JUI-LIANG YANG, DING-TSAIR SU,

WSEAS TRANSACTIONS on SYSTEMS YING-SHING SHIAO

Research on MPPT and Single-Stage Grid-Connected for Photovoltaic

System
JUI-LIANG YANG, DING-TSAIR SU, YING-SHING SHIAO
Department of Electrical Engineering
National Chang-hua University of Education
Bao-Shan Campus, Address: No.2, Shi-Da Road, Changhua City 520
REPUBLIC OF CHINA.(TAIWAN)
y5111@mail.elvs.chc.edu.tw, sdt@cc.ctu.edu.tw, shiaoing@cc.ncue.edu.tw
http://www.ncue.edu.tw
Abstract: - This paper proposes a grid-connected photovoltaic (PV) system that is consisted of a boost
converter with maximum power tracking, battery charge controller, inverter, and the related control circuits. A
sliding mode controller is designed for controlling the boost converter to output the maximum power of the
solar cells. A control strategy is proposed of a Maximum Power Point Tracking (MPPT) impendence adapter
coupling a photovoltaic generator to a battery. The proposed strategy is based on the sliding mode control
theory of continuing systems to permit a direct control of power converter. The PV generator price is even
relatively high, so the user is interested of the optimal coupling of PV generator to electromechanical loads. In
fact, this system uses the optimal coupling in order to have maximum power during the whole operating period.
A grid-connected inverter is designed to transform the maximum output power into the load and utility. The
real power control is using zero-cross and phase-lock circuits that control the inverter output voltage which is
phase lead and synchronized with the utility. A digital circuit is used to detect the phase angles of the load
voltage and current to control the inverter output amplitude to achieve the reactive power load compensation. A
monitor circuit is designed to detect the voltage and frequency variation in the utility. When the photovoltaic
system encounters a voltage and frequency deviation, the monitor circuit immediately disconnects the
photovoltaic system from the utility to prevent the island effect. The proposed system is implemented by using
OP amplifiers and micro controller (PIC16F877A). Both the simulation and experimental results using
MATLAB show the sliding mode controller for maximum power tracking has good performance. The phase
and amplitude control experiments achieve the real and reactive power control. The island effect experiments
demonstrate evidence of the solar cell system protection ability illustrating the proposed grid-connected
photovoltaic system has good operating performance.

Key-Words: - MPPT, Sliding mode control, Battery charger, Lyapunov function, Boost converter

1 Introduction the PV energy which is suitable for low-power

Due to depletion of fossil energy and system. On the other hands, grid-connected system
environmental contamination, a renewable energy does not require the battery bank and has become the
application such as photovoltaic (PV) system has primary PV application for high power applications.
been widely used for a few decades. The PV energy The main purpose of the grid-connected system is to
is free, abundant and distributed through the earth. transfer maximum solar array energy into grid with a
Photovoltaic (PV) systems are solar energy supply unity power factor.
systems, which either supply power directly to The output power of PV cell is changed by
electrical equipment or feed energy into the public environmental factors, such as illumination and
electricity grid. Generally, photovoltaics are temperature. Since the characteristic curve of a solar
considered as an expensive method of producing cell exhibits a nonlinear voltage-current
electricity. Moreover, with the developing of PV characteristic, a controller named maximum power
technologies, application of photovoltaic in grid- point tracker (MPPT) is required to match the solar
connected situations has grown rapidly, which shows cell power to the environmental changes. Many
that photovoltaic are very attractive to pro- duce algorithms have been developed for tracking
environmentally benign electricity for diversified maximum power point of a solar cell. In recent years,
purpose [1-3]. Among the PV energy applications, the research of the PV MPPT control methods has
they can be divided into two categories: one is stand- been paid extensive attention by many specialist and
alone system and the other is grid-connected system. obtained some fruits such as: P&O and fuzzy control
Stand-alone system requires the battery bank to store etc. Because the output energy of the PV arrays

