RCEEE2018 Programme Book

The 10th International Conference on Robotics, Vision, Signal Processing & Power Applications
14th & 15th August 2018 Wembley St-Giles Hotel, Penang
TABLE OF CONTENTS
Messages
Vice Chancellor Universiti Sains Malaysia 1
Dean, School of Electrical and Electronic Engineering USM 2
Conference Chairman ROVISP2018 & RCEEE2018 3
Conference Organising Committee 4
Opening Ceremony Schedule 5
Conference Schedule 6
Keynote Speakers
Keynote Speaker 1: 11
Sustainability Energy System in Collaboration
with High Voltage and High Power System
- Professor Dr. Satoshi Matsumoto
Risk and VaR’s Role in Power System
Sustainability
- Professor Dr. Junzo Watada
Robotic Manipulator Control for Softly Catching
A Falling Object in Future Industrial Application
- Professor Dr. Naoki Uchiyama
Nanotechnology in Medical Diagnostics
- Professor Dr. Uda Hashim
Embrace Artificial Intelligence –
Don’t Fear the Terminator
- Patrick Klotz
Technical Session 16
List of Abstracts 18
Acknowledgements 59
AUN/SEED.Net Profile 60
IET Profile 61
Tekmark Group Profile 62
Message from,
Vice Chancellor
Universiti Sains Malaysia
1
Message from,
The Dean
School of Electrical & Electronic
Engineering, Universiti Sains Malaysia
First of all, I would like to extend a warm welcome to all guests, speakers and
delegates to the 10th International Conference on Robotics, Vision, Signal Processing
and Power Applications 2018 (ROVISP2018) and 11th Regional Conference on
Electrical and Electronic Engineering (RCEEE2018) organised by the School of
Electrical and Electronic Engineering, Universiti Sains Malaysia (USM) and sponsored
by Japan International Cooperation Agency (JICA) for AUN/SEED-Net.
The theme that has been chosen for this conference is “Enabling Research &
Innovation towards Sustainability”. The goal of the conference is to provide an open
forum for researchers to present and discuss their latest innovations from not only
the technological perspective but also from the research perspective. Technological
outputs from research and innovation will surely contribute directly to the well-
being of individuals and society in general. In essence, the sustainability elements of
technological advancement should be given proper emphasis and focus in order to
ensure all potential breakthroughs are positive in nature.
This year conference is unique since this is the first time ROVISP conference series is
organised together with RCEEE2018. I believe the potential of networking and
exchanging ideas and opinions via AUN/SEE-Net platform will be a very positive way
forward. I sincerely hope that all participants will enjoy the keynote speeches and
technical papers that will be presented in this conference amidst the natural settings
of the conference venue. I end my message with my sincere gratitude to all those
who have contributed to this conference including the organizing committee from
School of Electrical & Electronic Engineering-USM, JICA and AUN/SEED-Net,
International Advisory Committee members, Keynote speakers, sponsors, paper
presenters, paper reviewers and all attendees.
May God reward you with bountiful goodness to all.
“When sustainability is viewed as being a matter of survival for your business, I

believe you can create massive change” – Cameron Sinclair
Thank you.
PROF. MOHD RIZAL MOHD ARSHAD
2
Message from,
Conference Chairman
ROVISP2018 & RCEEE2018
It is my great pleasure to welcome you to the 10th International Conference on

Robotics, Vision, Signal Processing and Power Applications (ROVISP2018) which is
organized by the School of Electrical and Electronic Engineering, Universiti Sains
Malaysia. The ROVISP conference was started way back in 2003 whilst the RCEEE
conference under AUN/SEED-Net are in its 11th year. This year, we are very proud
that the ROVISP2018 and the RCEEE2018 are being held in the Wembley St-Giles
Hotel Penang where over 100 papers from various disciplines of Electrical and
Electronic Engineering will be presented.
“Empowering Research and Innovation towards Sustainability” is the conference

theme for this year. The scientists and the engineers are continuously working in the
cutting edge technology and research area which could produce numerous
breakthroughs for the betterment of human lives and the environment. By
empowering the scientists and engineers with appropriate financial support,
adequate infrastructure and research facilities, product innovations and emergence
of new technology could be created into various areas of engineering applications
such as biomedicine, industrial automations, communications, robotics, image
processing, wireless power transfer and nanotechnology.
It is our hope that the joint conference of ROVISP2018 & RCEEE2018 would provide
good opportunity for the academic faculty members, researchers, engineers and
professionals to meet and expound their research findings and discuss possible
future collaborative works. With the presence of large number of participants from
various countries in the ROVISP2018 & RCEEE2018, we hope that you can meet
other researchers in similar fields and have beneficial exchange of knowledge and
research outputs.
Last but not least, I am sure that you would find the Wembley St-Giles Hotel Penang
is a wonderful place for scientific conference such as the ROVISP2018 & RCEEE2018.
The beautiful heritage buildings around the George Town would make your
attendance and stay a worthwhile experience. Have a good and productive
conference.
Thank you.
ASSOC. PROF. Dr. MOHD FADZLI SALLEH

3
CONFERENCE ORGANIZING COMMITTEE
GENERAL CHAIRMAN
MOHD FADZLI MOHD SALLEH
SECRETARIAT FINANCE & FUND

NOR MUZLIFAH MAHYUDDIN RAISING
NUR SYAZREEN AHMAD
TEH JIASHEN
DZATI ATHIAR RAMLI
MOHD NAZRI MAHMUD
TECHNICAL & LOGISTIC &

PUBLICATION ACCOMMODATION
TEOH SOO SIANG DAHAMAN ISHAK
MOHAMAD ADZHAR JAWAWI
NORAMALINA ABDULLAH
MULTIMEDIA & EVENT PUBLICITY & WEBSITE

SUPPORT NOR RIZUAN MAT NOOR
MOHD KHAIRI ISHAK
ABDUL LATIFF ABDUL TAWAB
AHMAD AHZAM LATIB
KAMARULZAMAN ABU BAKAR
HASLAMI HAZWAN HAMZAH
4
OPENING CEREMONY SCHEDULE

Wembley St-Giles Hotel, Penang, Malaysia
Wembley Ballroom 2, 10th Floor
Day 1, Tuesday, 14th August 2018
TIME EVENT
8:00 – 8:20 Arrival of VIPs/guests/delegates
Opening Ceremony
 Negaraku/Menara ILMU
8:20 – 8:40
 SEEE Video
 Doa recitation
Opening Remarks by ROVISP/RCEEE Conference

8:40 – 8:50 Chairman
 Assoc Prof. Dr. Mohd Fadzli Mohd Salleh
Welcome Speech by SEEE Dean

8:50 – 9:00
 Prof. Ir. Dr. Mohd Rizal Arshad
9:00 – 9:10 AUN/SEED-Net Representative
9:10 – 9:30 Officiating Ceremony & Souvenir Presentation
9:30 – 10:00 Keynote 1
10:00 – 10:30 Opening Ceremony ends & Coffee Break
5
CONFERENCE SCHEDULE
Day 1
Tuesday, 14th August 2018
6
Time Activity
8.00 – 9.00 Registration
10th Floor
Opening Ceremony
8.20 – 9.30 Wembley Ballroom 2, 10th Floor
Keynote Speech 1 – Prof. Dr. Satoshi Matsumoto

“Sustainable Energy System in collaboration with High
9.30 – 10.00
voltage and High Power System”
10.00 – 10.30 Coffee Break
Keynote Speech 2 – Prof. Dr. Junzo Watada

10.30 – 11.00 “Risk and VaR's role in Power Systems' Sustainability”
11.00 – 11.30 Keynote Speech 3 – Prof. Dr. Naoki Uchiyama

“Robotic Manipulator Control for Softly Catching a Falling
Object in Future Industrial Applications”
7
Technical Session 1
B1 – DEC/EDAS1 B2 – VIS1 B3 – PHR1 B4 – RF1

11.30 – Wembley Ballroom Wembley Wembley Wembley
13.00 2, Room 7, Room 8, Room 9,
10th Floor 9th Floor 9th Floor 9th Floor
13.00 – Lunch
14.00 Wembley Ballroom 1, 10th Floor
Technical Session 2
F
B1 – EMD1 B2 – VIS2 B3 – PHR2 B4 – ROB
14.00 – Wembley Wembley Wembley Wembley M
15.30 Ballroom 2, Room 7, Room 8, Room 9, M
M
15.30 –
Coffee Break E
16.00
E
Technical Session 3 T
B1 – VIS3 I
B2 – DCE/EDAS2 B3 – RF2 N
16.00 – Wembley Ballroom
Wembley Room 7, Wembley Room 8,
17.00 2, G
9th Floor 9th Floor
10th Floor
20.00 – GALA DINNER [RCEEE PARTICIPANT & COMMITTEE ONLY]

22.30 Wembley Room 10, 9th Floor
Track Details
VIS Vision, Image and Signal Processing
PHR Power Systems, High Voltage Engineering and Renewable Energy
RF Telecommunications, RF, Antenna and Applications
BIO Biomedical, Bio-Engineering and Applications
EMD Electrical Machines, Drives and Power Electronics
DCE/EDAS Devices, Circuits and Embedded Systems/ Electronic Design and Applications
AI Artifical Intelligence and Computer Applications
ROB Robotics, Control, Mechatronics and Automation
8
CONFERENCE SCHEDULE
Day 2
Wednesday, 15th August 2018
9
Time Activity
8.30 – 9.00 Keynote Speech 4 – Prof. Dr. Uda Hashim

“Nanotechnology in Medical Diagnostics”
Keynote Speech 5 –Patrick Klotz

9.00 – 9.30 “Embrace Artificial Intelligence – Don’t Fear the Terminator”

Technical Session 4
B1 – AI1 B2 – EMD2
B3 – PHR3 B4 – BIO
10.00 – Wembley Wembley
Wembley Room 8, Wembley Room 9,
11.45 Ballroom 2, Room 7,
9th Floor 9th Floor
10th Floor 9th Floor
12.00 – AUN/SEED-Net Introduction and Closing Ceremony

13.00 Wembley Ballroom 2, 10th Floor
13.00 –
Lunch
14.00
Technical Session 5
B1 – AI2 B2 – VIS4 B3 – PHR4 B4 –RF3 RCEEE2018

14.00 – Wembley Wembley Wembley Wembley Industrial
15.30 Ballroom 2, Room 7, Room 8, Room 9, Visit
15.30 –
Coffee Break
16.00
Track Details
10
KEYNOTE SPEAKER 1
Professor Dr. Satoshi Matsumoto
Sustainable Energy System in Collaboration with

High Voltage and High Power System
Shibaura Institute of Technology
[email protected]
Electric power demand is increasing year by year as the world population

increases. High power electric energy is mainly generated by hydropower, thermal
power, and nuclear power. On the other hand, renewable energy is generated by
wind power, sunlight, geothermal power, tidal power, wood, waste materials, etc.
In order to maintain the electric power quality and to reduce the greenhouse
effect, which caused by the greenhouse gases including carbon dioxide, it is
important to harmonize the high voltage high power transmission grid with
renewable energy system.
In the lecture, we will introduce the current situation on development of high

efficiency power plant, low transmission loss wire, energy saving equipment,
power storage system together with control system. To maintain reliability, surge
protection equipment, insulation diagnosis, high sensitivity detection of partial
discharge signal, elucidation of discharge phenomenon, high voltage test method,
signal processing are important. Recent research activities related on impulse test
for UHV equipment, partial discharge pulse propagation characteristics using Hertz
vector for magnet wire insulation, transformer tank vibration measurement,
Dissolved Gas Analysis (DGA) for power transformer diagnosis and Electric Double
Layer Capacitor (EDLC) having high capacitance using the graphene for energy
recycle system are introduced. In addition to those, human resource development
is more important in the field of high voltage and high power system. Those are
main topics for my presentation.
11
KEYNOTE SPEAKER 2
Professor Dr. Junzo Watada
Risk and VaR’s Role in Power System’s

Sustainability
Waseda University, Japan
[email protected]
Risk can be defined as Variance. Most often, optimization problem targets

the solution to maximize or minimize a target objective function. However in
investment, it is impossible to obtain the optimal solution without some loss
in payment. Big optimal solutions require much payment. For example, in
daily life, if money is saved in deposit account at a bank, steady interest rate
can be obtained with less risk. However if it is regarding stock market, it can
give a bigger mean return but with low probability. This means, the higher
the amount of money that is being invested, the higher the risk. In this case,
the payment means risk, thus the variance in this case is the payment. In
1952, H. Markowitz published a paper on portfolio model to solve a problem
of maximization of return and minimization of risk. In 1952, he wrote its
model in single objective problem. Subsequently, in this talk we will discuss
on the research that has been carried out under uncertainty to build a model
with risk and value at risk in power systems.
12
KEYNOTE SPEAKER 3
Professor Dr. Naoki Uchiyama
Robotic Manipulator Control for Softly
Catching a Falling Object in Future Industrial
Applications
Toyohashi University of Technology, Japan
[email protected]
This talk presents a controller design of a robotic manipulator for softly catching a
falling object. If a robotic system is able to catch a falling object softly, there will be
many applications expected in human activities such as industry, nursing,
housework and office work, because this ability allows a human operator or
another robot system to move an object to a catching robot without any
transportation system such as a conveyor or a mobile structure. First, a controller
design is presented for catching a falling object with small impact force. In order to
achieve soft catching, precise motion control is required to achieve the same
velocity of a robotic end-effector with a falling object when they are in contact.
Hence, we employ an adaptive controller that consists of a feedback controller to
compensate for disturbance such as friction and a feed forward controller to
improve the tracking performance to a desired trajectory by adjusting controller
parameters in real time. Experimental results with a falling raw egg demonstrate
the effectiveness of the proposed approach. Next, we consider a planar robotic
manipulator, which is typically used in industries, for catching a falling object.
Because falling trajectories can be assumed to be approximately known in
industrial applications, we employ an inexpensive RGB-D camera with a relatively
slow sampling period to estimate an exact falling trajectory in real time. Trajectory
estimation using the Kalman filter and catching control strategies are explained.
Experimental results of catching a ball launched from a toy thrower demonstrate
the performance.
13
KEYNOTE SPEAKER 4
Professor Dr. Uda Hashim
Nanotechnology in Medical Diagnostics

