Multistring Five-Level Inverter With Novel PWM

IEEE TRANSACTIONS ON INDUSTRIAL ELECTRONICS, VOL. 57, NO. 6, JUNE 2010 2111

Multistring Five-Level Inverter With Novel PWMControl Scheme for PV Application

Nasrudin A. Rahim, Senior Member, IEEE, and Jeyraj Selvaraj

Abstract—This paper presents a single-phase multistringfive-level photovoltaic (PV) inverter topology for grid-connectedPV systems with a novel pulsewidth-modulated (PWM) controlscheme. Three PV strings are cascaded together in parallel con-figuration and connected to a five-level inverter to produce outputvoltage in five levels: zero, +1/2Vdc, Vdc, −1/2Vdc, and −Vdc.Two reference signals that were identical to each other with an off-set that was equivalent to the amplitude of the triangular carriersignal were used to generate PWM signals for the switches. DSPTMS320F2812 is used to implement this PWM switching schemetogether with a digital proportional–integral current control al-gorithm. The inverter offers much less total harmonic distortionand can operate at near-unity power factor. The validity of theproposed inverter is verified through simulation and implementedin a prototype. The experimental results are compared with aconventional single-phase multistring three-level grid-connectedPWM inverter.

Index Terms—Grid-connected, multilevel inverter, multistring,photovoltaic (PV), pulsewidth-modulated (PWM) inverter,proportional–integral (PI) current control.

I. INTRODUCTION

A S THE WORLD is concerned with fossil-fuel exhaus-tion and environmental problems caused by conventional

power generation, renewable energy sources, particularly solarand wind energy, have become very popular and demanding.Photovoltaic (PV) sources are used today in many applicationsbecause they have the advantages of being maintenance andpollution free [1]. Solar-electric-energy demand has grown con-sistently by 20%–25% per annum over the past 20 years, whichis mainly due to the decreasing costs and prices. This declinehas been driven by the following: 1) an increasing efficiencyof solar cells; 2) manufacturing-technology improvements; and3) economies of scale [2]. A PV inverter, which is an importantelement in the PV system, is used to convert dc power from thesolar modules into ac power to be fed into the grid.

A general overview of different types of PV inverters isgiven in [3] and [4]. This paper presents a multistring five-levelinverter for PV application. The multistring inverter shown inFig. 1 is a further development of the string inverter, whereseveral strings are interfaced with their own dc–dc converter toa common dc–ac inverter [5]. This is beneficial, compared with

Manuscript received February 23, 2009; revised May 26, 2009 andJuly 20, 2009; accepted August 26, 2009. Date of publication October 20, 2009;date of current version May 12, 2010.

The authors are with the Center of Research for Power Electronics, Drives,Automation and Control, Department of Electrical Engineering, Faculty ofEngineering, University of Malaya, Kuala Lumpur 50603, Malaysia (e-mail:[email protected]).

Color versions of one or more of the figures in this paper are available onlineat http://ieeexplore.ieee.org.

Digital Object Identifier 10.1109/TIE.2009.2034683

Fig. 1. Configuration of multistring inverters.

Fig. 2. Carrier and reference signals.

the centralized system, because every string can be controlledindividually. Thus, the operator may start his/her own PVpower plant with a few modules. Further enlargements areeasily achieved because a new string with a dc–dc convertercan be plugged into the existing platform. A flexible designwith high efficiency is hereby achieved [3]. In this paper, afive-level inverter is used instead of a conventional three-levelpulsewidth-modulated (PWM) inverter because it offers greatadvantages, such as improved output waveforms, smaller filtersize, lower electromagnetic interference, lower total harmonicdistortion (THD), and others [6]–[12].

This paper proposes a single-phase multistring five-levelinverter topology. It consists of three strings of PV arrays con-nected to their own dc–dc boost converter. An auxiliary circuitcomprising four diodes and a switch is configured together witha conventional full-bridge inverter to form this topology. Anovel PWM control scheme is introduced to generate switchingsignals for the switches and to produce five output-voltagelevels: zero, +1/2Vdc, Vdc, −1/2Vdc, and −Vdc (assuming thatVdc is the supply voltage). This inverter topology uses tworeference signals instead of one to generate PWM signals forthe switches. Both reference signals Vref1 and Vref2 are identicalto each other, except for an offset value that is equivalent to theamplitude of carrier signal Vcarrier, as shown in Fig. 2.

