Available online at www.sciencedirect.

com

ScienceDirect
Availableonline
Available onlineatatwww.sciencedirect.com
www.sciencedirect.com
Energy Procedia 00 (2017) 000–000

ScienceDirect
ScienceDirect
www.elsevier.com/locate/procedia

Energy
EnergyProcedia
Procedia143 (2017) 000–000
00 (2017) 641–648
www.elsevier.com/locate/procedia

World Engineers Summit – Applied Energy Symposium & Forum: Low Carbon Cities & Urban
Energy Joint Conference, WES-CUE 2017, 19–21 July 2017, Singapore

Implementation of a fuzzy
The 15th International controlled
Symposium buck-boost
on District Heating andconverter
Cooling for
photovoltaic systems
Assessing the feasibility of using the heat demand-outdoor
temperature function
Gerard Ang a,
*, Paul for a long-term
Jerome district
Arcibala, Llorien Marieheat demanda,forecast
R. Crisostomo
Conrado F. Ostia, Jr.a, Peter Joshua Cornelius S. Joaquina, Jose Emmanuel C. Tabutona
I. Andrića,b,c*, A. Pinaa, P. Ferrãoa, J. Fournierb., B. Lacarrièrec, O. Le Correc
Mapua University, Intramuros, Manila 1002, Philippines
a
IN+ Center for Innovation, Technology and Policy Research - Instituto Superior Técnico, Av. Rovisco Pais 1, 1049-001 Lisbon, Portugal
b
Veolia Recherche & Innovation, 291 Avenue Dreyfous Daniel, 78520 Limay, France
Abstract c
Département Systèmes Énergétiques et Environnement - IMT Atlantique, 4 rue Alfred Kastler, 44300 Nantes, France

Man’s growing demand for energy calls for an increase in energy supply. Since burning of fossil fuels produces harmful chemicals,
finding new sources of power such as renewable energy is encouraged. This study aims to implement a DC/DC Buck-Boost
Abstractwith fuzzy logic controller for photovoltaic systems. In this study, use of a solar panel was integrated in the design of a
converter
buck-boost converter with a fuzzy logic controller was considered. The researchers used a microcontroller which measures the
District
input and heating networksofare
output voltages the commonly
system andaddressed
conductedinexperiments
the literature as oneitofwas
in which thedetermined
most effective solutions
that when for decreasing
the voltage the
reading of
greenhouse
the gasis emissions
solar panel 20 V, buck from
mode theisbuilding
operatedsector. Thesethe
and when systems
voltagerequire high
reading investments
of the which
panel is at least are
3 Vreturned
– 10 V,through the heat
boost mode is
sales. Due
operated. to the
In the changed
actual climate
application conditions
of the buck-boostandconverter,
building therenovation policies,
researchers heat
verified thatdemand in the future
at a particular time ofcould decrease,
operation, the
prolonging
converter wasthe
notinvestment return
able to boost the period.
voltage higher due to very small input current. For the 6 V, 9 V and 12 V batteries, the recorded
The main
voltage scope
ranges forofstate
thisofpaper is to
charge assess
from 0% the feasibility
- 100% are 6.03of using
V – 7.2theV,heat
8.4 demand
V – 9.54–Voutdoor
and 11.9temperature
V – 12.92 function for heat demand
V, respectively.
forecast. The district of Alvalade, located in Lisbon (Portugal), was used as a case study. The district is consisted of 665
©buildings
2017 Thethat vary in
Authors. both construction
Published by Elsevierperiod
Ltd. and typology. Three weather scenarios (low, medium, high) and three district
renovation scenarios were developed (shallow, intermediate,
Peer-review under responsibility of the scientific committee deep).
of the World ToEngineers
estimate Summit
the error,– Applied
obtainedEnergy
heat demand values
Symposium & were
compared
Forum: Lowwith results
Carbon from&aUrban
Cities dynamic heat Joint
Energy demand model, previously developed and validated by the authors.
Conference.
The results showed that when only weather change is considered, the margin of error could be acceptable for some applications
(the errorBuck-boost
Keywords: in annualconverter;
demand fuzzy
was logic
lowercontrol;
than 20%
solarfor
cellall weather scenarios considered). However, after introducing renovation
scenarios, the error value increased up to 59.5% (depending on the weather and renovation scenarios combination considered).
The value of slope coefficient increased on average within the range of 3.8% up to 8% per decade, that corresponds to the
decrease in the number of heating hours of 22-139h during the heating season (depending on the combination of weather and
renovation scenarios considered). On the other hand, function intercept increased for 7.8-12.7% per decade (depending on the
coupled scenarios). The values suggested could be used to modify the function parameters for the scenarios considered, and
improve the accuracy of heat demand estimations.

