novel modulation method for multidirectional matrix converter...novel modulation method for...
TRANSCRIPT
-
Research ArticleNovel Modulation Method for MultidirectionalMatrix Converter
Saman Toosi,1 Norhisam Misron,1,2 Tsuyoshi Hanamoto,3 Ishak Bin Aris,1
Mohd Amran Mohd Radzi,1 and Hiroaki Yamada4
1 Department of Electrical & Electronic, Faculty of Engineering, Universiti Putra Malaysia (UPM), 43400 Serdang,Selangor, Malaysia
2 Institute of Advanced Technology (ITMA), Universiti Putra Malaysia (UPM), 43400 Serdang, Selangor, Malaysia3 Department of Biological Functions Engineering, Graduate School of Life Science and Systems Engineering,Kyushu Institute of Technology, 2-4 Hibikino Wakamatsu-ku, Kitakyushu 808-0916, Japan
4Graduate School of Science and Engineering, Yamaguchi University, 2-16-1 Tokiwadai, Ube-shi, Yamaguchi 755-8611, Japan
Correspondence should be addressed to Norhisam Misron; [email protected]
Received 21 May 2014; Revised 22 August 2014; Accepted 22 August 2014; Published 14 September 2014
Academic Editor: Fernando Lessa Tofoli
Copyright Β© 2014 Saman Toosi et al. This is an open access article distributed under the Creative Commons Attribution License,which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.
This study presents a newmodulation method for multidirectional matrix converter (MDMC), based on the direct duty ratio pulsewidth modulation (DDPWM). In this study, a new structure of MDMC has been proposed to control the power flow directionthrough the stand-alone battery based system and hybrid vehicle. The modulation method acts based on the average voltage overone switching period concept. Therefore, in order to determine the duty ratio for each switch, the instantaneous input voltages arecaptured and compared with triangular waveform continuously. By selecting the proper switching pattern and changing the slopeof the carriers, the sinusoidal input current can be synthesized with high power factor and desired output voltage. The proposedsystem increases the discharging time of the battery by injecting the power to the system from the generator and battery at thesame time. Thus, it makes the battery life longer and saves more energy. This paper also derived necessary equation for proposedmodulation method as well as detail of analysis and modulation algorithm. The theoretical and modulation concepts presentedhave been verified in MATLAB simulation.
1. Introduction
More than 1.3 billion people in the world are not connected toa national grid. Although extension of the conventional elec-tricity grid remains preferable mode of electrification, it isnot economical for areas where the grid extension is difficult.Currently the stand-alone power system (SAPS) supplieslocal villages or individual users with lack access to electricity.Typical SAPS may be powered by one or more methods suchas microhydroturbine, wind turbine, solar panel geothermalsource, and diesel or biofuel generator to generate theelectricity [1].
A major requirement for stand-alone power system isto ensure continuous power flow by storing excess energyfrom the energy sources. For example, hybrid systems withbattery storage are employed as an efficient and reliable stand-alone system for remote areas [2]. Battery based systems
(BBS) are amongst the SAPS models which a battery mayemploy in series or parallel with renewable energy source. Inbattery based systems, the input power of the system convertsto desirable voltage and frequency through power electronicconverters in order to supply the system loads and charge thebattery [3, 4].
In recent years, the matrix converter becomes popular inthe category of AC to AC converters due to the desirable fea-tures such as sinusoidal input and output current, generationof load voltage with arbitrary amplitude and frequency, andability to control input power factor for any load [5]. In theearly 1980s Venturini and Alesina proposed the principle ofMC control [6].They derived duty ratio functions that can bemodulated by carrier signal. In this method, the voltagetransfer ratio was limited to 0.5. Alesina and Venturini (1981)theoretically proved that the maximum voltage ratio, πmax, isequal to 0.866 for the three-phase MC when using balanced
Hindawi Publishing Corporatione Scientific World JournalVolume 2014, Article ID 645734, 12 pageshttp://dx.doi.org/10.1155/2014/645734
http://dx.doi.org/10.1155/2014/645734
-
2 The Scientific World Journal
input voltage [7]. In 1989, Alesina and Venturini extended thevoltage ratio from 0.5 to 0.866 by taking advantage of thirdharmonic injection methods [8].
The βindirect transfer functionβwas derived byRodriguezin 1983 [9]. In this method, the matrix converter wasdescribed as virtual configuration of pulse with modulation(PWM) rectifier and inverter with βfictitious dc link.β Theoperational and technological research on MC were contin-ued in different areas such as new topology of MC [10β13],input filter design [14, 15], unbalance operational conditions[16β18], safe and practical commutation strategies [10, 19],new control methods [20, 21], new modulation methods [22,23], and new application such as hybrid vehicles [24].
Yoon and Sul (2006) [23] proposed new carrier basedmodulation methods for conventional matrix converter. Thismethod is the same as conventional space vector pulse mod-ulation (SVPWM) which is used in voltage source inverter.Yoon and Sul synthesized the sinusoidal input currentwith unity power factor by changing the slope of carrier andthe proper offset voltage. The reference output voltages arecalculated and compared with a discontinuous carrier to gen-erate the gating signals. However, it is difficult to intuitivelyunderstand the modulation principle since it employs theoffset for references and discontinues carrier signal. Further-more, thismethod cannot be used for theMC typologies witha neutral connection.
The preliminary concepts of a new carrier based PWMstrategy, named direct duty ratio PWM (DDPWM), are pre-sented by Li et al. (2008). This method can be implementedwithout complex calculations and lookup tables and doesnot require the reference offset voltage. Based on the averagevalue of each output phase in one switching period, theduty ratio values may be updated at each switching cycle byemploying input phase voltages. Thus, the PWM signals aregenerated by comparison of these duty ratio values with acontinuous triangular carrier waveform. The πmax of 0.866also can be easily obtained in the three-phase system byapplying the third harmonic injection method to the outputvoltage references. Furthermore, the input power factor canbe controlled by changing the slope of the carrier whilemaintaining the sinusoidal input currents [22]. Li and Choi(2009) extended theDDPWMto various topologies ofmatrixconverter and derived the control schemes for alternativestructures such as single-phase and three-phase four-legmatrix converters [25].