Universiti Malaysia Perlis.,
Perlis, Malaysia
[email protected]
The manipulation of matter at the atomic and molecular scale to create materials
with remarkably varied and new properties is a rapidly expanding area of research
with huge potential in healthcare. The technology allows the possibility of creating
materials, devices and systems with fundamentally new properties and functions,
in medical, it promising platform to be used in therapy, diagnostics, and many
areas of healthcare research, development and clinical applications. The use of
nanotechnology in the field of medicine could revolutionize the way we detect and
treat disease in the future. With the presence of nanotechnology, the performance
of current device can improve drastically. As the consequences, now the cancer
can be detected earlier and the disease outbreak can be controlled. Many
techniques only imagined a few years ago are making remarkable progress
towards becoming realities. Recognizing the importance of nanotechnology, the
Institute of Nano Electronic Engineering was set up by UniMAP. Its main focus is in
developing applications for Nano Biosensor, Nano Biochip and Nano Biomaterials
applicable to human health and medicinal benefits. Among structure being
produced in INEE nanowire, nangap, nanoparticles, nano thin film etc with various
geometries and configuration which provide powerful general platforms for the
ultrasensitive direct electrical detection of biological and chemical species, a
device for detecting cancer, dengue etc cells at earliest stage is being developed
using various structures. The sensor designed bind to complementary targets in
the cells, and generate electrical signal when that particular target is found.
Moreover, various devices such as Nano fibers, magnetic nanoparticles, gold
nanoparticles and carbon nanotubes being used in a sensor that detects proteins
and DNA for indicative of specific disease, various tests have shown these sensors
to be accurate in detecting target molecules and provide results in less than half an
hour. These applications in medicine are undergoing rapid changes. To help
develop a coordinated effort, we must prioritize nanotechnology as the key
important factors to strive in today’s world scenario.
14
KEYNOTE SPEAKER 5
Patrick Klotz
Embrace Artificial Intelligence – Don’t Fear

the Terminator
Datalynx Sdn. Bhd., Malaysia
Self-driving cars, Chat-bots, Robo-servants in restaurants…
When it comes to Artificial Intelligence, the questions today focus around: “Will I
still have a job in 5 years” and even: “Is human kind still needed and existing in 20
years?” Too often we forget, that Artificial Intelligence has been around us for
many years. It helped us to improve our life – be it at work or at home, make life
simpler and safer. But how will it develop over the next years? How can we
benefit from AI even more? What will be the demands on Artificial Intelligence?
One of the interesting applications of AI is when it comes to Big Data. Big Data is
the source for many AI applications, and on the other way around AI also plays a
major role on how new and better algorithms are developed to get to predictive
analytics. The power of machine learning is tremendous. This keynote will discuss
the current trends and developments of AI and some ideas on how we could and
should embrace Artificial Intelligence to support improving our lives and solving
the worlds’ real problems.
15
TECHNICAL SESSIONS
14th August 2018
Time Technical Areas

B1 –
DAY 1 B2 – VIS1 B3 – PHR1 B4 – RF1
DEC/EDAS1
Prof. Hla Myo Assoc. Prof.

Chairperson Dr. Le Viet Prof. Kamal
Tun Dr. Dzati
11.30 – 11.45 C0020 R0084 C0003 C0012

SESSION 1 11.45 – 12.00 C0028 C0031 C0011 C0018
12.00 – 12.15 C0027 R0040 C0013 C0022
12.15 – 12.30 C0009 R0071 R0005 C0025
12.30 – 12.45 C0014 R0072 R0012 R0110
12.45 – 13.00 R0070 R0077 R0018 R0119
13.00 - 1400 Lunch
B1 – EMD1 B2 – VIS2 B3 – PHR2 B4-ROB
Assoc. Prof. Dr.

Chairperson Dr. Amalina Dr. Zuraini
Dr. Rosmiwati Syazreen
14.00 – 14.15 R0034 R0096 R0019 R0043

SESSION 2 14.15 – 14.30 R0035 R0097 R0031 R0061
14.30 – 14.45 R0050 R0107 R0033 R0098
14.45 – 15.00 R0051 R0108 R0046 R0104
15.00 – 15.15 R0075 R0112 R0047 R0106
15.15 – 15.30 R0055
B1 – VIS3 B2 – DCE/EDAS2 B3 – RF2
Chairperson Dr. Teoh Dr. Jiashen Prof. Fadzil Ain
SESSION 3
16.00 – 16.15 R0091 R0114 R0026
16.15 – 16.30 R0121 R0111 R0039
16.30 – 16.45 R0094 R0116 R0041
16.45 – 17.00 R0095 R0074 R0049
16
TECHNICAL SESSIONS
15th August 2018
Time Technical Areas

DAY 2 B1 – AI1 B2 – EMD2 B3-PHR3 B4 – BIO
Assoc. Prof.
Chair person Dr. Noraisyah Prof. David Roderick
Dr. Shahrel
10.00 – 10.15 C0017 C0015 C0016 C0005
10.15 – 10.30 R0017 C0026 C0021 R0025
SESSION 4 10.30 – 10.45 R0030 C0029 C0030 R0060
10.45 – 11.00 R0032 C0032 R0085 R0062
11.00 – 11.15 R0048 R0101 R0066 R0064
11.15 – 11.30 R0068 R0090
11.30 – 11.45 R0013
12.15 – 13.15 Lunch
B1 – AI2 B2 – VIS4 B3-PHR4 B4 – RF3
Assoc. Prof.
Chair person Dr. Khairi Dr. Asiah Dr. Muzlifah
Dr. Haidi
14.00 – 14.15 R0065 R0092 R0069 R0063
14.15 – 14.30 R0080 R0122 R0079 R0067
SESSION 5
14.30 – 14.45 R0087 R0123 R0082 R0081
14.45 – 15.00 R0099 R0131 R0093 R0089
15.00 – 15.15 R0120 R0132 R0135 R0036
15.15 – 15.30 R0124 R0022
15.30 – 16.00 Coffee break
Track Details
17
LIST OF ABSTRACTS
TECHNICAL SESSION 1
DCE/EDAS1: DEVICES, CIRCUITS AND EMBEDDED SYSTEMS /
ELECTRONIC DESIGN AND APPLICATIONS
Analysis on Band Diagram Design and Physical Properties of p-GaN/n-ZnO Light

C0020
Emitting Diodes (LEDs)
Hla Myo Tun, Moe Shweye Myint, Thida Than, Myint Myint Than, Zaw Min Aung, Kyawt Khin
There have been many problems to fabricate the high performance optoelectronic devices in
real world. Among them the estimations of two properties like optical properties and
electronic properties are selected candidates in the combination of II-VI and III-V
semiconductors for advanced optoelectronic devices. The paper presents the band diagram
design and physical properties analyses on p-GaN/n-MgZnO Light Emitting Diodes (LEDs) at
room temperature (300K). The proposed LEDs structure with modeling by using computer
aided design with the help of MATLAB and the outcomes of the recent researches have been
analyzed. The parameters for band diagram evaluation have been applied based on the
electrons and holes effective mass and the donor and acceptor concentrations and so on. In
this work, wide band gap materials of GaN material as a p-type layer and MgZnO material as
an n-type layer for the proposed LEDs design have been designed with various mole fractions
(x=0.1 to x=0.4). Based on the analyses, the better solution for estimation of the physical
properties (optical and electronics) depends only on the mole fraction values of 0.4. Those
wide band gap materials with high mole fraction value have many advantages and could
utilize for several optical electronic devices such as LED, laser diode, photo diode, etc. Because
of the wide band gap nature, short wavelength, and high mobility, those two materials can be
formulated the design for various optoelectronic devices which will be used in optical
communication system. The simulation results based on computer aided design for band
diagram design with electronic and optical properties point out the advancement of the novel
devices to fabricate.
C0028 Implementation of an analog-to-digital stochastic conversion circuit in FPGA

Astria Nur Irfansyah
This paper deals with the implementation of a direct analog-to-digital stochastic conversion
circuit almost entirely on a field-programmable gate-array for stochastic computing.
Stochastic computing uses numbers represented as probabilistic values of bits, therefore
require conversion from radix-2 binary number system. A direct analog-to-stochastic
conversion circuit may simplify the overall system hardware requirement, omitting a
conventional ADC to be included in the system. To evaluate the challenges and performance
of such direct conversion circuit, a prototype implemented in an FPGA is presented in this
work. Measurement results as well as challenges are discussed.
18
C0027 Reactive Command Shaping for Vibration Control

Manuel Jr Del Rosario, Kendrick Amezquita-Semprun, Peter C.Y. Chen, Wei Lin
The differing methods for basic control systems in counteracting vibration often relies on
meticulous and/or constrained working conditions. Simple controllers such as P-controllers are
left to academic exercise unable to cope with industrial demand for vibration control. The work
presented here tackles the problem by reintroducing a known technique in vibration control,
command shaping, maximizing its potential to handle vibrations. The gathered results have
indicated that the configuration of command shaping into a feedback loop, improves the
characteristics of a system without the necessity to change the existing structure of a
controller.
A Hardware Model for the Control and Datapath of a Convolutional Neural

C0009
network using VHDL Behavioral Modeling
Roderick Yap, Lawrence Materum, Miguel Sese, Zion Eric Chan, Mac Excel Fallar, Patrick Julian
Ramos
Convolutional Neural Network (CNN) has found numerous applications in image recognition. In
many researches, CNN is implemented using the software approach such as C or python.
Notably for this learning algorithm, many variable weights are constantly updated during the
training process. In running the python code, one normally would expect a slow process as the
variables are being updated one at a time throughout the training process. The training time,
depending on the image size, could be anywhere from few seconds to several hours. In this
paper, parallelism of updating CNN variables is explored using Very High Speed IC Hardware
Description Language (VHDL). A non-synthesizable VHDL behavioral hardware model
architecture is proposed for the control and data path of a CNN. The advantages and
limitations of using VHDL for the CNN implemented is also presented. The algorithm for the
hardware model is based on a python code that is initially written. When both the python and
VHDL simulations are run, it was observed that VHDL run time is faster for achieving the desired
output.
Exploring the use of VHDL Generate Statement for a Component Saving Behavioral
C0014
Model of a Variable K-Means Clustering Algorithm
Roderick Yap, Lawrence Materum
K-means clustering has found numerous applications in various fields of studies. In neural
networks, a way to minimize memory requirements is to resort to weight sharing using trained
quantization. In this case, the weights can be compressed using K-means clustering. Various
values of K must be tried experimentally to see its effect on the neural network performance.
This paper presents a VHDL behavioral model of a system that can perform K-means clustering
with K being the variable parameter in the model. When the value of K needs to be changed, all
the user must do is change one k parameter in the model and then the whole system will adopt
to the new value of grouping. The system also explored the use of the VHDL generate
statement to save components.
19
R0070 Internet of Things: Security Challenges and Its Future Direction

Tan Soo Fun, Azman Samsudin and Suraya Alias
Internet of Things (IoT) has emerged as a new wave of digital technologies that promise a
transformative experience in future, such as smart cities, smart health, intelligence transport,
etc. While the security always serves as a key factor of public adoption for the new technology,
this paper provides a state of art review on IoT security and its challenges. This paper discusses
the security requirements of IoT from the aspect of confidentiality, Integrity, Authentication,
Authorization, and Availability. Subsequently, several security challenges and privacy concerns
will be highlighted in this paper. Lastly, this paper discusses future works towards a secure IoT.
VIS1: VISION, IMAGE AND SIGNAL PROCESSING
An Improved Initialization Based Histogram of K-Mean Clustering Algorithm for

R0084 Hyperchromatic Nucleus Segmentation in Breast Carcinoma Histopathological
Images
Tan Xiao Jian, Nazahah Mustafa, Mohd Yusoff Mashor and Khairul Shakir Ab Rahman
Mitotic count assessment in breast carcinoma can be a considerable challenge especially when
involve with algorithm development. The challenges lie within the hyperchromatic nucleus
segmentation that served as a fundamental block in mitotic count assessment. In this study, we
proposed an improved initialization based histogram of K-Mean clustering algorithm for
hyperchromatic nucleus segmentation in breast carcinoma histopathological images. The focus
is to segment the hyperchromatic nucleus from the background using K-Mean clustering
algorithm. Conventional initialization method for K-Mean clustering was improved by
establishing a relationship between the hyperchromatic nucleus and the intensity histogram.
75 images captured from 15 histopathological slides were used as dataset. The overall
Sensitivity in ground truth segmentation of the proposed method was found to have a
percentage of 100.0%. The values of precision (AreaPre) and sensitivity (AreaSen) in mitotic
cells area segmentation were found to be promising with percentages of 95.2% and 89.2%,
respectively. The promising results perhaps could be used to enhance performance of the true
mitotic cell detection.
20
C0031 Parallelization of CCSDS Hyperspectral Image Compression using OpenMP

