11-11-2016, 10:17 AM
FPGA-based real time incremental conductance maximum power point tracking controller for photovoltaic systems
1468374917-FPGAbasedrealtimeincrementalconductancemaximumpowerpointtrackingcontrollerforphotovoltaicsystems1.pdf (Size: 1.18 MB / Downloads: 6)
Introduction
In recent years, the installation of renewable energy
generation systems is rapidly growing because of concern
about environmental issues and the decline in the fossil fuel
resources. Among all kinds of renewable energy
technologies, photovoltaic (PV) technology is one of the
most common systems because it has some advantages such
as cleanness, low maintenance cost, availability and
noiselessness.
Research on PV systems consists of different subjects
including PV modelling, maximum power point tracking
(MPPT) algorithms, power converter configuration and grid
connection issues. Different models were proposed for PV
cells in the literature. Single diode model is the simplest
and most widely used model for PV cells as it offers a
good compromise between simplicity and accuracy [1].
Double diode model provides a more accurate P−V
characteristic of PV cell whereas the equations of this
model are more complex [2]. Many different dynamic PV
models were presented in literature [3–5]. These models
should be implemented in real time applications by
considering their complexity.
PV cells rarely work in maximum power point (MPP)
because the maximum output power of PV cells depends on
various variables (temperature and irradiation). Considering the non-linear properties of PV cells we need to track the
maximum power by means of controllers to improve power
conversion efficiency of PV cells.
Different MPPT algorithms have been proposed to increase
the efficiency of PV systems [6–12]. These methods vary in
convergence speed, oscillations around the MPP,
complexity, cost and electronic equipment requirements
[13]. In recent years, various MPPT algorithms have been
implemented on the FPGA as the platform for its controller.
Some of the most useful algorithms are Perturb & Observe
[14, 15], constant voltage [16, 17], incremental conductance
[18] and artificial intelligence methods [9, 19, 20] which
are widely used in PV applications. P&O method is the
most widely used algorithm because of its simplicity of
implementation. This method is based on perturbation in
the operation voltage, thus the major drawback of this
method is oscillation around MPP and the amount of power
loss in this point. Constant voltage method is based on
approximately constant ratio between Vmpp and
open-circuit voltage. Although this method is quite simple,
but it is difficult to determine the optimum value of
constant ratio between Vmpp and Voc, and even more
important requirement of sudden interruption of PV power
to measure open-circuit voltage [8]. The artificial
intelligence methods such as fuzzy logic [21] and neural
network [22] provide more rapid and accurate solutions for the MPPT problem but are generally more complicated in
implementation which limits their use in real-time
applications.
Incremental conductance method is based on the fact that
the power–voltage curve of PV generator at constant solar
irradiance and cell temperature levels has normally only one
MPP [23]. In this work, a criterion is presented based on
dynamic model of PV cells to improve the performance of
incremental conductance (INC) method. It is based on using
Lambert W-function [24], which has been fruitfully used
for expressing the current–voltage characteristic of a solar
array [5, 25]. Based on this criterion, a modified adaptive
INC method with a variable perturbation step size is
introduced to improve tracking speed and power fluctuation
around MPP. In the following, the modelling and the
hardware implementation of the MPPT system as well as
considerations in practical implementation are investigated.
In this paper: Section 2 presents a dynamic model of PV
cell. In Section 3, dynamic behaviour of buck converter is
explained and Section 4 describes the proposed MPPT
algorithm. The simulation results are presented in Section
5. Hardware implementation of the MPPT system and
experimental results are discussed in Sections 6 and 7,
respectively. Finally, conclusion is presented in Section 8.
2 PV cell modelling
To evaluate the performance of PV and MPPT systems, it is
necessary to model the current–voltage characteristics of the
PV cell in different operating conditions. Fig. 1a shows the
dynamic model of PV cell that it is used in this research.
As shown in this figure, the proposed dynamic model
includes a current source that its current is directly
proportional to irradiation. Rs resistance represents the
equivalent series resistance of the PV array, where the
equivalent parallel resistance is considered as Rp. Cp is an
embed transition and diffusion capacitance [26].
In this work, to verify the performance of the proposed
modified variable step size INC MPPT algorithm, a
MATLAB-Simulink model of PV system is initially
developed. The parameters of a commercial solar cell
LORENTEZ LC80-12M model are extracted and used for
the PV array model in simulation and the experimental
results. The related parameters and the electrical
specifications of PV module are listed in Table 1.
Some parameters in Table 1 are inserted according to data
sheet of LORENTEZ LC80-12M solar array, for example IMP,
VMP, PMAX, ISC, VOC, KV, KI and NS.
Description of the proposed MPPT
algorithm
In this section, the proposed algorithm is described based on
the INC method. This algorithm controls the duty cycle ratio
of MOSFET and redound a fast and accurate response of
MPPT problem. The INC is one of the MPPT methods that
it has different convergence speeds and oscillations around
MPP depending on the step size of the duty cycle ratio [29,
30]. In other words, reducing the step size of duty cycle,
the convergence speed and also the oscillations around the
MPP are decreased and vice versa. In simple INC, a fast
response and small oscillations cannot exist simultaneously.
In the proposed algorithm, the step size of the duty cycle
ratio is not considered as a constant. The higher step size is
used when the system operates far from MPP, whereas the
step size is decreased for the area around MPP. Also we
require a criterion to determine how a system operates
closely to the MPP.
In literature, several analytical models have been proposed
to describe the behaviour of solar cells under different
environmental conditions [5, 25, 30, 31]. Here, we analyse
the proposed model. According to (6) and disregard of
series and parallel resistors and capacitances, the output
power of PV cell is calculated by the following equation