This document summarizes a study on controlling DC/DC converters for maximum power tracking in solar energy systems. The study analyzed buck, boost, and buck-boost converter performance using a discrete-time proportional-integral control algorithm to track the maximum power points of a solar array. Simulation and experimental results showed the converters successfully tracked maximum power points under changing solar irradiance. The buck converter achieved the highest efficiency of around 90%. The control method and converter configurations allow solar energy systems to effectively transfer power from solar arrays.

1. Control of DC/DC Converters for Solar Energy System with
Maximum Power Tracking
Chihchiang Hua and Chihming Shen
Department of Electrical Engineering
National Yunlin University of Science & Technology, Taiwan
Abstract - Solar power converters were used to convert the
electrical energy from the solar arrays to a stable and
reliable power source. The object of this paper is to analyze
and design DC/DC converters of different types in a solar
energy system to investigate the performance of the
converters. A simple method which combines a discrete
time control and a PI compensator is used to track the
maximum power points (MPP's) of the solar array. The
system is kept to operate close to the MPPT's, thus the
maximum possible power transfer from the solar array is
achieved. The implementation of the proposed converter
system was based on a digital signal processor (DSP).
Experimental tests were carried out for buck, boost and
buck-boost converters using a simple maximum power
point tracking (MPPT) algorithm. The efficiencies for the
system with different converters are compared and
presented.
I . INTRODUCTION
In late years, the problem of energy crunch is more
and more aggravating. Very much exploitation and
research for new power energy are proceeded around the
world. In particular, the solar energy attracts lots of
attention. In recent years, The development of power
semiconductor technology results in easier conversion
between AC and DC. Therefore, the use of solar energy
is emphasized increasingly and regarded as an important
resource of power energy in the next century.
Solar cells represent the fundamental power
conversion unit of a photovoltaic system [1,2].
Crystalline silicon cell technology is well established.
The solar modules have a long lifetime (20 years or more)
and their best production efficiency is approaching 18%.
Solar energy can be utilized in two ways: solar
heating/cooling and solar electricity. Some appliances
can be connected directly because they work on dc at the
system voltage. Other appliances may need a voltage
adaptor to adjust the voltage [5] or a power inverter to
increase the voltage and change it to the ac forms[3].
The applications of solar arrays for residential or storage
systems have been addressed in [4].Solar arrays were
developed for power satellites in the space program [6].
In high power applications, parallel connected converters
are often used to provide electrical power [7].
As the power supplied by solar arrays depends
upon the insolation, temperature and array voltage, it's
necessary to draw the maximum power of the solar array.
Some papers [5-91 had proposed different maximum
power point (MPPT) controls in the past years. A DSP-
3ased simple MPPT algorithm that adjusts the solar array
foltage with a discrete PI control to track the MPP for
:he converter system is used in this paper to achieve the
naximum power transfer and high efficiency for the
solar energy system. The tracking efficiencies are
:on firmed by simulations and experimental results.
11. CHARACTERISTICSOF THE SOLAR ARRAY
Solar array characteristics profoundly influences the
:onverter and control system, therefore it will be briefly
eeviewed here. A solar cell is a nonlinear device and can
?e represented as a current source model as shown in Fig.
1.
Io
Fig. 1 Simplified equivalent circuit of a solar cell.
The traditional I-V characteristics of a solar cell,
when neglecting the internal shunt resistance, is given by
the following equation [11:
where 1, is the light generated current, I,, is the reverse
saturation current, q is the electronic charge, A is a
dimensionless factor, K is the Boltzmann constant, T is
the temperature in OK, Rqisthe series resistance of the
cell.
In literature, instead of the I-V characteristic given
by Eq (1) the V-I characteristic which is given below is
used in many cases:
Equation (1) was used in computer simulations to obtain
the output characteristics of a solar cell as shown in Fig.
2. From these figures, it is observed that the output
characteristics of a solar cell is non-linear and vitally

2. affected by the solar radiation, temperature and load
condition. Each curve has a maximum power point
(Pmax), which is the optimal operation point for the
efficient use of the solar array. When the temperature
rises, the open circuit voltage and the maximum power
fall, but the short circuit current increases slowly as
shown in Fig. 2(C).