Multidirectional converter has recently been proposedas an alternative power conversion concept which has bothrectifier and inverter capability [26, 27].Most desired featuresof multidirectional converter can be fulfilled by using MatrixConverter structures. In theMDMC, a bidirectional switch isused, coupled between the power source and load, to provideboth AC and DC properties, which cannot be achieved withconventional converters. This converter has ability to controlthe power flow and synthesise the desired output voltageby developing the space vector pulse width modulation(SVPWM) methods. In the SVPWM method, the modu-lation task of the multidirectional matrix converter can beresolved into the different imaginary stages of transforma-tion including inverter and rectifier stage which are linked
together by an imaginary DC link. However, the MDMC isnot being able to inject power from generator and batteryat the same time, since several vectors are utilized in oneswitching period [28]. Previous studies [29β31] show that, inconventional battery based system, generator should be disconnected from system when it is not being able to supplythe demand power.
Based on the literature highlighted above, this study aimsto inject power from battery and generator at the same timeby changing theMDMC structure and increasing the numberof time intervals of direct duty pulse width modulationmethod. In this study, the proposed modulation methoddetermines the switching state of each output phase byemploying the input DC phase voltages and input AC phasevoltages based on per-output-phase average concept over oneswitching period. At the first step of each switching period, inorder to generate the corresponding PWM signals, the dutyratio values for each output phase were calculated and theresults compared the continuous triangular waveform. Thisnew topology and new modulation method can increase thedischarging time of battery in the battery based systemswhenthe discharging time is directly proportional to the generatoroutput voltage. Therefore, the multidirectional matrix con-verter with a new modulation method is expected to break-through towards new technological advancements in the areaof sustainable energy and power electronics.
2. Proposed MDMC Structure forBattery Based System
Batteries are not efficient as a whole. Some energy is lost asheat and chemical reactions when charging and discharging.In common, the lead acid batteryβs efficiency is around 85%when state of charge (SOC) is varied from 0 to 100% [32]. Inbattery based stand-alone power system (BBSAPS), when thebattery is connected in series, total electricity generation fromsystem will be stored in battery before transmitting to theloads, while in system with parallel battery connection onlythe excess electricity will be saved in battery. Hence, a systemwith parallel battery connection ismore efficient compared toa system with series battery connection. In addition, the par-allel battery based system can be modified by combining allconverters as a single converter which is indicated as multi-directional matrix converter. Figure 1 shows the comparisonof block diagram of BBSs with matrix converter and theproposed MDMC system.
The multidirectional matrix converter is a single-stageconverter which has aπΓπmatrix (or array) of bidirectionalpower switches to connect anπ-phase voltage source to an π-phase load directly. In general, the proposedMDMCneeds 15bidirectional switches that is one switch between each inputand output phases. Figure 2 shows the circuit configuration ofproposed MDMC including the positive DC input voltage(battery), negative DC input voltage (battery), three-phaseinput voltages (AC generator), multidirectional matrix con-verter and resistor-inductor (π -πΏ) load, and second-order πΏ-πΆ filter which is used at the input terminals to filter out thehigh frequency harmonics of the input currents. In this study
-
The Scientific World Journal 3
Rectifier
Generator
Battery Inverter
Matrix converter AC load
(a) BBS with matrix converter
Generator
Battery
AC loadMultidirectional matrix converter
(MDMC)
(b) BBS with multidirectional matrix converter
Figure 1: Block diagram of system with parallel battery connection.
Generator
Battery
Vsa
Vsb
Vsc
s
s
VsP
VsN
isa Lf
Lf
Lf
Lf
Lf
+
+ β
β
+
β
Cf/2
Cf Cf Cf
R
S
T
RLβac LLβacioR
o
o
οΏ½oR
RLβdc LLβdc
RLβdc LLβdc
SaR
SbR
ScR
SPR
SNR
SaS
SbS
ScS
SPS
SNS
SaT
SbT
ScT
SPT
SNT
Source Input filter Bidirectional switches
AC and DC load
βΌ
βΌ
βΌ
Figure 2: Multidirectional matrix converter circuit.
the π πΏ-dc and πΏπΏ-dc and the π πΏ-ac and πΏπΏ-ac are considered as
DC and AC load, respectively.In the MDMC, the switching method should have sinu-
soidalwaveforms at the arbitrarymagnitude, frequency inACside, and clean DC voltage at the DC side.The input currentsalso should be sinusoidal at the desired power factor. In orderto achieve this target, a proper switching pattern should beapplied to the switches of the MDMC in each switchingperiod.The general switching function for the switches of theMDMC can be described as follows:
πππ (π‘) = {
1, πππclosed,
0, πππopen,
π = π, π, π, π,π π = π , π, π,
(1)
where the πππrefers to the switch on input line βπβ and output
line βπ.βMoreover, input phases should not be short circuited and
output phases should never be opened due to the inductive
nature of typical loads. In this study, these two constraints canbe expressed as below:
πππ+ πππ+ πππ+ πππ+ πππ= 1, π = {π , π, π} . (2)
By considering two states for each switch in (1) and by apply-ing the limitation of (2) to the switching algorithms of theproposed MDMC, allowable combinations will be derivedbased on the DDPWM technique. Voltages and currents ofsources and voltages and currents of load in Figure 2 can beexpressed as vectors that are defined by (3)where theπ can beinput and output phase-to-neutral voltage vectors or theMDMC input and output current vectors:
ππ ππ
= [
[
ππ (π‘)
ππ (π‘)
ππ (π‘)
]
]
, ππππππ
=
[[[[[
[
ππ (π‘)
ππ (π‘)
ππ (π‘)
ππ (π‘)
ππ (π‘)
]]]]]
]
. (3)
-
4 The Scientific World Journal
SaR = 1 SbR = 1 SbR = 1 SPR = 1 SNR = 1
SaS = 1 SbS = 1 ScS = 1 SPS = 1 SNS = 1
SaT = 1 SbT = 1 ScT = 1 SPT = 1 SNT = 1
tbR tcR tPRtNR
taS tbStcS tPS tNS
taT tbT tcT tPT tNT
Ts
TAC TDC
taR
Out
put
phas
eRO
utpu
tph
aseS
Out
put
phas
eT
Figure 3: The switching pattern in a sequence period.