ISSN: 1109-2777 1117 Issue 10, Volume 7, October 2008

 JUI-LIANG YANG, DING-TSAIR SU,
WSEAS TRANSACTIONS on SYSTEMS YING-SHING SHIAO

changes frequently by the surroundings, improving

the speed of tracking the PV power system could 2 The Boost Converter of Analysis and
obviously improve the system performance.
The existing tracking control methods for the Mathematical Modeling
MPPT can be classified into five categories: (i.) hill- The output energy of the PV arrays is influenced
climbing [1] / perturb and observe (P&O) [2]-[4]. by the surrounding, such as the surrounding
The system then oscillates about the MPP. The temperature, the solar radiation and the terminal
oscillation can be minimized by reducing the voltage of PV arrays etc, the PV arrays characteristic
perturbation step size. However, a smaller curve is shown as Fig.1. Mathematical model of the
perturbation size slows down the MPPT. Hill- solar array can be expresses as Equ. (1) [17][18].
climbing can fail under rapidly changing atmospheric  q 
I = I g − I sat  exp( V ) − 1 (1)
conditions. (ii.) incremental conductance [5-6]; (iii.)  AKT 
open-circuit voltage and short-circuit current [7-9];
(iv.) fuzzy logic control [10]; and (v) neural network where I g denotes a current of a solar array,
control [11]. X. Li et al. introduced the fuzzy V denotes an output voltage of a solar array,
tracking control approaches [10] in 2002. But it is q denotes an electron charge (C) , I sat denotes a cell
very difficult to formulate the fuzzy rules, which are
reverse saturation current (A), K is the Boltzman’s
usually obtained from the trial-and-error procedure.
In 2006, the computational expensive neural network constant (J/K), T is a cell temperature (K), A
was adopted by Tariq et al. [11]. The algorithm denotes the ideality factor。
requires on-line learning in order to make the robot In order to charger the battery, the PV MPPT
perform properly. system adopt step up type DC-DC converter
In this paper, we propose a grid-connected topology system. Fig. 2 presents its system
photovoltaic system that is consisted of a boost structure. Fig. 2 shows the Boost Converter to
converter with maximum power tracking, battery transfer power to load from solar array. In the
charge controller, inverter, and the required control Fig.2, D represents the switch function of the
circuits. we come up with sliding mode control power switch device to control the output energy
strategy [12-16] to solve the problem of MPPT under of the solar array. When D = 0 , power will be
the condition of all kinds of radiation levels. A grid-
switched to open. When D = 1 , power will be
connected inverter [4] is designed to transform the
maximum output power into the load and utility. The switched to close. From Fig.2, we can draw the
real power control is realized using zero-cross and system dynamic model as follows.
dP
phase-lock circuits that control the inverter output 25 dV
=0

voltage which is phase lead and synchronized with 1000w

the utility. A digital circuit is used to detect the phase

20
angles of the load voltage and current to control the 800w dP
>0
dV
inverter output amplitude to achieve the reactive dP
<0
Output Power (w)

power load compensation. A monitor circuit is 15 dV

600w
designed to detect the voltage and frequency
variation in the utility. When the photovoltaic system 10
400w
encounters a voltage and frequency deviation, the
monitor circuit immediately disconnects the 5
photovoltaic system from the utility to prevent the
island effect. This paper is organized as follows. The
0
DC-DC analysis and mathematical modeling are 0 5 10 15 20 25
Output Voltage (V)
presented in Section 2, and Section 3 aims at the
MPPT controller design. Section 4 is focused on the
charge controller design. Section 5 is devoted to the Fig.1 The PV arrays characteristic
grid connected Photovoltaic system. simulation and
experimental results are presented in section 6. In
Section 7 concludes this paper.