1 q
0.9 lOOmr/cn2 I
0.8
I
071
Pmax
r2
I Solar terminal voltage (Unit P.U.) I
! -1
1 4
1 2
0
0 0 1 0 2 0 3 0 4 05 06 07 08 0 9
Solar terminal voltage (Unit P U )
(b)
1 4 .......L.....4........................... I............. :..
I I . , - , , , ,
, , ~ . , . , ,
I , , , , , , ,
0 6 L.
.
.
.
.2..............
.
.
.
.
.
:.
.
.
.
.
.
:.
.
.
.
.
.:
..
.
.
.
.:...x
.
.
.
.
.
.
. . , . . . . . .
. . . , , , . . .
I I I O I I , , ,
0 5 : 1 1 ' 1 1 1 1 t ~
-80 -60 -40 -20 0 M 40 60 80 100
Solar Temperature ( "(')
(c)
Fig. 2 Output characteristicsof a solar array.
(a) Insolation characteristics.
(b) Temperature characteristics.
(c) Solar array parameters variation vs. Temperature.
1
1
1
.MPPT CONTROL APPROACH
As the power supplied by the solar array depends
upon the insolation, temperature and array voltage, an
important consideration in the design of efficient solar
array systems is to track maximum power point correctly.
The purpose of the MPPT is to move the array operating
voltage close to the MPP under changing atmospheric
conditions.
Many methods for tracking maximum power point
had been proposed [5-111. So far, two algorithms often
used to achieve the MPPT are: 1. Perturbation and
observation method; 2. Incremental conductancemethod.
Although the incremental conductance method offers
good performance under rapidly changing atmospheric
conditions, but four sensors are required to perform the
computations [111. If sensors or system require more
conversion time then it will result in a large amount of
power loss. On the contrary, if the execution speed of the
perturbation and observation method is increased, then
the system loss will be decreased. Moreover, this method
only requires two sensors, which results in the reduction
of hardware requirement and cost. Therefore, the
Perturbation and Observation method was used in this
paper to control the output current and voltage of the
solar arrays.
IV. CONFIGURATION OF THE SOLAR SYSTEM
The power circuit of the proposed solar energy
storage system is shown in Fig. 3 The system consists of
a non-linear current source as the power source, a
DCiDC converter power stage circuit as the power
processing unit, a battery set as the load and a control
circuit based on a DSP.
DC / DC
so
arr CONVERTER
Fig. 3 The solar energy storage system circuit.
A .DC/DC Converter
Fig. 4 shows the circuit of the buck converter,
whose output voltage (battery) is less than or equal to the
input voltage yn (solar array voltage). The switch
Soperates at a high frequency to produce a chopped
output voltage. This converter is suitable for use when
the array voltage is high and the battery voltage is low.
The power flow is controlled by adjusting the odoff
duty cycle of the switching. The average output voltage
is given by :
Fig. 5 shows the circuit of the boost converter.
When S is on, the current builds up in the inductor L

3. due to the positive inductor voltage, V, = V,. When S
is off, the voltage across L reverses (V, = Y,, - V,) and
adds to the input voltage, thus makes the output voltage
(battery) greater than the input voltage (solar array
voltage). This converter is suitable for use when the
battery voltage is high and the array voltage is low. For
steady state operation, the average voltage across the
inductor over a full period is zero:
is
+
-
K
v,, to, -(V, - v,, toff = 0 (4)
Therefore
S
0
+
C
vi, .D-T=(V, -V,)(l-D>T (5)
And
Fig. 6 shows the circuit of the buck-boost
converter. When S is on, the voltage acrossL is VL = V,, .
When S is off, the voltage across L reverses(V, = V,).
For steady state operation, the average voltage across L
is zero.
Therefore
And
v*- D
v;,, 1- D
If the solar energy system provides power to a load,
the system often operates away from maximum power
points of the soiar array. Fig. shows the solar array I-V
characteristics and the load curve, together with constant
power curves (P = VI = const) . It is observed that the
delivered output power, which is represented by the
operating point 1, is significantly smaller than the
maximum output power, which is represented by point 2.