© 2017 The Authors. Published by Elsevier Ltd.

Peer-review under responsibility of the Scientific Committee of The 15th International Symposium on District Heating and
Cooling.
* Corresponding author. Tel.: +63-915-701-2925
E-mail address: gang@mapua.edu.ph
Keywords: Heat demand; Forecast; Climate change
1876-6102 © 2017 The Authors. Published by Elsevier Ltd.
Peer-review under responsibility of the scientific committee of the World Engineers Summit – Applied Energy Symposium &
Forum: Low Carbon Cities & Urban Energy Joint Conference.
1876-6102 © 2017 The Authors. Published by Elsevier Ltd.
Peer-review under responsibility of the Scientific Committee of The 15th International Symposium on District Heating and Cooling.
1876-6102 © 2017 The Authors. Published by Elsevier Ltd.
Peer-review under responsibility of the scientific committee of the World Engineers Summit – Applied Energy Symposium & Forum: Low
Carbon Cities & Urban Energy Joint Conference.
10.1016/j.egypro.2017.12.740
642 Gerard Ang et al. / Energy Procedia 143 (2017) 641–648
2 Gerard Ang/ Energy Procedia 00 (2017) 000–000

1. Introduction

Energy is one of the major needs of humanity. For centuries, fossil fuels have been the main source of energy in
the world. Since burning of fossil fuels produce carbon emissions that contribute to global warming and other
pollutions, the need to find new sources of energy preferably renewable sources of energy becomes inevitable.
Alternative sources of energy such as renewable and sustainable energy like solar, wind, geothermal, hydro and
biomass are being considered. Today’s challenge is to design and build devices to harness solar energy effectively.
Fossil fuels which at present are the primary source of energy are slowly being depleted; thus, energy shortage will
take place when all fossil fuels have been completely depleted. Scientists, engineers and other researchers tend to find
ways to address this problem and create efficient ways to harness energy from the sun. This paper suggests that a buck-
boost converter can be controlled by a fuzzy logic controller. The main focus of the study is to integrate buck-boost
converter with a fuzzy logic controller and be implemented in a solar charging system.
The main objective of the study is to design and implement a DC/DC Buck-Boost converter with fuzzy logic
controller for harnessing solar energy. The specific objectives of this research are to design and construct a buck-boost
converter for PV cell, create a fuzzy logic controller for the charging circuit and to integrate a microcontroller to
observe the state of charge of the battery.
The significance of the study is that it will serve as a basis of any study regarding DC-DC converter applied to
renewable energies more specifically solar energy. This research will also serve as benchmark of any further study on
buck-boost DC-DC Converter and lastly this research will be an additional knowledge on applications of DC-DC
converters and fuzzy logic controller.
The scope of the study is that buck-boost converter and fuzzy logic controller are to be integrated to create a solar
charging system. The solar panel has a rating of 10 W and the battery ratings used were 6 V, 9 V and 12 V so that
buck and boost of input voltage can be applied. Triangular membership function was used for the fuzzy logic
controller. However, overcurrent protection, overvoltage protection, over temperature protection and maximum power
tracking were not included in the design of the circuit.

