3.

6 Modulation Techniques for Inverter

In order to achieve conversion from DC signal to AC signals, Pulses are provided at the gates of
the switches present in the inverter in an on and off manner. Many methods are used in order to
provide and generate this type of pulses on the gates. But the common way is Space Vector Base
Pulse Width Modulation as SVPWM and Sine Pulse Width Modulation as SPWM. As in this work
presented in this thesis SPWM is used so there need of introduction to SPWM is necessary here:

As the figures show that the pulses are supplied to the switches that are being in connection with
the positive dc bus rail in each leg and with negative side a complimentary signal is supplied [30].
This type of formation ensures that the DC bus will not get shorted as only one switch is on in a
leg in single time.

3.7 Control Structure of the Inverter

The most important and difficult stage of the inverter is its control part. Here we are using PLL to
synchronize the output voltage, frequency and phase with the same parameters on grid side. We
are basically feeding the grid voltage into the built in module of the PLL in mat-lab Simulink. Its
synchronized output is fed to the mat-lab function to obtain desired amplitude level. We called this
output as reference which is fed to PWM generator to generate the desired output which we are
giving at gate of our bridge inverter which generates synchronized output.

In order to synchronize Grid, phase locked loop is used. PLL gives the matched value of voltage,
frequency and phase with the voltage, frequency and phase on the grid side.

3.8 Inverter in a PV system

Most of the inverters we are using these days are self-commutated. Depending upon the system
frequency and power density the switching devices used in them are mostly IGBTs, MOSFETs or
some time BJTs. There are two types of inverters as current source inverter and voltage source
inverter. They both are explained below.

3.8.1 Current Source Inverter(CSI)

The current source for the current source inverter is the input side. The power flow is determined
through input voltage and the polarity of the input current does not change. An alternating current
wave form is produced by CSIs with a fixed magnitude and have adjustable time period. To
maintain the current, the input side of the CSI is fixed with an inductor [14].

3.8.2 Voltage Source Inverter (VSI)

The input of this type of inverter is voltage source that is achieved by a big capacitor. The DC
current input determines the direction of the flow of power and input voltage polarity does not
change. Using VSIs we can obtain constant amplitude alternating voltage, which is contrary to
CSIs. VSIs can operate in both current control mode and voltage control mode.
The CSI input is varying current from a DC voltage source which has a high resistance capacity.
The following table gives us the comparative study between VSI and CSI.

VSI CSI
VSI is fed from a DC voltage source having
CSI is fed with adjustable current from a DC
small or negligible impedance. voltage source of high impedance.
Input voltage is maintained constant The input current is constant but adjustable.
Output voltage does not dependent on the The amplitude of output current is
load independent of the load.
The waveform of the load current as well as
The magnitude of output voltage and its
its magnitude depends upon the nature of waveform depends upon the nature of the
load impedance. load impedance.
VSI requires feedback diodes The CSI does not require any feedback
diodes.
The commutation circuit is complicated Commutation circuit is simple as it contains
only capacitors.
Power BJT, Power MOSFET, IGBT, GTO They cannot be used as these devices have
with self-commutation can be used in the to withstand reverse voltage.
circuit.

Table 3.1
 Figure 3.2: Two level voltage source inverter [31]

Load Dependency:

load is not the criteria on which Voltage amplitude depends. O/P current waveform is dependent
upon load impedance. Current amplitude is independent of load. O/P voltage waveform depends
on load impedance.
Associated Loss:
Total power loss is low when conduction loss is low while switching loss is high and vice versa.

3.9 Control Loops

Balance of the power flow on DC and AC side is attained and quality of power being fed into the
grid is enhanced by two control loops. The loop on outer side is voltage loop and the loop on inner
side is the current loop. Modelling of these loops is visible in [32].
The DC link voltage is kept persistent by using voltage loop. As the input energy from the solar
beam changes because of difference in irradiance or temperature, the voltage of DC link would
also vary because the power received from solar array would not synchronize the power supplied
to the grid [33]. The key role of voltage loop controller is to alter the main power reference current
to ensure that power received from the solar array can be equalized to the power which is supplied
to the grid. The reference signals of desired voltage are provided by the current control loop, basing
on the active power current reference and reactive power current reference (taken as 0 in this
model, as the system only supplies active power to the grid).

