IEEE TRANSACTIONS ON CIRCUITS AND SYSTEMSI: REGULAR PAPERS, VOL. 55, NO.

3, APRIL 2008

953

Grid-Connected Photovoltaic Generation System

Rong-Jong Wai, Senior Member, IEEE, and Wen-Hung Wang

AbstractThis study addresses a grid-connected photovoltaic

(PV) generation system. In order to make the PV generation
system more flexible and expandable, the backstage power circuit
is composed of a high step-up converter and a pulsewidth-modulation (PWM) inverter. In the dcdc power conversion, the high
step-up converter is introduced to improve the conversion efficiency of conventional boost converters and to allow the parallel
operation of low-voltage PV modules. Moreover, an adaptive total
sliding-mode control system is designed for the current control of
the PWM inverter to maintain the output current with a higher
power factor and less variation under load changes. In addition,
an adaptive step-perturbation method is proposed to achieve the
objective of maximum power point tracking, and an active sun
tracking scheme without any light sensors is investigated to make
PV plates face the sun directly in order to capture maximum
irradiation and enhance system efficiency. Experimental results
are given to verify the validity of the high step-up converter, the
PWM inverter control, the ASP method, and the active sun tracker
for a grid-connected PV generation system.
Index TermsActive sun tracking scheme, adaptive step-perturbation (ASP) method, high step-up converter, photovoltaic (PV)
generation system, pulsewidth-modulation (PWM) inverter, total
sliding-mode control (TSMC).

I. INTRODUCTION

N THE PAST century, global surface temperatures have

increased at a rate near 0.6 C/century because of the
global warming taking place due to effluent gas emissions
and increasing CO [1], [2]. Problems with energy supplies
and use are related not only to global warming but also to
such environmental concerns as air pollution, acid precipitation, ozone depletion, forest destruction, and radioactive
emissions. To prevent these effects, some potential solutions
have evolved including energy conservation through improved
energy efficiency, a reduction in fossil fuel use and an increase
in environmentally friendly energy supplies. Recently, energy
generated from clean, efficient and environmentally friendly
sources has become one of the major challenges for engineers
and scientists. Among them, photovoltaic (PV) application
has received a great attention in research because it appears
to be one of the most efficient and effective solutions to this
environmental problem [3][7].
Dcdc converters with high voltage gain are required in many
industrial applications, such as front-end stages for clean-energy sources, dc back-up energy systems for uninterruptible
Manuscript received January 5, 2006; revised April 27, 2007, June 26, 2007,
and July 6, 2007. This work was supported in part by the National Science
Council of Taiwan, R.O.C. through Grant Number NSC 95-2221-E-155-070MY3. This paper was recommended by Associate Editor D. Masimovic.
The authors are with the Department of Electrical Engineering, Yuan Ze University, Chung Li 32003, Taiwan, R.O.C. (e-mail: rjwai@saturn.yzu.edu.tw;
s927104@mail.yzu.edu.tw).
Digital Object Identifier 10.1109/TCSI.2008.919744

power supplies (UPS), high-intensity discharge lamps for

automobile headlamps, and the telecommunications industry
[8][11]. The conventional boost converters cannot provide
such a high dc voltage gain, even for an extreme duty cycle.
It also may result in serious reverse-recovery problems, and
increase the rating of all devices. As a result, the conversion efficiency is degraded and the electromagnetic interference (EMI)
problem is severe under this situation [12]. In order to increase
the conversion efficiency and voltage gain, many modified
boost converter topologies have been investigated [13][17].
Although voltage-clamped techniques are manipulated in the
converter design to overcome the severe reverse-recovery
problem of the output diode in high-level voltage applications,
there still exist switch over-voltage stresses, and the voltage
gain is limited by the turn-on time of the auxiliary switch [13],
[14]. Wai and Duan [17] investigated a novel coupled-inductor
converter strategy to increase the voltage gain of a conventional boost converter with a single inductor, as well to deal
with the problem of the leakage inductor and demagnetization
of the transformer in a conventional coupled-inductor-based
converter. In this study, the high step-up converter topology in
[17] is introduced to boost and stabilize the output dc voltage
of PV modules for the utilization of a dcac inverter.
Developments in microelectronics and power devices have
provided widespread applications of pulsewidth-modulation
(PWM) inverters to industries. A PWM inverter used for a
grid-connected scheme is controlled in order to produce an
output current in phase with the utility voltage for obtaining a
unity power factor (PF). The performance is evaluated by the PF,
the transient response, and the efficiency. Thus, much attention
has been paid to the closed-loop regulation of PWM inverters to
achieve good dynamic response for the grid-connected scheme
in the past decade [18][21]. Variable structure control (VSC)
with sliding mode, or sliding-mode control (SMC), is one of the
effective nonlinear robust control approaches since it provides
system dynamics with an invariance property to uncertainties
once the system dynamics are controlled in the sliding mode
[22], [23]. The insensitivity of the controlled system to uncertainties exists in the sliding mode, but not during the reaching
phase, i.e., the system dynamic in the reaching phase is still
influenced by uncertainties. Recently, some researchers have
adopted the idea of total SMC (TSMC) to get a sliding motion
through the entire state trajectory [24][26]. Since there is no
reaching phase in TSMC, the motion of the controlled system
is never influenced by uncertainties. This study attempts to
extend an adaptive TSMC (ATSMC) from [25] to the current
control of a PWM inverter. Up to now, this is the first time to
investigate the application of TSMC to the power electronics
control.
In general, PV modules have nonlinear voltage-current characteristics, and there is only one unique operating point for a PV