2. Methodology

Figure 1 shows the conceptual framework of the whole system. The prototype was mainly tasked to charge batteries
with different voltage ratings of 6 V, 9 V and 12 V. The design was composed of two main systems: the fuzzy
controller and DC converter system. These systems were activated in sequential order starting from the detection of
system voltages done by the fuzzy controller. The system works by comparing the solar panel voltage to a reference
voltage to produce an error signal which is analyzed by the fuzzy controller to determine the required PWM duty
cycle.
As shown in Figure 2, the beginning of the procedure was to design and construct a buck-boost converter for
photovoltaic cells. The design of buck-boost converter includes sizing of components that will be used. After testing
the converter, the fuzzy logic controller was developed using MATLAB software. The MATLAB fuzzy inference
system (.fis) file was then converted to Gizduino code. After a successful program compilation, the whole system was
integrated by connecting the microcontroller to the buck-boost converter together with the solar panel and the batteries
to be charged. The voltage method was used in this research in determining the state of charge of the battery. By
knowing the minimum and maximum voltage of each battery, the percentage state of charge of the battery was
computed using the formula below.

Vbat  Vmin (1)

% SOC  100%
Vmax  Vmin

Lastly, data has been gathered and analyzed when the whole system is already functioning.
 GerardAng/
Gerard Ang Energy
et al. / Procedia
Energy Procedia 143000–000
00 (2017) (2017) 641–648 6433

Fig. 1. Conceptual Framework.

Fig. 2. Methodology Process.

Figure 3 shows the microcontroller flowchart. The program works by first detecting the voltage of the battery to
be charge. After the detection process, the microcontroller will set the target output voltage of the converter. The error
will be computed by comparing the reference voltage to the solar panel voltage measured by the sensor. The error
signals will be the input of the fuzzy controller. Duty cycle will be computed by the fuzzy controller based on the
fuzzy rules and the error signals. After computing for the required duty cycle, the system will now initiate whether to
buck or boost the input voltage generated by the solar panel.
644 Gerard Ang et al. / Energy Procedia 143 (2017) 641–648
4 Gerard Ang/ Energy Procedia 00 (2017) 000–000

Start

Read solar and battery

voltages

Determine system
voltage and charging
voltage set points

Compute for error

signals

Determine duty cycle

based on fuzzy
algorithm

Yes No
Enable buck mode of Is the duty cycle Enable boost mode of
the dc-dc converter less than 0.5? the dc-dc converter

Fig. 3. Microcontroller Flowchart.

3. Schematic Diagram

Fig. 4 shows the schematic diagram of the charger circuit. It consists of the buck-boost converter, gate driver circuit
and the voltage sensor. The non-inverting buck-boost converter was used in this research because the microcontroller
cannot read negative voltages and non-inverting buck-boost converters has wider output voltage range compared to
the inverting buck-boost converters. Gate driver was used to switch the high-side P-channel MOSFET. The gate driver
circuit was constructed using a small signal MOSFET and a push-pull Amplifier. The voltage sensor was created using
a voltage divider circuit.

Fig. 4. Charger Schematic Diagram.

 Gerard Ang et al. / Energy Procedia 143 (2017) 641–648 645
Gerard Ang/ Energy Procedia 00 (2017) 000–000 5

4. Fuzzy Logic Controller

The fuzzy logic controller of the system was developed using MATLAB software and was then implemented on a
Gizduino microcontroller. The inputs were crisp values from some measuring devices which were converted into
fuzzy values for each input fuzzy set with the fuzzifier. In this research, a two-input fuzzy controller was used. Inputs
were named “error” and “deltaerror”. The crisp values from the input were converted to degrees of membership by
looking up in the membership functions created by the researchers.

Fig. 5. Structure of Fuzzy Logic Controller.

Table 1 shows the fuzzy rule table. The rules were in “If then” format and formally the “If” side was called the
condition and the “Then” side was called the conclusion. In fuzzy logic control, the control strategy is stored in a
natural language. The linguistic variables used for the input membership functions were positive big (PB), positive
small (PS), zero (Z), negative big (NB) and negative small (NS). Fuzzy control rules were derived by observance of
the system behavior, perception and general knowledge of the system. However, fuzzy rules were not perfect so they
can achieve the exact value required by the system and thus; trial and error method was also used in developing the
control rules.

