A novel approach of maximizing energy harvesting in

photovoltaic systems based on bisection search theorem

Peng Wang, Haipeng Zhu, Weixiang Shen, Fook Hoong Choo and Poh Chiang Loh and Kuan Khoon Tan
School of Electrical and Electronic Engineering
Nanyang Technological University, Singapore
Email: {epwang, HPZHU, wxshen, efhchoo, epcloh, TA0008ON}@ntu.edu.sg
Abstract This paper presents a new approach of maximizing
energy harvesting in photovoltaic (PV) systems using bisection
search theorem (BST). The fundamental of the BST and its
application into maximum power point tracker (MPPT) in PV
systems are described. A microcontroller is used to control a
DC/DC boost converter to realize the MPPT function.
Experimental results from solar array simulator show that the
proposed technique can track maximum power point very fast
within a few steps. The feasibility of the proposed MPPT is also
verified in natural environment condition with two solar
modules in parallel. Since the proposed technique is simple in
computation, cheap in implementation and fast in tracking, it is
expected to be widely used to replace conventional MPPT
techniques in PV systems.

the most efficient techniques. For these two techniques, the

derivatives of voltage and power measured from the solar
module are still required. In this paper, a novel MPPT
technique based on bisection search theorem (BST) is
proposed without the necessity of derivative computation.
Thus, the new technique is even simpler in computation,
cheaper in implementation and faster in tracking. The
experimental results from solar array simulator in the
laboratory show that the proposed technique can track
maximum power point very fast within a few steps. The
feasibility of the proposed technique is also verified under real
solar modules at the presence of natural environmental
conditions. Thus, it is expected to be widely used to replace
conventional MPPT techniques in PV systems.

I. INTRODUCTION

II.

With ever-increasing concerns on environment and energy

conservation, the research and development of photovoltaic
(PV) system technology has been accelerated recently due to
its free of pollution, silent operation, long life time and low
maintenance [1]. Although the improvement of solar cell
technologies and the increasing demand for PV systems have
led to a reduction of the price of PV module [2], the costs of
PV systems are still too high. Therefore, it is an important to
design the PV systems which can maximize energy harvesting
from Sun through solar modules. The output power of solar
module varies as a function of solar radiance, temperature and
operating point because of its nonlinear current-voltage (I-V)
relationship [3]. Therefore, the maximum power point tracker
(MPPT) is widely used to maximize the power output of the
solar module. As such, many MPPT techniques have been
developed and implemented. The techniques vary in sensors
required, tracking speed, complexity and cost of hardware
implementation, such as Hill Climbing/P&O and its variants
[4-7], Incremental Conductance and its variants [8-9],
Factional Open-Circuit Voltage [10], [11] and Fractional
Short-Circuit Current [12]. For those techniques, the
qualitative comparison and the quantitative comparison in
terms of simulation and experiment have been conducted
[13-15], respectively. The results show that Hill
Climbing/P&O and Incremental Conductance are in general

A. Prinple of bisection search theorem

The bisection search theorem is one of the bracketing
methods for finding roots of equations [16], [17]. Assume
that function y = f (x) and an interval [a, b] which contains
a root x* of f (x) that lies somewhere in the interval as

shown in Fig. 1 such that f (c ) = 0 .

(a) if f (c) and f (b) have

opposite sign, then squeeze
from left

This work is supported by The National Research Foundation (NRF) of

Singapore through the research project NRF-G-CRP-2007-02.

978-1-4244-4783-1/10/$25.00 2010 IEEE

BISECTION SEARCH THEOREM AND MPPT APPLICATION

2143

(b) if f (a) and f (c) have

opposite sign, then squeeze
from right

Fig. 1 Decision process for bisection search

The BST systematically moves the endpoints of the interval

closer and closer together in the pace of halving interval for
each step until an interval of arbitrarily small width that
brackets the zero is obtained. The decision step for this
process is first to choose the midpoint c = (a + b) / 2 and
then to analyze the three possibilities that might rise: 1). If
f (a) and f (b) have opposite signs, a zero lies in [a, c] . 2) If
f (a) and f (b) have opposite signs, a zero lies in [c, b] . 3) if
f (c ) = 0 , then the zero is c . If either case (1) or (2) occurs,
an interval half as wide as the original interval that contains
the root is found as shown in Fig.1. If the process continues,
c1 , c2 , c3 ,...cn represents the sequence of midpoints which
converges to the root x * within a certain degree of accuracy
(DOA), namely
*

cn x (b a) / 2 for n = 1, 2, ,

(1)

From (1), it can be seen that increasing n can make cn more

closer to x * at the DOA in terms of the value of (b a) / 2 n .