Because the inverter is used in a PV system, aproportional–integral (PI) current control scheme is employed

0278-0046/$26.00 © 2010 IEEE

2112 IEEE TRANSACTIONS ON INDUSTRIAL ELECTRONICS, VOL. 57, NO. 6, JUNE 2010

Fig. 3. Single-phase multistring five-level inverter topology.

to keep the output current sinusoidal and to have high dynamicperformance under rapidly changing atmospheric conditionsand to maintain the power factor at near unity. Simulation andexperimental results are presented to validate the proposedinverter configuration.

II. MULTISTRING FIVE-LEVEL INVERTER TOPOLOGY

The proposed single-phase multistring five-level invertertopology is shown in Fig. 3. It consists of three dc–dc boostconverters connected to a common dc bus, an auxiliary circuit,and a full-bridge inverter configuration. Input sources, PVstring 1, PV string 2, and PV string 3 are connected to theinverter via the dc–dc boost converters. Because the proposedinverter is used in a grid-connected PV system, the utility gridis used instead of a load. The dc–dc boost converters are usedto track the maximum power point (MPP) independently and tostep up inverter output voltage Vinv to be more than

√2 of grid

voltage Vg to ensure power flow from the PV arrays into thegrid [13], [14]. As a step-up transformer with a ratio of 1 : 2 isused, Vinv should be

Vinv >

√2Vg

2(1)

or

Vinv >Vg√

2. (2)

Therefore, the dc-bus voltage is assumed to be approximately200 V.

In this paper, the multistring approach is adopted becauseeach dc–dc converter can independently perform MPP tracking(MPPT) for its PV strings. This will compensate for mis-matches in panels of like manufacture, which can be up to 2.5%[15]. It offers the further advantage of allowing panels to begiven different orientations and so open up new possibilities inarchitectural applications. Furthermore, a greater tolerance tolocalized shading of panels can be achieved. Another advantageof multistring configuration is that the mixing of differentsources becomes possible, i.e., existing PV panel strings couldbe extended by adding new higher output panels without com-promising the overall string reliability or performance. Otherthan that, greater safety during installation and maintenanceadds to the advantages of multistring configuration. Dependingon the design, each converter module may be able to isolateits connected power source so that the wiring of series or

parallel connection of these strings can be performed safely.The power-source-converter connection is a safe low-voltageconnection [16].

The dc–dc boost converters are connected in parallel to avoidhigh dc-bus voltage, which will eventually increase the sizeof the capacitors and the inverter’s cost. Therefore, only twocapacitors with equal capacitance rating are used as the dc bus,and the other dc–dc boost converters are connected to this dcbus, as shown in Fig. 3.

A filtering inductance Lf is used to filter the current injectedinto the grid. The injected current must be sinusoidal with lowharmonic distortion. In order to generate sinusoidal current, asinusoidal PWM is used because it is one of the most effectivemethods. A sinusoidal PWM is obtained by comparing a high-frequency carrier signal with a low-frequency sinusoid signal,which is the modulating or reference signal. The carrier has aconstant period; therefore, the switches have constant switchingfrequency. The switching instant is determined from the cross-ing of the carrier and the modulating signal.

III. PWM MODULATION

Modulation index Ma for a five-level PWM inverter is givenas [17]

Ma =Am

2Ac(3)

where Ac is the peak-to-peak value of carrier and Am is thepeak value of voltage reference Vref . Because, in this paper,two reference signals that are identical to each other are used,(3) can be expressed in terms of the amplitude of carrier signalVc by replacing Ac with Vc, and Am = Vref1 = Vref2 = Vref

M =Vref

2Vc. (4)

If M > 1, a higher harmonic in the phase waveform is ob-tained. Therefore, M is maintained between zero and one. Ifthe amplitude of the reference signal is increased higher thanthe amplitude of the carrier signal, i.e., M > 1, this will leadto overmodulation. Large values of M in sinusoidal PWMtechniques lead to full overmodulation [18]. Fig. 4 shows thecarrier and reference signals for different values of M .

From the PWM modulation, the analysis of harmonic com-ponents in the proposed inverter can be preformed. The outputvoltage produced by comparison of the two reference signalsand the carrier signal can be expressed as [7]

Vo(θ) = A0 +∞∑

n=1

(An cos nθ + Bn sinnθ). (5)

If there are P pulses per quarter period, and it is an oddnumber, coefficients Bn and Ao would be zero, where n is aneven number. Therefore, the (5) can be rewritten as

Vo(θ) =∞∑

n=1,3,...