Table 1. Fuzzy Rule Table

deltaerror
NB NS ZZ PS PB
PB ZZ PS PS PB PB
PS NS ZZ PS PS PB
error
ZZ NS NS ZZ PS PS
NS NB NS NS ZZ PS
NB NB NB NS NS ZZ

5. Actual Testing

The prototype was tested in a residential home in Manila. The solar panel was placed in a location with good solar
irradiance. It was expected that the prototype would produce electricity to charge the battery under normal condition.
The data gathered from the testing was recorded using the microcontroller. The actual testing was done for three days
for about 6 to 7 hours per day, and data were gathered every minute. The values that were gathered by the
microcontroller was saved as a notepad text file and was easily transferred to excel in order to make a graphical
representation of it.
646 Gerard Ang et al. / Energy Procedia 143 (2017) 641–648
6 Gerard Ang/ Energy Procedia 00 (2017) 000–000

6. Results and Discussion

After successfully constructing the buck boost converter. It was tested by using a dc power supply with varying
duty cycle from 0 to 1. Table 2 shows the actual and theoretical output voltage response with respect to duty cycle
change with an average input voltage of 8 V.

Table 2. Buck Boost Converter Pre-testing.

Duty cycle Input voltage Output voltage Theoretical Mode
output voltage)
0.00 0.00 0.00 0.00 Buck
0.10 8.48 4.73 1.70 Buck
0.20 8.25 6.14 3.30 Buck
0.30 8.25 6.83 4.95 Buck
0.40 8.39 6.69 6.71 Buck
0.50 8.23 7.56 8.23 Buck
0.60 8.14 8.92 10.18 Boost
0.70 7.89 11.43 13.15 Boost
0.80 7.17 14.89 17.93 Boost
0.90 5.79 17.33 28.95 Boost
1.00 4.35 1.93 ∞ Boost

Based on Figures 6, 7 and 8, the researchers verified that the output voltage and the state of charge (SOC) is directly
proportional to time, while percent voltage regulation (%VR) is inversely proportional to time. In order to demonstrate
the functionality of the prototype, the batteries were discharged and charged to its full capacity. During this period,
input and output voltages were measured per minute. Through this data, %VR of the system can now be computed.
The duty cycle recorded depends upon the amount of sunlight present and the connected battery that is being charged.
The duty cycle ranges from 0 to 0.5 when the system is in buck mode, and from 0.51 to 1.00 when it is in boost mode.
The duty cycle determines when will the converter buck or boost the output voltage. Based on the recorded data, the
duty cycle ranges from 0.4 to 0.8. The state of charge of the battery was also recorded. In order to get an accurate
reading, the batteries were discharged and voltage was measured. The discharged battery was then charged to its full
capacity. In order to determine whether the battery is full, the recorded %VR must be approximately equal to 0%. The
voltage of the fully charged battery was then recorded. The researchers then verified that for the 6 V, 9 V and 12 V
batteries, the recorded voltage ranges for state of charge from 0% – 100% are 6.03 V – 7.2 V, 8.4 V – 9.54 V and 11.9
V – 12.92 V, respectively. Figures 6, 7 and 8 shows the plot of the state of charge and voltage regulation for each
battery.

Fig. 6. Graph of %VR and %SOC against time for the 6 V Battery
 Gerard Ang et al. / Energy Procedia 143 (2017) 641–648 647
Gerard Ang/ Energy Procedia 00 (2017) 000–000 7

Fig. 7. Graph of %VR and %SOC against time for the 9 V Battery

Fig. 8. Graph of %VR and %SOC against time for a 12 V Battery.

Based on the data gathered, the 12 V battery was fully charged for roughly 6 hours and 39 minutes while the 6 V
battery was fully charged for roughly 6 hours and 56 minutes and the 9 V battery was fully charged for roughly 1 hour
and 32 minutes. The charging time depends on the power generated by the solar panel and the total capacity of each
battery. The 9 V battery was charged fastest compared to the other battery because it has lower capacity of 220 mAh
compared to the 12 V and 6 V batteries which has capacity of 4 Ah.