1549-8328/$25.00 2008 IEEE

954

IEEE TRANSACTIONS ON CIRCUITS AND SYSTEMSI: REGULAR PAPERS, VOL. 55, NO. 3, APRIL 2008

Fig. 1. Configuration of grid-connected PV generation system.

generation system with a maximum output power under a particular environmental condition. However, the maximum power
point varies with irradiation and temperature, so that the maximum power point tracking (MPPT) at all atmospheric situations is a challenging problem. In the past decade, many MPPT
methods have been applied to PV generation systems for exacting maximum available powers from PV modules, e.g., incremental conductance method [4], perturbation and observation (P&O) method [27][29], and voltage-and current-based
method [30], [31], etc. The P&O method, which measures the
variations of power and voltage to judge the momentary region and change the reference voltage for operating close to
the maximum power point, is often used because of its simple
structure and fewer measured parameters [28], [29]. Although
the P&O method is easy to realize, the reference voltage still
varies periodically when the MPPT is reached, so that it may
cause oscillation phenomena around the maximum power point,
causing extra power losses. For this reason, an adaptive step-perturbation (ASP) method modified from the conventional P&O
method is proposed to obtain faster tracking response and stable
operation by perturbing the voltage of PV modules with an adaptive voltage step.
According to different irradiations, the output power of a PV
module is substantially changed. For example in Taiwan, the
direction with maximum average irradiations during one year
is the South, and the corresponding angle of inclination is 23.5
so that many PV modules are installed in this posture. However,
it could not capture maximum irradiations persistently by this
method so that the performance of the PV generation system
can not be improved effectively. Nowadays, many researchers
have focused on sun tracking investigations [32][35]. Conventional sun tracking strategies have light sensors equipped
on the terminals of PV plates. When the feedback signals from
light sensors are equal, it means that the PV plate directly faces
the sun and has the maximum irradiation at the corresponding
position. Unfortunately, the initial proofreading and correcting
of light sensors are time consuming and the devices properties are easily varied under different operational conditions. In
order to overcome the aforementioned drawbacks, this study
investigates an active sun tracking scheme without light sensors
via the property of open-circuit voltage of PV modules proportional to the corresponding irradiation, to follow the trail of the
sun.

II. SYSTEM DESCRIPTION

In this study, the configuration of a grid-connected PV generation system is depicted in Fig. 1. The system connected with
a utility power is mainly composed of PV modules, an active
sun tracker, a high step-up converter, a full-bridge inverter, and
a system controller. Due to the photo-voltaic effect, the voltage
of a PV cell is not very high. Because PV panels in a series
string are constrained to all conduct the same current, the least
efficient cell sets this string current, which may spell failure
when one cell of a string is inactive. The overall efficiency of
the PV array is reduced to the efficiency of this least efficient
cell. It means that PV panels in a series string must be given the
same orientation and be of identical size for obtaining a higher
output voltage. Besides, the corresponding output voltage
is varied easily with respect to the variation of loads. In order to
satisfy the requirement of high-voltage demand, a dcdc converter with high voltage gain is one of the essential mechanisms
in the grid-connected PV generation system. In this study, a high
step-up converter [17] is implemented to reduce the series-connected numbers of PV modules, to maintain a constant dc bus
for the inverter utilization, and to decouple and
voltage
simplify the inverter control design.
A unipolar PWM full-bridge inverter, including four power
semiconductors and an output inductor, is regarded as the
dcac power conversion circuit to meet the requirement of grid
connection. Since the PWM inverter is expected to produce an
output current in phase with the utility voltage for obtaining
a unity PF, an ATSMC system [25] is introduced by way of
switching four power semiconductors in this inverter to mainwith a higher PF and less variation
tain an output current
under load changes.
PV modules exhibit nonlinear voltage-current characteristics, and the maximum power point varies with the irradiation
and temperature. At a particular environmental condition, only
one unique operating point exists to provide the PV generation
system with maximum output power. In this study, an ASP
method is proposed to inject an adaptive step perturbation
into the voltage of PV modules. Since the step perturbation is
persistently adjusted in accordance with the variations of the
voltage and power of PV modules, the ASP method boosts the
MPPT speed and stability much higher than the conventional
methods, to further reduce extra power losses in the tracking
process.

WAI AND WANG: GRID-CONNECTED PV GENERATION SYSTEM

955

Fig. 2. Architecture of high step-up converter.

Generally speaking, the output power of PV modules is substantially changed according to different irradiations. To further
enhance the capability of the PV generation system, an active
sun tracker actuated by a synchronous motor is investigated on
the basis of the open-circuit voltage of a PV module, for maintaining the PV plate in the face of the sun in order to improve
the generation efficiency of the fixed-installation PV module,
and to lower the cost of conventional sun trackers with light sensors. Incidentally, it is unnecessary to modify the original circuit
framework of the grid-connected PV generation system because
of the sole requirement of the open-circuit voltage of PV modules in the active sun tracking scheme.
In this study, the PWM inverter control, the ASP method, and
the active sun tracking scheme are carried out using Turbo C language inserted into a system controller, i.e., a digital-signal-processor (DSP) development module. This development module
has: Texas Instruments TMS320LF2407A central processing
unit with an evaluation module (EVM), 16 channel 10-bit
analog-to-digital, 4 channel 12-bit digital-to-analog converters
and programmable I/O ports. The central processing unit has: a
40MIPS 16-bit fixed point DSP core, 16 PWM channels, four
general purpose timers and two encoder channels. The detailed
functions of the main components in the grid-connected PV
generation system are described in the following sections.
III. HIGH STEP-UP CONVERTER
The architecture of a high step-up converter introduced from
[17] is depicted in Fig. 2, where it contains seven parts: a PV
module input circuit, a primary-side circuit, a secondary-side
circuit, a passive regenerative snubber circuit, a filter circuit,
a dc output circuit, and a feedback control mechanism. In this
strategy, a coupled inductor with a lower-voltage-rated switch
is used for raising the voltage gain whether the switch is turned
on or turned off. Moreover, a passive regenerative snubber is
utilized for absorbing the energy of stray inductance so that the
switch duty cycle can be operated under a wide range; the re-

