 25 30

0.1 0.11 0.12 0.13 0.14 0.15 0

0 5 10 15 20 25 30
Time (s) Harmonic order

Time (s) Harmonic order (a) (b)

(a) (b) Fig. 8. Simulation results of three phase to three phase sparse matrix
Fig. 5. Simulation results, (a) input current of three phase to three phase converter after filtering, (a) Input current, (b) Harmonic Spectrum
sparse matrix converter before filtering, (b) Harmonic Spectrum
Fundamental (50Hz) = 266.1 , THD= 0.30%
120

M a g (% o f F u n d a m e n ta l)
Fundamental (50Hz) = 19.61 , THD= 2.74% 200 100
120
v s (V )
80
M a g (% o f F u n d a m e n ta l)

100
20 0 60
80
is ( A )

40
0 60
-200
20
40
-20 0.1 0.11 0.12 0.13 0.14 0.15 0
20 0 5 10 15 20 25 30

0.1 0.11 0.12 0.13 0.14 0.15 0

Time (s) Harmonic order

Time (s)
0 5 10 15
Harmonic order
20 25 30
(a) (b)
(a) (b) Fig. 9. Simulation results of three phase to three phase sparse matrix
Fig. 6. Simulation results of three phase to three phase sparse matrix converter converter after filtering (a) Output voltage, (b) Harmonic spectrum
before filtering, (a) Output current, (b) Harmonic spectrum 40 Fundamental (50Hz) = 25.38 , THD= 0.18%
120

20

M a g ( % o f F u n d a m e n ta l)
100
Fundamental (50Hz) = 205.6 , THD= 92.02%
is ( A )

120
400 80
0
M a g (% o f F u n d a m e n ta l)

100
60
200
80 -20
v s (V )

40
0 60 20
-40
-200 40 0.1 0.11 0.12 0.13 0.14 0.15 0
0 5 10 15 20 25 30

-400
0.1 0.11 0.12 0.13 0.14 0.15
20
Time (s) Harmonic order
0
0 5 10 15 20 25 30
Time (s) Harmonic order
(a) (b)
(a) (b)
Fig. 7. Simulation results, (a) Output voltage of three phase to three phase
Fig. 10. Simulation results of three phase to three phase sparse matrix
sparse matrix converter before filtering, (b) Harmonic Spectrum
converter after filtering (a) Output current, (b) Harmonic spectrum

Fig. 11. Sparse matrix converter

414
 When the output frequency is 25Hz, the measured THD of the 40 120
Fundamental (100Hz) = 24.83 , THD= 0.69%

input current in phase 1 is reduced to 2.04%. The input current is kept 20

M a g ( % o f F u n d a m e n ta l )
100

is ( A )
free of harmonics. So, the THDs of the output voltage and output 0
80

current in phase 1 are reduced to 0.33% and to 0.22%, respectively. -20

60

40

-40 20

40 120
Fundamental (50Hz) = 23.75 , THD= 2.04% 0.1 0.105 0.11 0.115 0.12 0.125 0
0 5 10 15 20 25 30

Time (s)

M a g ( % o f F u n d a m e n ta l)
20 100 Harmonic order

(a) (b)
ie ( A )

80
0
60 Fig. 17. Simulation results of three phase to three phase sparse matrix
-20 40 converter after filtering (a) output current, (b) Harmonic spectrum.
-40 20
0.1 0.11 0.12 0.13 0.14 0.15
0
0 5 10 15 20 25 30 The Vdc output voltage of the rectifier is fixed arround 500V, which
Time (s) Harmonic order

(a) (b) expresses the top notch function of the rectifier.

