IEEE Power Electronics Magazine - September 2021 - 50

charges rapidly until it attains open circuit voltage condition.
Useful points for tracing the I-V curve are obtained in this
process. Apart from being a fast option, with ideal capacitor
sizing, dynamic capacitor charging can be a cost-efficient
way to determine GMPP and achieve MPP.
Proposed Solution
SPV systems are connected to the utility grid in two ways: a
two-stage system or a single-stage system as shown in Figure
2. In the two-stage system, the SPV panel is connected
to a dc-dc converter where the MPPT algorithm is utilized
and implemented. The output from the dc-dc converter is
connected to the inverter either with isolation using a transformer
or direct connection to the grid. If the output voltage
from the SPV panel/array is high relative to grid voltage, it
can be directly connected to the inverter to feed power to
the utility grid. The MPPT technique in this case is implemented
in the inverter itself.
As mentioned earlier, prior intelligent MPPT techniques
oscillate around the maximum point continuously. These
oscillations depend on the duty cycle step size provided by the
MPPT algorithm. With a large step size, the MPPT algorithm
reaches the MPP relatively fast but with more oscillations at
the steady-state and vice versa for a small step size. These
oscillations directly affect the maximum power generation of
the SPV panel/array. Therefore, there is a definite requirement
to identify a MPPT solution that has minimal or no oscillations
whilst being capable of rapidly tracking the MPP.
Proposed MPPT Method
The MPPT technique proposed in this article is built with an
additional hardware module (AHM) composed of three gallium
nitride (GaN) power semiconductor devices (PSD)
(GaN-1, GaN-2 and GaN-3), an electrolytic capacitor 'C' and
a dc bus capacitor 'UC' as shown in Figure 3(a). PSDs such
A
vPV
B
Solar PV Panel
MPPT Based
dc-dc Converter
(a)
iPV
vPV
B
Solar PV Panel
(b)
MPPT Based
dc-ac Inverter
A
dc-ac Inverter
iPV
as MOSFET, IGBT, SiC etc., may be utilized in the place of
GaN depending on the rating of the SPV array. This PSD
selection also depends on the application of the applied
SPV voltage and current. Traditional switching devices
have a limit on how quickly they can switch high voltages,
or more appropriately, the dV/dt ability of the device may
be limited. This slow ramp up and down increases conduction
loss because the device spends more time in the
switching state. The increased switching time also
increases the amount of dead time required in the control
system to prevent shoot-through and shorts. The architecture
presented in this article utilizes GaN high electron
mobility transistor (HEMT) because of its distinctive
advantages such as low switch on/off time, and compact
size compared with other PSD options.
The intrinsic property of the capacitor and its charging
characteristics are used to track the I-V and P-V curve of
a SPV panel/array. The architecture of the proposed single
point MPPT technique is shown in Figure 3. Here the nomenclature
vPV , iPV , iC , vC , iUC , iUC refers to the SPV panel/array
voltage, SPV panel/array current, capacitor current, capacitor
voltage, dc bus-capacitor current and DC bus-capacitor
voltage, respectively.
The PSDs GaN-1 and GaN-3 are complementary to each
other. When GaN-3 is switched ON, the current iPV will
flow through the capacitor 'C' as shown in Figure 3(b). The
designed voltage and current sensors across the capacitor
will capture the information and pass it through the control
circuitry to find the MPP.
During this charging state, the capacitor 'C' will be
Utility Grid
Utility Grid
FIG 2 Connection of solar photovoltaic panel to grid. (a) Two-stage grid connected
inverter and (b) Single-stage grid connected inverter.
50 IEEE POWER ELECTRONICS MAGAZINE z September 2021
charged to its open circuit point. After the charging state is
complete, before facilitating the next sweep, it is necessary
to discharge the capacitor. Unlike other methods, in the proposed
architecture the capacitor will be discharged through
a PSD, GaN-2 in this case, by controlling its gate voltage as
is shown in Figure 3(c). GaN-2 along with a
heat sink will act as an electronic load to dissipate
the energy from the capacitor. To facilitate
total energy utilization, it is possible to
use ultra-lift charge pump circuits during the
capacitor discharging state [9]. This electronic
load-based technique is specifically chosen to
increase the life of the capacitor and to reduce
the oscillations during discharging. At this
instant GaN-3 is turned OFF and the circuit
operates as a normal SPV system connected
with junction box terminals A and B.
The time between two successive
switching transitions between GaN-1 and
GaN-2 of the SPV array is typically around
18-24 s. It is important to maintain continuous
operation while the SPV system is connected
to the capacitor 'C'. Hence, depending on the
SPV system capacity, a large capacity capacitor
or a supercapacitor must be connected at
the terminals of the AHM. The capacitor 'UC'
IEEE Power Electronics Magazine - September 2021

Table of Contents for the Digital Edition of IEEE Power Electronics Magazine - September 2021