lated voltage gain is higher than that of other coupled-inductorbased converters. In addition, all devices in this scheme also
have voltage-clamped properties and their voltage stresses are
relatively smaller than the output voltage. Thus, it can select
low-voltage low-conduction-loss devices, and there are no reverse-recovery currents within the diodes in this circuit. Furthermore, the closed-loop control methodology is utilized to
overcome the voltage drift problem of the power source under
the load variations. As a result, this converter topology can increase the voltage gain of a conventional boost converter with
a single inductor, and deal with the problem of the leakage inductor and demagnetization of the transformer for a coupled-inductor-based converter.
The major symbol representations are summarized as follows.
and
denote dc input voltage and current, and
is an
and
input filter capacitor in the PV module input circuit.
represent individual inductors in the primary and secondary sides
, respectively. is a switch in the
of the coupled inductor
and
are the output voltage command
primary-side circuit;
and the trigger signal in the feedback control mechanism, respecand
denote a clamped capacitor, a clamped
tively.
diode and a rectifier diode in the passive regenerative snubber ciris a high-voltage capacitor in the secondary-side circuit.
cuit.
and
are the output diode and the filter capacitor in the
filter circuit, respectively.
and describe dc output voltage
and current, respectively, in the dc output circuit.
The coupled inductor in Fig. 2 is modeled as an ideal trans, and a leakage inductor
former, a magnetizing inductor
. The turns ratio
and coupling coefficient
of this
ideal transformer are defined as
(1)
(2)
where
and
are the winding turns in the primary and secondary sides, respectively. The voltages across the switch, the

956

IEEE TRANSACTIONS ON CIRCUITS AND SYSTEMSI: REGULAR PAPERS, VOL. 55, NO. 3, APRIL 2008

primary and secondary winding of the ideal transformer, and

and
, rethe leakage inductor are denoted as
of the coupled
spectively. Moreover, the primary current
inductor is composed of the magnetizing current
and the
. The secondary current
is
primary induced current
through the ideal
formed by the primary induced current
. In
transformer, and its value is related to the turns ratio
addition, the conductive voltage drops of the switch
and all
and
) are neglected in order to simplify cirdiodes (
cuit analyses.
According to the detailed circuit analyses in [17], the voltage
of the high step-up converter and the corresponding
gain
can be represented as
switch voltage

Fig. 3. PWM inverter framework.

(3)
(4)
where is the duty cycle of the switch
. Because the voltage
is less sensitive to the coupling coefficient
, (3)
gain
as
and (4) can be rewritten with
(5)
(6)

Fig. 4. Equivalent dynamic model of PWM inverter.

the state-space average method [12] and the linearization technique as

(8)

According to (5) and (6), one can obtain

(7)
is not related to the
By analyzing (7), the switch voltage
and the switch duty cycle
if the
input power source
and the turns ratio
are
values of the output voltage
fixed. Thus, it can ensure that the maximum sustainable voltage
is constant. As long as the input voltage is not
of the switch
higher than the switch voltage-rated, the high step-up converter
can be applied well to low-voltage PV power sources even with
large voltage variations.
IV. PWM INVERTER CONTROL
A. Dynamic Model Description
Fig. 3 illustrates the PWM inverter framework including four
power semiconductors and an output inductor. In Fig. 3,
is
the equivalent series resistor of the output inductor
and
are the output voltage of the full-bridge inverter and
is the output current of the full-bridge
the utility voltage;
emulates the disturbance
inverter; and the voltage source
incurred by load variations. In order to analyze conveniently,
the following assumptions are made in this study: i) the value of
is small enough to ignore; ii) the conduction and switching
losses are zero since all power switches are assumed to be ideal
devices; iii) the delay time between the switch turn-on and
turn-off states is small enough to neglect; and iv) the control
signal and input/output voltages are taken as constant values
because the switching frequency is greater than the system
dynamic frequency.
Due to the symmetry property of the positive-half and negative-half period in the unipolar PWM switching, the dynamic
equation during the positive-half period can be represented via

where
is the duty cycle of the switches
and
during
one switching period. Define the duty cycle and the power gain
and
, where
is a
as
sinusoidal control signal and
is the amplitude of a triangular
, then the dynamic equation of the PWM
carrier signal
inverter can be given by
(9)
By way of the Laplace transformation of (9), the equivalent dynamic model of the PWM inverter is depicted in Fig. 4, where
is the Laplace operator.
as the system state
By choosing the ac output current
as the control input, (9) can be
and the control signal
rearranged as

(10)
where
and
and
denote
and
the nominal values of and , respectively;
represent the system parameter variations;
is called the
lumped uncertainty and defined as
(11)
Here, the bound of the lumped uncertainty is assumed to be
given by
(12)

WAI AND WANG: GRID-CONNECTED PV GENERATION SYSTEM

957

Fig. 5. ATSMC system for PWM inverter.

where is the operator of an absolute value, and

positive constant.

