Page 131 - Renewable Energy Devices and System with Simulations in MATLAB and ANSYS
P. 131
118 Renewable Energy Devices and Systems with Simulations in MATLAB and ANSYS ®
®
solar irradiation on each PV module of the PV array are fed in the off-line-trained ANN, which then
produces an estimation of the approximate position of the global MPP. This information is used as
an initial operating point by a P&O MPPT algorithm for deriving the true global MPP.
In [89], the current–voltage curve of the PV source is initially traced by disconnecting it from the
power converter and connecting it to a parallel RC circuit containing a discharged capacitor. This
action causes the PV source voltage to sweep in the range of 0 −V oc . The current and voltage mea-
surements acquired during that tracing process are used to move the PV source operating point to a
region close to the global MPP. Then, a P&O algorithm is applied for converging to the global MPP.
This technique has the disadvantage that during the execution of the current–voltage curve tracing
process, the PV source power is not transferred to the PV system load.
In an alternative MPPT method, the total output voltage range of the PV source (i.e., 0 −V oc ) is
divided into intervals and the corresponding output power is measured at each interval [90]. The P&O
MPPT method with variable step size (i.e., reducing the step size as the MPP is approached) is then
applied for tracking the global MPP at the particular interval, which gave the highest power measure-
ment. The number of intervals is selected to be higher than the number of PV modules connected in
series in the PV source, thus requiring the knowledge of the PV source operational characteristics.
The output voltage of each PV module of the PV string is measured in [91] for calculating the
(
number, q, of different levels of solar irradiation, G j G 1 < G 2 < < G q), which are received by the
(
PV string and the corresponding number of PV modules receiving each such level, M j j = 1,..., q).
Then, a P&O MPPT process is applied to the entire PV string, which is initiated at the open-circuit
voltage of the PV string, V oc , as well as at each of the distinct PV string voltage levels defined by the
following equation for j = 1,..., q −1:
∑ j M i
⋅
i=1
.
V pv j, = 0851 − N ⋅ V oc (5.22)
where N is the total number of PV modules of the PV string.
By comparing the power produced at the individual MPPs tracked by the P&O algorithm, the
position of the global MPP is derived. However, the number of voltage sensors and accompany-
ing signal-conditioning circuits is significantly high when applying this technique in PV sources
composed of strings with a high number of PV modules, thus increasing the complexity and cost of
the control unit.
In [92], the ESC MPPT method is sequentially applied at individual segments of the voltage
range of the PV array in order to detect the positions of local MPPs. Among them, the local MPP
where the maximum power is produced corresponds to the global MPP. A similar process is also
applied in [93], but in this case the segments that do not contain the global MPP are identified using
information about the gradient of the power–voltage curve. By this technique, convergence to the
(local) MPP of these segments is avoided, thus speeding up the global MPP detection process. In the
ESC-based global MPPT method, the individual segments of the PV source output voltage range are
selected using the values of the PV module electrical characteristics, which must be known when
developing the corresponding MPPT system. Additionally, the complexity of the control circuit,
which implements the ESC-based global MPPT technique, is relatively high.
5.4.7 Comparison of Global MPPT Methods for
Nonuniform Solar Irradiation Conditions
The global MPPT methods for nonuniform solar irradiation conditions, which have been described
previously, are compared in terms of their operational characteristics in Table 5.3. With the excep-
tion of the DMPPT techniques, the remaining global MPPT algorithms, which have been presented