Nur Ashiqin Nur Shaharim, Tan Lit Chez, Ahmad Zarif Zainul and Nor Rizuan Mat Noor
The advent of space technologies ease the collection information from earth surface through
remote sensing. However, the bandwidth and storage limitation impose on spaceborne devices
have increased the need of data compression technique. As the response, Consultative
Committee for Space Data System (CCSDS) have released Lossless Multispectral and Hyperspectral
Image Compression standard (CCSDS-MHC, also referred as CCSDS-123) as the standard to
losslessly compress the hyperspectral image taken by spaceborne/airborne imagers. Currently,
most implementation of the CCSDS-MHC algorithm utilize single processor core/thread for
compression process. However, CCSDS-MHC have the potential to operate on multi-core/multi-
thread system with the use of parallelization. With the introduction of multi-core/multi-thread
processing system on space satellite, this could further decrease the execution time of the system.
In this research, the aim is to design a parallelization algorithm on CCSDS-MHC using Open Multi-
Processing (OpenMP), an open-source C++ application programming interface (API). The first step
of the research is converting the CCSDS-MHC algorithm into a full program in C++, with both
compression and decompression features. Next, the parallelizable section of the algorithm is
identified and coded using OpenMP. The algorithm has been parallelized by dividing the bands of
hyperspectral image into several continuous chunks and running them concurrently. The program
is then tested in several systems with different number of threads. The execution of parallelized
CCSDS-MHC algorithm shows significant speedup for all the system and hyperspectral image
tested.
R0040 Hand Gesture Detection and Its Application to Virtual Reality Systems
M. Fikret Ercan and Allen Qiankun Liu
Detecting hand gestures can provide a useful non-contact interaction tool with machines and
systems and it has been employed for a wide range of applications. Recently, smart glasses and
Virtual Reality (VR) headsets become viable solutions for various training applications ranging
from surgical training in medicine to operator training for heavy equipment. A major challenge in
these systems is to interact with the training platform since user’s view is blocked. In this paper,
we present hand gesture detection using deep learning as a means of interaction with the VR
system. Real world images are streamed by a camera mounted on the VR headset. User’s hand
gestures are detected and blended into the virtual images providing more immersive and
interactive user experience.
21
AUTOMATED TESTING OF VEHICLE INSTRUMENT CLUSTER BASED ON COMPUTER
R0071
VISION
Tan Wei Ren, Wan Shahmisufi bin Wan Jamaludin, Kueh Ying Lin, Muhammad Nasiruddin
Mahyuddin and Bakhtiar Affendi Bin Rosdi
With advancement of technologies, instrument cluster is become more complex. Conventional
manual testing and validation is difficult to cover all the test cases to provide flawless delivery of
product within restricted development timescale. As for now, we are proposing to use computer
vision system in automotive manufacturing for automated design validation testing process. The
purpose of the inclusion of computer vision system is to replace the conventional design
validation testing process which is time consuming and extremely labor intensive. The
speedometer, tachometer, fuel gauge and temperature gauge are inspected by comparing the
accuracy of the pointer detected using the developed algorithms with the pointer position
displayed on the meter. Besides that, the signal indicators status can be inspected using pixel
intensity test. Under the assumption of controlled light surrounding and fixed position of the
camera, pixel intensity can produce accurate results. The deviation error of needle gauge test
and signal indicator test are 2.3%, those error occur while there is some noises influence the
threshold value. Besides that processing time of computer vision is within 0.5 seconds which is
quite efficiency in testing process. In conclusion, the machine vision system is able to help for
spectating the automated instrument cluster testing process.
R0072 Intelligent Mobile Dictionary and Thesaurus

Sasitharan Mahalingam, Nur Intan Raihana Ruhaiyem, Sharifah Mashita Syed-Mohamad
This project is a mobile dictionary application which takes advantage of camera with capability of
Optical Character Recognition (OCR) to scan and extract words from the captured images then
search offline dictionary database. The motivation that drives us toward this implementation is to
contribute to any mobile users that prefer a dictionary application with OCR capabilities that is
easy to use. The objectives of this project are to offer a mobile dictionary that provides OCR
assistance to scan words from captured image and to allow users to search terms for synonym,
nearest and antonym, and to connect with any dictionary databases. The results of this
application are promising and show the potential for further model improvement. The target
language for this application is English.
A Literature Review on Data Conversion Methods on EEG for Convolution Neural

R0077
Network Applications
Chi Qin Lai, Haidi Ibrahim, Mohd Zaid Abdullah, Jafri Malin Abdullah, Shahrel Azmin Suandi and
Azlinda Azman
Convolution Convolution neural network (CNN) presents high robustness in computer vision
applications. In state-of-the-art methods, CNN is being used in EEG processing for various
classification and problem solving. To enable EEG to fit in the CNN architecture, data conversion
of EEG has to be done. The ways of data conversion need to be investigated in order to fully
utilize the information. From the study, it was found that the simplest way of re-arranging the
signal is by creating a two dimensional matrix of channels versus time points. There are
approaches that
compute Pearson correlation coefficients and fit them into a two dimensional matrix to represent
the input signal. There are also methods which extract frequency components and fit them in
matrix structure as channels versus frequency components, such as power spectral density. Other
approaches includes graph representation and wavelet components.
22
PHR1: POWER SYSTEMS, HIGH VOLTAGE ENGINEERING AND RENEWABLE

ENERGY
C0003 Investigation of Grid Faults on Performance of Grid Tied Photovoltaic System

Masoud Ahmadipour, Hashim Hizam, Mohammad Lutfi Othman, Mohd Amran Radzi
In this paper, the influence of various types of faults on grid tied photovoltaic system is
examined in Matlab Simulink environment for a 250-kW grid-connected photovoltaic array
system, considering the aspects of ambient temperature and solar irradiance, grid codes, power
control strategies and utility grid conditions. In order to simulate grid fault types, various
distances from point of common coupling (PCC) of photovoltaic system is taken into account.
Obtained simulation’s results illustrate that symmetrical faults such as short circuit faults in
power network have disturbing influences on ideal procedure of grid tied photovoltaic system.
Furthermore, effect of grid faults on the PV systems performance depend on grid fault types and
is less affected by the distance of the fault.
LVAC Distribution System Design with PV and Centralized Battery Energy Storage
C0011
Integration
Vannak Vai, Long Bun, Marie-Cécile Alvarez, Bertrand Raison
This paper presents a design of AC Low-Voltage (LVAC) distribution system integrated with
Photovoltaic (PV) and Centralized Battery Energy Storage (CeBES) for an urban village. This work
aims at searching for an optimal topology and sizing of PV-CeBES. Firstly, mixed integer
quadratically constrained programming (MIQCP) is developed to find out the optimal topology
which minimize the conductor use and load balancing. Next, a maximum PV penetration is sized
by using iterative technique. The sizing strategy of CeBES is established to remove the reverse
power flow and to increase also autonomous operation according to time-varying solar radiation
and demand curves. The results obtained on an example of LV test system show the validity of
novel algorithm.
A Method for Determining Loss of Energy Expectation for Customer by Using the
C0013
Load Duration Curve
Le Viet Tien
Loss of energy expectation is an important index for evaluation the reliability of electricity
supply for the customers. This paper presents the method for determining loss of energy
expectation for the customer by using the linear matching load duration curve (LMLDC) in
combining with supply capacity of the distribution network. Results of this paper shown that
LMLDC is useful to determine loss of energy expectation for loads.
23
R0005 Load Management Benefits of Electrical Power Grids Performance: A Review

Hussein Jumma Jabir, Jiashen Teh, Dahaman Ishak, Hamza Abunima
The production cost of electricity varies continuously over the year, leading to the price
uncertainty at the customer side. This often results to less equitable fare and lack of fairness
among costumers. The balance between supply and demand-side is a basic requirement of
stable electrical power grids. Load shaping usually is implemented either when demand is more
than supply or when electricity price is high. Load management increases the operation
flexibility as well as offers promising solutions for the reliability and financial performance of
electrical power grids. This paper presents a review of various benefits on electrical power grids
performance achieved by implementing load management activities.
An Investigation of Reactive-Active Power Control approach for Grid-Connected PV

R0012
Arrays in A Low Voltage Distribution System
Hamza Abunima, Jiashen Teh, Hussein Jumma Jabir
The intermittent nature of solar power poses a considerable challenge to the development of
this industry. One of the implications of this characteristic is limiting the penetration capacity of
PV system into the grid. This intermittent power source affects the voltage behavior at the point
of common coupling (PCC) representing voltage rise along the distribution system feeder. This
phenomenon may damage devices in the distribution system and customer side. Although it
usually occurs for short time, it limits the level of solar power penetration in order to protect the
distribution network. Reactive power control approach is used to maintain the voltage at PCC
within the allowable bounds. However, this control approach is sometimes inadequate,
especially in case of high PV power penetration level. This paper investigates combining two
approaches, reactive power control and active power curtailment, in order to maintain the
voltage level in case of high PV power penetration. Open Distribution System Simulator
(OpenDSS) and MATLAB had been used. The IEEE 34-bus distribution test system was examined
to demonstrate the effectiveness of this approach. The results showed that combining the two
control approaches is effective to limit the voltage rise during high PV penetration. Adopting this
control approach would enhance the grid safety and enlarge the hosting capacity for more PV
power injection into the distribution network.
Power Quality Improvements in A Novel 24-Pulse Line Commutated Converter HVDC

R0018
Transmission System
Sabah Ramadhan Mohammed, Jiashen Teh and Mohamad Kamarol Mohd Jamil
The paper reports the investigation of the dynamic operation of novel 24-pulse HVDC
transmission system for eliminating the 11th and 13th harmonic currents from spreading out to
the AC system and improving power quality. The complete simulation of the 24-pulse HVDC
system is performed using the MATLAB/Simulink. Simulation results show that the Total
Harmonic Distortion (THD) of AC system in 12-pulse system is 8.51%. Whereas, the THD of AC
system in 24-pulse system is 4.49%. The waveforms of AC current have improved without
employing 11th and 13th tuned filters at AC systems.
24
RF1: TELECOMMUNICATIONS, RF, ANTENNA AND APPLICATIONS
C0012 Wideband Monopole Antenna with Rotational Circular SRR

Murtala Aminu-Baba, Mohammad Kamal A Rahim, Farid Zubir, Mohd Fairus Mohd Yusoff, Noor
Asmawati Samsuri
This paper presents a technique to achieve two narrow bands and wideband metamaterial
monopole antenna with Rotational Split Ring Resonator (R-SRR). To realize the additional
bands and widen the impedance bandwidth of the conventional monopole antenna,
parametric analysis of the gaps and SRR positions are performed. Three bands are achieved
covering the operating bands of GSM, WLAN WiMAX and 5G applications. The proposed
antenna has the overall size of 14.6 x 11 mm2 and designed on an FR4 substrate due to its low
profile, light weight and low cost. Simulated results are presented and discussed. The proposed
design can be applied to GSM, WLAN WiMAX and 5G applications and have advantages of
compactness, miniaturization and simple design.
C0018 Switchable X-band Reflector/ Wideband Absorber

M. M. Gajibo, M. K. A. Rahim, N. A. Murad, O. Ayop, M. A. Baba, H. A. Majid
A unit cell of a Metamaterials structure capable of switching between AMC reflector and
Absorber is presented. It operates at 11.50 GHz as a reflector and across X-band range (9.50
GHz to 12.50 GHz) as an absorber. Technique of adding neighboring resonances was used to
achieved wideband absorption. An FR4 substrate was used and the incidental wave angles
were varied from 00 to 600. The structure demonstrated above 80% reflection and almost a
unit absorption across X-band region. It was further tested with a microstrip patch antenna.
The structure with respect to the reflector proved to increase the realized gain and directivity
of the antenna. The structure also demonstrated largest bandwidth to be 3.60 GHz and the
lowest to be 1.79 GHz with percentage bandwidth of 32.14% and 15.98% respectively.
C0022 Analysis on Multi-Beam Microstrip Phased Array Antenna For 12 Ghz Applications
Mohamadariff Othman, Wong Soon Kent, Tan Teng Hwang, Maisarah Abu, Tarik Abdul Latef,
Tengku Faiz Tengku Mohmed Noor Izam
This paper presents an analysis on multi beam microstrip phased array antenna at 12 GHz by
using Computer Simulation Technology (CST) simulation tool. The proposed multi beam array
antenna employs five elements of 4×1 array structure excited by 50 Ω corporate feed.
Experimental result of a single 4×1 array antennas are in agreement to the simulation result.
Extensive optimization for optimum antenna performance is generated and beam forming
pattern of 5G antenna is analyzed using CST software. The phased array antenna provides
steerable multi beam with comparable high directivity of at least 16.5 dBi within the range of
16˚ scanning angle.
25
C0025 Multi-Band Frequency Reconfigurable Metamaterial Antenna Design