ISSN: 1109-2777 1118 Issue 10, Volume 7, October 2008

 JUI-LIANG YANG, DING-TSAIR SU,
WSEAS TRANSACTIONS on SYSTEMS YING-SHING SHIAO

Fig. 2 The circuit diagram of PV DC-DC converter system

di L 3 The DC-DC Controller Design

Vi = L + (1 − D)V0
dt Based on the solar array characteristic curve shown
dV in Fig.1, when the solar array is operating in its
iL = ii − C1 i (2) maximum output power state, we can get
dt
dV0 ∂P / ∂Vi = 0 (6)
i0 = (1 − D)i L − C 2 Where (6) represents maximum power achieved.
dt
∂P ∂ (Vi ii ) ∂ii
V&i = (ii − i L ) / C1 = = Vi + ii = 0 (7)
∂Vi ∂Vi ∂Vi
1 1
i&L = − (1 − D )V0 + Vi (3) From (7), the switch function can be selected,
L L ∂P ∂i
V&0 =
1
(1 − D )i L −
1
i0
S ( x) = = ( i Vi + ii ) (8)
C2 C2 ∂Vi ∂Vi
Based on the two states of PV arrays in Fig.1 and
V&i  the system circuit diagram shown as Fig.2, the switch
 
x& =  i&L  control signal can be selected as
V&0  0 S ≥0
  D= (9)
1 S <0
 
 (i − i ) / C  Let
 i L 1
 ∂S ∂S ∂S
S& = T x& = T f ( x) + T g ( x) Deq = 0 (10)
f ( x ) =  (V1 − V0 ) 
 1
(4) ∂x ∂x ∂x
L Therefore, the equivalent control variable is shown
 
 1 (i − i )  below.
 C2 L 0  i0
Deq = (11)
iL
 0 
g ( x) =  V0 / L 
Theorem: For the system shown in (5) and the
switch function (8), if the expression (9) is adopted,
 −iL / C2  they could make the system eventually stabilize at
x& = f ( x) + g ( x) D (5)
the status that switch function is equal to zero from
any initial state.
Proof:
Let Lyapunov function as

ISSN: 1109-2777 1119 Issue 10, Volume 7, October 2008

 JUI-LIANG YANG, DING-TSAIR SU,
WSEAS TRANSACTIONS on SYSTEMS YING-SHING SHIAO

1 2 And the switch functions as:

V= S >0 (12) ∂P ∂i
2 S ( x) = = ( i Vi + ii ) (19)
dS dP d dP ∂Vi ∂Vi
V& = S = ⋅ ( )
dt dDeq dt dDeq
Substituting equation (1) into (8), then
∂P ∂i 4 The Charge Controller Design
S= = ( i Vi + ii ) This paper uses pulse charge model to save the
∂Vi Vi power on the lead-acid battery. The pulse width of
 qVi  qI V qVi power transition switch is by the voltage of lead-acid
= I g − I sat  exp( ) − 1 − sat i exp( ) battery to control the charge current. The voltage of
 AKT  AKT AKT
battery full-charged for every cell is about 2.43V so
 qVi  qI satVi  qVi  that total full-charged battery is about 29.16V.Fig. 3
= I g + I sat − I sat exp( ) − exp( AKT ) 
 AKT  AKT   shows the measurement circuit of lead-acid voltage
qVi  qVi  is measured by current sensor.
= ( I g + I sat ) − I sat (1 + ) exp( (13)
AKT 
The measurement of Battery voltage is put into
AKT  microchip, PIC16F877A, to control the pulse width
of power transition switch and also to control the
charge current. As Fig.4 shows, during the current
3.1 When S > 0 charged, the current generated by booster converter
Base on (8) (9) and Fig. 1, the system is operating is provided to battery via pulse charge mode.
in left, the switch function D = 0 , and Vi is increase. According to the voltage of lead-acid battery and
dVi maximum power of solar cell, the battery charged
>0 (14) will be the best adjusted.
dt
dS qI  qVi  dVi
= − sat exp( )
dt AKT  AKT  dt
5 The Grid-Connected Photovoltaic
qVi  qV i  q dVi
− I sat (1 + ) exp( ) (15) System
AKT  AKT  AKT dt This paper proposes the grid-connected
dS photovoltaic system as shown in the Fig.5 which
Bring (14) into (15), then < 0 , so includes solar cell, boost converter and grid-
dt
connected inverter. In the grid-connected city power
dS
S <0 (16) mode, city power is a constant AC and inverter is a
dt voltage source. Therefore, by the technology of
Phase-Locked Loop, it will lock the frequency and
3.2 When S < 0 phase of city power to achieve the synchronization
The system is operating in right, the switch with the city power. The phase and amplitude of
function D = 1 , Vi is decreasing, so output voltage is controlled by inverter to achieve the
dVi real power and the reactive power. Besides, in
<0 (17) normal condition, SW2 and SW3 are shorted. As the
dt city power is out of order, SW2 and SW3 are off to
Finally, substituting equation (17) into (15), then prevent island effect.
dS dS
> 0 and S > 0.
dt dt
Obviously, the system could reach global stability
and the switch function is tending to zero whether
the system is operating in left or in right.