The MDMC instantaneous switching function matrix can beexpressed as follows:
π = [
[
π ππ π ππ π ππ π ππ
π ππ
π πππ πππ πππ ππ
π ππ
π πππ πππ πππ ππ
π ππ
]
]
. (4)
Equation (5) shows the relation between load, input voltages,and currents, where ππ is the transpose matrix of πmatrix:
ππ ππ
= π β Vπππππ
, ππππππ
= ππβ ππ ππ. (5)
Modulation rules can be derived by applying the differentswitching pattern to the power switches (see Figure 3).
As indicated in Figure 3, the output phase βπ β is con-nected to the input phase βπβ during π‘
ππ and when π
π is the
sequence period of switching for MDMC system. It is alsoconnected to phase βπ,β βπ,β βπ,β and βπβ during timeperiods π‘
ππ , π‘ππ , π‘ππ , π‘ππ
, respectively. Arbitrary amplitudeand frequency can be generated bymodulating the duty cycleof the switches using their respective switching functions.
If πππ(π‘) = π‘
ππ/πseq, the restrictions of the duty cycle (based
on (2)) can be represented as below:
0 β€ πππβ€ 1, π
ππ+ πππ+ πππ+ πππ+ πππ= 1,
π β (π, π, π, π,π) , π β (π , π, π) .
(6)
The matrix π can be replaced by matrix π· (3 Γ 5) and finallythe low frequency transfer matrix is defined as below:
Vπ ππ
= π· β Vπππππ
, ππππππ
= π·πβ ππ ππ, (7)
where ππ ππ
and Vπ ππ
are a set of sinusoidal currents andarbitrary amplitude, frequency output voltages, and V
πππππ
and ππππππ
are sinusoidal input voltages and input currents atthe MDMC terminals.
3. Modulation Method
In this modulation method, reference output phase voltagecan be synthesized by utilizing all five input phase voltagesover one switching period in the average sense.Therefore, theswitching period π
π is divided into two time periods, π
πand
π3. Duringπ
π, the input phases ofACgenerator are connected
to a corresponding output terminal, and during π3the input
phases ofDCbattery are connected to a corresponding output
dR1
1 1
0
T1 T2
Tc
T3
Ts
MN
MX MX
MD
TR1 TR2 TR3 TR4 TR5 TR6
Neg
Pos
Figure 4: Output π -phase switching state in switching pattern-I.
terminal. In addition, the time interval ππis divided into two
periods π1and π
2. Also, the MX, MD, and MN denote the
instantaneous values of maximum, medium, and minimuminput voltages of AC generator. Furthermore, POS and NEGdenote the instantaneous values of positive andnegative inputvoltages of DC battery, respectively. During π
1, the line-
to-line voltage between MX and MN is used, which is themaximum line-to-line voltage among three line-to-line inputvoltages of generator at the sampling instant. During π
2, the
second maximum line-to-line voltage is used which is MX toMD for switching pattern-I and MD to MN for switchingpattern-II. Finally, during π
3the line-to-line voltage between
POS and NEG is employed.In this method, the three line-to-line input voltages of
the generator and the input voltages of batteries are readcontinuously at the sampling instant. Then, duty ratio values(range between 0 and 1) are predetermined for each outputphase at the beginning of each switching period. Also, theduty ratio of each phase is compared with a common con-tinuous triangular carrier waveform, in order to generate thecorresponding six time subintervals (see Figure 4). These sixtime subintervals determine the connection time of the cor-responding output terminal to the input phases during oneswitching cycle. Therefore, the desired output voltage canbe synthesized by updating the duty ratio value during eachswitching period. In addition, the input power factor can becontrolled by manipulating the slopes of the triangularcarriers. Due to the time subintervals extension, this methodis called extended direct duty pulse width modulation(EDDPWM).
3.1. Switching Pattern-I. Figure 4 shows the switchingpattern-I, where the π -phase duty ratio value (π
π 1) is com-
pared with triangular carrier waveform to generate the π -phase output voltage.The output phase is changed during theswitching pattern-I from MNβMXβMXβMDβNEGβPOS, consequently. The actual output voltage mergeof π -phase is illustrated in Figure 6 when applying switchingpattern-I. As illustrated in Figures 4 and 6, the output phaseβπ β is connected to the input phase βMNβ during π
π 1
-
The Scientific World Journal 5
dR2
1 1
0
Tc
T1 T2 T3
Ts
MN MN
MX
MD
TR1 TR2 TR3 TR4 TR5 TR6
Neg
Pos
Figure 5: Output π -phase switching state in switching pattern-II.
and when ππ is the sequence switching period. And it is
connected to phases βMX,β βMX,β βMD,β βNEG,β and βPOSβduring time periods π
π 2, ππ 3, ππ 4, ππ 5, and π
π 6, respectively.
These six time subintervals can be represented as (8), whereππ 1
is the π -phase duty ratio value and carrier slops aredefined asπ = π
1/ππand π = π
π/ππ :
ππ 1= ππ 1ππππ ,
ππ 2= (1 β π
π 1)πππ
π ,
ππ 3= (1 β π
π 1) (1 β π) πππ ,
ππ 4= ππ 1 (1 β π) πππ ,
ππ 5= ππ 1 (1 β π) ππ ,
ππ 6= (1 β π
π 1) (1 β π) ππ .
(8)
The fluctuation of the input voltage is negligible during theswitching periods.Thus, the integration of the output voltageVππ
over ππ can be expressed in
β«
ππ
0
Vππ ππ‘ β π
π 1β MN + (π
π 2+ ππ 3) β MX
+ ππ 4MD + π
π 5β NEG + π
π 6β POS.
(9)
Based on (8) and (9), the average output voltage can beexpressed in terms ofπ and π as follows:
Vππ =1
ππ
β«
ππ
0
Vππ ππ‘
β ππ 1β (β (1 β π)POS β π β MX + (1 β π) β π β MD
+π β π β MN + (1 β π) β NEG)
+ π β MX β (1 β π) β POS.(10)
0t
Ts
οΏ½oR
MN
MX
MD
TR1 TR2 TR3 TR4 TR5 TR6
Neg
Pos
VoA
Figure 6: Output π -phase voltage synthesis in switching pattern-I.
For the present switching cycle, the duty ratio value, ππ 1, can
be written as follows:
ππ 1= (Vβππ β π β MX β (1 β π) β POS)
Γ (β (1 β π)POS β π β MX + (1 β π) β π β MD
+π β π β MN + (1 β π) β NEG)β1,
(11)
where the Vβππ
is the π -phase output voltage command whichis equal to the V
ππ .