Adamu. Y. Iliyasu, Mohamad Rijal Bin Hamid, Mohamad Kamal A Rahim, Noor Asmawati
Samsuri, Mohd Fairus Bin Mohd Yusoff
This paper presents the design of multi-band frequency reconfigurable metamaterial antenna.
The design is based on the idea of obtaining multi-band antenna by applying switches in a right
position through frequency reconfiguration in metamaterial antenna. The advantage of
proposed antenna includes compactness and multi-bands frequency operation simultaneously.
This is achieved by eliminating the horizontal ohm’s slot from the basic structure and
introduced H slot in the top patch, extension in the both side of bottom patch and three PIN
Diode switches at different position. Computer Simulation Technology (CST) software is used to
determine the operation and effectiveness of the proposed antenna. From the simulation
results, 1.6 GHz, 2.6 GHz, 3.3 GHz, 4.5 GHz and 5.4 GHz operating bands was obtained with
multi-band characteristics. The realized peak gain for 1.6 GHz, 2.6 GHz, 3.3 GHz, 4.5 GHz and 5.4
GHz are 1.94 dBi, 2.28 dBi, 3.2 dBi, 2.9 dBi, and 4.25 dBi respectively with average efficiency of
95%. From the results obtained, the bands covered the WLAN, WiMAX application and multi-
frequency application like cognitive radio to reduce spectrum congestion and interference.
Automatic Modulation Recognition in Wireless Communication Systems Using

R0110
Feature-Based Approach
Tarik Adnan Almohamad, Mohd Fadzli Mohd Salleh, Mohd Nazri Mahmud, Adnan Haider Yusef
Sa’d, Samir A. Al-Gailani
We propose a novel scheme for automatic recognition of four digital modulation types in the
domain of wireless communication system. The presented scheme exploits the distinct features
reflected by two dimensional asynchronous sampled In-phase-Quadrature data’ histograms
(2D-ASIQHs) for the recognition of various digital modulation types. The effect of the medium
between the transmitter and receiver is only limited to additive white Gaussian noise (AWGN).
The proposed work utilizes support vector machine (SVM) tool for the recognition of various
and popular digital modulation types. Simulation results from the proposed technique showed
accurate discrimination among four digital signals with various bit rates of 100% total
classification accuracy. The simulation was carried out over a broad range of signal-to-noise-
ratio (SNR) of 0–35 dB with a step of 0.5. The proposed method exploits the existed structure of
coherent receivers to construct two-dimensional histograms without the need for any added
hardware devices. Therefore, it provides a promising solution and cost-effective identification
technique for modulation types in next wireless communication systems.
26
Performance Evaluation of MANETs Routing Protocols in Non-Uniform Node

R0119
Density Topology
Mohammed A. Mahdi, Tat-Chee Wan and Rosni Abdullah
This paper studies three reactive mobile ad hoc routing protocols, Ad hoc On Demand Distance
Vector (AODV), Dynamic Source Routing (DSR) and Cluster Based Routing Protocol (CBRP) to
compare their performance in Mobile Ad-Hoc Network (MANET) environments with varying
(non-uniform) node density. For this purpose the efficiency of each protocol was analyzed to
verify their behavior in non-uniform node density environments to determine the protocol
which is more efficient on the basis of average end to end delay and normalized routing load
(NRL). Network Simulation (NS2) is applied to evaluate the protocols performances. The
experiment results show that: CBRP protocol outperformed DSR and AODV in terms NRL when
traffic sources greater than 20 sources and AODV outperforms DSR and CBRP in terms Average
Delay for all traffic sources where AODV achieved the lowest delay. In addition, when node
speed increases, the performance of (AODV, DSR and CBRP) protocols decreases.
27
TECHNICAL SESSION 2
EMD1: ELECTRICAL MACHINES, DRIVES AND POWER ELECTRONICS
Bank Division Topology for Existing Hierarchical Balancing Control in a Grid-scale

R0034
Battery Energy Storage System
Chia Ai Ooi, Jiashen Teh
Lithium-ion (Li-ion) cells are becoming increasingly attractive for use in grid-scale Battery
Energy Storage Systems (BESSs). A key problem with BESSs is the potential for poor utilization
of mismatched cells and reliability issues resulting from the use of large series strings of cells.
This paper investigates the close integration of a cascaded H-bridge multi-level inverter and a
large number of Li-ion cells interfacing with an AC electrical grid. The cells are organized in a
hierarchical structure consisting of modules, sub-banks, banks and phases. The control strategy
includes four levels of balancing: balancing of cells within a module, balancing of modules
within a sub-bank, sub-banks within banks, and banks within phases. The system is validated in
simulation for a 380 kWh BESS using 2835 Li-ion cells. Charge balancing is demonstrated for
mismatched cells by varying the parameters such as ampere-hour capacity, internal resistance,
and initial State-of-Charge. Bank division topology has been implemented to look for the
required cells to achieve the reference voltage by keep dividing the cells into two banks. It
reduces the simulation time by 73.5 % and thus improves the operational efficiency of the
balancing control while achieving cell balancing. This work is intended to address the challenges
of eventual scaling towards a 100 MWh+ BESS, which may be composed upwards of 100 k
individual cells.
R0035 Cell Balancing Topologies in Battery Energy Storage Sys-tems: A Review

Ashraf Bani Ahmad, Chia Ai Ooi, Dahaman Ishak, Jiashen Teh
The performance of a battery energy storage system is highly affected by cell imbalance.
Capacity degradation of an individual cell which leads to non-utilization for the available
capacity of a BESS is the main drawback of cell imbalance. Cell imbalance is common due to
internal and/or external sources such as manufacturing deviations, self-discharge rate variation
and discharging the cells in an unequal number of cycles. Accordingly, several cell balancing
topologies have been proposed by the researchers in the last decade. This paper presents a
review of the proposed cell balancing topologies for BESSs. Comparison among the topologies is
performed for four categories: balancing speed, charge/discharge capability, main elements
required to balance n cell, and application types.
28
R0050 Model Predictive Control of Off-Board PEV Charger

Adrian Soon Theam Tan, Dahaman Ishak, Rosmiwati Mohd-Mokhtar and Sze Sing Lee
This paper presents a three-phase off-board charger for plug-in electric vehicles (PEV) with
model predictive direct power control (MPDPC) for PEV battery charging. In the paper, the
system description of the PEV charger is explained and its control is thoroughly discussed. Then,
the steady-state characteristics of the charger is verified using MATLAB Simulink to confirm the
feasibility of the PEV charger using MPDPC. Results show good power delivery for battery
charging with respect to preset active and reactive power references while providing constant
battery charging voltage and current.
R0051 Grid-Connected PV Generator Using Three-Phase VSC with Model Predictive Control
Adrian Soon Theam Tan, Dahaman Ishak, Rosmiwati Mohd-Mokhtar and Sze Sing Lee
A grid-connected PV generator employing three-phase voltage source converter (VSC) controlled
using model predictive direct power control (MPDPC) is presented in this paper. The system
description of PV generator is explained and its control is thoroughly discussed. Then, the steady-
state characteristics of the system is verified using MATLAB Simulink to confirm the practicality
of controlling grid-connected PV generator using MPDPC. Maximum power is extracted from the
PV array using maximum power point tracking algorithm while the grid compensates the
remaining load active power that is required. Results show good power delivery to the load from
PV array and the grid.
R0075 Wireless Power Transfer

Talha Irshad, Dahaman Ishak, Nor Azura Samsudin
Wireless Power Transfer (WPT) system is an alternative and promising candidate for charging
electrical devices and equipment such as electric vehicles, biomedical implants, portable tools,
smartphones, network sensors, etc. Most widely used WPT techniques are: Inductive Power
Transfer (IPT) and Magnetic Resonance Coupling (MRC). In this paper, a comparative analysis of
rectangular and circular structures of four-coil Wireless Power Transfer System is presented. The
calculated results are verified by using ADS simulation. This paper achieves transmission
efficiency for the circular and rectangular coil structures of 84.045% and 79.525% at a distance of
13cm respectively.
29
Artificial neural network application for damages classification in fibreglass pre-

R0096
pregs laminated composites (FGLC) from ultrasonic signal
M. F. Mahmod, Elmi Abu Bakar, Raiminor Ramzi, Mohd Azhar Harimon, N. Abdul Latif,
Mohammad Sukri Mustapa and Al Emran Ismail
Ultrasonic testing (UT) is a major Non-Destructive Test (NDT) technique used in composite
laminates inspection. The traveling ultrasonic waves in various mode display is used to detect
any damage. A qualified NDT inspector who complies with ISO 9712 is required to interpret the
damages form the ultrasonic signal. However, the inspection performance is subjected to human
factors due to fatigue and lack of concentration. Therefore, a study of a damages detection
system is carried out to detect and classify the damages. In this study, the damage detection of
pre-impregnated laminated composites has been made using ultrasonic prototype machine
namely ISI i-InspeX TWO and the classification from the extracted features of A-scan mode
display has been performed using Back Proportional Network (BPN). The classification employs
two classification stages which is CLASS-1 and CLASS-2 for the first and the second phase
respectively. The results of the average performance of CLASS-1 concluded that the proposed
approach attained reliable results with the accuracy of 99.99% while the performance result of
CLASS-2 was 94.21%. Thus, these promising classification performances showed that the
proposed system is applicable to assist NDT inspectors in their quality inspection process.
R0097 Cross-talk Level of Mechanomyography Signal on Compartmental Forearm Muscle

Mohamad Razif Mohamad Ismail, Chee Kiang Lam, Kenneth Sundaraj and Mohd Hafiz Fazalul
Rahiman
Mechanomyography (MMG) signal is a technique for recording and interpreting mechanical
activity (vibration) in contracting muscle. In MMG, cross-talk refers to the contamination of the
signal from the muscle of interest by a signal from another muscle or muscle group in close
proximity. This study analyzed the cross-talk in MMG signals generated by the Palmaris Longus
(PL), Flexor Carpi Ulnaris (FCU), Extensor Digitorum (ED) and Extensor Carpi Ulnaris (ECU) of
forearm muscles during maximal voluntary contraction (MVC) force of hand exercise motions
(grip strength supinated, grip strength pronated, finger flexion and pinch grip) by using different
level of hand exercise balls (soft, medium and firm). Eight healthy subjects using dominant hand
(mean ± SD: age = 23.25±1.91 year) participated in this study to record MMG signals. During
each muscle contraction, four VMG sensor (TSD250A) sensitive accelerometer (32.64mm
diameter) type were used with BIOPAC Vibromyography Systems to obtain MMG signals from
flexor and extensor side of forearm muscles. Peak cross-correlation coefficients at zero time lags
were used for quantification of the cross-talk. The results of cross-talk level between flexor and
extensor side indicates that the magnitude range level of cross-talk for flexor side muscle groups
were higher than the extensor side muscle groups for all hand exercise motions exerted.
30
R0107 Biocoustic Sound Separation based on FastICA and Infomax Algorithms

Norsalina binti Hassan and Dzati Athiar Ramli
In bioacoustics technology, advances such as automated sound recognition based on animal
vocalization help in biological research and environmental monitoring. However, in a noisy
acoustic environment, where there will be interferences such as overlapping sounds made by
multiple species, may greatly hamper the automated sound recognizer performance to identify
the specific species. Hence, it is desirable to extract the sound made by the target species from
the interferences as a pre-process prior to the recognition to get more accurate results. This
paper exploits two Blind Source Separation (BSS) algorithms namely Info-max and FastICA to
obtain the target frog sounds from the mixtures. The comparison of algorithm performances is
expressed according to Signal-to-Interfere (SIR). The empirical simulation results show that
FastICA outperforms Infomax in terms of separation quality with 65.76dB and 49.75dB,
respectively.
Pedestrian Detection in Visual Images using combination of HOG and HOM

R0108
Features
Wei Chee Kok and Soo Siang Teoh
The ability to detect pedestrian is an important feature in autonomous driving vehicle and
advanced driver assistance system (ADAS). The detection can be very challenging due to the
complex scene and bad visibility on the road. In addition, it is difficult to achieve high accuracy
and good speed performance at the same time due to more processing power is required to
increase the accuracy. To address this constraint, we propose a framework to detect pedestrian
through fusion of image gradient and magnitude properties and the process is speed up with
integral image implementation. Both image gradient and magnitude properties were extracted
using Histogram of Oriented Gradient (HOG) and Histogram of Magnitude (HOM) features. From
the experiment results, we showed that the combination of HOG and HOM features can achieve
99.0% accuracy compared to HOG (98.6%) or HOM (95.5%) features when they are used
independently.
Performance Comparison of Recognition System using I-Vector based on Different

R0112
Conditioning Methods
Noor Salwani Ibrahim and Dzati Athiar Ramli
This paper focuses on the analysis of the i-vector paradigm, a compact representation of
speaker utterances that is used by most of the state of the art speaker verification systems. The
i-vector subspace modeling is one of the recent methods that has become the state of the art
technique in this domain. This method largely provides the benefit of modelling both the intra-
domain and inter-domain variabilities into the same low dimensional space. In this study, 2656
syllables bio-acoustic signals from 55 species of frog taken from 5 sources of database for frog
identification system. Parameters of the system are initially tuned such as GMM component
size (16, 32, 64 and 128 Gaussians) and experimented for 3 different conditioning methods
which are whitening, Linear Discriminant Analysis (LDA) and Within Class Covariance
Normalization (WCCN). This work was mainly motivated by the need to quantify the impact of
their steps on the final performance, especially their ability to model data according to a
theoretical Gaussian framework. To the end, we assess the effect of the parameter tuned and
observe the recognition rate. We observed that, the accuracy for smaller GMM component and
i-vector with WCCN outperform others with result 86.67%. These investigations allow
highlighting the key points of the approach, in particular a core conditioning procedure that
leads to the success of the i-vector paradigm. 31
PHR2: POWER SYSTEMS, HIGH VOLTAGE ENGINEERING AND RENEWABLE