7. Conclusion

The objectives of the study were achieved by the researchers. A DC/DC Buck-Boost converter with fuzzy logic
controller for harnessing solar energy was implemented. A buck-boost converter for PV cell, fuzzy logic controller
for the charging circuit and a microcontroller to monitor the state of charge of the battery was designed and
constructed. Based on the gathered data, the researchers were able to verify that output voltage is directly proportional
to time. Each battery has a specific characteristic wherein the voltage of a battery states its state of charge. The
researchers used 6 V, 9 V and 12 V batteries. The voltage of the 6 V battery was 6.03 V if discharged and 7.2 V if
fully charged. For the 9 V battery, its voltage was 8.4 V if discharged and 9.54 V if fully charged. For the 12 V battery,
its voltage was 11.9 V if discharged and 12.92 V if fully charged.
Buck and boost operations were also verified based on the acquired data. Buck operated within 0.00 – 0.50 duty
cycle (fuzzy output), while boost operated within 0.51 – 1.00 duty cycle (fuzzy output). In the actual application of
the buck-boost converter, the researchers verified that at a particular time of operation, the converter was not able to
boost the voltage higher due to very small input current. When the voltage reading of the solar panel is 20 V, buck
mode was operated since the batteries used were less than 20 V. On the other hand, when the voltage reading of the
solar panel was at least 3 V – 10 V boost mode was operated.
648 Gerard Ang et al. / Energy Procedia 143 (2017) 641–648
8 Gerard Ang/ Energy Procedia 00 (2017) 000–000

8. Recommendation

This paper focused on the implementation of a DC/DC Buck-Boost converter with fuzzy logic controller for
photovoltaic (PV) systems. In terms of fuzzy logic membership functions, the researchers used triangular membership
functions for easier computations. For better accuracy, the researchers recommend using other membership functions
such as Gaussian membership functions and trapezoidal membership functions but it will require a high speed
microprocessor. In order to determine a more accurate state of charge of the battery, researchers recommend using
hydrometer, coulomb counting, or impedance spectroscopy. Overcurrent, overvoltage and over-temperature
protection circuits can be integrated to the buck-boost converter. It is recommended to use microcontrollers with faster
processors to improve performance, though it may cost more. In harvesting solar energy, the researchers recommend
adding additional solar panels or a higher power rating of solar panel in order to charge batteries faster and sufficiently.
Also, maximum power tracking can be integrated to determine the maximum power that can be harnessed from the
sun. Servo motors installed can rotate the solar panel and photodiodes for light sensor.

Acknowledgements

The authors would like to thank the Lord God for giving us the wisdom and strength to finish this study and the
management of Mapua University for the financial support for this research study.

References

[1] Mertens, K., (2014), Photovoltaics Fundamentals, Technology and Practice

[2] Lynn, P., (2010) Electricity from Sunlight: An Introduction to Photovoltaics
[3] Tiwari G N., Dubey S., (2010) Fundamentals of Photovoltaic Modules and their Applications
[4] Wang, W, A Speed Regulation System of DC Motor Based on PWM Technology, Applied Mechanics and Materials, Vols. 29-32 (2010), pp.
2194-2199
[5] Dewangan, AKN, Chakraborty, S. Shukla and V. Yadu, PWM Based Automatic Closed Loop Speed Control of DC Motor, International Journal
of Engineering Trends and Technology, Vol. 3, Issue 2, 2012, pp. 110-112.
[6] Weber, HF, (2011) Pulse Width Modulation DC Motor Control
[7] Choi B., (2013) Pulse Width Modulated DC-to-DC Power Conversion Circuits, Dynamics and Control Designs
[8] Sivanandam, SN, Sumathi, S and Deepa, SN, (2007) Introduction to Fuzzy Logic using MATLAB
[9] Huang, CH., Huang, CC, Ou, TC, Lu, KH and Hong, CM, (2009) Intelligent Fuzzy Logic Controller for a Solar Charging System
[10] Jaber, FS, (2011) Development of a DC-DC Buck Boost Converter using Fuzzy Logic Control
[11] Hsieh, G.C., et al., (2001) Fuzzy-Controlled Li-Ion Battery Charge System with Active State-of-Charge Controller