is a given

B. ATSMC System
The objective of the PWM inverter control is to force the
to track a reference output current
system state
, which is designed by the ASP method introduced later.
An ATSMC system, as shown in Fig. 5, is introduced for the
current control of the PWM inverter, where the control error is
. Define a sliding surface [25]
chosen as
as
(13)
where
is the initial value of
and is a nonzero positive
constant.
The ATSMC system is divided into three main parts. The first
part addresses performance design. The objective is to specify
the desired performance in terms of the nominal model, and it
. Following the
is referred to as the baseline model design
baseline model design, the second part is the curbing controller
to totally eliminate the unpredictable perturbation
design
effect from the parameter variations and external disturbance
so that the baseline model design performance can be exactly
ensured. Finally, the third part is the adaptive observation deto estimate the upper bound of the lumped uncertainty
sign
for alleviating the chattering phenomenon caused by the inappropriate selection of a conservative constant control gain in
the curbing controller. The entire control methodologies of the
ATSMC system are summarized in the following theorem.
Theorem 1: If the PWM inverter scheme shown in (10) is controlled by the three-part ATSMC system described by (14)(16)
with the adaptive observation design shown in (17), then the stability of the ATSMC system for the current control of the PWM
inverter can be guaranteed
(14)
(15)
(16)
(17)
where

is a positive constant.

Fig. 6. Control flowchart of ASP method.

Proof:

Consider

Lyapunov function candidate

with
and take the
with respect to time; there exists an
derivative of
ATSMC system described by (14)(16) with the adaptive
. According
observation design shown in (17) such that
to the Lyapunov stability theorem and Barbalats lemma [22],
will converge to zero as
[23], it can imply that the function
. Moreover, the parameter estimation error can be
guaranteed to be bounded. The detailed proof of Theorem 1
can be referred to [25]. As a result, the stable behavior for the
current control of the PWM inverter can be ensured.
V. ASP METHOD
Due to the characteristic of PV modules with nonlinear
voltage-current relation and the variation of the maximum
power point with respect to irradiation and temperature, an
ASP method is proposed in this study to adjust the reference
voltage by injecting a step perturbation into the voltage of
PV modules and judging the momentary operation region for
raising the operating point close to the maximum power point.
In this way, the major drawback concerned with the tradeoff
between tracking speed and response in the conventional P&O
method can be effectively improved. The corresponding control
flowchart of the proposed ASP method is illustrated in Fig. 6.
and
represent the terminal
In Fig. 6,
voltage, current and power of PV modules, respectively, in
and
represent
which denotes the iteration number;
the reference voltage and its step perturbation, respectively;

958

IEEE TRANSACTIONS ON CIRCUITS AND SYSTEMSI: REGULAR PAPERS, VOL. 55, NO. 3, APRIL 2008

Fig. 7. Block diagram of current command for PWM inverter.

and
denote the variations of
, respectively. Define an ASP control law as

and
(18)

where is a positive constant. Because the step perturbation

, is proportional to the ratio of
of the reference voltage,
, it means that
PV power with respect to voltage
the ASP method has the ability to adjust the reference voltage
according to instantaneous operational conditions.
and
When the control process starts, the terminal voltage
of PV modules are first measured, then the power of
current
can be obtained from the product of
and
PV modules
. If the variations of
and
both are not equal to zero,
the reference voltage could be produced by (18), otherwise, the
reference voltage will be maintained at the previous value. The
block diagram of the current command produced by the indirect
voltage control of PV modules for the PWM inverter is depicted
in Fig. 7, where the voltage control error is defined as
is a proportional gain;
is a time delay;
is the amplitude of the current command,
, in
which
is the utility frequency sensed by a phase-lock loop
(PLL) circuit [4]. In other words, the current command
for the PWM inverter is composed of the product of
and
so that the inverter current will be in phase with the
grid voltage when the PWM inverter control error
converges to zero by the proposed ATSMC system.
VI. ACTIVE SUN-TRACKING SCHEME
Because the movement of the sun is slow and monotonous,
and the variation range of the climbing angle is within
,
it is unnecessary to adjust the inclined angle of the PV plate
for simplifying the mechanical framework. Only by way of the
single axis direction control, it can immediately achieve the goal
of collection of maximum irradiation. In this study, an active sun
tracking scheme actuated by a synchronous motor is used for the
sun tracking via the information of the open-circuit voltage of
the PV module, and the corresponding control flowchart of the
active sun tracker is depicted in Fig. 8.
and
represent the present and preIn Fig. 8,
denotes the varivious open-circuit voltages, respectively;
ation of the open-circuit voltage. Because the sun only moves
from the East to the West during one day, the PV plate is rotated by the unit angle for a time at clockwise (CW) in the
beginning of the control process to disturb the corresponding
open-circuit voltage. In this way, it can adjust the rotating direction by observing the variation trend of the open-circuit voltage
to capture more irradiations because the open-circuit voltage of
the PV module is proportional to the corresponding irradiation.
holds, the PV plate is rotated by
If the condition of
the unit angle for a time
at counterclockwise (CCW), i.e.,

Fig. 8. Control flowchart of active sun tracker.