ENERGY
R0019 Effect of Distributed Generation (DG) to Earth Fault

Wan Syaza Ainaa Wan Salman, Mohd Rafi Adzman, Muzamir Isa, Haziah Abdul Hamid, Noor
Hasnizam Hanafi
Demand of electrical power is growing rapidly and this lead to the introduction of distributed
generation (DG) into the local system. DG can helps in supporting the power demand and
provide a cleaner energy as it uses renewable sources. However, the addition of DG into our
power system may change the control and protection system of the distribution system. By
using the IEEE 34 nodes test feeder as reference, the circuit was modelled with ATPDraw
software. The aim of this paper is to study the effect of DG to the faults when it were located in
several places. Besides that, the effect of location of DG was also investigated.
TEMPERATURE DISTRIBUTION PATTERN IN PILC CABLE WITH VOID DEFECT BY

R0031
USING COMSOL MULTIPHYSICS
Siti Faridah Satar, Nor Asiah Muhamad and Nurun Najah Abdul Rahim
A power system with high voltage transmission is necessary to transmit a large amount of
electrical energy from generating stations to consumers. Insulation is the most important part
of high voltage system together with cable and machine. In order to keep the equipment in
good operating condition, it is necessary to maintain the insulation in the equipment. In
particular condition, heat and other form of energy are released during the discharge activities
which may degrade the performance of the insulation. In this paper, the temperature
distribution pattern in the insulation of PILC cable is studied in order to help in determining the
location of discharge activities. In particular, the effect of the size and location of the void in the
insulation are investigated by using COMSOL Multiphysics. The result shows the amount of
heat pattern is affected by the size and location of the void in the insulation. The finding shows
that, the smaller size of void and the nearer void to the conductor made the insulator more
heated and leading the cable to breakdown.
PHOTOVOLTAIC BOOST DC/DC CONVERTER FOR POWER LED WITH ADAPTIVE P&O-
R0033
FUZZY MAXIMUM POWER POINT TRACKING
Muhammad Ammirrul Atiqi Mohd Zainuri, Mohd Amran Mohd Radzi and Nor Farahaida Abdul
Rahman
This paper presents the development of photovoltaic (PV) dc/dc converter to operate power
LED with an improved MPPT technique. The proposed MPPT technique is the combination and
integration of Perturb and Observe (P&O) and fuzzy logic control (FLC) to form a single
technique, named as adaptive P&O-fuzzy control MPPT. The proposed MPPT is aimed to
enhance MPPT performance and it is compared with conventional P&O and FLC MPPT
techniques respectively. Laboratory prototype was developed, and the setup includes PV test
panel of Kyocera KD210GH-2PU, boost dc/dc converter, and power LED as a load. Behavior of
the adaptive P&O-fuzzy MPPT was tested and compared with two mentioned techniques. The
proposed MPPT shows the best performance in achieving fast time response.
32
Performance Characterization of Different Configuration of Piezoelectric Raindrop

R0046
Energy Harvester
I.M. Izrin, Zuraini Dahari and Asrulnizam Abd Manaf
This paper proposed a study of the performance of various configurations of AC-DC converter
in piezoelectric raindrop energy harvester (PREH). The output voltage generator from PREH
produced a damping behaviour with variable amplitude AC voltage. To overcome this problem,
AC-DC converter is used to produce a steady output voltage. In this study, an individual
component in PREH that contributed to the performance of AC-DC converter was investigated.
Firstly, the performance of a different type of configurations for Polyvinylidene fluoride (PVDF)
piezoelectric in PREH was analysed. Then, the study on different types of diode used in the
power converter circuitry was conducted as well as various capacitance value to investigate the
output VDC generated. The results demonstrated that a consistent VDC above 1.2 V was
observed for the proposed PREH. Hence, PREH system has the potential to be implemented in
low power application.
Application of Three Independent Sources to Mobile Phone Charging During

R0047
Emergencies
Azuka Affam and Rosmiwati Mohd-Mokhtar
In this paper, the amalgamation of three input power sources as a charging tool for phones and
low power communication devices is presented. The developed system comprises a human
powered dynamo, solar panels to harvest energy from the sun and a rechargeable battery
supply. A simple switching mechanism is incorporated to switch between respective sources. A
common dc-dc regulation circuit adjusts the vacillating input voltages to satisfy phone and
communication device charging voltage specification. The multi-source system eliminates the
effect of periodic variation and unavailability of any of the input sources in event of a need for
emergency device charging and demonstrates the potential to power or charge a phone.
Electrical Surface Tracking Behaviour of XLPE Containing Untreated Zinc Oxide

R0055
Nano-Filler
M. Fairus, M. Hafiz, N. S. Mansor, D. Ishak, M. Mariatti, H. S. A. Halim, A. B. A. Basri and M.
Kamarol
XLPE has been widely used as an insulating material especially in high voltage underground
cables due to an excellent dielectric, mechanical and thermal properties. One of the insulation
failures of the power cable is due to poor cable jointing whereby it will cause the occurrence of
electrical surface tracking. The frequent occurrence of electrical tracking activities on the
surface of XLPE insulation will reduce the insulation strength and eventually breakdown occurs
and damages the cable. Thus, in this paper, the electrical surface tracking behaviour in XLPE
containing untreated ZnO nano-filler was investigated. The specimens were prepared with 1
wt% to 4 wt% concentrations of untreated ZnO nano-filler in XLPE composite. The experiment
was conducted according to the IEC 60587 standard. The result revealed that the XLPE
containing 1 wt% to 4 wt% ZnO nano-filler has poor electrical surface tracking profile.
33
ROB: ROBOTICS, CONTROL, MECHATRONICS AND AUTOMATION
R0043 Time Synchronization in WSAN using Sliding Mode and PID Control
Yeong Chin Koo and Muhammad Nasiruddin Mahyuddin
This paper discusses time synchronization in wireless sensor and actuator networks. An
enhanced time synchronization protocol which using distributed control scheme, incorporates
with sliding mode control and PID control is presented in this paper. The presented protocol
besides having the advantages of a distributed time synchronization protocol, it also has faster
convergence rate and lower energy consumption, compared to a protocol from literature.
Optimization of High Efficiency Permanent Magnet Synchronous Machine Using

R0061
Multi-Objective Differential Evolution
M. Rezal, Dahaman Ishak and Tiang Tow Leong
This paper presents an optimization of surface-mounted permanent magnet synchronous
machine (PMSM) based on analytical sub-domain model (ASM) together with differential
evolution (DE). A three-phase, 15-slot/10-pole, PMSM is selected in the design with initial
motor parameters which have been determined from sizing equations. Five motor parameters
are to be optimized i.e. magnet thickness, airgap length, slot-opening width, magnet arc, and
stator inner radius. Three objective functions are chosen i.e. to have the lowest total harmonic
distortions in the induced phase back-emf, highest output torque and highest efficiency. The
optimization of 15-slot/10-pole PMSM is further analyzed by comparing with other
optimization algorithms i.e. genetic algorithm (GA), and particle swarm optimization (PSO).
From the results, it is observed that PSO has the fastest computing time compared to GA and
DE. Whereas, DE is approximately 55% faster than GA. The design work for PMSM can
potentially become computationally intelligent without compromising the accuracy. While
repetitive changes in motor parameters in finite element modeling could be avoided after
applying this Differential Evolution.
R0098 Modeling of Torpedo-shaped Micro Autonomous Underwater Vehicle

Muhammad Azri Bin Abdul Wahed and Mohd Rizal Arshad
We present a mathematical model for a torpedo-shaped micro autonomous underwater
vehicle (µAUV) designed and developed at Underwater, Control and Robotics Group (UCRG),
Universiti Sains Malaysia. µAUV is a small and portable vehicle operated by a single operator.
We propose a mathematical model that does not takes account the effect of lifts force
subjected by the AUV. Mathematical model used in the control law is derived based on
theoretical and empirical methods adapted from similar AUVs while parameters of the AUV
taken from 3d model data in SolidWorksTM. This mathematical model used to represent this
AUV for fabrication of a working AUV prototype.
34
R0104 Seabotix Thruster Modeling for a Remotely Operated Vehicle (ROV)

Ahmad Faris Bin Ali and Mohd Rizal bin Arshad
This paper presents the Seabotix BTD150 thruster model for a Remotely Operated Vehicle
developed by Underwater, Control, and Robotics Group, USM. The bollard-pull experiments to
find the relationship between th input signal from Arduino Mega 2560 in Pulse Width
Modulation (PWM) and the thrust force for Seabotix thrusters by using spring balance in
kilograms unit. Result shows left and right thrusters lose 10% thrust force where the thruster
attached outside ROV. A center thruster loses 30% where thruster attached at the center ROV.
The parameters identified through a low-cost bollard pull experiment which done in
laboratory water tank.
R0106 Effect of Yaw Rate and Sideslip to Lateral Vehicle Dynamic Stability
Zainab Zainal and Wan Mohd Yusof Rahiman
This research presents the study on the effect of yaw rate and sideslip to vehicle lateral
dynamic stability using the simulation of Sine with Dwell (SWD) test as an actual working
condition. The comparison was made between the result of various constant longitudinal
speeds and road surfaces. The yaw rate, ψ˙ increased significantly comparing to the sideslip, β.
In addition, the outputs contribute to the instability of the vehicle towards higher speed and
slippery road surface.
35
TECHNICAL SESSION 3
A Survey on Infrared Thermography based Automatic Electrical Fault Diagnosis

R0091
Techniques
Shin Yee Lee and Soo Siang Teoh
Infrared thermography is a non-contact and non-destructive technique for electrical

equipment monitoring and fault diagnostics. It has been widely used since it can inspect the
condition of electrical equipment and detect possible faults without needing to disconnect the
equipment from its normal operation. Fault diagnosis is performed through the analysis of
thermal images captured by a thermal camera. Manual analysis of thermogram for diagnosing
the status of equipment needs to be carried out by a trained personal and this may take a lot
of time. It may also prone to human error in the diagnosis. To overcome this, there are several
researches focus on the development of methods for automatic electrical fault diagnostics.
Most of the methods combine image processing and computational intelligence techniques in
the diagnosis. Due to the large variability of equipment and diverse fault conditions, the
diagnosis task could be very challenging. There are different techniques being proposed in the
literature. This paper presents a survey on the current techniques for automatic electrical fault
diagnostics based on infrared thermography.
Liver Tumor Segmentation using Triplanar Convolutional Neural Network: A Pilot

R0121
Study
Sheng Hung Chung, Keng Hoon Gan, Anusha Achuthan and Rajeswari Mandava
In this work, a liver tumor segmentation approach is proposed using triplanar views, consisting
of axial, sagittal and coronal planes. These three planes are integrated as the input streams for
the Convolutional Neural Network (ConvNet). The main objective of including the input
patches from triplanar views is to enrich the ConvNet with more context information to aid in
the classification of liver tumor. The input patches are extracted from liver Computed
Tomography (CT) dataset using center pixel of interests from the triplanar views. These
patches are fed into the proposed Triplanar ConvNet. Pilot experiments were conducted to
evaluate the efficiency of using triplanar views in comparisons with single view from axial
plane. The preliminary results achieved in this study revealed that triplanar approach yield
better results than using patches only from single view.
36
Drill Bit Flank Wear Monitoring System in Composite Drilling Process using Image
R0094
Processing
Raiminor Ramzi, Elmi Abu Bakar and M.F Mahmod
Composite drilling is a hole making operation that is mainly involved in aircraft manufacturing
industry. The poor machinability of the composite materials causes the cutting tool to wear
faster and increasing the production cost. Ignoring the tool condition would be a bad idea for
the production as worn cutting tool tends to damage the highly expensive composite panel of
the aircraft. Tool condition monitoring (TCM) is required to keep the process in balance
between cost and quality. This paper presents a system to perform tool condition monitoring
of a drill bit flank wear using image processing approach. The real industrial sample of carbide
drill bit which was used to drill carbon fibre composite panel is obtained directly from the
manufacturing assembly line. The images of the drill bit are acquired from the top view for
every 100 holes using the developed hardware system. Edge detection is used to detect the
boundary of the cutting lips and the images are compared with the reference image of the
brand new drill bit using image registration method. The wear rate of the erosion flank wear
measured is recorded at average rate of 0.0198% per hole and considered worn at maximum
amount of 25.48% wear.
Improving Classification of Microembolus and Artefact of HITS Event by Feature