An cos nθ (6)

An = − 2Vdc

nπ

P∑m=0

4∑i=1

[(−1)int(i/2) sin(nαm+i)

](7)

RAHIM AND SELVARAJ: FIVE-LEVEL INVERTER WITH NOVEL PWM CONTROL SCHEME FOR PV APPLICATION 2113

Fig. 4. Carrier and reference signals for different values of modulationindex (M). (a) M = 0.3. (b) M = 0.5. (c) M = 0.7. (d) M = 1.2.

where m is a pulse number and α is the phase displacementangle. The Fourier series coefficients of the conventional single-phase full-bridge inverter by sinusoidal PWM is given as

An =4Vdc

nπ

P∑m=1

[(−1)m sin(nαm)] . (8)

IV. OPERATING PRINCIPLES OF MULTISTRING

FIVE-LEVEL INVERTER

Combinations of PV strings are used as the input voltagesources. The voltage across the strings are known as Vpv1, Vvp2,and Vpv3. Referring to (1) and (2), Vpv1, Vvp2, and Vvp3 areboosted by the dc–dc boost converters to exceed grid voltageVg , and the voltage across the dc bus is known as Vpv. Theoperating principle of the proposed inverter is to generate fiveoutput-voltage levels, i.e., 0, +Vpv/2, +Vpv, −Vpv/2, −Vpv, asin Fig. 5. As shown in Fig. 3, an auxiliary circuit that consistsof four diodes and a switch S4 is used between the dc-buscapacitors and the full-bridge inverter. Proper switching control

Fig. 5. Inverter output voltage (Vinv) and switching pattern for the single-phase five-level inverter.

TABLE IINVERTER OUTPUT VOLTAGE DURING S4–S8 SWITCH ON AND OFF

of the auxiliary circuit can generate half level of PV supplyvoltage, i.e., +Vpv/2 and −Vpv/2 [7].

Two reference signals Vref1 and Vref2 will take turns to becompared with the carrier signal at a time. If Vref1 exceedsthe peak amplitude of carrier signal Vcarrier, then Vref2 will becompared with the carrier signal until it reaches zero. At thispoint onward, Vref1 takes over the comparison process untilit exceeds Vcarrier. This will lead to a switching pattern, asshown in Fig. 5. Switches S4–S6 will be switching at the rateof the carrier signal frequency, while S7 and S8 will operateat a frequency that is equivalent to the fundamental frequency.Table I illustrates the level of Vinv during S4–S8 switch onand off.

If one of the PV strings is disconnected from the dc bus,the operation of the other dc–dc boost converters will not beaffected because they are connected in parallel. As the dc–dcboost converters are used to track the MPPT point, it can beconcluded that the MPPT of the PV strings is done indepen-dently. Later expansion of the PV strings is also possible byadding a dc–dc boost converter, as shown in Fig. 6.

V. CONTROL SYSTEM ALGORITHM AND IMPLEMENTATION

One of the problems in the PV generation systems is that theamount of electric power generated by the solar arrays is alwayschanging with weather conditions, i.e., the intensity of solarradiation. An MPPT method or algorithm, which has quick-response characteristics and is able to make good use of the


Fig. 6. PV string extension for existing configuration.

electric power generated in any weather, is needed to solve theaforementioned problem [19]. Various MPPT control methodshave been discussed in detail in [20]–[22].

In this paper, the perturb-and-observe algorithm is used toextract maximum power from the PV arrays and deliver it to theinverter. The feedback controller used for the inverter is the PIalgorithm. As shown in Fig. 7, the current injected into the grid,which is also known as grid current Ig , is sensed and fed backto a comparator, which compares it with reference current Iref .Iref is obtained by sensing utility grid voltage Vg . The sensed Vg

signal is converted into a reference signal before it is multipliedwith variable m. Therefore

Iref = Vg × m. (9)

Variable m is the sum of m1, m2, and m3, i.e.,

m = m1 + m2 + m3. (10)

Variables m1, m2, and m3 are obtained from the MPPT algo-rithm, as shown in the flowchart in Fig. 8. Variables m1, m2,and m3 correspond to the MPPT algorithm for strings 1, 2,and 3, respectively. The values of m1, m2, and m3 change withrespect to the irradiance level. If the irradiance level is high, thecorresponding values of m1, m2, and m3 are also high. Thus,by referring to (10), it will lead to high value of m. BecauseIref is proportional to m, a high value of Iref is obtained. As aresult, the inverter’s output current Ig will be high as it followsIref to minimize the instantaneous error between Ig and Iref .