it is returned to the previous location. After that, the control

to further ensure whether or not
process will wait for a time
disappears. If the condition of
the reason for decreasing
holds, the control process also waits for a time
for the next CW rotation. Note that, the function of the waiting
time is helpful for alleviating the extra power consumption in
back and forth motion. According to the aforementioned action
principle, the control target of the active sun tracking scheme
can be achieved. In this study, the implementation of the MPPT
control is faster than in the case of the sun tracker. Because the
sun moves slowly in general, it is reasonable to ignore the interaction of the ASP control and the active sun tracker.
VII. EXPERIMENTAL RESULTS
The validity of the high step-up converter, the PWM inverter
control, the ASP method, and the active sun tracker in the gridconnected PV generation system are verified by the following
experimental results.
A. Experimental Results of High Step-Up Converter
In order to verify the effectiveness of the high step-up
converter, the input side consists of six 75-W PV modules
manufactured by the MOTECH Company (F-MSN-75W-R-02)
connecting in parallel as a low-voltage power source. The
specifications of a single PV module for the standard condi76.78 W, rated
tion (100 mW/cm , 25 C) are rated
17.228 V, rated
4.4567 A, open-circuit
21.61 V, short-circuit
4.9649 A, and PV
%.
In the experimentation, the high step-up converter is designed
initially to operate from the variability dc input of PV modules,
200 V. Assume that the
to deliver a constant dc output,
maximum value of the switch voltage is clamped at 34 V; the
according to (7).
turns ratio is
, is reasonable in pracFrom (6), the related duty cycle,
tical applications if the minimum input voltage is assumed to be
10 V. In order to solve the problem of the output voltage of PV
modules varying with the load variations, this converter with
dc voltage feedback control is utilized to ensure the system stability, and a PWM control IC TL494 is adopted to achieve the

WAI AND WANG: GRID-CONNECTED PV GENERATION SYSTEM

959

goal of feedback control. The prototype with the following specifications is designed to illustrate the design procedure given in
Section III.
Switching frequency:
kHz
Coupled inductor:
H
H

Capacitor:
F

V
V

F
F
F

V
V

Switch:
V
Diode:

V
V
The experimental voltage and current responses of the high
are
step-up converter operating at 320 W-output power
depicted in Fig. 9. From Fig. 9(a), the switch voltage
is clamped at 34 V, which is much smaller than the output
V, and the curve of the switch current
voltage
is similar to a square wave so that it can further reduce the
conduction loss of the switch
. By observing Fig. 9(b) and
keeps about 30A; thus, only a
(c), the primary current
H. According to
smaller core capacity is necessary for
Fig. 9(d)(j), the reverse-recovery currents in all of the diodes
and ) can be alleviated effectively, and the voltages
(
and the high-voltage capacitor
of the clamped capacitor
are close to constant values. Therefore, it can alleviate
the reverse-recovery problem and exhibit the voltage-clamped
effect for further raising the conversion efficiency. From
Fig. 9(i), the selection of the output diode with 200-V blocking
voltage is enough in this application because the voltage of the
200 V)
output diode is limited below the output voltage (
at all times. Fig. 10 summarizes the experimental conversion
efficiency of the high step-up converter under different output
powers. As can be seen from this figure, the conversion efficiency at light powers is over 95% and the maximum efficiency
is over 96.5%, which is comparatively higher than conventional
converters with the same voltage gain.
By the same assumption of the maximum value of the switch
voltage to be clamped at 34 V in this study, the turns ratio is

Fig. 9. Experimental voltage and current responses of high step-up converter

with P
320 W and V
200 V.

according to (7). From (6), the

, is reasonable in practical applirelated duty cycle,
cations if the minimum input voltage is assumed to be 8.614 V
is decreased by a factor of two). Even though the
(i.e., when
solar array voltage changes from 17.228 V to less than 8.614 V
due to clouding, snow or dirt, the step-up converter can compensate for such a drop and provide the inverter with rated voltage
of 200 V regardless of irradiation changes as long as the solar
array voltage is not higher than the switch voltage rating and
the clamped switch voltage is appropriately preset. Although the
combination of a step-down/step-up converter also can provide
an alternative solution handling the voltage variation problem

960

IEEE TRANSACTIONS ON CIRCUITS AND SYSTEMSI: REGULAR PAPERS, VOL. 55, NO. 3, APRIL 2008

Fig. 10. Conversion efficiency of high step-up converter with

under different output powers.

200V

due to irradiation changes, it may fail under excessive voltage

variations and the multistage conversion loss is a latent problem
to be overcome.
The experimental inductor current and switch voltage curves
of the high step-up converter under the condition of 24-V
input voltage, 200-V output voltage, and 1-kW output power
are shown in [17, Fig. 14] to examine the coupled-inductor
capability for a high power application. As can be seen from
this figure, the proposed converter with the same size of the
coupled inductor in [17] can be operated over 1-kW power.
Moreover, the conversion efficiency of the high step-up converter for 24-V input voltage and 200-V output voltage under
different output powers are shown in [17, Fig. 15] to verify
its superiority over that of the coupled-inductor in [16]. According to these comparisons in [17], one can conclude that this
high step-up converter with the appropriate design of circuit
component specifications is still suitable for the application of
kilowatt-level power conversion. Although the efficiency of
the high step-up converter would be reduced because of the
losses in the inductors, the decrease of winding turns and the
increase of air gap in the coupled inductor are helpful to solve
this problem efficiently.
B. Experimental Results of PWM Inverter Control
The circuit components of the PWM inverter scheme
are
IRFP264 (250 V/38 A) and
mH; the switching frequency is
kHz.
Moreover, the parameters of the ATSMC systems for the PWM
inverter scheme are given as follows:
(19)
All the parameters in the ATSMC system are chosen to achieve
the best transient control performance by considering the requirement of stability. The experimental results of the ATSMC
for the PWM inverter of the grid-connected PV generation
system under light load and heavy load are depicted in Fig. 11.
It can be noticed that the output current is almost in phase with
the utility voltage, and the PF of the PWM inverter is higher
than 0.98 that satisfies the PF demand in industrial applications.
The experimental results of the grid-connected PV generation
system with the ATSMC system for the PWM inverter under

Fig. 11. Experimental results of grid-connected PV generation system with

ATSMC for PWM inverter. (a) Light load. (b) Heavy load.