3.3 Control Algorithm

The system control rules can be present as follows.
0 S ≥0
D= (18)
1 S <0 Fig.3 Circuit of Charger Control

ISSN: 1109-2777 1120 Issue 10, Volume 7, October 2008

 JUI-LIANG YANG, DING-TSAIR SU,
WSEAS TRANSACTIONS on SYSTEMS YING-SHING SHIAO

Fig. 4 Flow Chart of Charger Control

Fig.5 Architecture of Grid Connected Photovoltaic System

5
ISSN: 1109-2777 1121 Issue 10, Volume 7, October 2008
 JUI-LIANG YANG, DING-TSAIR SU,
WSEAS TRANSACTIONS on SYSTEMS YING-SHING SHIAO

5.1 Grid Connection Inverter on Real Power Vs (t ) = Vm sin ωt (21)

Reactive Power Controller Design
In the Fig.5 grid connection inverter is a single where Vm is the peak of the city power
phase full-bridge inverter. When M1 and M4 is voltage, ω is the angle frequency of the city power
shorted, the load of the current is passed from M1 to voltage.
M4. Similarly, as M2 and M3 are shorted, the load of The output voltage of grid connection inverter
the current is passed from M2 to M3. By the power Vout (t ) is shown below.
switch, M1 ~ M4, the output DC voltage of boost
Vout (t ) = Vc sin(ωt + δ ) (22)
converter is transformed to AC output. The selection
on PWM switches frequency of M1~M4 is by the Where Vc is the peak of the grid connection
phase-lock circuit and frequency doubler. It uses the inverter. The ω is the angle frequency of grid power.
frequency, 60Hz, of the city power to raise the The δ is the phase difference of the grid connection
frequency to 24kHz and makes the output AC of inverter output voltage and city voltage.
inverter synchronize with the city power. However,
The real power of inverter PS is shown below.
the PWM modulation signal of 24kHz can reduce the
value of inductor, L2, and capacitor, C2, of filter. In VmVc / 2
order to filter the high frequency of PWM to achieve PS = sin σ (23)
Xf
60Hz, the L2 and Io need to be adjusted.
If the value of the inductor is too small, the value of The reactive power of inverter QS is shown below.
ripple current gets larger. These will influent both
increasing switch and inductor power lost. Also, if QS =
VC / 2
Xf
[
VC / 2 − (Vm / 2 ) cos σ ] (24)
the value of the inductor is too large, the dynamic
response of inverter will be reduced especially for where X f is inductive reactance of L f .
the non-linear load. Therefore, the selection of
capacitor depends on the operation current and By Equ. (23), to control the angle, δ , of VS (t ) can
switching frequency. control the PS , the real power.
Moreover, the equivalent Series Resistor, ESR, will
By Equ. (24), to control the amplitude, VC , can
directly influent the value of output voltage ripples.
That is because ESR is one of consumption factor of control the reactive power, QS
capacitor. The inner power consumption generates
heat and the lifetime of the capacitor is reduced.
Therefore, the selection of the capacitor depends on
.6 Simulation and Experiment Results
the value of the serial resistor.
So, the frequency selection of inductor and 6.1 Simulation
This paper used Matlab/Simulink to simulate P&O,
1 fuzzy and sliding mode control. Fig. 7 shows
capacitor is less than of f pwm and is greater than
10 comparison of the tracking maximum power which
10 times of as shown in Equ. (20). reveals the sliding mode controller only needs 0.06s,
1 f PWM fuzzy controller needs 0.24s and P&O controller
10 ∗ f output ≤ ≤ (20) needs 1.74s. Therefore, the proposed sliding mode
2π L f C f 10
controller is improved its performance of MPPT. Fig.
Parameters for this paper is f output =60Hz, 8 shows the tracking condition of the sliding mode
controller as the illumination changes. As the time
f pwm =24kHz, L2=2mH and C2=5µF. changes from 1 second to 1.2 second and the
After the solar cells are traced by boost converter isolation changes from 0.6 kW / m 2 to 1 kW / m 2 .
with MPPT, the inverter and the city power will be Hence, the MPPT changes from 14 W to 24.5 W .
paralleled. Therefore, output amplitude and phase of Fig.9 shows the tracking condition of the controller
inverter will directly influent the real power and as the temperature changes. As the time changes
reactive power of the city power. from 2 second to 2.2 second and the temperature
Fig. 5 shows the system circuit for the output
changes from 25 °C to 50 °C . Hence, the MPPT
voltage control mode of inverter is paralleled with
city power. changes from 25 W to 23 W . Fig.10 shows the
tracking condition of the controller as the
The city voltage Vs (t ) is shown below.
illumination changes.