The instantaneous current error is fed to a PI controller.The integral term in the PI controller improves the trackingby reducing the instantaneous error between the reference andthe actual current. The resulting error signal u, which formsVref1 and Vref2, is compared with a triangular carrier signal,and intersections are sought to produce PWM signals for the

inverter switches. This is to ensure that Ig is in phase with gridvoltage Vg and always at near-unity power factor.

A. Mathematical Formulation

The PI algorithm can be expressed in the continuous timedomain as

u(t) = Kpe(t) + Ki

t∫

τ=0

e(τ) dτ (11)

where

u(t) control signal;e(t) error signal;t continuous-time-domain time variable;τ calculus variable of integration;Kp proportional mode control gain.Ki is the integral mode control gain

Implementing this algorithm by using a DSP requires oneto transform it into a discrete-time domain. Trapezoidal sumapproximation is used to transform the integral term into adiscrete-time domain because it is the most straightforwardtechnique. The proportional term is directly used without ap-proximation

P term: Kpe(t) =Kpe(k) (12)

I term:

Ki

t∫

τ=0

e(τ) dτ ∼=Ki

k∑i=0

h

2[e(i) + e(i − 1)] . (13)

Time relationship: t = k ∗ h, where h is the sampling periodand k is the discrete-time index, with k = 0, 1, 2, . . .. Forsimplification, it is convenient to define the new controllergains as

K ′i = Ki

h

2(14)

from which one can construct the discrete-time PI controllaw as

u(k) = Kpe(t) + K ′i

k∑i=0

[e(i) + e(i − 1)] . (15)

To eliminate the need to calculate the full summation at eachtime step (which would require an ever-increasing amount ofcomputation as time goes on), the summation is expressed as arunning sum

sum(k) = sum(k − 1) + [e(k) + e(k − 1)] (16)

u(k) =Kpe(k) + K ′isum(k). (17)

These two equations, which represent the discrete-time PIcontrol law, are implemented in DSP TMS320F2812 to controlthe overall operation of the inverter.


Fig. 7. Multistring five-level inverter with control algorithm implemented in DSP TMS320F2812.

Fig. 8. MPPT flowchart.

B. Algorithm Implementation

Control signal saturation and integral-mode antiwindup lim-iting are easily implemented in software. In this paper, thecontrol signal itself takes the form of PWM outputs from theDSP. Therefore, the control signal is saturated at the value thatcorresponds to 100% duty cycle for the PWM. An undesirableside effect of saturating the controller output is the integral-mode windup. When the control output saturates, the integral-

Fig. 9. PI control algorithm implemented in DSP TMS320F2812.

mode control term (i.e., the summation) will continue to in-crease but will not produce a corresponding increase in thecontroller output (and hence will not produce any additionalincrease in the plant response). The integral can become quitelarge, and it can take a long time before the controller is ableto reduce it once the error signal changes sign. The effect ofwindup on the closed-loop output is larger transient overshootand undershoot and longer settling times.

One approach for overcoming the integral-mode windup isto simply limit in the software the maximum absolute valueallowed for the integral, independent of the controller outputsaturation [23], as shown in Fig. 9.

VI. SIMULATION AND EXPERIMENTAL RESULTS

A. Simulation Results

Simulations were performed by using MATLAB/Simulink toverify that the proposed inverter can practically be implementedin a PV system. It helps one to confirm the PWM switch-ing strategy for the multistring five-level inverter. Then, this


Fig. 10. PWM switching strategy and PWM signal for S4–S8.

strategy is implemented in a real-time environment, i.e., theDSP to produce PWM switching signals for the switches.Fig. 10(a) shows the way the PWM switching signals aregenerated by using two reference signals and a triangular carriersignal. The resulting PWM signals for switches S4–S8 areshown in Fig. 10(b)–(f).