3.2. Switching Pattern-II. Theprocedure to drive the equationfor switching pattern II is the same as the previous switchingpattern. Figure 5 illustrates the case of switching patternII where the π -phase duty ratio value (π
π 2) is compared with
triangular carrier waveform to generate the π -phase outputvoltage. The output phase is changed during the switchingpattern-II from MNβMXβMDβMNβNEGβPOS,consequently. The actual output voltage merge of π -phase isillustrated in Figure 7when the output phase βπ β is connectedto the input phases during the time subintervals sequentially.Similarly, the integration of the output voltage V
ππ and the
average output voltage Vππ
is presented as below:
β«
ππ
0
Vππ ππ‘ β (π
π 1+ ππ 4) β MN
+ ππ 2β MX + π
π 3MD + π
π 5β NEG + π
π 6β POS,
Vππ =1
ππ
β«
ππ
0
Vππ ππ‘
β ππ 2β (β (1 β π)POS β ππ β MX
β (1 β π) β π β MD + π β MN
-
6 The Scientific World Journal
+ (1 β π) β NEG)+ (1 β π) β POS β π β π β MX + (1 β π) β π β MD.
(12)
By letting the Vππ
be equal to Vβππ
the duty ratio value ππ 2
canbe written as follows:
ππ 2= (Vβππ β π β MX β (1 β π) β POS)
Γ (β (1 β π)POS β π β MX + (1 β π) β π β MD
+π β π β MN + (1 β π) β NEG)β1.
(13)
3.3. Outputs Voltage Merged for MDMC. Five bidirectionalswitches are used for each output phase to apply the switchingpattern-I and II. The POS and NEG input phase are alwaysconstant while the MX, MD, and MN are selected byinstantaneous comparison of the AC input phases. When theswitching state for output phase βπ β is POS, NEG, MX, MD,or MN, the output phase βπ β is connected to the input phasewhich the voltage is POS, NEG, MX, MD, or MN, respec-tively. This modulation control method can be applied totheMDMC as amodular structure for each phase where eachoutput phase has the independent reference control signal.This reference control signal can be different in terms offrequency, waveform shape, and amplitude.
0t
οΏ½oR
MN
MX
MD
TR1 TR2 TR3 TR4 TR5 TR6
Neg
Pos
TsVoA
Figure 7: Output π -phase voltage synthesis in switching pattern-II.
The duty ratio of phases π and π is indicated as ππand π
π
and can be derived in the same way of phase π by letting theVππ
and Vππ
be equal to the π and π phase voltage com-mand Vβ
ππand Vβ
ππ, respectively. Duty ratio of phases can be
expressed as follows:
ππ =
{{{
{{{
{
Vβππ β π β MX β (1 β π) β POS
β (1 β π)POS β π β MX + (1 β π) β π β MD + π β π β MN + (1 β π) β NEG, for Pattern-I,
Vβππ β π β MX β (1 β π) β POS
β (1 β π)POS β π β MX + (1 β π) β π β MD + π β π β MN + (1 β π) β NEG, for pattern-II,
ππ=
{{{
{{{
{
Vβππβ π β MX β (1 β π) β POS
β (1 β π)POS β π β MX + (1 β π) β π β MD + π β π β MN + (1 β π) β NEG, for Pattern-I,
Vβππβ π β MX β (1 β π) β POS
β (1 β π)POS β π β MX + (1 β π) β π β MD + π β π β MN + (1 β π) β NEG, for pattern-II,
ππ=
{{{
{{{
{
Vβππβ π β MX β (1 β π) β POS
β (1 β π)POS β π β MX + (1 β π) β π β MD + π β π β MN + (1 β π) β NEG, for Pattern-I,
Vβππβ π β MX β (1 β π) β POS
β (1 β π)POS β π β MX + (1 β π) β π β MD + π β π β MN + (1 β π) β NEG, for pattern-II.
(14)
In the proposed method, the output voltages have been wellsynthesised by using two out of five line-to-line input voltagesduring each switching period, while the input currents aredistorted. In order to improve the input current quality andreduce the input currents distortion, five input phases con-ducted the current during each switching period.
3.4. Inputs Current Merged for MDMC. The π and π canproperly be adjusted to reduce the input current distortion in(14).This current distortion can be reduced by controlling theinput power factor which is directly depending on the slopeof the triangular carrier. By maintaining the π
π at a constant
value and adjusting the value of π andπ to the desired value,the input current is synthesized.The output voltagewaveform
will not be disturbed since π and π are considered in thederivation of (14).
The output currents are almost constant during theswitching cycles; thus the input current can be merged basedon the PWM switching pattern. These six time subintervalsfor each phase can be expressed as follows:
ππ1= ππππππ = πππ1,
ππ2= (1 β π
π)πππ
π = (1 β π
π) π1,
ππ3= (1 β π
π) (1 β π) πππ = (1 β ππ) π2,
ππ4= ππ(1 β π) πππ = πππ2,
-
The Scientific World Journal 7
ππ5= ππ (1 β π) ππ = πππ3,
ππ6= (1 β π
π) (1 β π) ππ = (1 β ππ) π3,
π = π , π, π.
(15)
3.4.1. Switching Pattern-I. Five inputs are connected to theoutput terminal through the bidirectional switches. Accord-ing to the switching state as shown in Figures 4 and 6, theoutput phases π , π, and π during π
π 1, ππ1, and π
π1, are con-
nected to the input phase whose voltage is MN. In the sameway, the input phase MX is connected to the output phasesπ , π, and π during (π
π 2+ ππ 3), (ππ2+ ππ3), and (π
π2+ ππ3),
respectively, and the input phase MD is connected to theoutput terminals π , π, and π during π
π 4, ππ4, and π
π4,
respectively. In addition, the input phaseNEG is connected tothe output terminals π , π, and π during π
π 5, ππ5, and π
π5,
respectively, and input phase POS is connected to the outputterminalsπ , π, andπ duringπ
π 6,ππ6, andπ
π6, respectively. By
applying the average concept to each input phase, the inputcurrent can be presented as follows:
ππ MXππ = (ππ 2 + ππ 3) πππ + (ππ2 + ππ3) πππ + (ππ2 + ππ3) πππ,
ππ MDππ = ππ 4πππ + ππ4πππ + ππ4πππ,
ππ MNππ = ππ 1πππ + ππ1πππ + ππ1πππ,
ππ NEGππ = ππ 5πππ + ππ5πππ + ππ5πππ,
ππ POSππ = ππ 6πππ + ππ6πππ + ππ6πππ.