B. MPPT technique based on BST

In order to apply the BST into the MPPT technique in PV
systems, the function of y = f (x) and the variable x should be

power point is essential to find the root in the function P by

regulating the voltage of solar module or solar array. As a
result, the function y = f (x) can be regarded as the change
in power P , where the variable x is the voltage of solar
module or solar array and can be written as:

y = P(Vm )

(2)

where Vm is the voltage across the solar module. To regulate

the voltage Vm , converters are conventionally required as
interface between solar modules and loads.

160

140

120

A DC/DC booster converter is adopted to implement the

proposed MPPT technique in this paper. Fig. 4 shows the
schematic diagram of the system. When the converter works
at continuous mode, the relationship between the duty cycle
and the voltage of solar modules at steady-state is written as:

100

Vm = (1 D )Vout

80

60

40

20

Power (W)

Current (A)

chosen carefully. Fig. 2 shows a real current-voltage ( I V )

curve of a solar module and its corresponding power-voltage
( P V ) curve tested in a sunny day in Singapore.

Fig. 3 Typical change in power versus voltage curve

where Vout is the output voltage of the converter. Thus, the

duty cycle D ( 0 D 1 ) can be used to regulate the voltage
of solar module or solar array [18].

0
0

12

16

20

24

28

(3)

32

Voltage (V)

Fig. 2 Typical current-voltage and power-voltage curves

From the P V curve, it can be observed that the change in

power P with respect to voltage approaches zero at the
maximum power point as illustrated in Fig. 3. Obviously, the
powers at short circuit voltage (0 V) and open circuit voltage
( Voc ) are zero, so maximum power should not happen in
these two particular points even though the changes in power
at these two points are also zero, which is caused by the small
powers around these two points. Thus, tracking the maximum
Fig. 4 DC-DC boost converter for MPPT

2144

The working principle of the system can be described as

follows. The entire system is controlled by a microcontroller.
The voltage and current of solar modules are continuously
sampled and the duty cycle of the converter is calculated by
the microcontroller based on the proposed MPPT technique
using the BST. The flowchart of the program embedded in
the microcontroller is shown in Fig. 5.

III.

EXPERIMENTAL SETUP

To verify the feasibility and effectiveness of the proposed

MPPT technique, the hardware of the system has been set up.
Two sources were used to test the proposed MPPT. One is a
real solar module working under natural environmental
conditions while the other is a solar array simulator which
simulates different sizes of solar modules under different
conditions.

A. Solar modules
The solar module is a power source of PV systems. In this
experimental setup, solar array consists of two solar modules
connected in parallel. The specifications of the solar module
are shown in Table 1.
TABLE I.

SPECIFICATIONS OF SOLAR MODULE

Item
Nominal Maximum Output (Pin)
Nominal Open Circuit Voltage (Voc)
Nominal Short Circuit Current (Isc)
Nominal Maximum Output Voltage (Vmpp)
Nominal Maximum Output Current (Impp)
Nominal Weight

Value
45W
18V
3.45A
14.5V
3.11A
5.3kg

These two solar modules have almost the same I-V

characteristics, then the output voltage of the solar array is
equal to the output voltage of one solar module, and the
output current is twice as much as the output current of one
solar module. Thus, for this solar array at a sunny day, the
voltage at the maximum power should be around 14.5 V
while the current at the maximum power should be around
6.2 A, as shown in Fig. 6.

Fig. 6 Two solar modules for testing MPPT

Fig. 5 Flowchart of program for MPPT implementation

B. Solar array simulator

The solar array simulator (SAS) is the important tools to
investigate the PV systems. It can create the I-V curve of
different sizes of solar array under various environmental
conditions. It can also test the response time of the MPPT
when solar radiation changes from one into another. Fig. 7

2145

shows the Agilent SAS model of E4360. It can output power

up to 1200W with maximum open circuit voltage of 130V
and maximum short circuit current of 10 A. Therefore, any
size of solar array within the above-mentioned ranges can be
simulated by using this SAS.