Fig. 12. Simulation results of three phase to three phase sparse matrix
converter after filtering, (a) Input current, (b) Harmonic spectrum 600

400

V dc
Fundamental (25Hz) = 261.5 , THD= 2.33%
120
200
200
M a g ( % o f F u n d a m e n ta l)

100
v s (A )

80 0
0 0.1 0.11 0.12 0.13 0.14 0.15
60
Time (s)
-200 40

20
Fig. 18. Simulation results of the Vdc output voltage of the rectifier
0.1 0.12 0.14 0.16 0.18 0.2 0
0 5 10 15 20 25 30

Time (s) Harmonic order TABLE I. PECIFICATIONS ELEMENTS OF THE CIRCUIT

(a) (b)
Fig. 13. Simulation results of three phase to three phase sparse matrix Circuit Value
converter after filtering (a) Output voltage, (b) Harmonic spectrum Specifications

Fundamental (25Hz) = 24.04 , THD= 0.22%

Input side 220. √2 , 50
40 120
M a g ( % o f F u n d a m e n ta l )

20 100
Load 10 , 10
is ( A )

80
0
-20
60
Input filter 0.1 , 11 ,
40

-40 20
100
0.1 0.12 0.14 0.16 0.18 0.2 0
0 5 10 15 20 25 30
Output filter 0.1 , 4.7 ,
Time (s) Harmonic order 200
(a) (b) Switching frequency 10
Fig. 14. Simulation results of three phase to three phase sparse matrix
converter after filtering (a) Output current, (b) Harmonic spectrum.
The Table II summarizes the features of input current, output
If the output frequency is 100Hz, the measured THD of the input voltage and current with different output frequencies. These
current in phase 1 is reduced to 1.66%. So, the THDs of the output characteristics proves that the ratio voltage is improved when output
voltage and output current in phase 1 are reduced to 0.65% and to frequency increase from 25Hz to 100Hz. The THD of the input
0.69%, respectively. The input current output voltage and current are current is also improved with increasing output frequency, but THD
kept free of harmonics. of output voltage and current are more mediocre.

40 Fundamental (50Hz) = 23.97 , THD= 1.66%

120
M a g ( % o f F u n d a m e n ta l )

20 100 TABLE II. RESULTS SIMULATION FOR DIFFERENT OUTPUT FREQUENCIES

ie ( A )

80
0
60 THD (%)
-20 40
Frequencies Ratio Input Output Output
20
-40 (%)
0.1 0.11 0.12 0.13 0.14 0.15 0
0 5 10 15 20 25 30
current current voltage
Time (s) Harmonic order 25Hz 84 2.04 0.22 0.33
(a) (b) 50Hz 85.56 1.8 0.18 0.3
Fig. 15. Simulation results of three phase to three phase sparse matrix 100Hz 94.2 1.66 0.69 0.65
converter after filtering, (a) Input current, (b) Harmonic Spectrum.

Fundamental (100Hz) = 293.2 , THD= 0.65%

120
VII. CONCLUSION
M a g (% o f F u n d a m e n ta l)

100
200
80 In this paper a sparse topology of indirect matrix converter is
v s (V )

0 60 developed and results simulation are obtained for an RL load.

-200
40
Further, the operation of a three phase to three phase matrix converter
20

0.1 0.105 0.11 0.115 0.12 0.125 0

synthesized using the PWM control has been detailed. This paper is
Time (s)
0 5 10 15
Harmonic order
20 25 30
helpful to study different topologies of matrix converter with many
(a) (b) other different loads and synthesize the sine wave input current and
output voltage using various control technique. The study offer us a
Fig. 16. Simulation results of three phase to three phase sparse matrix very wide field of research especially in the study of reliability,
converter after filtering (a) output voltage, (b) Harmonic spectrum maintainability, availability; faults tolerances and stability of these