(16)
By substituting (15) into (16), the instants value of the inputphase current during one switching cycle can be obtained asfollows:
ππ MXππ = β ππ β (ππ πππ + πππππ + πππππ) , (17)
ππ MDππ = π2 β (ππ πππ + πππππ + πππππ) , (18)
ππ MNππ = π1 β (ππ πππ + πππππ + πππππ) , (19)
ππ NEGππ = π3 β (ππ πππ + πππππ + πππππ) , (20)
ππ POSππ = βπ3 β (ππ πππ + πππππ + πππππ) . (21)
By substituting (19) into (17), theπ can be obtained as follows:
π β‘π1
ππ
= βππ MNππ MX
. (22)
Also, by calculating the ππ MX β ππ based on (ππ MX + ππ POS) β ππΆ,
the π can be obtained as below:
π β‘ππ
ππ
=ππ MX
(ππ MX + ππ POS)
. (23)
3.4.2. Switching Pattern-II. Like switching pattern-I, by apply-ing the switching pattern-II which is indicated in Figures 5and 7, input phaseMX is connected to the output phasesπ , π,
π during time subintervalππ 2,ππ2, andπ
π2, respectively. Sim-
ilarly, the input phase MD is connected to the output phasesπ , π, and π during π
π 3, ππ3, and π
π3, respectively and the
input phase MN is connected to the output phases π , π, andπ during (π
π 1+ππ 4), (ππ1+ππ4), and (π
π1+ππ4), respectively.
Finally, the input phase NEG is connected to the outputterminals π , π, and π during π
π 5, ππ5, and π
π5, respectively,
and input phase POS is connected to the output terminalsπ , π, and π during π
π 6, ππ6, and π
π6, respectively. The input
currents can be presented as follows:
ππ MXππ = ππ 2πππ + ππ2πππ + ππ2πππ, (24)
ππ MDππ = ππ 3πππ + ππ3πππ + ππ3πππ, (25)
ππ MNππ = (ππ 1 + ππ 4) πππ + (ππ1 + ππ4) πππ + (ππ1 + ππ4) πππ,
(26)
ππ NEGππ = ππ 5πππ + ππ5πππ + ππ5πππ, (27)
ππ POSππ = ππ 6πππ + ππ6πππ + ππ6πππ. (28)
By considering the time intervals for ππ and substituting (26)
into (24), theπ can be represented as follows:
π β‘π1
ππ
= βππ MXππ MN
. (29)
In addition, by calculating the ππ MX β ππ based on (ππ MX+ππ POS) β
ππΆ, the π can be obtained as below:
π β‘ππ
ππ
=ππ MN
(ππ MN β ππ POS)
. (30)
Furthermore, when the power factor is one in balance system,the currents π
π MN, ππ MX, and ππ POS can be replaced by voltagesVπ MN, Vπ MX, and Vπ POS (see (22), (23), (29), and (30)). As theinput voltage is directly sensed from the power circuit, themodulation calculation becomes easier.
To achieve unity power factor, load can be assumed ascurrent source in one switching cycle; accordingly, inputcurrent can be synthesized based on the switching state of theoutput phase current. The magnitude of each input currentvaried based on the ratio betweenπ
1andπ2and ratio between
ππand π
3while π
π is constant. On the other hand, due to the
missing of the energy storage component in MDMC, theinput and output powers should be kept balanced all the timeat any load. The practical selection of π and π for switchingpattern-I and II can be determined by the input voltage angleπΌπ, to synthesize sinusoidal input current.The three-phase input voltages signal shown in Figure 8 is
divided into the 6 segments and each segment is divided intotwo sectors which correspond to either switching pattern-I orII. By letting the power factor of system equal to π, the inputcurrent angle can be obtained as follows:
π½π= πΌπ+ π. (31)
In the proposed modulation, input power factor is controlledby adjusting the value of π and π (Table 1). Based on theavailable maximum line-to-line voltage, the range of duty
-
8 The Scientific World Journal
Table 1: Determination of π andπ based on the input voltage sector.
Sector π Uniformity π Uniformity
I-2 β sin (π½πβ 2π/3)
β3
2π· toπ· (β sin(π½
π+ 2π/3))/ sin(π½
πβ 2π/3) 1β0.5 down
II-2 β sin (π½πβ 2π/3)
β3
2π· toπ· (β sin(π½
π))/ sin(π½
πβ 2π/3) 0.5β1 up
II-1 sin (π½π)
β3
2π· toπ· (β sin(π½
πβ 2π/3))/ sin(π½
π) 1β0.5 down
III-1 sin (π½π)
β3
2π· toπ· (β sin(π½
π+ 2π/3))/ sin(π½
π) 0.5β1 up
III-2 β sin (π½π+ 2π/3)
β3
2π· toπ· (β sin(π½
π))/ sin(π½
π+ 2π/3) 1β0.5 down
IV-2 β sin (π½π+ 2π/3)
β3
2π· toπ· (β sin(π½
πβ 2π/3))/ sin(π½
π+ 2π/3) 0.5β1 up
IV-1 sin (π½πβ 2π/3)
β3
2π· toπ· (β sin(π½
π+ 2π/3))/ sin(π½
πβ 2π/3) 1β0.5 down
V-1 sin (π½πβ 2π/3)
β3
2π· toπ· (β sin(π½
π))/ sin(π½
πβ 2π/3) 0.5β1 up
V-2 sin (π½π)
β3
2π· toπ· (β sin(π½
πβ 2π/3))/ sin(π½
π) 1β0.5 down
VI-2 sin (π½π)
β3
2π· toπ· (β sin(π½
π+ 2π/3))/ sin(π½
π) 0.5β1 up
VI-1 sin (π½π+ 2π/3)
β3
2π· toπ· (β sin(π½
π))/ sin(π½
π+ 2π/3) 1β0.5 down
I-1 sin (π½π+ 2π/3)
β3
2π· toπ· (β sin(π½
πβ 2π/3))/ sin(π½
π+ 2π/3) 0.5β1 up
I II III IV V VI
0
1 1 1 1 1 12 2 2 2 2 2
Input voltage
πΌi
Figure 8: Interval voltage sector based on the input voltage (πΌπ).
ratio is changed during time intervals π1and π
2. The range
of available voltage ratio during the π1is higher than π
2.