ISSN: 1109-2777 1122 Issue 10, Volume 7, October 2008

 JUI-LIANG YANG, DING-TSAIR SU,
WSEAS TRANSACTIONS on SYSTEMS YING-SHING SHIAO

Fig. 6 The achievement of real power and reactive power

Fig. 7 The system output power (1 kW / m 2 , 25 °C )

ISSN: 1109-2777 1123 Issue 10, Volume 7, October 2008

 JUI-LIANG YANG, DING-TSAIR SU,
WSEAS TRANSACTIONS on SYSTEMS YING-SHING SHIAO

Fig. 8 The system output power (illumination change)

Fig.9 The system output power (temperature change)

8
ISSN: 1109-2777 1124 Issue 10, Volume 7, October 2008
 JUI-LIANG YANG, DING-TSAIR SU,
WSEAS TRANSACTIONS on SYSTEMS YING-SHING SHIAO

Fig. 10 Sliding algorithm of tracking maximum power point tracker

Fig. 11 Diagram of photovoltaic charger system

ISSN: 1109-2777 1125 Issue 10, Volume 7, October 2008

 JUI-LIANG YANG, DING-TSAIR SU,
WSEAS TRANSACTIONS on SYSTEMS YING-SHING SHIAO

6.2 Experiment current is 0.5A. By adjust the pulse width module it

can get the best charger method. Fig.15 shows
Fig.11 shows the control architecture, the control
zero-cross circuit of output plot. The phase-lock
system comprises Microcontroller PIC16F877A,
circuit demonstrates the circuit can track for the city
MOSFET, L etc. Fig.12 shows the photovoltaic
power of frequency 63Hz (in Fig. 16). Fig. 16 , 17,
system. DC/DC is the charge tracking the maximum
18 show the phase-lock circuit the circuit can track
power. The storage battery is as the load. Fig.12
the city power of 60 Hz, 57Hz, 63Hz.. Fig.12 shows
shows the photovoltaic system of
the photovoltaic system of
MPPT. TP &O = 85.9 sec ,
MPPT. TP &O = 85.9sec , TFuzzy = 16.3sec ,
TFuzzy = 16.3 sec , Tsliding = 8.2 sec . Fig.13 shows
Tsliding = 8.2sec
pulse width module is 0.2 and charge current is 0.1A.
Fig.14 shows pulse width module is 0.8 and charge

