INTRODUCTION

PV grid-connected system is one of the most commonly found distributed generations. Such a system typically comprises a photovoltaic generator (PVG), a DC-bus, a pulse width modulation (PWM) Inverter, a grid filter and a grid utility. A power electronic interface is developed to convert the available direct current generated by the PV panels and feed it into the utility grid. This power electronic interface mainly consists of an inverter. The inverter has to fulfill three main features in order to feed energy from a PV array into the utility grid: 1. To shape the current into a sinusoidal waveform 2. To invert the current into an AC current and 3. If the PV array voltage is lower than the grid voltage, the PV array voltage has to be boosted up with a further element

PE CONVERTER TOPOLOGIES FOR PHOTOVOLTAIC SYSTEMS

There are single phase inverter topologies as well as three phase topologies. Three-phase inverter topologies generally need more semiconductor switches, but their advantage is the possibility to drastically reduce energy storage in the DC-link due to the constant power flow from the generator to the grid. In single-phase systems large energy storage components are necessary to compensate the low frequency power pulsation of the grid. Evolution of Single-Phase Grid Connected PV Inverters A. The Past - Centralized Inverters The past technology was based on centralized inverter as shown in Fig. 1(a). Here the PV modules were divided into series connections (called a string), each generating a sufficiently high voltage to avoid further amplification. These series connections were then connected in parallel, through string diodes, in order to reach high power levels. Limitations: Need for high-voltage dc cables between the PV modules and the inverter Power losses due to a centralized MPPT Mismatch losses between the PV modules, losses in the string diodes Due to nonflexible design the benefits of mass production could not be reached. The grid-connected stage was usually line commutated by means of thyristors, involving many current harmonics and poor power quality. B. The Present - String Inverters The present technology consists of the string inverters which id depictrd in Fig. 1(b). Here a single string of PV modules is connected to the inverter. Advantages: The input voltage may be high enough to avoid voltage amplification.

The possibility of using fewer PV modules in series also exists, if a dc-dc converter or line-frequency transformer is used for voltage amplification. There are no losses associated with string diodes. Separate MPPTs can be applied to each string. Increased overall efficiency compared to the centralized inverter, and reduced price, due to mass production.

C. The Future - Multi-String Inverters The multi-string inverter is shown in Fig. 1(c). Here several strings are interfaced with their own dc-dc converter to a common dc-ac inverter. This is beneficial, compared with the centralized system, since every string can be controlled individually. Thus, the operators may start their own PV power plant with a few modules.

Figure 1. Historical overview of PV inverters (a) Past centralized technology. (b) Present string technology. (c) Present and future multi-string technology. (d) Present and future acmodule and ac cell technologies.

Multilevel Converters
Multilevel voltage source converters are emerging as a new breed of power converter options for high power applications. The multilevel voltage source converters typically synthesize the staircase voltage waveform from several levels of dc capacitor voltages. The voltage source inverters produce an output voltage or a current with levels either 0, + or Vdc. They are known as the two-level inverter.

To obtain a quality output voltage or a current waveform with a minimum amount of ripple content, they require high-switching frequency along with various pulse-width modulation (PWM) strategies. In high-power and high voltage applications, these two-level inverters have some limitations in operating at high frequency mainly due to switching losses and constraints of device ratings. Increasing the number of voltage levels in the inverter without requiring higher ratings on individual devices can increase the power rating. As the number of voltage levels increases, the harmonic content of the output voltage waveform decreases significantly. A multilevel converter has several advantages over a conventional two-level converter that uses high switching frequency pulse width modulation (PWM). The attractive features of a multilevel converter can be briefly summarized as follows. Staircase waveform quality: Multilevel converters not only can generate the output voltages with very low distortion, but also can reduce the dv/dt stresses; therefore electromagnetic compatibility (EMC) problems can be reduced. Common-mode (CM) voltage: Multilevel converters produce smaller CM voltage; therefore, the stress in the bearings of a motor connected to a multilevel motor drive can be reduced. Input current: Multilevel converters can draw input current with low distortion. Switching frequency: Multilevel converters can operate at both fundamental switching frequency and high switching frequency PWM. It should be noted that lower switching frequency usually means lower switching loss and higher efficiency.

Unfortunately, multilevel converters do have some disadvantages. One particular disadvantage is the greater number of power semiconductor switches needed. Although lower voltage rated switches can be utilized in a multilevel converter, each switch requires a related gate drive circuit. This may cause the overall system to be more expensive and complex. Multilevel Concept Multilevel inverters include an array of power semiconductors and capacitor voltage sources, the output of which generate voltages with stepped waveforms. The commutation of the switches permits the addition of the capacitor voltages, which reach high voltage at the output, while the power semiconductors must withstand only reduced voltages. Fig. 2 shows a schematic diagram of one phase leg of inverters with different numbers of levels, for which the action of the power semiconductors is represented by an ideal switch with several positions. A two-level inverter generates an output voltage with two values (levels) with respect to the negative terminal of the capacitor [see Fig. 2(a)], while the three-level inverter generates three voltages, and so on.