In order to ensure a maximum power transfer, DC/DC
converters are used to adjust the voltage at the load to the
value of V, = m,where R is the equivalent
resistance of the load.
t /
vo VR Vm voc
Fig. 7 The operation of the MPPT.
B . System Control Discretization
The block diagram of the MPPT control is shown in
Fig. 8 The proposed control consists of two loops, the
maximum power point tracking loop is used to set a
corresponding Vrcf to the charger input, the regulating
voltage loop is used to regulate the solar array output
voltage according to Vrerwhich is set in the MPPT loop.
The functions of the two loops are performed by a DSP-
based controller. The control:,. senses the solar array
current and voltage to calculate the solar array output
power, power slope and VrcJ for maximum power
control.
-
-
-
I
Fig. 4 The buck converter.
Kp+Ki,S Drive DCDC z; A
circuit converter
I Vf
I I
Fig. 5 The boost converter.
I r
-
I
Fig. 6 The buck-boost converter.
Fig. 8 Block diagram of the control loop.
The algorithm can be expressed as the following
equation :
C is the amount of disturbance and the sign of C is
determined by the power slope. In the voltage loop, the
PI compensator is used to make the system stable.
Therefore, the discretization of the compensator transfer
function is required for system implementation. The
transfer function of a traditional compensator is
T&+l)= Y&)+C (9)
8 3Q

4. converters operate with high efficiencies under the
maximum power point tracking.
where K, is the proportional gain, and K, is the integral
gain.
Rearranging equation (10) in finite-difference form
gives
-
Y(n+l)-Y(n)= K,U(n)+K,[ u(n+1)-U (n)
T T
L J
where T is the sampling time .
Taking the Z-transform of equation (11)yields
U ( Z ) = K p + -
y(z) KIT
z-1
Equation (12) can be expressed in state variable
form as
x ( n + 1 ) = AX(n)+BU(n) (13)
Y(n)= cx(n)+ou(n) (14)
where A=l ,B=K,T, C=l ,D=K,, and X ( n ) is the state
variable.
for digital implementation .
Fig. 9 shows the block diagram of the compensator
I I
Fig. 9 Implementationof the digital compensator.
V. SIMULATIONSAND EXPERIMENTALRESULTS
The converter circuit topology is designed to be
compatible with a given load to achieve maximum
power transfer from the solar arrays. Fig. 10(a)-(c)
shows the tracking waveforms for buck, boost and buck-
boost converters under rapidly changing atmospheric
conditions(about 40-85 mw/cm2). Fig. 10(d) shows the
simulated tracking waveforms. Fig. 1l(a)-(c) shows the
tracking waveforms for buck, boost and buck-boost
converters under slowely changing atmospheric
conditions(about 80 mw/cm2). Fig. 1l(d) shows the
simulated tracking waveforms. From Figs. 10 and 11, it
is observed that the designed dc/dc converters
successfully followed the variations of solar insolation.
Fig. 12 shows the solar array current and the converter
duty cycle waveforms for buck, boost and buck-boost
converters under the MPPT control. The switching
signals are also shown in the figures for comparison. Fig.
13 shows the curves of efficiency vs. output power for
buck, boost and buck-boost converters. The efficiency
curves show the difference among buck , boost and
buck-boost converters. The efficiency of buck converter
is a little bit higher than those for boost and buck-boost
converters. With the proposed MPPT control, these
VI. CONCLUSION
In this paper, a simple MPPT algorithm based on a
DSP is presented to deliver the highest possible power to
the load from the solar arrays. D O C converters of
different types were used in the solar energy system to
investigate the performance of the converters.
Experimental results show that the system can track the
MPP’s correctly under rapidly changing atmospheric
conditions. The simulated and experimental results show
excellent performance (efficiencies about 90% for
DC/DC converters). In the near future, the resonant
conversion techniques will be used for DC/DC
converters to achieve higher efficiency, small size, light
weight, low component stress and low noise for solar
energy systems.