Fig. 12 P&O, Fuzzy, Sliding charge curve

ISSN: 1109-2777 1126 Issue 10, Volume 7, October 2008

 JUI-LIANG YANG, DING-TSAIR SU,
WSEAS TRANSACTIONS on SYSTEMS YING-SHING SHIAO

Fig. 13 D=0.2 and charge current=0.1A

Fig. 14 D=0.8 and charge current=0.5A

ISSN: 1109-2777 1127 Issue 10, Volume 7, October 2008

 JUI-LIANG YANG, DING-TSAIR SU,
WSEAS TRANSACTIONS on SYSTEMS YING-SHING SHIAO

Fig. 15 The zero-cross circuit of output plot

Fig. 16 The phase-lock circuit tracking for 60 Hz plot

12
ISSN: 1109-2777 1128 Issue 10, Volume 7, October 2008
 JUI-LIANG YANG, DING-TSAIR SU,
WSEAS TRANSACTIONS on SYSTEMS YING-SHING SHIAO

Fig. 17 The phase-lock circuit tracking for 57 Hz plot

Fig. 18 The phase-lock circuit tracking for 63Hz plot

ISSN: 1109-2777 1129 Issue 10, Volume 7, October 2008

 JUI-LIANG YANG, DING-TSAIR SU,
WSEAS TRANSACTIONS on SYSTEMS YING-SHING SHIAO

7 Conclusions References:
This paper presents the sliding mode controller [1] W. Xiao and W. G. Dunford, “A modified
from VSC that controls Booster type converter of adaptive hill climbing MPPT method for
solar array and battery charger to compete the photovoltaic power system,,” in 35th Annual
tracking of MPP. After using Matlab/Simulink to IEEE Power Electron. Specialists Conf., 2004,
simulate P&O control, fuzzy control and sliding pp. 1957-1963.
control, compare its response time. According to the [2] N. Femia, G. Petrone, G. Spagnuolo, and M.
fig.3, the PV maximum power point tracking (MPPT) Vitelli, “Perturb and Observe MPPT Technique
speed is faster apparently comparing the P&O and Robustness Improved,” IEEE International
fuzzy control method. Seen from the simulating Symposium on Industrial Electronics, vol.
result, figure 4, 5 and 6, it is obvious that the 2, May 2004,pp.845 – 850.
efficiency of the PV MPPT is greatly improved the [3] N. Femia, G. Petrone, G. Spagnuolo, and M.
system shown implements the sliding control Vitelli, “Optimization of Perturb and Observe
algorithm. In order to prove the feasibility of the Maximum Power Point Tracking Method,
sliding control method; a system of PV arrays was “IEEE Transactions On Power Electron., vol.
designed as fig.7 and 8. In the control system 20, July 2005, pp. 963-973.
comprises Micro-controller, IC16F877A, IGBT, L [4] N. Kasa. T, Iida. and L, Chen, “Flyback
etc. Experimental results show the response time of Inverter Controlled by Senseless Current MPPT
the proposed sliding control is better than P&O and for Photovoltaic Power System,” IEEE Trans.
fuzzy control. This paper, consisting of a Boost type Iind. Electron., vol, 52, pp, 1145-1152,AUG,
dc/dc converter, which micro-processor, 2005.
PIC16F877A, is used to implement the sliding mode [5] K. Kobayashi, I. Takano, and Y. Sawada, “A
controller. Comparing with other techniques used in study on a two stage maximum power point
the past, the use of the proposed MPPT control tracking control of a photovoltaic system under
improves the PV system performance. The results of partially shaded insolation conditions, “in IEEE
simulation and experiment are present. Power Eng. Society General Meeting, 2003,
A grid-connected inverter is designed to transform pp.2612-2617.
the maximum output power into the load and utility. [6] W. Wu, N. Pongratananukul, W. Qiu, K.
The real power control is using zero-cross and phase- Rustom, T. Kasparis, and I. Batarseh, “DSP-
lock circuits that control the inverter output voltage based multiple peak power tracking for
which is phase lead and synchronized with the utility. expandable power system, “in Eighteenth
A digital circuit is used to detect the phase angles of Annual IEEE Appl. Power Electron. Conf. and
the load voltage and current to control the inverter Exposition, 2003, pp.525-530.
output amplitude to achieve the reactive power load [7] T. Noguchi, S. Togashi, and R. Nakamoto,
compensation. A monitor circuit is designed to detect “Short-current photovoltaic power generation
the voltage and frequency variation in the utility. system, “in Proc. 2000 IEEE International
When the photovoltaic system encounters a voltage Symp. on Ind. Electron., 2000, pp. 157-162.
and frequency deviation, the monitor circuit [8] N. Mutoh, T. Matuo, K. Okada, and M. Sakai,
immediately disconnects the photovoltaic system “Prediction-dada-based maximum-power-point-
from the utility to prevent the island effect. The tracking method for photovoltaic power
proposed system is implemented by using OP generation system,” in 33rd Annual IEEE
amplifiers and micro controller (PIC16F877A). Both Power Electron. Specialists Conf., 2002, pp.
the simulation and experimental results using 1489-1494.
MATLAB show the sliding mode controller for [9] M. A. S. Masoum and H. Dehbonei. E. F. Fuchs,
maximum power tracking has good performance. “Theoretical and Experimental Analyses of
The phase and amplitude control experiments Photovoltaic Systems with Voltage and Current-
achieve the real and reactive power control. The based Maximum Power-point Tracking,” IEEE
island effect experiments demonstrate evidence of Transactions on Energy Conversion. vol. 17,
the solar cell system protection ability illustrating the Issue 4, Dec. 2002, pp. 514-522.
proposed grid-connected photovoltaic system has [10] B. M. Wilamowski and X. Li, “Fuzzy System
good operating performance. Based Maximum Power Point Tracking for PV
System,” IEEE Annual Coufereuce of the
Industrial Electronics Society, Vol4, Nov.
2002,pp. 3280-3284.