415
 types of converters. The possibility of drawing developed strategies [11] J. W. Kolar, F. Schafmeister, S. D. Round, and H. Ertl, “Novel Three-
controls ensuring the optimization of conduction and switching losses Phase AC–AC Sparse Matrix Converters,” IEEE Trans. Power
with high performance operation of the converter.  The matrix Electron., vol. 22, no. 5, September 2007.
[12] C. Klumpner, M. Lee, and P. Wheeler, “A new three-level sparse
converter has advantages
indirect matrix converter,” in Proc. IECON, pp. 1902–1907, 2006.
The possibility of reducing the number of switches forming [13] M. Y. Lee, C. Klumpner, and P. Wheeler, “Experimental evaluation of
the converter thus reducing the number of semi-conductors the indirect three-level sparse matrix converter,” in Proc. Power
and subsequently losses. Electron., pp. 50–54, .Mar. 2008.
These difficulties switching are reduced, switches of the [14] H. Hojabri, H. Mokhtari, and L. Chang, “A Generalized Technique of
input stage (rectifier) are reduced and the second stage of Modeling, Analysis, and Control of a Matrix Converter Using SVD”,
the converter switches as a standard inverter. IEEE Transactions on Industrial Electronics, vol. 58, no. 3, pp. 949-959,
March 2011.
ACKNOWLEDGMENT [15] S. Bhim, N. Singh, A. Chandra, K. Al-Haddad, A. Pandey and P.
Kothari, “A Review of Single-Phase Improved Power Quality AC–DC
The authors gratefully thank the Canada Research Chair in Converters”, IEEE Transactions on Industrial Electronics, Vol. 50, no.
Electric Energy Conversion and Power Electronics at the École de 5, pp. 962-981, October 2003.
Technologie Supérieure for their financial support. [16] J.-K. Kang, H. Hara, E. Yamamoto, M. Swamy, and T. Kume, “Output
voltage distortion in matrix converter by commutation of bi-directional
REFERENCES switches,” in IEEE Int. Elec. Machines and Drives Conf. IEMDC, vol. 1,
[1] R. Bucknall and K. Ciaramella, “On the conceptual design and pp. 55–59, Jun. 2003,
performance of a matrix converter for marine electric propulsion,” IEEE [17] A. Arias, L. Empringham, G. M. Asher, P. W. Wheeler, M. Bland, M.
Trans. Power Electron., vol. 25, no. 6, pp. 1497–1508, Jun. 2010. Apap, M. Sumner, and J. C. Clare, “Elimination of waveform distortions
[2] P. Wheeler, J.C. Clare, L. Empringham, M. Apap, K. Bradley, C. in matrix converters using a new dual compensation method,” IEEE
Whitley and G. Towers, “A matrix converter based permanent magnet Trans. Ind. Electron., vol. 54, no. 4, pp. 2079–2087, Aug. 2007.
motor drive for an aircraft actuator system”, Proceedings of IEEE [18] J. Mahlein, M. Bruckmann and M. Braun, “Passive protection strategy
International Electric Machines and Drives Conference, vol. 2, pp. 1295 for drive system with matrix converter and induction machine”, IEEE
– 1300, June 2004. Transactions and Industrial Electronics, vol. 49, no. 2, pp. 297-303,
[3] Z. Daning, K. Sun, L. Huang, K. Sasagawa, “A Novel Commutation April 2002.
Method of Matrix Converter Fed Induction Motor Drive Using [19] J. Choi and S. Sul, “A new compensation strategy reducing
RBIGBT”, Fourtieth. IEEE IAS Industry Applications Society Annual voltage/current distortion in PWM VSI systems operating with low
Meeting, pp. 2347-2354, 2005. voltages”, IEEE IAS Annual Meeting, vol.31, pp. 1001-1008, 1995.
[4] X. Huang, A. Goodman, C. Gerada, Y. Fang, and Q. Lu, “A Single [20] F. Schafmeister and J. Kolar, “Novel Hybrid Modulation Schemes