The first type of test is to prove its feasibility and show the
steps of how the proposed MPPT tracks the maximum power
point under constant solar radiation, where the SAS repeat
generating one I-V curve. The results of tracking process
which shows variation of voltage, current and power are
illustrated in Fig. 9. It can be seen that seven steps are
approximately required for the proposed MPPT to track the
maximum power which is equivalent to about 1.5 seconds.

Fig. 7 Solar array simulator for testing MPPT

C. Hardware implementation of MPPT

The proposed MPPT based on the BST is realized by a
Freescale MC9S08AW60 Microcontroller. The inputs of the
microcontroller are the voltage and current sensed from solar
array and the output of the microcontroller is the pulse width
modulation (PWM) pulse which is used to control the dutycycle of the IGBT in the DC-DC boost converter through a
gate driver circuit as shown in Fig. 4, where the parameters of
major components are illustrated in Table II. The C-language
program is employed in this controller. The experimental
setup of the proposed MPPT is shown in Fig. 8.
TABLE II.

Fig. 9 Verification of feasibility and steps to track maximum power for the
proposed MPPT

PARAMETERS OF MAJOR COMPONENTS

Item
Booster inductor, L (mH)
Smoothing capacitor, C (uF)
Switching frequency, fs (kHz)
IGBT (type)

The second type of test is to investigate its tracking

capability under slow variation of solar radiation, where the
SAS generates two I-V curves repetitively with a defined
interval in between. The results of tracking process which
shows the variation of voltages, currents and powers under
the conditions of the changing solar radiations are illustrated
in Fig. 10. It indicates that the proposed MPPT can still track
the maximum power point despite the changing solar
radiation.

Value
0.2mH
470uF
40khz
IRG4PC50UD

Fig. 8 Experimental setup of MPPT based on BS theorem

IV.

EXPERIMENTAL RESULTS AND DISCUSSIONS

A. Experimental results
Three types of tests have been conducted on the
experimental setup of the MPPT technique based on BST.

Fig. 10 Investigation of tracking capability of the proposed MPPT under

variation of solar radiation

2146

The third type of test is to operate the proposed MPPT

under natural environmental conditions, where the MPPT
controller is connected to two solar modules in parallel as
shown in Fig. 6. The experimental results are illustrated in
Table III. Under different times (or solar radiations), the duty
cycle of power converter can stabilize at 40.6% and 32.7%
(see Figs. 11 and 12), respectively, which are very close to
the values of duty cycles: 41% and 33%. These two values
are corresponding to the maximum powers for the solar
modules at the times indicated in Table III.
TABLE III EXPERIMENTAL RESULTS FOR PROPOSED MPPT TO OPERATE
UNDER TWO SOLAR MODULES IN PARALLEL AT DIFFERENT TIMES (DIFFERENT
SOLAR RADIATIONS)
15:05, 8 May, 2009
Duty cycle
80%
70%
60%
50%
41%
40%
30%
20%

Output Voltage
12.05V
13.16V
14.17V
15.29V
16.03V
15.83V
14.66V
11.43V

Output Power
4.84W
5.77W
6.69W
7.79W
8.57W
8.35W
7.16W
4.35W

16:34, 8 May, 2009

Duty cycle
80%
70%
60%
50%
40%
33%
30%
20%

Output Voltage
13.74V
15.31V
17.13V
19.27V
21.13V
21.56V
21.30V
16.92V

Output Power
6.29W
7.81W
9.78W
12.38W
14.88W
15.49W
15.12W
9.54W

Fig. 12 Power converter at duty cycle of 32.7%

B. Discussions
According to the description of the proposed MPPT
technique and its experimental results, some observations can
be made and discussed here.
Firstly, there is no any requirement of derivatives of
voltage and power measured from solar array, which reduces
the complexity in computation and hence implementation.
Therefore, the proposed MPPT technique is very suitable for
the use of PV systems.
Secondly, the control signal of IGBT can be easily
generated due to the fact that every time the new duty cycle is
simply taken by halving the sum of its previous value and
current value. Thus, tracking maximum power point can be
very fast within a few steps.
Finally, although some ripples of voltage, current and
power can be seen from the experimental data which is not
evitable for the DC/DC booster converter, the proposed
MPPT technique can still work well, which show a certain
degree of robust and reliability of the proposed MPPT
technique.
V.