different step load changes are given to examine the load variation effect. In Fig. 12(a), the load is changed from light load
to heavy load; reversely, the load is changed from heavy load to
light load in Fig. 12(b); and in Fig. 12(c)(d) the load changed
from no load to heavy load and heavy load to no load, respectively. As can be seen from this figure, the control performance
of the ATSMC system for the PWM inverter is insensitive to
the abrupt load changes. Generally speaking, inverters may generate small dc voltages/currents due to somewhat asymmetric
gating. In conventional way, transformers are generally connected between inverter and power system to prevent these dc
currents from entering the power system. Due to the powerful
control ability of the proposed ATSMC system, a transformer
is omitted in this study to reduce the energy transformation loss
and lower manufacturing cost. If it is desirable to use a transformer for electric isolation under safety considerations, the proposed control strategies still can work well by additionally considering the turn ratio of the transformer.
C. Experimental Results of ASP Method
Figs. 1315 are the experimental results for the purpose of
verifying the effectiveness of the ASP method. The atmospheric
condition is the irradiation level 88 mW/cm and the module
temperature 53 C dated on the afternoon of November 6, 2005.
The control flowchart as shown in Fig. 6 is implemented via a
DSP with 0.166 ms time step and the parameters of the ASP
method are given as follows:
(20)
The experimental results of conventional P&O method with
fixed step perturbations 0.15 and 0.3 V are illustrated in Figs. 13
and 14, respectively. It is obvious that the smaller step perturbation (0.15 V) results in slower tracking response, but it has stable

WAI AND WANG: GRID-CONNECTED PV GENERATION SYSTEM

961

Fig. 13. Experimental results of grid-connected PV generation system with

conventional P&O method (0.15-V step). (a) Tracking response. (b) Transient
VP and VI curve. (c) Steady VP and VI curve.

D. Experimental Results of Active Sun Tracker

In order to verify the validity of the sun tracking scheme by
way of realistic experimentations, a single PV module actuated
by a synchronous motor manufactured by TUSHING Company
(GL-301) is used to form the active sun tracker, and the cor. The atmospheric circumresponding rotational angle is
stance is the irradiation level 67 mW/cm and the module temperature 30 C dated on October 5, 2005 (Local time PM 3:00,
Taiwan). The control flowchart shown in Fig. 8 is implemented
using a DSP with 1-ms sampling interval, and the parameters of
the active sun tracking scheme are given as follows:
(21)
Fig. 12. Experimental results of grid-connected PV generation system with
ATSMC for PWM inverter under load change. (a) Light load to heavy load.
(b) Heavy load to light load. (c) No load to heavy load. (d) Heavy load to no
load.

operation around the maximum power point. Although the larger

step perturbation (0.3 V) can provide faster tracking response,
oscillations around the maximum power point occurred. For
comparison, Fig. 15 shows the experimental results of the ASP
method. By observing Fig. 15, the operating point could be controlled to locate at the maximum power point rapidly and stably.
As a result, the proposed ASP method indeed yields superior
performance to that of the conventional P&O method.

Two conditions of irradiation are examined here: one is the

nominal condition, and the other is the shading condition by
placing a plastic plate abruptly above a PV plate. The experimental results of the grid-connected PV generation system with
the active sun tracker at nominal and shading conditions are
depicted in Fig. 16. In Fig. 16(a), the open-circuit voltage is
V to
V when the acincreased from
tive sun tracker was started to rotate the PV plate, so that it
would result in the increasing of output powers. In Fig. 16(b),
the open-circuit voltage suddenly decreased because the shading
condition occurred at 34 s such that the PV plate was returned
to its previous location for waiting a span. When the shading
condition was removed at 62s, the PV plate was rotated again
after the waiting time to track the suns direction, so that the

962

IEEE TRANSACTIONS ON CIRCUITS AND SYSTEMSI: REGULAR PAPERS, VOL. 55, NO. 3, APRIL 2008

Fig. 14. Experimental results of grid-connected PV generation system with

conventional P&O method (0.3V-step). (a) Tracking response. (b) Transient
VP and VI curve. (c) Steady VP and VI curve.

Fig. 15. Experimental results of grid-connected PV generation system with

ASP method. (a) Tracking response. (b) Transient VP and VI curve. (c) Steady
VP and VI curve.

open-circuit voltage increased to a steady state. According to

the experimental results in Fig. 16, the expected goal of the active sun tracker can be realized perfectly, and this simple active sun tracking mechanism could be taken as a supervisor
to further provide the adjustable command for PV plates in a
large-scale PV generation system. However, the proposed active
sun tracking scheme can not handle well more than one power
maximum (i.e., power versus voltage functions with two power
peaks due to partially shaded solar arrays with partial bypassing
of solar cells by diodes). How to jump over the local maximum
power point is worthy of investigation in the future research.
VIII. CONCLUSION
This study has successfully developed a grid-connected PV
generation system. The effectiveness of the high step-up converter, the PWM inverter control, the ASP method, and the active sun tracker for a grid-connected PV generation system was
verified by realistic experimentations. According to the experimental results, the conversion efficiency of the high step-up converter at rated power is over 95%, and the overall efficiency including the inverter and losses in the sun tracker is over 85%.
Moreover, the output current of the PWM inverter can almost be
maintained in phase with the utility voltage. The corresponding
PF under different loads are higher than 0.98, satisfying the PF
standards in industrial applications. In addition, the realization
of the ASP method provides faster tracking response with 3 s

Fig. 16. Experimental results of grid-connected PV generation system with active sun tracker. (a) Nominal condition. (b) Shading condition.