R0095
Selection
Najah Ghazali, Maz Jamilah Masnan and Dzati Athiar Ramli
In monitoring system using transcranial Doppler ultrasound for stroke detection, the
occurrence of high intensity transient signal can happen at different branch of arteries, i.e
internal cerebral artery (ICA), middle cerebral artery (MCA) and posterior cerebral artery (PCA).
The representations of features can sometimes be redundant and not useful, which can
degrade the classification performance. Thus, feature selection is studied and presented in this
paper. The applied selection criteria are based on the unbounded Mahalanobis distance
(referred as A) and single-feature-accuracy measure (referred as B). The result indicates that
kinematic descriptor (SMV) is the most significant feature to predict HITS with 85.8% correct.
However, the classification accuracy further improved when SMV is combined with other
features in different feature subsets.
37
DCE/EDAS2: DEVICES, CIRCUITS AND EMBEDDED SYSTEMS /

ELECTRONIC DESIGN AND APPLICATIONS
Physical Modelling of Gallium Nitride (GaN) Based Double Barrier Quantum Well
R0114
Device
W. N. N. Zaharim, N. Z. I. Hashim, M. F. Packeer Mohamed, A. A. Manaf and M. A. Md Zawawi
This paper describes the physical simulation of Al0.3Ga0.7N/GaN-based double barrier

resonant tunnelling diode (RTD) with cubic-GaN structure using Silvaco ATLAS. Cubic-GaN is
interesting for vertical transport devices due to the absence of internal polarization. Proper
selection of material parameters is vital in order to optimize the device performance. The
relationship between current-voltage (I-V) characteristic, particularly the negative differential
resistance (NDR) with respect to variation in barrier composition, well width and barrier
thickness are studied in detail. RTDs with clear NDR are demonstrated with peak-to-valley-
current ratio (PVCR) of 7.19 and peak current density of 3.77 mA/cm2 at room temperature.
The optimum barrier thickness for a given composition is determined by using the ratio of
current magnitude at resonant peak to the width of the resonance, and also used as figure of
merit for the device. Finally, optimum parameters for the promising cubic Al0.3Ga0.7N/GaN
RTD are proposed.
Place and Route Optimization for High Coverage Multi-corner Multi-mode Timing
R0111
Fix
J.S.K. Lee, M.F. Packeer Mohamed, M.A. Md Zawawi and Nur Z. Hashim
Physical design convergence becomes complicated as the number of gates is increasing with
decreasing size of the transistors. Multi-corner multi-mode timing convergence is introduced to
compensate for the manufacturing variation, raising the number of signoff scenarios for static
timing analysis. As a result, timing violations in all scenarios need to be closed and the timing
convergence process is prolonged due to the “Ping Pong Effect”. The place and route
implementation tool might not have the visibility to all violations in all scenarios since not all
scenarios are taken into optimization process. Some new violations might be induced due to
the fixing of the other violations. Therefore, it is important to improve the visibility of the
implementation tool in the actual worst slack or worst margin of all timing paths. In this paper,
a methodology is proposed to create the worst-case scenario in timing slack and used in
optimization process. The worst-case scenario is successfully modeled with generated worst
timing margin across all the timing paths in all sign-off scenarios. With this strategy, the timing
results for hold violations in terms of the worst number of slack, total number of slack and total
violated path has improved by 53.2 %, 95.5 % and 75.9 %, respectively. No significant result is
recorded for setup violations. Nonetheless, it is seen as a promising approach to minimize the
timing closure process.
38
R0116 8-Bit Hybrid DAC with Rail to Rail Buffer Amplifier

Alia Rosli, Zulfiqar Ali Abd Aziz and Shukri Korakkottil Kunhi Mohd
This paper presents 8-bit differential output of hybrid Digital-to-Analog Converter (DAC) with
rail-to-rail buffer using Silterra 0.18µm CMOS process technology for mixed-signal application.
The proposed design combines binary-weighted resistor and thermometer coding, which is able
to enhance the speed of the data converter by implementing a small die size area as compared
to the conventional architecture. The design achieved standard accuracy of 8-bit resolution with
INL and DNL of ±0.5LSB and ±0.7LSB, respectively, and overall power consumption of 56mW.
The performance is accomplished with an effective design area of 133µm x 172µm.
Along-Track Interferometry TanDEM-X Satellite Data For Modelling Sea Surface

R0074
Level Change and Sea Surface Current Velocity
Maged Marghany and Narimah Samat
This investigation is the novel study, which is contacted along the Malaysian coastal water to
implement the potential of TanDEM-X satellite data for retrieving sea surface current. In this
view, the main objective of this investigation is to exam the multichannel MAP height estimator
for retrieving sea surface current pattern as a function of the tidal elevation. The investigation
reveals that the MAP algorithm has excellent performance with the lowest RMSE of ±0.0074
m/s. Additional, MAP algorithm can authenticate the spatial variability of coastal water from
TanDEM-X satellite data. Concluding, TanDEMX data exposes a fabulous assurance for modelling
the sea surface current flow with X-band.
A Review on Performances Evaluation of Low Power Wide Area Networks

R0026
Technology
I.S. Ismail, N.A. Abdul Latiff, F.Z Rokhani and S. Abdul Aziz
Low Power Wide Area Networks technology is considered as a new technology which enables a
long-range communication where it targets the low power applications. There are many
available solutions for Low Power Wide Area Networks technology in the market which among
them are LoRa from Semtech and NB-IoT from 3GPP. This paper provides a review of the
performances evaluation for Low Power Wide Area Networks Technology. It is shown that the
parameters of PHY layer such as frequency, bandwidth, transmit power, spreading factor and
code rate able to affect the performances of Low Power Wide Area Networks. Meanwhile, the
received signal strength indicator and number of successful received packets data are the
common performance parameter studied. In addition, the performance also conducted in the
various location either in line-of-sight or not such as outdoor, indoor and underground.
39
R0039 Characterization of Low Noise Amplifier (LNA) for mm-Wave Wireless Systems
Mohd Fadzil Ain, Mohamad Faiz Omar, Roslina Hussin, Mohd Zaid Abdullah, Intan Sorfina
Zainal Abidin and Zainal Arifin Ahmad
This paper reports the technical design of millimeter wave (mm-Wave) low noise amplifier
(LNA). The fast-growing wireless communication market and the necessity of a higher data rate
certainly drive the radio frequency (RF) system towards the mm-Wave technology. The 28 GHz
channel band possesses a large amount of raw bandwidth space with a spectrum allocation
that exceeds the bandwidth of 1 GHz. The mm-Wave cellular wireless system is developed to
address with a worldwide wireless bandwidth shortage. The 28 GHz LNA is used to amplify a
very low signal at 28 GHz and minimize the present of unwanted additional noises. The
performances of single and cascaded LNAs are analyzed and presented. The footprint design of
LNA is fabricated on ROGERS 4003C with permittivity of 3.38. The off-the-shelf commercial
MMIC LNA is mounted and matched to 50 ohm. The fabricated LNA is analyzed based on the
output power levels and signal gains. The minimum and maximum gains for a single LNA are
recorded at 11dB and 14.83dB respectively. 22.05dB is verified as the lowest gain for the
cascaded LNA while the highest gain is indicated at 28.83dB when the RF signal input power is
quoted at -74.54dBm.
R0041 A New Double-Negative Material for Multi-band Satellite Applications
Sikder Sunbeam Islam, Ahmed Mahfuz Tamim and Mohammad Rashed Iqbal Faruque
In this study, a compact new double-negative (DNG) material unit cell structure for multi-band
satellite applications has been presented and proposed. A π-shape design has been developed
which is splitted at two ends. The structure of unit cell has been designed using CST Microwave
Studio and this simulation tool is also used to find out the transmission and reflection
parameters for this DNG material unit cell. This unit cell shows resonance in the microwave
multi-band (S-band, C-band and X-band) frequency range. It also shows double-negative
characteristics in all these above frequency bands.
40
Parametric Study of Parallel Inhomogeneous Cylindrical Dielectric Resonator at C-

R0049
band Frequency
Rashidah Che Yob, Nor Muzlifah Mahyuddin and Mohd Fadzil Ain
Dielectric resonator has been implemented in many frequencies and microwave front-end
design, where the main focus is on homogeneous (single) permittivity that performed only one
purpose at one time. In this article, the parallel inhomogeneous cylindrical dielectric resonator
at C-band resonant frequency is presented. The parallel inhomogeneous cylindrical resonator
modeled by combining two different dielectric permittivities; Magnesium Titanate and Cobalt
Tinatane that have high and low permittivity, respectively. Each permittivity material is divided
90 degrees apart. A parallel microstrip-line coupling is used to excite the inhomogeneous
cylindrical dielectric resonator. The parametric study of the parallel inhomogeneous cylindrical
dielectric resonator is highlighted to be investigated purposely to identify the impact on any
changing parameter behavior and obtain their best configuration of performances. These
parametric studies evaluated in-term of configuration the parallel microstrip-line and the
effect of angular position inhomogeneous cylindrical dielectric resonator. Different types of
configuration microstrip-line are tested, i.e. coupling spacing and curves configuration at λ/4 of
wavelength. The return loss at the input port of parametric study in-terms of coupling
coefficients is satisfactory at 17 and 19 (mm) of parallel microstrip-line for same-port positions.
Meanwhile, the investigation into parallel microstrip-line curve configuration at λ/4 of
wavelength indicates the best performance at curve configuration of 155 degree with 9.77%
improvement on 5 GHz resonant frequency. In conclusion, the optimum performance of
inhomogeneous cylindrical dielectric resonator at 5 GHz is performed when coupling distance
of 19 mm with 155 degree curves positions of parallel microstrip-line without the angular
position.
41
TECHNICAL SESSION 4
AI1: ARTIFICAL INTELLIGENCE AND COMPUTER APPLICATIONS
American Alphabet Hand Sign Language Detection and Recognition Using Haar
C0017
Cascades and Convolutional Neural Networks
Kevin Richard G. Operiano, Pann Mya Hmue, Wanchalerm Pora, Suree Pumrin
Sign Language is an important means of communication for the hearing-impaired individuals. It

is extensively used to express actions, words, emotions, etc. However, there are words that do
not have specific hand gestures such as proper names. Therefore, fingerspelling is used. Recent
developments in hand gesture detection and recognition made it possible for computers to
locate and identify a hand signal in an image or stream of images using vision-based algorithms.
The relative quickness of Haar Cascades to detect objects in an image combined with the
proficiency of Convolutional Neural Networks to recognize objects is a promising approach to
contribute in this development. The Convolutional Neural Networks in this system achieved an
average accuracy rate of 89.38%.
Neurocontrol Design for an Aerodynamics System: Simple Backpropagation

R0017
Approach
N. M. H. Norsahperi and K. A. Danapalasingam
This paper proposed a Neurocontroller (NNC) for Twin Rotor Aerodynamics System (TRAS) by
simple backpropagation approach to attain the pitch position accuracy. A concept known as the
gradient descent method was applied to adjust the weights adaptively. The approach has several
notable merits namely low computational cost, simple and promising controller. The viability of
NNC was verified by using MATLAB to simulate the controller performance and effort. PID
controller was benchmarked against the proposed NNC to determine the effectiveness of the
controller. From the simulation work, it was discovered that NNC was superior then PID
controller by reducing about 14%, 23% and 97% in the value of the overshoot, settling time and
steady-state error respectively. The promising part of NNC was the improvement shown in the
controller effort by significantly eliminating the fluctuation and chattering in the control signal.
By looking into the future, this work will be a foundation for future improvement due to the fact
that there are numerous types of approaches could be embedded in the Neural Network
algorithm.
42
R0030 A Study of Artificial Neural Network (ANN) in Power System Dynamic Stability
N. Rosle, N. F. Fadzail, M. N. K. H. Rohani
The stability of a power system refers to the ability of a system to return to its steady state
when subjected to a disturbance. The dynamic stability had experienced to poorly damped low
frequency oscillation (LFO) after disturb by a small disturbance. This disturbance makes an
entire of a power system become unstable and takes a longer time to become a steady state,
thus a stabilizer is needed to stabilize the power system. In this research, an optimum gains
from Riccati equation are represent as a stabilizer combined with power system model to
become power system stabilizer (PSS). Continuos-time algebraic Riccati equation (CARE) in the
tool function of MATLAB is used due to fast and efficient to get an optimum gains. The
performance of PSS to stabilize the power system is compared to another tool in the MATLAB
software known as an artificial neural network (ANN). ANN is selected due to the ability to get
data samples rather than the entire data sets to arrive at solutions, fast and though
approximate but acceptable solutions in real time. This research deals with the development
of ANN systems which provide solutions to stabilize the power system based on PSS
performance. The step response from simulation results are presented to show the
effectiveness and robustness of the designed ANN system to get the same performance as PSS
with minimal steps in order to stabilize the oscillation damping of the power system.
R0032 Hand Posture Detection of Smartphone Users Using LSTM Networks

Tan Song Lim, Ng Hui Fuang, Ooi Boon Yaik, Tan Hung Khoon, Jacqueline Ang Lee Fang
Automatic hand posture detection of smartphone users is important for adaptive user
interface design, context aware application development, and activity analysis. This paper
presents a method for hand posture and phone placement detection from data produced by
accelerometer, magnetometer and gyroscope of a smartphone using LSTM networks. Real-
time testing results indicated that LSTM network is effective in hand posture and phone
placement prediction, and the proposed method outperformed existing methods by significant
margins.
R0048 Enhanced Simulated Annealing for Constrained Design Problems