Note that one leg of the inverter is operating at a highswitching rate that is equivalent to the frequency of the carriersignal, while the other leg is operating at the rate of fundamentalfrequency (i.e., 50 Hz). The switch at the auxiliary circuit (S4)also operates at the rate of the carrier signal. As mentionedearlier, modulation index M will determine the shape of in-verter output voltage Vinv and grid current Ig. Fig. 11 showsthe simulation results of Vinv and Ig for different values of M .

Referring to (1) and (2), the dc-bus voltage is set to 200 V(> Vg/

√2; in this case, Vg is 240 V) to inject current into the

grid. Fig. 11(a) shows that Vinv is less than Vg/√

2 because Mis less than 0.5. The inverter should not operate at this conditionbecause the current will be injected from the grid into theinverter, as shown in Fig. 11(b). The overmodulation condition,

which happens when M > 1.0, is shown in Fig. 11(c). It has aflat top at the peak of the positive and negative cycles becauseboth reference signals exceed the maximum amplitude of thecarrier signal. This will cause Ig to have a flat portion at thepeak of the sine waveform, as shown in Fig. 11(d). To optimizethe power transferred from the PV arrays to the grid, it isrecommended to operate at 0.5 ≤ M ≤ 1.0. Vinv and Ig for theoptimal operating condition are shown in Fig. 11(e) and (f). AsIg is almost a pure sinewave, THD can be reduced comparedwith that under other values of M .

B. Experimental Results

The simulation results are verified experimentally by usingDSP TMS320F2812. Three PV strings with different types ofsolar modules and locations are connected to the five-level in-verter via a common dc bus. Table II illustrates the PV modules’characteristics and their location, while Table III shows themultistring five-level inverter’s specifications and its controllerparameters. The prototype inverter is shown in Fig. 12. PWM


Fig. 11. Inverter output voltage (Vinv) and grid current (Ig) for different values of M . (a) Vinv for M < 0.5. (b) Ig for M < 0.5. (c) Vinv for M > 1.0.(d) Ig for M > 1.0. (e) Vinv for 0.5 ≤ M ≤ 1.0. (f) Ig for 0.5 ≤ M ≤ 1.0.

switching signals for the switches are generated by comparinga triangular carrier signal with two reference signals, as shownin Fig. 13.

Code Composer Studio (CCS), the programming platformfor DSP TMS320F2812, programs the control algorithm forthe proposed multistring five-level inverter. CCS offers theadvantage of graph displaying, which can be used to investigatethe results as in Figs. 14–18. As the input voltage of each stringis converted into floating-point values for DSP manipulation,the values corresponding to Figs. 14–18 do not represent theactual values of the input voltage but are good enough forinvestigation and analysis. Fig. 14 shows the input of PV strings1, 2, and 3. Here, PV string 1 is on, while PV string 3 is off.When PV string 2 is turned from off to on, m increases toincrease the amplitude of Iref because more power is generatedat the input, subsequently increasing the current injected intothe grid. As PV string 2 shuts down, its voltage goes to zero,while PV string 1 maintains its MPP, as shown in Fig. 15.m decreases to decrease the amplitude of Iref as the currentinjected into the grid is less because PV string 2 stops producingpower.

The same phenomena happen when PV string 1 is on, PVstring 2 is off, and PV string 3 is turned from off to on, asshown in Fig. 16. When PV string 3 is turned off, m decreasesas in Fig. 17 to decrease Iref . As a result, less current is injectedinto the grid compared with the previous condition when PVstring 3 was on. Fig. 18 is captured when all three strings are oninitially. Then, PV string 3 is turned off followed by PV string2. m decreases when PV string 3 is turned off, and it decreasesfurther when PV string 2 is turned off. This shows that thestrings are working independently and that later expansion ofthe strings is possible.

Fig. 19 shows the experimental results for grid voltage Vg

and the inverter’s output voltage Vinv. As the grid voltage hadbeen stepped down to half the actual voltage by using a 1 : 2-ratio transformer, the magnitude of Vg is now 120 V. To injectcurrent into the grid, Vinv >

√2Vg; Vinv is thus set at 200 V.