ISSN: 1109-2777 1130 Issue 10, Volume 7, October 2008

 JUI-LIANG YANG, DING-TSAIR SU,
WSEAS TRANSACTIONS on SYSTEMS YING-SHING SHIAO

[11] A. Tariq and M. S. Jami Asghar, “Development

of an Analog Maximum Power Point Tracker
for Photovoltaic Panel,” International IEEE
Annual Conference on Power and Electronics
and Drives Systems, Vol.1, Jan. 2006, pp.251-
255.
[12] R. A. Decarlo and S. H. Zak. G. P. Matthews,
“Variable Structure Control of Nonlinear
Multivariable System: A tutorial,” Proceedings
of the IEEE, Vol.76, March. 1988,pp. 212-232.
[13] J. J. Negroni; C. Meza; D. Biel; F.
Guinjoan; ”Control of a buck inverter for grid-
connected PV systems: a digital and sliding
mode control approach,” Industrial Electronics,
2005. ISIE 2005. Proceedings of the IEEE
International Symposium on Volume 2, 20-23
June 2005 Page(s):739 - 744 vol. 2
[14] A. F. Filippov, Differential Equation with
Discontinuous Rightsides, Kluwer Academic,
Dordrecht, Boston, London, 1988.
[15] E. Baily and A. Arapostathis, “Simple sliding
mode control scheme applied to robot
manipulator,” Int. J. of Systems Science, Vol. 17,
pp. 876-885, 1986.
[16] I. Lagrat, H. Ouakka, and I.Boumhidi,” Sliding
Mode PI Controller for Nonlinear Systems,”
Proceedings of the 6th WSEAS International
Conference on Simulation, Modelling and
Optimization, Lisbon, Portugal, September 22-
24, 2006.
[17] T. J.Mabnoto, K. Sopian, W. R. W. Daud, M.
Agoul and A. Zaharim,” Mathematical Model
for Determining the Performance Characteristics
of Muti-Crytalline Photovotaic Modules”, Proc.
of the 9th WSEAS Int. Conf. on Mathematical
and Computational Methods in Science and
Engineering, Trinidad and Tobago, November
5-7, 2007.
[18] D. Wu and Z. Wang, “ Study on Model and
Control System of Variable-Speed Pitch-
Controlled Wind Turbine,” Proceedings of the
7th WSEAS international Conference on
Simulation, Modeling, and Optimization,
Beijing, China,, September 15-17, 2007

ISSN: 1109-2777 1131 Issue 10, Volume 7, October 2008