WAI AND WANG: GRID-CONNECTED PV GENERATION SYSTEM

settling time and overcomes the oscillation problem in the conventional P&O method for reducing extra power losses. Furthermore, the implementation of the active sun tracking scheme
on the basis of the open-circuit voltage of PV modules, is for
improving the generation efficiency of the fixed-installation PV
array, and lowering the cost of the conventional sun tracker
with light sensors. This system-integration research provides
designers with an alternative choice to convert the PV energy
efficiently, and it also can be extended easily to a large-scale PV
generation system.
REFERENCES
[1] S. R. Bull, Renewable energy today and tomorrow, Proc. IEEE, vol.
89, no. 8, pp. 12161226, Aug. 2001.
[2] S. Rahman, Green power: What is it and where can we find it?, IEEE
Power Energy Mag., vol. 1, no. 1, pp. 3037, Jan. 2003.
[3] J. A. Gow and C. D. Manning, Photovoltaic converter system suitable
for use in small scale stand-alone or grid connected applications, Proc.
IEE Electr. Power Appl., vol. 147, no. 6, pp. 535543, Jun. 2000.
[4] T. J. Liang, Y. C. Kuo, and J. F. Chen, Single-stage photovoltaic energy conversion system, Proc. IEE Electr. Power Appl., vol. 148, no.
4, pp. 339344, 2001.
[5] S. J. Huang and F. S. Pai, Design and operation of grid-connected
photovoltaic system with power-factor control and active islanding detection, Proc. IEE Gen. Trans. Distrib., vol. 148, no. 3, pp. 243250,
2001.
[6] Y. K. Chen, C. H. Yang, and Y. C. Wu, Robust fuzzy controlled photovoltaic power inverter with Taguchi method, IEEE Trans. Energy
Conv., vol. 38, no. 3, pp. 940954, Mar. 2002.
[7] T. Shimizu, O. Hashimoto, and G. Kimura, A novel high-performance
utility-interactive photovoltaic inverter system, IEEE Trans. Power
Electron., vol. 18, no. 2, pp. 704711, Feb. 2003.
[8] J. E. Harry and D. W. Hoare, Electronic power supplies for high-density discharge (HID) lamps, IEE Eng. Sci. Educ. Jo., vol. 9, no. 5, pp.
203206, 2000.
[9] I. Barbi and R. Gules, Isolated dcDC converters with high-output
voltage for TWTA telecommunication satellite applications, IEEE
Trans. Power Electron., vol. 18, no. 4, pp. 975984, 2003.
[10] O. Abutbul, A. Gherlitz, Y. Berkovich, and A. Ioinovici, Step-up
switching-mode converter with high voltage gain using a switched-capacitor circuit, IEEE Trans. Circuits Syst. I, Fundam. Theory Appl.,
vol. 50, no. 8, pp. 10981102, Aug. 2003.
[11] K. C. Tseng and T. J. Liang, Novel high-efficiency step-up converter,
Proc. IEE Electr. Power Appl., vol. 151, no. 2, pp. 182190, Feb. 2004.
[12] N. Mohan, T. M. Undeland, and W. P. Robbins, Power Electronics:
Converters, Applications, and Design. New York: Wiley, 1995.
[13] M. M. Jovanovic and Y. Jang, A new soft-switched boost converter
with isolated active snubber, IEEE Trans. Ind. Appl., vol. 35, no. 2,
pp. 496502, Mar. 1999.
[14] C. M. C. Duarte and I. Barbi, An improved family of ZVS-PWM active-clamping dc-to-dc converters, IEEE Trans. Power Electron., vol.
17, no. 1, pp. 17, Jan. 2002.
[15] E. S. Da Silva, L. Dos Reis Barbosa, J. B. Vieira, L. C. De Freitas, and
V. J. Farias, An improved boost PWM soft-single-switched converter
with low voltage and current stresses, IEEE Trans. Ind. Electron., vol.
48, no. 6, pp. 11741179, Dec. 2001.
[16] Q. Zhao and F. C. Lee, High-efficiency, high step-up dcDC converters, IEEE Trans. Power Electron., vol. 18, no. 1, pp. 6573, Jan.
2003.
[17] R. J. Wai and R. Y. Duan, High step-up converter with coupled inductor, IEEE Trans. Power Electron., vol. 20, no. 5, pp. 10251035,
May 2005.
[18] P. G. Barbosa, L. G. B. Rolim, E. H. Watanabe, and R. Hanitsch, Control strategy for grid-connected dcac converters with load power factor
correction, Proc. IEE Gen. Trans. and Distri.b, vol. 145, no. 5, pp.
487491, 1998.
[19] T. Erika and D. G. Holmes, Grid current regulation of a three-phase
voltage source inverter with an LCL input filter, IEEE Trans. Power
Electron., vol. 18, no. 3, pp. 888895, Mar. 2003.
[20] S. B. Kjaer, J. K. Pedersen, and F. Blaabjerg, A review of single-phase
grid-connected inverters for photovoltaic modules, IEEE Trans. Ind.
Appl., vol. 41, no. 5, pp. 12921306, Sep. 2005.