Therefore, the desired value of π is varied from 0.5 to 1 ineach sector to reach the maximum power of input source. InTable 1, theπ· = V
π MX/(Vπ MX + Vπ POS) indicates the magnitudevariation of π, where V
π MX is nominal value of generator inputvoltage.
4. Simulation Result
Simulation of the proposed modulation method for MDMCis performed by using MATLAB software. The system has
Table 2: Simulation parameter.
π -πΏ load π = 5Ξ©, πΏ = 10mHInput voltage(line-to-neutral) π
π -RMS56V
Battery voltage(line-to-neutral) π
π -dcΒ±48V
Output voltage(line-to-neutral) π
π-RMS28V
Voltage ratio π 0.5Input frequency π
π 60Hz
Output frequency ππ
50Hz
been investigated to synthesize the voltage and current in twoconditions, when the input voltage of generatorπ
π -RMS is big-ger than the battery voltageπ
π -dc and vice versa.The switchingperiod π
π is assumed to be 200πs in all conditions.
4.1. Condition I. In this condition the line-to-neutral voltageof generatorπ
π -RMS is bigger than the battery voltageππ -dc andbatteryβs charging (SOC) is equal to 50%. Thus, the power isinjected from generator to the AC and DC loads. Simulationparameters for condition (I) are shown in Table 2.
In point of view of power transferring, when the batteryvoltage π
π -dc is less than the line to-neutral voltage of gen-erator π
π -RMS whole loadβs power demand is supplied by thegenerator. Thus, the time interval π
3is equal to zero and
voltage ratio is varied based on the value of generatorβsvoltage.
-
The Scientific World Journal 9
100
50
0
β50
β100
Switching pulse
οΏ½oR
ioR
0.00 0.01 0.02 0.03 0.04 0.05 0.06 0.07 0.08 0.09 0.10
10
5
0
β10
β5
AC o
utpu
t vol
tage
,E(V
)AC
out
put c
urre
nt,I
(A)
Time, t (s)
100
50
0
β50
β100
0.0190 0.0192 0.0194
Figure 9: AC voltage and current output waveform by voltagesynthesizing.
0.0190 0.0192 0.0194
10
5
0
οΏ½oST
ioS
0.00 0.01 0.02 0.03 0.04 0.05 0.06 0.07 0.08 0.09 0.10
Time, t (s)
100
150
50
0
100
150
50
0
DC
outp
ut v
olta
ge,E
(V)
DC
outp
ut cu
rren
t,I
(A)
Figure 10: DC voltage and current output waveform by voltagesynthesizing.
4.1.1. Voltage Synthesizing. In voltage synthesizing situation,the slopes of triangular waveforms are constant. Therefore,the time intervals are considered to be 100πs for π
1and
π2. Figure 9 illustrates the output line-to-neutral voltage V
ππ
which is connected to the AC load and AC output current πππ .
Figure 10 represents the output line-to-line voltage Vπππ
which
0
10
5
0
β10
β5
Inpu
t vol
tage
,E(V
)
Inpu
t cur
rent
,I(A
)
0.01 0.02 0.03 0.04 0.05 0.06 0.07 0.08 0.09 0.10
Time, t (s)
100
50
β50
β100
οΏ½sa
isa
Figure 11: Input voltage and current waveform by voltage synthesiz-ing.
100
50
0
β50
β100
Switching pulse
οΏ½oR
ioR
0.00 0.01 0.02 0.03 0.04 0.05 0.06 0.07 0.08 0.09 0.10
10
5
0
β10
β5
AC o
utpu
t vol
tage
,E(V
)AC
out
put c
urre
nt,I
(A)
Time, t (s)
100
50
0
β50
β100
0.0190 0.0192 0.0194
Figure 12: AC voltage and current output waveform by currentsynthesizing.
is connected to the DC loads and DC output current πππ . The
red line in the expand graph indicated the switching pulsewaveform.
Figure 11 shows the input phase voltage Vπ π
and filteredinput current π
π πin steady state. The simulation results
revealed that the proposed modulation method is capable ofsynthesizing the output voltages while the distortion on theinput current is visible in Figure 11.
4.1.2. Current Synthesizing. By changing the slope of thetriangular carriers which is related to the value of π, powerfactor has been controlled and the sinusoidal input currentshave been synthesised. Figures 12 and 13 show the output ACoutput voltage/current waveform and DC output voltage/current waveform, respectively.
Figure 14 shows the input phase voltage Vπ π
and filteredinput current π
π πin current synthesizing mode. The simula-
tion result in Figure 14 shows that the current input is wellmerged when there is no additional distortion in outputvoltage.The simulation result for outputwaveform also shows
-
10 The Scientific World Journal
0.01900.01920.01940.0196
10
5
0
οΏ½oST
ioS
0.00 0.01 0.02 0.03 0.04 0.05 0.06 0.07 0.08 0.09 0.10
Time, t (s)
100
150
50
0
100
150
50
0DC
outp
ut v
olta
ge,E
(V)
DC
outp
ut cu
rren
t,I
(A)
Figure 13: DC voltage and current output waveform by currentsynthesizing.
0
10
5
0
β10
β5
Inpu
t vol
tage
,E(V
)
Inpu
t cur
rent
,I(A
)
0.01 0.02 0.03 0.04 0.05 0.06 0.07 0.08 0.09 0.10
Time, t (s)
100
50
β50
β100
οΏ½sa
isa
Figure 14: Input voltage and current waveform by current synthe-sizing.
that the power factor has been controlled and unity powerfactor almost achieved.
4.2. Condition II. In the second condition when the ππ -RMS
is less than battery voltage (ππ -dc) and state of charge of
batteryβs is SOC = 95%, power will be injected from generatorand battery to the AC load at the same time. The switchingsequence is assumed to be π
π = 200 ππ . All simulation
parameters are the same as parameters in Table 2, exceptthe input voltage (line-to-neutral) π
π -RMS, which is equal to20πRMS in this condition. In condition II, the voltage ratio (π)is independent of the generator input voltage. The power ofthe generator which is transferred to the load will be deter-mined by the differences between π
π -RMS and ππ -dc.