Figs. 20–22 show the experimental results for Vinv and Ig for8, 5, and 3 A, respectively. It can be seen that Vinv consists offive levels of output voltage, and Ig has been filtered to resemblea pure sinewave. The magnitude of Vinv did not change, butit maintained at 200 V as the current injected into the grid


TABLE IICHARACTERISTICS OF PV MODULES

TABLE IIIPV MULTISTRING FIVE-LEVEL INVERTER SPECIFICATIONS

AND CONTROLLER PARAMETERS

Fig. 12. Prototype of the multistring five-level PWM inverter.

decreased when the irradiance level decreased. Modulationindex M is 0.8. For M that is less than 0.5, Vinv is less thanVg/

√2. Therefore, current will be injected from the grid into

the inverter, as shown in Fig. 23. This condition should beavoided to protect the PV system from damage.

Fig. 13. PWM switching signals for S4–S8. (a) S4. (b) S5 and S6.(c) S7 and S8.

In the case of M being more than one, the results arenot shown because the PV system is designed to operate atconditions of M being less than one. This is done by calculatingthe input current and input voltage corresponding to the outputvoltage and output current. Then, M is varied accordinglyfor the inverter to operate at minimum- and maximum-powerconditions. Below the minimum-power condition (for example,during heavy clouds or nighttime) or above the maximum-power condition (for example, overrating of PV arrays, whichexceeds the rating of the inverter), the inverter should not oper-ate to ensure the safety of the PV system and the environment.

To prove that the proposed multistring five-level inverterhas advantages over the conventional multisting three-levelinverter in terms of THD and power factor, the corresponding


Fig. 14. Conditions during PV string 1’s being switched on and PV string 2’s being switched from off to on.

Fig. 15. Conditions during PV string 1’s being switched on and PV string 2’s being switched off.

measurements were made on both inverters. A FLUKE 43BPower Quality Analyzer was used for this purpose. The con-ventional multistring three-level inverter for grid-connected PVapplication is shown in Fig. 24. The same current controltechniques were used to control the overall performance ofthe inverter. The only difference between both inverters is the

elimination of the auxiliary circuit, and therefore, only one dc-bus capacitor is used. Fig. 25 shows the THD measurementfor the multistring five-level inverter, while Fig. 26 showsthe THD measurement for the multistring three-level inverter.The %THD for the five-level inverter is recorded as 5.7%,while the %THD for the three-level inverter is 9.5%. From


Fig. 16. Conditions during PV string 1’s being switched on and PV string 3’s being switched from off to on.

Fig. 17. Conditions during PV string 1’s being switched on and PV string 3’s being switched off.

both illustrations, the THD measurement for the multistringfive-level inverter is much lower than that for the multistringthree-level inverter. The power-factor measurement is shown inFig. 27. It is notable that both grid voltage Vg and the currentinjected into grid Ig are in phase with a power factor of 0.99.

Fig. 28 shows the relationship between Ig and the THDmeasurement. It shows that, as Ig increases, the THD decreases.Because Ig is increased by increasing modulation index M toforce more current injected into the grid, it can be concludedthat M is proportional to Ig .


Fig. 18. Conditions during PV string 1’s being switched on and PV strings 2’s and 3’s being switched off.

Fig. 19. Experimental results of Vg and Vinv for M = 0.8.

Fig. 20. Experimental results of Vinv and Ig at Ig = 8 A for M = 0.8.

Efficiency measurements were carried out to compare theefficiency of the multistring three-level PWM inverter withthe multistring five-level PWM inverter for PV application.Table IV illustrates the measured efficiency of both invertersoperating at different output powers. At 960- and 600-W op-



erating conditions, the measured efficiency of the multistringthree-level PWM inverter was approximately 90%, while themeasured efficiency for the multistring five-level PWM inverterwas 86%. For the 360-W operating condition, the efficiency


Fig. 23. Experimental results of Vinv and Ig for M = 0.2.

Fig. 24. Conventional multistring three-level PWM inverter for PVapplication.

Fig. 25. THD result of the multistring five-level PV inverter.

decreased to 89% and 84% for the three- and five-level PWMinverters, respectively.

As expected, the efficiency of the multistring five-level PWMinverter is lower compared to the conventional multistringthree-level PWM inverter. The main reason is the addition of

Fig. 26. THD result of the multistring three-level PV inverter.

Fig. 27. Grid voltage Vg and grid current Ig at near-unity power factor.

Fig. 28. Relationship between Ig and THD measurement.