Hussein Samma, Junita Mohamad-Saleh, Shahrel Azmin Suandi
Real-world design problems such as welded beam design, pressure vessel design, and three-
bar truss design were recognized as challenging tasks due to the associated constraints. This
work aims to develop an Enhanced Simulated Annealing (ESA) optimizer that embeds the Q-
learning algorithm in order to control its execution at run time. Specifically, the Q-learning
algorithm is used to guide SA toward the best performing value of the annealing factor at run-
time. To assess the performance of ESA, a total of four popular constrained engineering design
problems were conducted. The outcomes reveal the ability of ESA to significantly overcome
the standard SA as well as other optimization algorithms such as GWO, PSO, and CLPSO.
43
EMD2: ELECTRICAL MACHINES, DRIVES AND POWER ELECTRONICS
A New Hybrid Multilevel Inverter Topology with Reduced number of Switches and
C0015
dc Voltage Sources
Marif Daula Siddique, Saad Mekhilef, Noraisyah Binti Mohamed Shah
Multilevel inverters are finding great attention in medium and high power applications. A new
hybrid multilevel inverter topology based on cascade connection is suggested in this paper.
Higher number of levels can be realized at the output with additional half H-bridge connected
to the proposed basic unit. The proposed topology is designed with the purpose of reducing
the number of switches and dc voltage sources. The results are verified by a detailed
comparison with different multilevel inverter topologies. Fundamental frequency switching
modulation technique is used for the generation of pulses for switches used in the proposed
topology. The simulation results are carried out by using MATLAB software. The simulation
results are validated by the experimental results obtained from the prototype setup at the
laboratory.
FCS-Model Predictive Control of Induction Motors fed by MultilLevel Cascaded H-

C0026
Bridge Inverter
Ha Thanh Vo, Nam Thanh Hoang, Phuong Hoang Vu, Minh Trong Tran, Dich Nguyen Quang,
Hoang Trung Thong
This paper presents a finite control set – model predictive control (FCS-MPC) of induction
motor fed by a multilevel cascaded H-bridge inverter. The control scheme has been based on
discretized model of the IM motor and also on the discrete nature of power electronic
converter. In each predict time period the optimization procedure estimates cost function
under a finite control set which is among the most effective switching voltage vectors of the
multilevel converter. To ease the future experimental prototype building, this work has
purposely developed a MPC algorithm for 2 step prediction time and the IM fed by a 3-levels
cascaded H-bridge inverter. The simulation results show all good attractive performance and
potential implementation.
44
C0029 Implementation of PMSM Servo Drive Using Digital Signal Processing

Hien Nguyen Nhu, Truong Tran Minh, Phuong Vu Hoang, Ngoc Nguyen Dinh, Quang Bui Dang,
Phuong Nguyen Huy
This paper presents the implementation of vector control drive of Permanent Magnet
Synchronous Motor (PMSM) using increment encoder. The algorithm can control motor in the
transitioning process smoothly and help the current always in the safety zone. Simulation and
experiment are two steps to verify the control algorithm. The simulation process bases on
Matlab Simulink and the experimental system use a controller board based on TMS320F28377s
DSP. The simulation and experimental results show that both speed and position have good
responds and steady-state accuracy, the current is smooth and its overshoot is very small, that
demonstrate the the effectiveness of the control algorithm.
C0032 Performance Analysis of Microwave Power Generator

Leong Wen Chek, Saad Mekhilef and Harikrishnan Ramiah
This paper presents several techniques of developing Microwave Power Generator (MPG)
system in effort to realize the needs of Wireless Power Transmission (WPT) to eliminate bulky
battery and long charging time over device’s operation time. However the implementation of
Microwave Power Transmission (MPT) for commercial use is still in doubt due to the effect
towards other electronics devices and transmission efficiency. Hence, several types of
microwave power generator with more than 80% efficiency as well as the element that
probably affects the generator efficiency are compiled and discussed. The types of microwave
power generator includes magnetron as the most common generator used, magnetic
superlattices, coaxial virtual cathode oscillator using velvet cathodes and graphite, picosecond
opto-electronic technique and backward wave oscillator which is similar to travelling wave
tube (TWT). The listed microwave power generator then classified into low, medium, high and
extremely high power microwave generator. However the performance is magnetic
superlattices microwave power generator by using metal oxide field effect transistor (MOSFET)
element as halls bar is more superior. In the end of this paper, also recommend additional
semiconductor material technology can be used in most microwave power generator to obtain
better performance.
R0101 Switching Circuit and Drive for Cascaded Multilevel Inverter

Tarmizi, M. K. Mat Desa, W.A.F.W. Othman and S. Taib
A reduce device count multilevel inverter or RDC-MLI can generate high levels of sinusoidal
output waveforms with minimum number of switches. In this paper, switching signals and drives
for RDC-MLI inverter are design and verify through software and prototype circuit. The switching
signals were first design and simulate using PSIM environment. For prototype circuit, PIC16F877
microcontroller was used to generate the switching signal. TL250 optocoupler was used to drive
the level generation switches while IR2101 was used to drive the H-bridge. The simulation output
waveform signals were identical to experimental prototype results. RDC-MLI inverter
implemented with the designed switching signals and drives circuit, successfully produced high
levels of output waveform with four asymmetry sources.
45
PHR3: POWER SYSTEMS, HIGH VOLTAGE AND RENEWABLE ENERGY
C0016 Power Delivering on Single-Phase Grid–Connected Photovoltaic System

Syafrudin Masri, Khairunaz Mat Desa, Muhammad Hafeez
This paper presents a simple method for delivering of power at single-phase grid-connected
photovoltaic (PV) system. The active and reactive power are sequentially can delivered to grid
by control the inverter output voltage magnitude and power angle. The maximum active and
reactive power into grid can achieved at power angle 90o. The phase sifting of power angle has
implemented by using analog technique. The simulation and experimental evaluation of the
interface were presented.
Optimal Maintenance Planning of Substation Equipment Considering Life Cycle

C0021
Cost and System Reliability
Surawit Laohanan and David Banjerdpongchai
The important issue of smart grid and condition monitoring is maintenance scheduling. The
maintenance schedule affects equipment operation as well as reliability indices of substation
and life cycle cost (LCC). This paper focuses on long term maintenance planning with
consideration of system reliability. The reliability indices of substation depend on failure event
which can be approximated from devices that severely impact on system reliability.
Maintenance can improve the reliability indices and LCC of system. Optimal maintenance
scheduling can be formulated as discrete optimization. However, finding the exact solution is
complicated. We first approximate LCC by associating the cost to individual devices. Then, we
employ dynamic programming (DP) to find suboptimal maintenance scheduling and satisfy
reliability of system. We apply the DP approach to Main and Transfer, and Breaker and a Half
configuration to demonstrate the effectiveness of proposed method and categorize
maintenance devices in substation.
C0030 ENERGY SELECTION WITH ANP & SWOT METHODS: KALIMANTAN CASE
Farizal, M. Dachyar, Nyi Mas Asri
Kalimantan has enormous energy potency. However, until now its electrification ratio is still
relatively low compare to the average national electrification ratio and its availability is sparse.
Fossil energy utilization, which currently the backbone of electrical energy in Indonesia, will be
reduced in the future. To fulfill future electricity need, this energy source have to be combined
with new and renewable energy. To ensure the sustainability of future energy supplies, the
selection of types of energy sources and their quantities must be determined by including
various criteria. This study aims to find the most suitable energy from various non renewable
and new renewable energy sources in Kalimantan. For the purpose ANP and SWOT were used
with six criteria. i.e. availability, economics, technology, social politics, environment, and
technical. The results found that of the ten types of energy, hydro is selected. To promote its
successful utilization, aggressive strategy is recommended.
46
R0085 Study of Electronic Energy Meter Performance under Harmonics Current Condition
Syafrudin Masri, Khairunaz M.D and M.N.Mamat
This paper presents the experimental results on the effect of total current harmonic distortion
(THDi) on the deviation readings of a residential customer's digital energy meter (kWh).
Experimental results were in accordance to theoretical analyses. Electrical energy measurement
results are obtained under harmonic current wave conditions based on IEEE-159 std
recommendations. The study focused on how the percentage THDi from non-linear load
influence to EDS2560 and Fluke-43B electronic energy meters (kWh) reading performances,
while the voltage supply constant and sinusoidal without harmonics content. Both electronic
energy meters show good measurements under the linear load condition and they have
significant deviation on high percentage THDi.
Multi Objective Optimization of DG Allocation and Sizing in Distribution Systems

R0066
Using Non-dominated Sorting Genetic Algorithm II
Norainon Mohamed and Dahaman Ishak
Nowadays, dispersed generators (DG) has been used in order to achieve the better system
performance. Losses minimization and voltage profile enhancement are the main target of DG
optimal placement. In this paper, the DG allocation and sizing considering stability index
improvement, DG penetration and reduction in power losses has been investigated by using
Non-dominated Sorting Genetic Algorithm II (NSGA II). Modal analysis has been used to evaluate
the stability index. The utilized method has been applied to IEEE 30 bus test system.
Partial Discharge Behaviour and The Movement of Two Different Shapes of Metal
R0068
Particle in Mineral Oil
Kiasatina Azmi, Dahaman Ishak, Ahmad Zuhairi and M. Kamarol
Partial discharge (PD) due to different shapes of free metal particle in mineral oil has been
measured in order to study the PD behaviour induce by metal particle in mineral oil. PD under
influence of spherical metal particle has been compared with cylindrical metal particle under AC
applied voltage in various gaps. Partial discharge inception voltage (PDIV) for both shapes of
metal particle appeared to increase with increasing of gaps. PDIV of cylindrical metal particle
demonstrate a lower value compared to spherical metal particle with increasing gaps. PD due to
sphere shape metal particle revealed to be less compared to PD due to cylinder shape metal
particle. The observation on spherical and cylindrical metal particle movement revealed that
there were different stages of motion process involved. The electric field distribution analysis
shows a consistency with PDIV results. Shape of metal particle revealed to influence the PD
behaviour in mineral oil.
47
BIO:BIOMEDICAL, BIO-ENGINEERING AND APPLICATIONS

C0005 Patterning of microparticles/cells through the acoustic-assisted nozzle for 3D printer
Yannapol Sriphutkiat and Yufeng Zhou
Acoustic excitation is capable of accumulating or focusing of particles or cells during the additive
manufacturing process. In general, printed part’s efficiency, performance, and functionality
could be increased by optimal orientation or arrangement of the microstructure. This could be
utilized in various applications such as fibers reinforced polymer matrix, hydrogel scaffold, and
biomedical samples. In this work, acoustic manipulation of microparticles in the cylindrical glass
nozzle is demonstrated. With the acoustic excitation, microparticles were accumulated at the
center of the nozzle and subsequently printed part at the fundamental frequency of 871 kHz.
The distribution of particles and cells fits well with a Gaussian distribution. Afterward, muscle
cell lines were also accumulated at the center of the cylindrical glass tube and in the printed part
(biomaterials). In short, the proposed acoustic approach is capable of focusing the particles and
cells in the printed part with biocompatible, highly-tunable, and cost-effective.
R0025 Human Sperm Tracking using Improved Anti-collision Mean Shift Tracking Method
Weng Chun Tan, Nor Ashidi Mat Isa and Mahaneem Mohamed
Sperm tracking is challenging in sperm motility assessment. Most of the existing methods are
not able to track the object while the collision between sperms are occurred. This paper
introduces an anti-collision method to detect the collision and track sperm robustly. By using the
pixel weight and moment features, the new non-occluded region is extracted to perform the
tracking under collision condition. Based on the results, the proposed method is able to solve
the existing drawbacks and producing low Bhattacharyya distance results compared with
standard mean shift tracking method. In future, this method is expected to be implement on
multiple sperm tracking and classifying sperm motility categories according to the latest WHO
manual.
Muscle Oxygen Saturation Correlates with Muscle Mechanomyography during
R0060
Prolonged Electrical Stimulation-Evoked Wrist Extension Exercise
Nurul Salwani Mohamad Saadon, Nur Azah Hamzaid, Nazirah Hasnan, Muhammad Afiq Dzulkifli,
Mira Teoh, Gan Kok Beng and Glen M. Davis
This study assessed electrically-evoked sustained muscle contractions with real-time changes in
near-infrared spectroscopy and mechanomyography signals. Twenty healthy volunteers
performed electrical-evoked wrist extension for 10 minutes. Root mean square derived from
MMG (%RMS-MMG) and tissue oxygen saturation (%StO2) from NIRS of the extensor carpi
radialis muscle were monitored throughout the sessions. The correlated responses of these two
measures comprised 7 consecutive sets of 10 contractions each. %StO2 revealed an initial
decline from muscle contraction 1 to 10 and then an increase from contraction 11 to 70. For
%RMS-MMG, a significant decrease was observed from contraction 1 to 10 as well as from
contraction 41 to 60 while an increase was seen from contraction 11 to 40 and from contraction
61 to 70. Positive low-order correlations were found between %StO2 and %RMS-MMG during
the start of the exercise while later on, the two variables were negatively correlated. As the
number of contractions increased over time, increased %StO2 revealed good oxygen saturation
aligned with decreased %RMS-MMG values, suggesting decreased muscle fiber activation and
strength. In this study, the physiological muscle responses were associated with its mechanical
characteristics.
48
R0062 Electrical Pulsed Artocarpus Altilis on Cervical Cancer cell