963

[21] M. Armstrong, D. J. Atkinson, C. M. Johnson, and T. D. Abeyasekera,

Low order harmonic cancellation in a grid connected multiple inverter
system via current control parameter randomization, IEEE Trans.
Power Electron., vol. 20, no. 4, pp. 885892, Apr. 2005.
[22] J. J. E. Slotine and W. Li, Applied Nonlinear Control. Englewood
Cliffs, NJ: Prentice-Hall, 1991.
[23] K. J. Astrom and B. Wittenmark, Adaptive Control. New York: Addison-Wesley, 1995.
[24] R. J. Wai, Adaptive sliding-mode control for induction servomotor
drive, Proc. IEE Electr. Power Appl., vol. 147, no. 6, pp. 553562,
2000.
[25] R. J. Wai and K. M. Lin, Robust decoupled control of direct fieldoriented induction motor drive, IEEE Trans. Ind. Electron., vol. 52,
no. 3, pp. 837854, Jun. 2005.
[26] C. C. Kung and K. H. Su, Adaptive fuzzy position control for electrical
servodrive via total-sliding-mode technique, Proc. IEE Electr. Power
Appl., vol. 152, no. 6, pp. 14891502, 2005.
[27] C. Hua, J. Lin, and C. Shen, Implementation of a DSP-controlled photovoltaic system with peak power tracking, IEEE Trans. Ind. Electron.,
vol. 45, no. 1, pp. 99107, Feb. 1998.
[28] E. Koutroulis, K. Kalaitzakis, and N. C. Voulgaris, Development of
a microcontroller-based, photovoltaic maximum power point tracking
control system, IEEE Trans. Power Electron., vol. 16, no. 1, pp. 4654,
Jan. 2001.
[29] T. T. Yun, D. S. Kirschen, and N. Jenkins, A model of PV generation
suitable for stability analysis, IEEE Trans. Energy Conv., vol. 19, no.
4, pp. 748755, 2004.
[30] M. Veerachary, T. Senjyu, and K. Uezato, Voltage-based maximum
power point tracking control of PV system, IEEE Trans. Aero. Electron. Syst., vol. 38, no. 1, pp. 262270, Jan. 2002.
[31] M. A. S. Masoum, H. Dehbonei, and E. F. Fuchs, Theoretical and experimental analyses of photovoltaic systems with voltage- and currentbased maximum power-point tracking, IEEE Trans. Energy Conv.,
vol. 17, no. 4, pp. 514522, Apr. 2002.
[32] A. Konar and A. K. Mandal, Microprocessor based automatic sun
tracker, Proc. IEE Sci. Meas. Technol., vol. 138, no. 4, pp. 237241,
1991.
[33] B. Koyuncu and K. Balasubramanian, A microprocessor controlled
automatic sun tracker, IEEE Trans. Consumer Electron., vol. 37, no.
4, pp. 913917, Apr. 1991.
[34] A. Ferriere and B. Rivoire, An instrument for measuring concentrated
solar-radiation: A photo-sensor interfaced with an integrating sphere,
Sol. Energy, vol. 72, no. 3, pp. 187193, 2002.
[35] J. D. Garrison, A program for calculation of solar energy collection by
fixed and tracking collectors, Sol. Energy, vol. 72, no. 4, pp. 241255,
2002.

Rong-Jong Wai (M99A00M02SM05) was

born in Tainan, Taiwan, R.O.C., in 1974. He received
his B.S. degree in electrical engineering and the
Ph.D. degree in electronic engineering from Chung
Yuan Christian University, Chung Li, Taiwan,
R.O.C., in 1996 and 1999, respectively.
Since 1999, he has been with the Department of
Electrical Engineering, Yuan Ze University, Chung
Li, Taiwan, R.O.C., where he is currently a Professor.
He is also the Director of the Electric Control and
System Engineering Laboratory at Yuan Ze University, and the Energy Conversion and Power Conditioning Laboratory at the Fuel
Cell Center. He is a chapter-author of Intelligent Adaptive Control: Industrial
Applications in the Applied Computational Intelligence Set (CRC, 1998) and
the coauthor of Drive and Intelligent Control of Ultrasonic Motor (Tsang-Hai
Press, 1999), Electric Control (Tsang-Hai Press, 2002) and Fuel Cell: New Generation Energy (Tsang-Hai Press, 2004). He has authored numerous published
journal papers in the field of control system applications. His research interests
include power electronics, motor servo drives, mechatronics, energy technology,
and control theory applications.
Dr. Wai received the Excellent Research Award in 2000, and the Wu Ta-You
Medal and Young Researcher Award in 2003 from the National Science
Council, R.O.C. In addition, he was the recipient of the Outstanding Research
Award in 2003 and 2007 from the Yuan Ze University, R.O.C.; the Excellent
Young Electrical Engineering Award in 2004 from the Chinese Electrical
Engineering Society, R.O.C; the Outstanding Professor Award in 2004 and

964

IEEE TRANSACTIONS ON CIRCUITS AND SYSTEMSI: REGULAR PAPERS, VOL. 55, NO. 3, APRIL 2008

2008 from the Far Eastern Y. Z. Hsu-Science and Technology Memorial

Foundation, R.O.C.; the International Professional of the Year Award in 2005
from the International Biographical Centre, U.K., the Young Automatic Control
Engineering Award in 2005 from the Chinese Automatic Control Society,
R.O.C., and the Yuan-Ze Lecture Award in 2007 from the Far Eastern Y. Z.
Hsu-Science and Technology Memorial Foundation, R.O.C.His biography
was listed in Marquis Whos Who in Science and Engineering in 20042009,
Marquis Whos Who in 20042007, and Leading Scientists of the World (International Biographical Centre) in 2005, Marquis Whos Who in Asia, Marquis
Whos Who of Emerging Leaders in 20062009, and Asia/Pacific Whos Who
(Rifacimento International) in Vol. VII and VIII.

Wen-Hung Wang was born in Taichung, Taiwan,

R.O.C., in 1981. He received his B.S. and M.S.
degrees in electrical engineering from Yuan Ze
University, Chung Li, Taiwan, R.O.C., in 2003 and
2006, respectively.
His research interests include photovoltaic generation system, power electronics, and adaptive control.