Sided Matrix Converter Drive for a Brushless DC Motor in Aerospace Significantly Extending the Reactive Power Control Range of All
Applications”, IEEE Transactions on Industrial Electronics, vol. 59, Matrix Converter Topologies With Low Computational Effort”, IEEE
no. 9, pp.3542-3552, September 2012. Transactions on Industrial Electronics, vol. 59, no. 1, pp. 194-210,
[5] A. Garcés and M. Molinas, “A Study of Efficiency in a Reduced Matrix January 2012.
Converter for Offshore Wind Farms”, IEEE Transactions on Industrial [21] N. Celanovic and D. Boroyevich, “A comprehensive study of neutral-
Electronics, Vol. 59, no. 1, pp. 184-193, January 2012. point voltage balancing problem in three-level neutral-point-clamped
[6] L. Rmili, S. Rahmani, H. Vahedi, and K. Al-Haddad, “Comprehensive voltage source PWM inverters,” IEEE Trans. Power Electron., vol. 15,
Analysis of Matrix Converters: Indirect Topology”, Proc. 15th no. 2, pp. 242–249, Mar. 2000.
international conference on Sciences and Techniques of Automatic [22] X. Changliang, Y. Yan, P. Song, and S. Tingna, “Voltage Disturbance
control & computer engineering - STA'2014, pp. 679-684, December Rejection for Matrix Converter-Based PMSM Drive System Using
2014. Internal Model Control”, IEEE Transactions on Industrial Electronics,
[7] P. Wheeler, J. Rodriguez, J. Clare, L. Empringham, and A. Weinstein, vol. 59, no. 1, pp. 361-372, January 2012.
“Matrix converters: A technology review”, IEEE Trans. Ind. Electron., [23] R. Rojas, T. Ohnishi, and T. Suzuki, “Simple structure and control
vol. 49, no. 2, pp. 276–288, Apr. 2002. method for a neutral-point-clamped PWM inverter,” in Proc. PCC, pp.
[8] L. Rmili, S. Rahmani, H. Vahedi and K. Al-Haddad, “A Comprehensive 26–31, 1993.
Analysis of Matrix Converters: Bidirectional Switch, Direct Topology, [24] J. Rodriguez, M. Rivera, J. W. Kolar, P. W. Wheeler, “A Review of
Modeling and Control”, In Proc. 23rd IEEE International Symposium on Control and Modulation Methods for Matrix Converters,” IEEE Trans.
Industrial Electronics, IEEE-ISIE, pp. 313-318, June 2014. Ind. Electron., vol. 59, no. 1, pp. 58–70, Janu.2012.
[9] E. Ibarra, I. Kortabarria, J. Andreu, I. M. de Alegria, J.L. Martin, and P. [25] J. W. Kolar, T. Friedli, J. Rodriguez, and P. W. Wheeler, “Review of
Ibañez, “Improvement of the Design Process of Matrix Converter Three-Phase PWM AC–AC Converter Topologies,” IEEE Trans. Ind.
Platforms Using the Switching State Matrix Averaging Simulation Electron., vol. 58, no. 11, pp. 4988–5006, Nov. 2011.
Method”, IEEE Transactions on Industrial Electronics, vol. 59, no. 1, [26] S. Kouro, P. Lezana, M. Angulo, and J. Rodriguez, “Multicarrier pwm
pp. 220-234, January 2012. with dc-link ripple feedforward compensation for multilevel inverters,”
[10] L. Rmili, S. Rahmani, F. Fnaiech, and K. Al-Haddad, “Space Vector IEEE Trans. Power Electro., vol. 23, no. 1, pp. 52–59, Jan. 2008.
Modulation Strategy for a Direct Matrix Converter”, In Proc. 14th [27] K. Mohapatra, P. Jose, A. Drolia, G. Aggarwal, and S. Thuta, “A novel
international conference on Sciences and Techniques of Automatic carrier-based PWM scheme for matrix converters that is easy to
control & computer engineering - STA'2013, pp. 1-6, December 2013. implement”, in Proc. 36th IEEE Power Electron. Spec. Conf., pp. 2410–
2414, Jun. 2005.

416

Powered by TCPDF (www.tcpdf.org)