TABLE IVMEASURED EFFICIENCY OF THREE- AND FIVE-LEVEL PWM INVERTERS

AT DIFFERENT OUTPUT POWER VALUES

the auxiliary circuit between the dc–dc boost converters andthe full-bridge inverter configuration. The switching losses ofswitch S4 in the auxiliary circuit caused the efficiency of the


multistring five-level PWM inverter to be approximately 4%less than that of the multistring three-level PWM inverter.However, the simulation and experimental results show that theTHD of the proposed inverter is lower compared to that of theconventional three-level PWM inverter, which is an importantelement for grid-connected PV systems.

VII. CONCLUSION

This paper has presented a single-phase multistring five-levelinverter for PV application. A novel PWM control scheme withtwo reference signals and a carrier signal has been used to gen-erate the PWM switching signals. The circuit topology, controlalgorithm, and operating principle of the proposed inverter havebeen analyzed in detail. The configuration is suitable for PVapplication as the PV strings operate independently and laterexpansion is possible. Furthermore, the experimental resultsindicate that the THD of the multistring five-level inverter ismuch less than that of the conventional multistring three-levelinverter. In addition, both grid voltage and grid current are inphase at near-unity power factor.

REFERENCES

[1] N. A. Rahim and S. Mekhilef, “Implementation of three-phase grid con-nected inverter for photovoltaic solar power generation system,” in Proc.IEEE PowerCon, Oct. 2002, vol. 1, pp. 570–573.

[2] J. M. Carrasco, L. G. Franquelo, J. T. Bialasiewicz, E. Galvan,R. C. PortilloGuisado, M. A. M. Prats, J. I. Leon, and N. Moreno-Alfonso,“Power-electronic systems for the grid integration of renewable energysources: A survey,” IEEE Trans. Ind. Electron., vol. 53, no. 4, pp. 1002–1016, Aug. 2006.

[3] S. B. Kjaer, J. K. Pedersen, and F. Blaabjerg, “A review of single-phasegrid connected inverters for photovoltaic modules,” IEEE Trans. Ind.Appl., vol. 41, no. 5, pp. 1292–1306, Sep./Oct. 2005.

[4] S. Daher, J. Schmid, and F. L. M. Antunes, “Multilevel inverter topologiesfor stand-alone PV systems,” IEEE Trans. Ind. Electron., vol. 55, no. 7,pp. 2703–2712, Jul. 2008.

[5] M. Meinhardt and G. Cramer, “Past, present and future of grid-connectedphotovoltaic- and hybrid-power-systems,” in Proc. IEEE-PES SummerMeeting, Jul. 2000, vol. 2, pp. 1283–1288.

[6] S. Kouro, J. Rebolledo, and J. Rodriguez, “Reduced switching-frequency-modulation algorithm for high-power multilevel inverters,” IEEE Trans.Ind. Electron., vol. 54, no. 5, pp. 2894–2901, Oct. 2007.

[7] S. J. Park, F. S. Kang, M. H. Lee, and C. U. Kim, “A new single-phase five-level PWM inverter employing a deadbeat control scheme,” IEEE Trans.Power Electron., vol. 18, no. 18, pp. 831–843, May 2003.

[8] L. M. Tolbert and T. G. Habetler, “Novel multilevel inverter carrier-basedPWM method,” IEEE Trans. Ind. Appl., vol. 35, no. 5, pp. 1098–1107,Sep./Oct. 1999.

[9] Y. Liu, H. Hong, and A. Q. Huang, “Real-time calculation of switchingangles minimizing THD for multilevel inverters with step modulation,”IEEE Trans. Ind. Electron., vol. 56, no. 2, pp. 285–293, Feb. 2009.

[10] N. S. Choi, J. G. Cho, and G. H. Cho, “A general circuit topology ofmultilevel inverter,” in Proc. IEEE 22th Annu. PESC, Jun. 24–27, 1991,pp. 96–103.

[11] G. Carrara, S. Gardella, M. Marchesoni, R. Salutari, and G. Sciutto, “Anew multilevel PWM method: A theoretical analysis,” IEEE Trans. PowerElectron., vol. 7, no. 3, pp. 497–505, Jul. 1992.

[12] R. Gonzalez, E. Gubia, J. Lopez, and L. Marroyo, “Transformerlesssingle-phase multilevel-based photovoltaic inverter,” IEEE Trans. Ind.Electron., vol. 55, no. 7, pp. 2694–2702, Jul. 2008.