Figure 2. One phase leg of an inverter with (a) two levels, (b) three levels, and (c) n levels Considering that m is the number of steps of the phase voltage with respect to the negative terminal of the inverter, then the number of steps in the voltage between two phases of the load k is K = 2m + 1 and the number of steps p in the phase voltage of a three-phase load in wye connection is p = 2k 1 There are mainly three types of multilevel voltage source inverters which are recently developed: (1) Diode Clamp; (2) Flying Capacitor; and (3) Cascaded Inverters with separate dc sources.

1. Diode Clamped multilevel Converter

The neutral point converter proposed by Nabae, Takahashi, and Akagi in 1981 was essentially a three-level diode-clamped inverter. An m-level diode-clamp converter typically consists of m-1 capacitors on the dc bus and produces m levels of the phase voltage. Fig. 3 shows a single phase full bridge five-level diode clamp converter in which the dc bus consists of four capacitors, C1, C2, C3 and C4. For a dc bus voltage Vdc, the voltage across each capacitor is Vdc/4, and each device voltage stress will be limited to one capacitor voltage level, Vdc/4, through clamping diodes. To explain how the staircase voltage is synthesized, the negative dc rail, 0, is considered as the output phase voltage reference point. Using the 5-level converter shown in Fig. 3 as an example, there are 5 switch combinations to synthesize five level voltages across a and 0. (1) For voltage level Va0 = Vdc, turn on all upper switches Sa1 through Sa4. (2) For voltage level Va0 = 3Vdc /4, turn on three upper switches Sa2 through Sa4 and one lower switch Sa1. (3) For voltage level Va0 = Vdc /2, turn on two upper switches Sa3 and Sa4 and two lower switches Sa1 and Sa2.

(4) For voltage level Va0 = Vdc /4, turn on one upper switch Sa4 and three lower switches Sa1 through Sa3. (5) For voltage level Va0 = 0, turn on all lower half switches Sa1 through Sa4. Fig. 4 shows phase and line voltage waveforms of the example 5-level converter. The line voltage consists of a positive phase-leg a voltage and a negative phase-leg b voltage. Each phase voltage tracks one- half of the sinusoidal wave. Te resulting line voltage is a 9-level staircase wave. This implies that an m-level converter has an m-level output phase voltage and a (2m-1)-level output line voltage.

Figure 3. Single phase full bridge five-level diode clamp converter

Figure 4. Phase and line voltage waveforms of the 5-level converter

Advantages: All of the phases share a common dc bus, which minimizes the capacitance requirements of the converter. For this reason, a back-to-back topology is not only possible but also practical for uses such as a high-voltage back-to-back interconnection or an adjustable speed drive. When the number of levels is high enough, harmonic content will be low to avoid the need for filters. The capacitors can be pre-charged as a group. Efficiency is high because all devices are switched at the fundamental frequency. Reactive power flow can be controlled. The control method is simple for a back-to-back intertie system.

Disadvantages: The number of clamping diodes required is quadratically related to the number of levels, which can be cumbersome for units with a high number of levels. Real power flow is difficult for a single inverter because the intermediate dc levels will tend to overcharge or discharge without precise monitoring and control.

2. Flying Capacitor Multilevel Converter

The structure of this inverter is similar to that of the diode-clamped inverter except that instead of using clamping diodes, the inverter uses capacitors in their place. Fig. 5 illustrates the fundamental building block of a single-phase full-bridge flying capacitor based 5-level converter. Each phase-leg has an identical structure. Assuming that each capacitor has the same voltage rating, the series connection of capacitors in Fig. 5 is to indicate the voltage level between the clamping points. Three inner loop balancing capacitors for phase leg a, Ca1, Ca2, and Ca3 are independent from those for phase leg b. All phase legs share the same dc link capacitors, C1-C4. The voltage level defined in the flying-capacitor converter is similar to that of the diodeclamp type converter. The phase voltage of an m-level converter has m levels including the reference level, and the line voltage has (2m - 1) levels. Assuming that each capacitor has the same voltage rating as the switching device, the dc bus needs (m - 1) capacitors for an mlevel converter. The voltage synthesis in a flying-capacitor converter has more flexibility than a diode-clamp converter. Using Fig. 5 as the example, the voltage of the 5-level phase-leg a output with respect to the negative dc rail, Va0, can be synthesized by the following switch combinations. 1) For voltage level Va0 = Vdc, turn on all upper switches Sa1 through Sa4. 2) For voltage level Va0 = 3Vdc/4, there are four combinations: a) Sa1, Sa2, Sa3, Sa4 (Va0 = Vdc Vdc/4),

b) Sa2, Sa3, Sa4, Sa1 (Va0 = 3Vdc/4 c) Sa1, Sa3, Sa4, Sa'2 (Va0 = Vdc 3Vdc/4 + Vdc/2), and d) Sal, Sa2, Sa4, Sa'3 (Va0 = Vdc - Vdc/2+ Vdc/4).