Fig. 11 Power converter operating at duty cycle of 40.6%

ONCLUSIONS AND FUTURE WORK

In this paper, a novel MPPT technique based on bisection

search theorem has been presented. A microcontroller
together with a DC-DC boost converter is applied to
implement the proposed MPPT control system. Experimental
results from solar array simulator show that the proposed
MPPT technique can track the maximum power very fast
under slow variation of solar radiation within a few steps. The
feasibility of the proposed MPPT control system has also
been verified in natural environmental conditions.
Further research can be conducted on how the proposed
MPPT can track the maximum power at fast change of solar
radiation and how it can identify the global maximum and
local maxima under the partially-shaded solar modules.

2147

REFERENCES
[1]
[2]
[3]
[4]

[5]
[6]
[7]
[8]
[9]
[10]
[11]

[12]
[13]
[14]
[15]
[16]
[17]
[18]

Bialasiewicz, J.T., Renewable Energy Systems with Photovoltaic

Power Generators: Operation and Modeling, IEEE Trans. on Industrial
Electronics, vol. 55, no. 7, pp. 2752-2758, July 2008.
D. Poponi, Analysis of diffusion paths for photovoltaic technology
based on experience curves, Solar Energy, vol. 1, no. 74, pp.331-349,
2003.
Tomas Markvart, Solar electricity, John Wiley & Sons Inc. 2000.
E. Koutroulis, K. Kalaitzakis, and N. C. Voulgaris, Development of a
microcontroller-based, photovoltaic maximum power point tracking
controlsystem, IEEE Trans. Power Electron., vol. 16, no. 21,
pp. 4654, Jan. 2001.
O. Wasynczuk, Dynamic behavior of a class of photovoltaic power
systems, IEEE Trans. Power App. Syst., vol. 102, no. 9,
pp. 30313037, Sep. 1983.
S. Jain andV.Agarwal, A newalgorithm for rapid tracking of
approximatemaximum power point in photovoltaic systems, IEEE
Power Electron. Lett., vol. 2, no. 1, pp. 1619, Mar. 2004.
N. Femia, G. Petrone, G. Spagnuolo, and M. Vitelli, Optimization of
perturb and observe maximum power point trackingmethod, IEEE
Trans. Power Electron., vol. 20, no. 4, pp. 963973, Jul. 2005.
K. H. Hussein and I. Mota, Maximum photovoltaic power tracking:
An algorithm for rapidly changing atmospheric conditions, in IEE
Proc. Generation Transmiss. Distrib., 1995, pp. 5964.
Y.-C. Kuo, T.-J. Liang, and J.-F. Chen, Novel maximum-power-point
tracking controller for photovoltaic energy conversion system, IEEE
Trans. Ind. Electron., vol. 48, no. 3, pp. 594601, Jun. 2001.
G. W. Hart, H. M. Branz, and C. H. Cox, Experimental tests of open
loop maximum-power-point tracking techniques, Solar Cells, vol. 13,
pp. 185195, 1984.
M. A. S. Masoum, H. Dehbonei, and E. F. Fuchs, Theoretical and
experimental analyses of photovoltaic systems with voltage and
current-based maximum power-point tracking, IEEE Trans. Energy
Convers., vol. 17, no. 4, pp. 514522, Dec. 2002.
S. Yuvarajan and S. Xu, Photo-voltaic power converter with a simple
maximum-power-point-tracker, in Proc. 2003 Int. Symp. Circuits
Syst., 2003, pp. III-399III-402. R. M. Hilloowala and A. M.
T. Esram,
P.L. Chapman, Comparison of photovoltaic array
maximum power point tracking techniques, IEEE Trans. on Energy
Conversion, vol. 22, pp. 439-449, June 2007.
R. Faranda, S. Leva, Energy comparison of MPPT techniques for PV
systems, WSEAS Transaction on Power systems, vol. 3, no. 6,
pp. 446-455, June 2008
M. Berrera, A. Dolara, R. Faranda, S. Leva, Experimental test of
seven widely-adopted MPPT algorithms, in Proc. 2009 Int. Conf.
Power Tech, 2009, June 28th - July 2nd, Bucharest, Romania.
Horst A. Eiselt, Carl-Louis Sandblom, Linear programming and its
applications, Springer, 2007.
John H. Mathews and Kurtis K. Fink, Numerical method using Matlab,
Prentice-Hall Inc., Upper saddle river, New Jersey, USA, 2004.
N. Mohan, T. M. Undeland, and W. P. Robbins, Power ElectronicsConverters, Applications, and Design, John Wiley & Sons, 2002.

2148