VI1 REFERENCES
[l] T. Markvart, “ Solar Electricity”, John Wiley &
[2] S.Roberts,“Solar
~ - _ _ _ _ _ _
Electricity”, Prentice Hall, 1991.
[3] F. Harashima and H. Inaba, “Microprocessor-
Controlled SIT Inverter for Solar Energy System”,
IEEE Trans. on Industrial Electronics, vol. IE-34,
no. 1, Feb. 1985, pp50-55.
[4] B.K. Bose, P.M. Szczesny and R.L. Steigerwald,
“Microcomputer Control of a Residential
Photovoltaic Power Condictioning System”, IEEE
Trans. on Industry Applications, vol. IA-21, no. 5,
Sep. 1985, ppll82-1191.
[5] Z. Salameh, F.Dagher and W.A.Lynch, “Step-
Down Maximum Power Point Tracker for
Photovoltaic System”, Solar Energy, vol. 46, no.
[6] S. J. Kim, J. R. Lee and B. H. Cho, “Large Signal
Analysis of Space craft Power Systems”, IEEE
[7] K.Siri, T.-F.Wu and C.Q.Lee, “Maximum Power
Tracking in Parallel Connected Converter
System”, IELEC‘9 1, 1991, pp. 128-133.
[8] C.R. Sullivan and M.J. Powers, “A High-
Efficiency Maximum Power Point Tracking for
Photovoltaic Arrays in a Solar-Power Race
Vehicle”, IEEE PESC‘93, 1993, pp.574-580.
[9] K. Siri, V.A. Caliskan and C.Q. Lee, “Peak Power
Tracking in Parallel Connected Converters”,
[101 S.M. Alqhuwainem, “Matching of a dc motor to a
photovoltaic generator using a step-up converter
with a current-locked loop”, IEEE Trans. on
Energy Conversion, vol. 9, no. 1, March 1994.
[111K.H. Hussein and G. Zhao, “MaximumPhotovolatic
Power Tracking:An Algorithm for Rapidly
Changing Atmospheric Conditions”, IEE Proc.-G,
vol. 142,no. 1, Jan. 1995.
Sons, 1994.
1, 1991, pp.278-282.
PESC, 1989,pp.2873-2880.
PESC‘92, 1992, pp.1401-1406.

5. - 08
D
07
2
2 08
h
2 05
2 04
03
a
8
d 02
0 1
0
02 0 4 06 08 1
Solar terminal voltage (Unit P.U.)
(a)
h 08
D
PI 07
& 06
0 5
04
903
?a
P
3 0 2
0 1
0
0 02 04 0 6 0 8 1
Solarterminal voltage (UnitP.U.)
(b)
02 0 4 06 0 8 1
Solar terminal voltage (Unit P U,)
(c)
0.91
Solar terminal voltage (Unit P.U.)
(4
h
D O 8
F4 07
$ 0 6
b 05
a
!$O 4
0 3
0 2
0 1
0
-
2
0 0 2 04 06 0 8 1
Solarterminal voltage (Unit P U,)
(d)
Solar terminal voltage (Unit P.U.)
(b)
0.91 85mw/cn? JI
0.91
Fig. 10 Tracking waveforms of dcldc converts under
rapidly changing atmospheric conditions. (about 40-85
mw/cm2). (a) Buck converter. (b) Boost converter. (c)
Buck-boost converter. (d) Simulation.
Solar terminal voltage (Unit P.U.)
(c)
O.’
00
1
0
.
2 04 0.6 0.8 1
Solar terminal voltage (Unit P.U.)
(4
Fig. 11 Tracking waveforms of dcldc converts under
rapidly changing atmospheric conditions. (about 80
mw/cm2). (a) Buck converter. (b) Boost converter. (c)
Buck-boost converter. (d) Simulation.

6. rmCStopped: 371 AWUlSItlom
::1,,IFF;;,!,
, , , , , I , , , , , , , ,,
82
100 130 160 190 220
Outputpower
Fig. 13 Efficienciesof DCDC converters.
Fig. 12 The waveforms of dc/dc converters under the
MPPT control.(a) The buck converter. (b) The boost
converter. (c) The buck-boostconverter.
832