3.10 PWM Generation

The signal we receive from PLL module is given to the Mat lab function and multiplied with
modulation index to get the required amplitude. Then it is fed to PWM generator and output of
PWM generator is given to the gate of bridge rectifier. The PWM signal is shown as below.

Figure 3.3: PWM generated signal

3.11 Half bridge inverter

A bipolar switching mode is widely used in PWM Inverter. The switching circuits for a bipolar
switching mode are commonly used the full-bridge (FB) and half-bridge inverter (HBI) circuits.
A full-bridge inverter consists of two leg of half-bridge inverter and is preferred over the half-
bridge in higher power ratings. With the same DC. input voltage, the maximum output voltage of
the full-bridge inverter is twice that of the half-bridge inverter. This implies that for the same
power, the output current is one-half of those for a half-bridge inverter. With respect to the same
output current of the two circuits, the DC input voltage (Vs) of a half-bridge inverter must be twice
of a full-bridge inverter.
The benefits of the circuit, when compared to a full (H) bridge circuit, are:

 Only two switching devices T1, T2 (MOSFETs) are needed.

 Current is flowing through only one MOSFET at a time. This reduces voltage drops and
power dissipation.
 The drive circuits for the MOSFETs are comparatively simpler.
  The required voltage rating of the MOSFETs is lower than that used in a full bridge at
the same supply voltage.

As we do not have any concern related to high power rating and seeking for simple design so we
are going to use half bridge inverter.

3.12 Grid
The grid is modelled as a single-phase voltage source Inverter with a Line to Line RMS voltage of
220V. Load is linked between the inverter and the grid. Practically we have used single phase
RMS value of 220V coming from the grid.

Figure 3.4 Grid 220V

3.13 Filter

The inverter is interconnected with the grid by using LCL filter in this design. As an inverter works
on the basis of switching devices and gating signals in the shape of pulses must be available for
the switches, the power quality of the output current is deteriorated due to considerable harmonic
distractions.
The criteria used to calculate the values of the components is given below. All the calculations are
carried out using per phase circuit [34].
The inverter side inductor is sized as,

VDC Link
Li  (3.1)
16. f s I L

Where,

Li = inverter side inductor per phase

f s = switching frequency of the inverter = 3 kHz

VDC  Link = 400V,

and ΔIL is the ripple current of the inductor which is chosen to be 10% of the phase current and
can be computed using (3.2).

2P
I L  0.1 (3.2)
V ph grid 

Where,
P = is nominal power of the system

Vph grid  = the single-phase voltage of grid

After substituting these values, it is found that, Li  1.74 mH

The grid side inductance is computed using,

The capacitance is worked out basing on the formula given in [36]. The capacitance is taken as 5%
of base capacitance and calculated using,

P
C f  0.05  Cb  (3.3)
grid .V ph2  grid 

Where,

P is the single-phase power,

ω grid is the rotational frequency of grid,

Vph grid  = 220V.

The obtained per phase capacitance for the filter is, C f  6F

Per [35], a damping resistor should be included in series with the LCL filter capacitor to enhance
the efficiency of the filter. It is obtained using,

1
Rd  (3.4)
30C f

C f is the filter capacitor,

o is the resonance frequency of the LCL filter which can be obtained using,
 1 Li  Lg
f0  (3.5)
2 Li C f Lg

This results in f 0  3000 Hz , and the value of damping resistor needed is 30 ohms per phase.
However, in the simulation, a lower value is used.

3.14 Filter as Grid Interface

To the grid as the switching devices cannot be connected directly. The reason to this is, inverter
generates harmonics which reduces quality of power. The harmonics that can be inserted into the
grid are on limitation by different standards in place [22], [23] and [35]. Further a transformer is
needed to be connected between the system and the grid [36]. To reduce the harmonics found in
the current wave the winding of transformer becomes a source of inductance. As the transformer
are high priced, heavy and bulky in size so it leads to the system being expensive and costly.
Hence, there comes an idea of a topology to connect inverter to the grid known as transformer-less
topology. This happens by using a filter circuit that serves like an interface. There are many types
of filters but only three of them are used. These filters include L, LC and LCL filters as shown in
figure 3.3.