[13] M. Calais, V. G. Agelidis, and M. S. Dymond, “A cascaded inverterfor transformerless single phase grid-connected photovoltaic systems,” inProc. IEEE 31st Annu. PESC, Jun. 18–23, 2001, vol. 3, pp. 1173–1178.

[14] J. Selvaraj and N. A. Rahim, “Multilevel inverter for grid-connectedPV system employing digital PI controller,” IEEE Trans. Ind. Electron.,vol. 56, no. 1, pp. 149–158, Jan. 2009.

[15] T. Noguchi, S. Togashi, and R. Nakamoto, “Short-circuit pulse-basedmaximum-power-point tracking method for multiple photovoltaic-and-converter system,” IEEE Trans. Ind. Electron., vol. 49, no. 1, pp. 217–223,Feb. 2002.

[16] G. R. Walker and P. C. Sernia, “Cascaded dc–dc converter connectionof photovoltaic modules,” IEEE Trans. Power Electron., vol. 19, no. 4,pp. 1130–1139, Jul. 2004.

[17] V. G. Agelidis, D. M. Baker, W. B. Lawrance, and C. V. Nayar, “Amultilevel PWM inverter topology for photovoltaic application,” in Proc.IEEE ISIE, Guimaraes, Portugal, 1997, pp. 589–594.

[18] M. H. Rashid, Power Electronics: Circuits, Devices, and Applications,3rd ed. Englewood Cliffs, NJ: Prentice-Hall, 2004, p. 267.

[19] N. Mutoh and T. Inoue, “A control method to charge series-connectedultraelectric double-layer capacitors suitable for photovoltaic generationsystems combining MPPT control method,” IEEE Trans. Ind. Electron.,vol. 54, no. 1, pp. 374–383, Feb. 2007.

[20] T. Esram and P. L Chapman, “Comparison of photovoltaic array maximumpower point tracking techniques,” IEEE Trans. Energy Convers., vol. 22,no. 2, pp. 439–449, Jun. 2007.

[21] N. Femia, G. Petrone, G. Spagnuolo, and M. Vitelli, “Optimizing duty-cycle perturbation of P&O MPPT technique,” in Proc. IEEE 35th Annu.PESC, Jun. 20–25, 2004, vol. 3, pp. 1939–1944.

[22] X. Liu and L. A. C. Lopes, “An improved perturbation and observationmaximum power point tracking algorithm for PV arrays,” in Proc. IEEE35th Annu. PESC, Jun. 20–25, 2004, vol. 3, pp. 2005–2010.

[23] D. M. Aletr, “Thermoelectric cooler control using TMS320F2812 DSPand a DRV592 power amplifier,” Texas Instrum., Dallas, TX.

Nasrudin A. Rahim (M’89–SM’08) was born inJohor, Malaysia, in 1960. He received the B.Sc.(Hons.) and M.Sc. degrees from the University ofStrathclyde, Glasgow, U.K., and the Ph.D. degree in1995 from Heriot–Watt University, Edinburgh, U.K.

He is currently a Professor with the Departmentof Electrical Engineering, Faculty of Engineering,University of Malaya, Kuala Lumpur, Malaysia,where he is also the Director of the Center of Re-search for Power Electronics, Drives, Automationand Control. His research interests include power

electronics, real-time control systems, and electrical drives.Dr. Rahim is a Fellow of the Institution of Engineering and Technology,

U.K., and a Chartered Engineer. He is also the Chairman of the Working GroupWG-8, covering reluctance motor, of the IEEE Motor Subcommittee under theIEEE Power Engineering Society/Electric Machinery Committee.

Jeyraj Selvaraj was born in Kedah, Malaysia, in1980. He received the B.Eng. (Hons.) degree fromMultimedia University, Cyberjaya, Malaysia, in2002, the M.Sc. degree in power electronics anddrives jointly from the University of Birmingham,Birmingham, U.K., and the University ofNottingham, Nottingham, U.K., in 2004, and thePh.D. degree from the University Malaya, KualaLumpur, Malaysia, in 2009.

He is currently with the Center of Research forPower Electronics, Drives, Automation and Control,

Department of Electrical Engineering, Faculty of Engineering, UniversityMalaya. His research interests are single- and three-phase multilevel invert-ers, digital PI current control techniques, photovoltaic inverters, and dc–dcconverters.

Multistring Five-Level Inverter With Novel PWM

Documents

Transcript of Multistring Five-Level Inverter With Novel PWM