4.2.1. Voltage Synthesizing. In this mode, the slopes of trian-gular waveforms are constant. Therefore, the time intervalsare considered to be π
1= π2= 50 πs, π
3= 100 πs. Figure 15
illustrated the average output line-to-neutral voltage Vππ
which is connected to the AC load, π -phase output current
οΏ½saisa
ioR
οΏ½oR
0.01 0.02 0.03 0.04 0.05 0.06 0.07 0.08 0.09 0.10
30
15
0
β15
β30
30
0
β30
3
0
β3
10
5
0
β5
β10
Time, t (s)
Out
put c
urre
nt,I
(A)
Inpu
t cur
rent
,I(A
)
Out
put v
olta
ge,E
(V)
Inpu
t vol
tage
,E(V
)
Figure 15: Simulation result by voltage synthesizing.
οΏ½sa
isa
ioR
οΏ½oR
30
15
0
0
β15
β30
30
15
β15
β30
3
0
1.5
β3
β1.5
10
5
0
β5
β100.01 0.02 0.03 0.04 0.05 0.06 0.07 0.08 0.09 0.10
Time, t (s)
Out
put c
urre
nt,I
(A)
Inpu
t cur
rent
,I(A
)
Out
put v
olta
ge,E
(V) I
nput
vol
tage
,E(V
)
Figure 16: Simulation result by current synthesizing.
πππ , input phase voltage V
π π, and filtered input current π
π π.
The simulation results showed that the proposed modulationmethod is able to well synthesize the output voltages, whilethe distortion on the input current is visible.
4.2.2. Current Synthesising. By changing the slope of the tri-angular carriers (π andπ) based on Table 2, power factor hasbeen controlled and the sinusoidal input currents have beensynthesised. The simulation result in Figure 16 shows thatthe current input is well merged when there is no additionaldistortion in output voltage. The result indicated that thepower factor has been controlled and unity power factoralmost achieved.
4.3. Current Control. In order to test the system stability,reference current is changed in AC and DC sides while theloads are constant. Figure 17 shows the simulated responsesofMDMC systemwhen the current is changed in AC andDCside at π‘ = 0.05 and π‘ = 0.07 s, respectively.
Figure 17 shows that the current in πππand πππ
is constantat π‘ = 0.05 regardless of the changing in π
ππ which is
increased by 0.2 pu. In addition, when the DC reference
-
The Scientific World Journal 11
ioRioS
ioT
0.00 0.01 0.02 0.03 0.04 0.05 0.06 0.07 0.08 0.09 0.10
6
3
0
β3
β6Out
put c
urre
nt,I
(A)
Time, t (s)
Figure 17: Output currents waveform during load variation.
current reduced by 0.2 pu, the current in AC side remainsconstant at π‘ = 0.07 s.
It can be clearly seen from Figure 17 that the system isable to track the variation of reference current, and currentoutput of each terminals is completely independent of otheroutput current terminals. Moreover, the simulation resultsexhibited that the undershoot/overshoot and steady-stateerror for output currents is acceptable for low power batterybased system.
5. Conclusions
This study represents a structure for multidirectional matrixconverter which is suitable for battery based system.This newMDMC configuration reduces the cost and size of the system.This study also presents a novel DDPWMmethod to controlthe power flow from generator and battery to the load. Thisnewmodulationmethod used the concept of the average volt-age per switching period and a continuous carrier formultidi-rectionalmatrix converters. By applying the proper switchingpattern and determining the duty ratio for each switch, thevoltage of each output terminal has been well synthesized. Inaddition, by changing the slope of the carriers whichis related to the value of π andπ, power factor has been con-trolled and the sinusoidal input currents have been synthe-sised.
The feasibility of the proposed MDMC structure andEDDPWM technique has been verified by MATLAB simula-tion. Results of this study revealed that the proposed carrierbased modulation technique can be used for the applicationwhere battery is essential. This method can easily control thepower factor andmerged the output voltage and input currentwithout any lookup tables. Since this new modulation hasa good flexibility and applicability, it can be effectivelyapplied in a system with a connection between the input andoutput neutrals with a desired output voltage and frequency.Furthermore, battery discharging time has been increased byletting power flow from both generator and battery to theload, simultaneously.The simulation results of this study alsorevealed that the current response in AC and DC of theMDMC system with proposed EDDPWMmethod is accept-able for low power battery based systems.
Conflict of Interests
The authors declare that there is no conflict of interestsregarding the publication of this paper.
References
[1] S. Bhattacharyya, Rural Electrification Through DecentralisedOff-Grid Systems in Developing Countries, Green Energy andTechnology, Springer, London, UK, 2013.
[2] S. Colle, S. Luna Abreu, and R. RuΜther, βEconomic evaluationand optimization of hybrid diesel/photovoltaic systems inte-grated to utility grids,β Solar Energy, vol. 76, no. 1β3, pp. 295β299,2004.
[3] M. J. Khan and M. T. Iqbal, βPre-feasibility study of stand-alone hybrid energy systems for applications inNewfoundland,βRenewable Energy, vol. 30, no. 6, pp. 835β854, 2005.
[4] P. K. Goel, B. Singh, S. S. Murthy, and N. Kishore, βIsolatedwind-hydro hybrid system using cage generators and batterystorage,β IEEE Transactions on Industrial Electronics, vol. 58, no.4, pp. 1141β1153, 2011.
[5] P. W. Wheeler, J. RodrΜΔ±guez, J. C. Clare, L. Empringham, and A.Weinstein, βMatrix converters: a technology review,β IEEETran-sactions on Industrial Electronics, vol. 49, no. 2, pp. 276β288,2002.
[6] M. Venturini and A. Alesina, βhe generalised transformerβanew bidirectional sinusoidal waveform frequency converterwith continuously adjustable input power factor,β in Proceedingsof the Power Electronics Specialists Conference (PESC β80), pp.242β252, 1980.
[7] A. Alesina and M. G. B. Venturini, βSolid-state power conver-sion: a fourier analysis approach to generalized transformersynthesis,β IEEE Transactions on Circuits and Systems, vol. 28,no. 4, pp. 319β330, 1981.
[8] A. Alesina and M. G. B. Venturini, βAnalysis and design ofoptimum-amplitude nine-switch direct AC-AC converters,βIEEE Transactions on Power Electronics, vol. 4, no. 1, pp. 101β112,1989.