Figure 5. Multilevel Converter using flying capacitor 3) For voltage level Va0 = Vdc/2, there are six combinations: a) Sa1, Sa2, Sa'3, Sa'4 (Va0= Vdc - Vdc/2), b) Sa3, Sa4, Sa'1, Sa2 (Va0 = Vdc/2), c) Sal, Sa3, Sa'1, Sa'4 (Va0 = Vdc - 3Vdc/4 + Vdc/2 - Vdc/4) d) Sal, Sa4, Sa2, Sa3 (Va0 = Vdc - 3Vdc/4 + Vdc/4), e) Sa2, Sa4, Sa1, Sa'3(Va0 = 3Vdc/4 - Vdc/2 + Vdc/4), and f) Sa2, Sa3, Sa1, Sa4(Va0 = 3Vdc/4 - Vdc/4). 4) For voltage level Va0 = Vdc/4, there are four combinations: a) Sal, Sa'2, Sa'3, Sa'4 (Va0 = Vdc 3Vdc/4), b) Sa4, Sa1, Sa'2, Sa'4 (Va0 = Vdc/4), c) Sa3, Sa1, Sa2, Sa4 (Va0= Vdc/2 - Vdc/4), and d) Sa2, Sa1, Sa3, Sa4 (Va0 = 3vd/4 - Vdc/2). 5) For voltage level Va0 = 0, turn on all lower switches Sal through Sa4.

Advantages: Large amount of storage capacitors provides extra ride through capabilities during short duration outages and deep voltage sags. Provides switch combination redundancy for balancing different voltage levels. When the number of levels is high enough, harmonic content will be low enough to avoid the need for filters. Real and reactive power flow can be controlled.

Disadvantages: The large numbers of capacitors are both more expensive and bulky than clamping diodes in multilevel diode-clamped converters. Packaging is also more difficult in inverters with a high number of levels. Control is complicated to track the voltage levels for all of the capacitors. Also, precharging all of the capacitors to the same voltage level and startup are complex. Switching utilization and efficiency are poor for real power transmission.

3. Cascaded-Inverter with Separate DC Sources

Fig. 6(a) shows the basic structure of the cascaded-inverters with SDCs, shown in a singlephase configuration. Each SDC is associated with a single-phase full-bridge inverter. The ac terminal voltages of different level inverters are connected in series. Fig. 6(b) shows the synthesized phase voltage waveform of a 9-level cascaded inverter with four SDCs. The phase output voltage is synthesized by the sum of four inverter outputs, i.e. van = v1 + v2 + v3 + v4 Each single-phase full bridge inverter can generate three level outputs, +Vdc, 0, and -Vdc. This is made possible by connecting the dc sources sequentially to the ac side via the four gate-turn-off devices. Each level of the full-bridge converter consists of four switches, SI, S2, S3, and S4. Using the top level as the example, turning on S1 and S4 yields v1 = +Vdc. Turing on S2 and S3 yields v1 = -Vdc. Turning off all switches yields v1 = 0. Similarly, the ac output voltage at each level can be obtained in the same manner. Minimum harmonic distortion can be obtained by controlling the conducting angles at different inverter levels. With the phase current ia, leading or lagging the phase voltage van by 90, the average charge to each dc capacitor is equal to zero over one line cycle, shown in Fig. 6(b). Therefore, all SDC capacitor voltages can be balanced. To comply with the definition of the previously mentioned diode-clamp and flying-capacitor multilevel converters, the level in a cascadedinverters based converter is defined by m = 2s + 1, where m is the output phase voltage level, and s is the number of dc sources. For example, a 9-level cascaded- inverters based converter will have four SDCs and four full bridges.

Figure 6. Cascaded multilevel converter using separate dc sources Advantages: Requires the least number of components among all multilevel converters to achieve the same number of voltage levels. Modularized circuit layout and packaging is possible because each level has the same structure, and there are no extra clamping diodes or voltage balancing capacitors. Soft-switching can be used in this structure to avoid bulky and lossy resistorcapacitor-diode snubbers. Disadvantages: Needs separate dc sources for real power conversions, and thus its applications are somewhat limited as already have multiple SDCSs readily available..

SUMMARY
The multilevel converters can immediately replace the existing systems that use traditional multi-pulse converters without the need for transformers. For a 3-phase system, the relationship between the number of levels, m, and the number of pulses, p, can be formulated by p = (m - 1) x 6.

Table I compares the power component requirements per phase leg among the three multilevel voltage source converters mentioned above. This comparison assumes that all devices have the same voltage rating, but not necessarily the same current rating, and that the cascaded-inverters type uses a full bridge in each level as compared to the half-bridge version used in the other two types. The multilevel converter using cascaded-inverters requires the least number of components and is most promising for utility interface applications because of its capabilities of modularization and soft-switching.