[9] J. Rodriguez, βA new control technique for AC-AC converters,βin Proceedings of the IFAC Control in Power Electronics andElectrical Drives Conference, pp. 203β208, 1983.
[10] D. Casadei, G. Serra, A. Tani, and L. Zarri, βMatrix convertermodulation strategies: a new general approach based on space-vector representation of the switch state,β IEEE Transactions onIndustrial Electronics, vol. 49, no. 2, pp. 370β381, 2002.
[11] C. Klumpner, P. Nielsen, I. Boldea, and F. Blaabjerg, βA newmatrix convertermotor (MCM) for industry applications,β IEEETransactions on Industrial Electronics, vol. 49, no. 2, pp. 325β335,2002.
[12] K. S. Muhammad, R. Baharom, and M. K. Hamzah, βAdvancebattery charger topology using SPMC incorporating activepower filter,β in Proceeding of the International Conference onScience and Social Research (CSSR β10), pp. 548β553, KualaLumpur, Malaysia, December 2010.
[13] N. Nguyen-Quang, D. A. Stone, C. M. Bingham, and M. P.Foster, βA three-phase to single-phase matrix converter forhigh-frequency induction heating,β in Proceedings of the 13thEuropean Conference on Power Electronics and Applications(EPE β09), pp. 1β10, September 2009.
[14] S. Ferreira Pinto and J. Fernando Silva, βInput filter design forslidingmode controlledmatrix converters,β inProceedings of theIEEE 32nd Annual Power Electronics Specialists Conference(PESC β01), pp. 648β653, June 2001.
[15] J.-K. Kang, H. Hara, E. Yamamoto, E. Watanabe, and J. Oyama,βInput Harmonics Damping for High Power Quality Appli-cations of Matrix Converter,β in Proceedings of the 38th IAS
-
12 The Scientific World Journal
Annual Meeting. Conference Record of the Industry ApplicationsConference, pp. 1498β1503, October 2003.
[16] F. Blaabjerg, D. Casadei, C. Klumpner, and M. Matteini, βCom-parison of two current modulation strategies for matrix con-verters under unbalanced input voltage conditions,β IEEETransactions on Industrial Electronics, vol. 49, no. 2, pp. 289β296, 2002.
[17] L. Wei, Y. Matsushita, and T. A. Lipo, βInvestigation of dual-bridge matrix converter operating under unbalanced sourcevoltages,β in Proceedings of the IEEE 34th Annual Power Elec-tronics Specialists Conference (PESC β03), pp. 1293β1298, June2003.
[18] Y.-D. Yoon and S.-K. Sul, βCarrier-based modulation methodfor matrix converter with input power factor control and underunbalanced input voltage conditions,β in Proceedings of the 22ndAnnual IEEE Applied Power Electronics Conference and Exposi-tion (APEC β07), pp. 310β314, March 2007.
[19] O. Simon, J. Mahlein, M. N. Muenzer, and M. Bruckmann,βModern solutions for industrial matrix-converter applica-tions,β IEEE Transactions on Industrial Electronics, vol. 49, no. 2,pp. 401β406, 2002.
[20] D. Casadei, G. Serra, A. Tani, and L. Zarri, βStability analysis ofelectrical drives fed by matrix converters,β in Proceedings of theIEEE International Symposium on Industrial Electronics (ISIEβ02), pp. 1108β1113, 2002.
[21] D. Casadei, G. Serra, A. Tani, and L. Zarri, βEffects of inputvoltage measurement on stability of matrix converter drivesystem,β IEE Proceedings: Electric Power Applications, vol. 151,no. 4, pp. 487β497, 2004.
[22] Y. Li, N. S. Choi, B. M. Han, K. M. Kim, B. Lee, and J. H. Park,βDirect duty ratio pulse width modulation method for matrixconverters,β International Journal of Control, Automation andSystems, vol. 6, no. 5, pp. 660β669, 2008.
[23] Y.-D. Yoon and S.-K. Sul, βCarrier-based modulation techniquefor matrix converter,β IEEE Transactions on Power Electronics,vol. 21, no. 6, pp. 1691β1703, 2006.
[24] K. T. Chau and C. C. Chan, βEmerging energy-efficient tech-nologies for hybrid electric vehicles,β Proceedings of the IEEE,vol. 95, no. 4, pp. 821β835, 2007.
[25] Y. Li and N.-S. Choi, βCarrier based pulse widthmodulation formatrix converters,β in Proceedings of the 24th Annual IEEEApplied Power Electronics Conference and Exposition (APECβ09), pp. 1709β1715, February 2009.
[26] D. Liu and H. Li, βA ZVS bi-directional DC-DC converter formultiple energy storage elements,β IEEE Transactions on PowerElectronics, vol. 21, no. 5, pp. 1513β1517, 2006.
[27] M. Qiang, W.-Y. Wu, and Z.-L. Xu, βA multi-directional powerconverter for a hybrid renewable energy distributed generationsystem with battery storage,β in CES/IEEE 5th InternationalPower Electronics and Motion Control Conference (IPEMC β06),vol. 3, pp. 1β5, Shanghai, China, 2006.
[28] S. Toosi, M. Norhisam, T. Hanamoto, I. B. Aris, M. A. M. Radzi,and H. Yamada, βConfigurations comparison of multi direc-tional matrix converter for low power application,β Journal ofEngineering Science and Technology, vol. 9, pp. 1β10, 2014.
[29] S. Ashok, βOptimised model for community-based hybridenergy system,β Renewable Energy, vol. 32, no. 7, pp. 1155β1164,2007.
[30] L. Barote, R. Weissbach, R. Teodorescu, C. Marinescu, and M.Cirstea, βStand-alonewind systemwith vanadium redox battery
energy storage,β in Proceedings of the 11th International Con-ference on Optimization of Electrical and Electronic Equipment(OPTIM β08), pp. 407β412, Bras,ov, Romania, May 2008.
[31] E. Koutroulis and K. Kalaitzakis, βDesign of a maximum powertracking system for wind-energy-conversion applications,βIEEE Transactions on Industrial Electronics, vol. 53, no. 2, pp.486β494, 2006.
[32] J. W. Stevens and G. P. Corey, βStudy of lead-acid batteryefficiency near top-of-charge and the impact on PV systemdesign,β in Proceedings of the 25th IEEE Photovoltaic SpecialistsConference, pp. 1485β1488, May 1996.