Suhassni Ganeson, Muhammad Mahadi bin Abdul Jamil, Mohamad Nazib bin Adon and
Muhammad Shukri bin Ahmad
This paper elucidates on a method using electrical pulse-mediated chemotherapy, known as

electrochemotherapy and artocarpus altilis extract to study the anti-proliferation effect on HeLa
cancer cell. Electrical field parameters used in this study ranging from 200V/cm, 400V/cm,
600V/cm, 800V/cm, 1000V/cm and pulse duration of 100µs, 500µs, 1ms, 5ms and 10ms with
40µg/ml of artocarpus fruit extract concentration. Results indicate that viable cells, as low as
30% was prompted by 600V/cm and 5ms of electrical field parameter after 72 hours.
Development of Health Monitoring System with Support Vector Machine Based

R0064
Machine Learning
Yoong Khang Ooi and Haidi Ibrahim
The world’s population is increasing since year 1950 till now. The large number of population
with age between 25 to 29 implied the importance of health care services to maintain this
population’s good health. However, there are a lot of health measuring devices only measure
one health parameter from each person. This is very inconvenience to most of the users.
Another problem encountered is that there is a large number of health data that are not
analysed by the system. Therefore, the purpose of this research is to develop a data acquisition
system that consists of three sensors, which are temperature sensor, pulse oximeter sensor and
heart rate sensor. Besides, this project also develops Support Vector Machine (SVM) based
machine learning algorithm to monitor health condition. All the sensors will measure respective
reading and read by Arduino microcontroller. The reading will then transfer to Raspberry Pi 3 via
serial communication for health prediction using machine learning. A classification model is
derived from 240 training data and tested with 60 testing data. The classification model gives an
overall accuracy of 93.33%. While looking at user’s accuracy on each class, all class except two
classes give 100% accuracy. However, both ROC of these two classes are 0.998, which are still
high. Therefore, the classification model is good and can be used to predict health condition.
Microfabrication and Electrical Characterization of Non-spiral Planar Microcoils for

R0090
Biomedical Applications
Krishnapriya S, Rama S Komaragiri and Suja.K.J
Microcoils are significant components for micro magnetic sensors and actuators especially in
biomedical devices such as micropumps. Non-spiral planar microcoil is an emerging type of
planar microcoil which can reduce the overall device dimensions with an easier fabrication
process. Coil inductance of the order of 1 mH and series resistance as low as 0.25Ω are obtained.
Non-spiral planar microcoils of the common square and circular geometries are studied here by
analyzing their electrical parameters. Comparison between the experimental and theoretical
model is also done. Inductance, series resistance and parasitic capacitance of circular and square
non-spiral coil geometries are compared along with their magnetic field distributions. Results
show that the circular non-spiral coil gives better performance over the square counterpart.
49
R0013 Breadfruit Peel Extract Impact on HeLa Cancer Cell Viability and Proliferation
Suhassni Ganeson, Muhammad Mahadi bin Abdul Jamil and Radzi Ambar
Cancer is referred to a type of diseases characterized by an abnormal growth of cells in bodies,
and it’s the second leading cause of death worldwide. While numerous treatments associated
with cancer therapy being practise, it also said to have serious permanent side effects due to
high usage of drugs in the treatment. Said that, a fine complementary method which can
eliminate the side effect of drugs is needed. Constituents obtained from plant extract have been
a potent source of anti-cancer agents. This present review is mainly to emphasize potentiality of
artocarpus altilis (peel part) in opposing the proliferation of cervical cancer cell over a period of
72 hours. Proliferation factor and viability of cervical cancer cell observed with different
concentration of peel part of artocarpus altilis ranging from control, 10µg/ml, 20µg/ml, 30µg/ml,
40µg/ml, and 50µg/ml. Result obtained from this research shows decrease in percentage of
cervical cancer cell viability as well as cell proliferation rate with the increase of extract
concentration. Thus the study result quantitatively proves the dependence of cell proliferation
on extract concentration. Findings of this study can be applied in electrochemotherapy field as
Artocarpus altilis/ breadfruit has the ability to inhibit cervical cancer cell proliferation and
viability.
50
TECHNICAL SESSION 5
AI2: ARTIFICAL INTELLIGENCE AND COMPUTER APPLICATIONS
Improving a Run Time Job Prediction Model for Distributed Computing Based on
R0065
Two Level Predictions
Hazem Al-Najjar, S S N Alhady and Junita Mohammad Saleh
Nowadays, distributed computing environment faces many difficulties because the number of
submitted jobs is increasing dramatically. One of the most used method to serve the jobs is to
find the accurate run time of the submitted jobs. This paper proposes a new job prediction
method, to predict on jobs' run time using two level prediction namely linear regression model
and fitting model. The proposed model uses six variables including user ID, group ID, executable
ID, number of CPUs, memory size and average CPU time, furthermore to solve the problem of
the categorical variables (i.e. user ID, group ID and executable ID) a dummy code is used. To
adjust and to find the best combination between linear regression model and fitting models,
different fitting models are used by combining linear and nonlinear fitting models. By simulation
the results show that the proposed model is better than previous models when smoothing spline
fitting is used, also the results indicate that proposed model is efficient with low error and high
prediction rate compared with previous models.
Distribution System Expansion Planning Using Enhanced Version Boltzmann

R0080
Machine
Siti Hajar Mohd Tahar, Shamshul Bahar Yaakob and Amran Ahmed
The Distribution System Expansion Planning (DSEP) problem is a highly complex problem. The
objective of the DSEP is to fulfill electricity load increment at the lowermost cost and consumers'
reliability desires with a level of satisfaction. This paper proposed a solution for DSEP problem by
using meta-heuristic technique. The proposed method will be formulated by using mean-
variance analysis approach in the form of mixed-integer quadratic programming problem. It
consists of two layers meta-heuristic which combining Hopfield network and Boltzmann machine
in upper and lower layer respectively named as Enhance Version Boltzmann Machine. The
originality of the proposed method is it will delete the unit of the lower layer, which is not
selected in upper layer in its execution. Then, the lower layer is restructured using the selected
units. Due to this feature, the proposed method will greatly reducing times consuming and
improve the accuracy of solution. Since the solution provided in variety option, thus it offered
flexibility for decision maker in solving planning problem.
R0087 Expressive Malay Online Speech Interface (EMOSI)

Ai-Dii Chai and Syaheerah Lebai Lufti
Speech Synthesis plays an important role in enhancing human-machine interaction. In recent

decades, researchers are paying more attention on the emotional expression in the synthetic
voice. This is because the appropriate emotion can help improve the naturalness of the synthetic
voice and thus increase its acceptability by the public. This project aims at developing a HMM-
based Malay emotional speech synthesizer that is practical to be deployed in real life
application. In order to make it applicable to the public, an Expressive Malay Online Speech
Interface (EMOSI) that is able to synthesize any form of Malay text input in different expression
will be created.
51
Artificial Fish Swarm-Inspired Whale Optimization Algorithm for Solving Multimodal
R0099
Benchmark Functions
Imran Rahman, Junita Mohamad-Saleh and Noorazliza Sulaiman
Multimodal benchmark function optimization has gained a growing interest exclusively in the
evolutionary computation research field which involves achieving all or most of the multiple
solutions contrasting a single best solution. A large number of real-world optimization problems
can be considered as multimodal function optimization. Recently introduced Whale
Optimization Algorithm (WOA) algorithm is inspired by the hunting behavior of humpback
whales. The performance of WOA is very promising but the robustness and convergence need
further improvement. In this paper, ‘step equation’ of Artificial Fish Swarm Algorithm (AFSA)
was incorporated to enhance the robustness and convergence of the original WOA considering
five multimodal test functions (F1-F5) for global numerical optimization. The proposed variant of
WOA showed improved performances compared to original WOA in terms of average best
fitness, robustness and convergence.
Pythagorean Theorem for Shortest Distance in CA Based Pedestrian Simulation: A

R0120
Case Study on the Closed Area
Najihah Ibrahim and Fadratul Hafinaz Hassan
Cellular Automata (CA) model approach has become a well-known approach in demonstrating
the microscopic movement of pedestrian for path finding and navigation to reach point of
interest. One of the features in the CA movement model is to find the shortest path distance
that will navigate the pedestrian movement towards point of interest in the fastest of time.
Nowadays, there are many path navigation systems that were constructed by using the pre-
defined movement simulation based on the static obstacles’ layout. However, the pre-defined
movement simulation is not able to highlight the near-realistic movement of the pedestrian due
to the dynamic obstacle’s avoidance and physical collision that will affect the pedestrian
movement direction’s selection and safety. Hence, this research has simulated the enhanced
transition direction of the CA based pedestrian movement using the optimal Moore
Neighborhood transition method and has proposed the Pythagorean Theorem for the shortest
distance path finding for assisting the pedestrian direction selection. This research has proven
that the enhanced movement transition direction and the implementation of Pythagorean
Theorem able to construct an intelligent and near-realistic pedestrian simulation that able to
evacuate in fastest time (< 25 seconds) from the selected spatial layout design compared to the
transition method based pedestrian movement simulation.
Defects and Components Recognition in Printed Circuit Boards Using Convolutional

R0124
Neural Network
Leong Kean Cheong, Shahrel Azmin Suandi and Saimunur Rahman
This paper introduces an automated components recognition system for printed circuit boards
using Convolutional Neural Network (CNN). In addition to that, localization on the defects of the
PCB components is also presented. In the first stage, a simple convolutional neural network-
based component recognition classifier is developed. Since training a convolutional neural
network from scratch is expensive, transfer learning with pre-trained models is performed
instead. Pre-trained models such as VGG16, DenseNet169 and InceptionV3 are used to
investigate which model suits the best for components recognition. Using transfer learning with
VGG-16, the best result achieved is 99% accuracy with the capability of recognizing up to 25
different components. Following that, object localization is performed using faster region-based
convolutional neural network (R-CNN). The best mean average precision (mAP) achieved for the
defects localization system is 96.54%. 52
R0092 Statistical Analysis of Image Quality Measures for Face Liveness Detection
Enas A. Raheem and Sharifah Mumtazah Syed Ahmad
Face recognition is essential for a wide range of technologies that requires person
identification. Due to the presence of spoof face attacks, an additional layer of security is
needed to protect the system, which can be provided by liveness detection. In this paper we
develop a technique for discriminating live from fake images. Our approach is based upon the
hypothesis that spoofing scheme leave statistical indication or structure in images which can be
utilized for detection by assistance of image quality features. To achieve this, image quality
measures (IQMs) statistical evaluation has been implemented using the analysis of variance
(ANOVA) technique. A feature set of measures with highest discrimination power to distinguish
between real and fake images was obtained. This ensures the simplicity of detection system
and improves its computational efficiency.
R0122 A Tunable Multiple Outputs FIR Filter Structure Realization

Pemmavit Sutthikarn, Sorawat Chivapreecha and Akraphon Trirat
This paper proposes a design of a new finite impulse response (FIR) digital filter structure which
gives 6-tunable frequency responses as 2 low-pass filters (LPF), 2 high-pass filters (HPF), 1
band-pass filter (BPF) and 1 band-stop filter (BSF) at the same time. The design procedure is
initiated from 2 LPFs design. However in order to obtain a tunable filter, the unit delay is
replaced by first-order single-multiplier structure for all-pass filter. Therefore, this new
structure can be achieved and called tunable multiple outputs FIR filter. Finally, the amplitude
responses that obtained from proposed structure can show the tunable capability using tuning
parameter.
A New Threshold using Gaussian Density Function for Gray Scale to Binary Image
R0123
and Its Application
Phuvin Kongsawat and Sorawat Chivapreecha
The quality assurance is crucial and requires for the Thai rice product. However, there is the
existing method to check quality by using human analysis, but this method consumes a lot of
time and also gives uncertainty in results in order to improve efficiency the machine vision is
required. This paper proposes the new method to find the threshold for converting grayscale
to the binary image which is an important process in image processing application for rice
geometrical measurement for Thai rice quality assurance. The result from a proposed method
is compared with the existing techniques such as Otsu's method and the adaptive threshold.
The experiment uses a flatbed scanner for input image acquisition and binary image from a
proposed new thresholding can give the best result by independent from the external light
condition.
53
Analysis of Local Binary Pattern for Facial Expression Recognition using Patch Local
R0131
Binary Pattern on Extended Cohn Kanade Database
Halina Hassan and Shahrel Azmin Suandi
Deriving an effective facial representation from original face images is a vital step for successful
facial expression recognition. In this paper, we explore the effect of representing the
information in facial expression recognition using local patch local binary pattern (LBP). The
processed information with accurate representation of different expressions can discriminate
and improve the overall facial expression recognition accuracy. The objective of this paper is to
analyze the effect of using local region patch to represent facial features with local binary
pattern as the feature extraction method. In our experiment, first facial landmark is being
detected using cascade linear regression, followed by alignment and normalization. The LBP
feature extraction is being performed on the holistic image, followed by patch of mouth and
eyes. At the final stage support vector machine (SVM) has been used as a classifier to examine
the recognition rate. The results are being validated using the extended Cohn-Kanade database.
From the analysis, it is found that using local region LBP can significantly reduce the number of
features to be fed into SVM. Hence the processing time is improved.
R0132 An Interactive Evolutionary Multi-objective Approach to Skin Lesion Segmentation

Ooi Woi Seng, Khoo Bee Ee and Lim Chee Peng
In this paper, a generic framework for skin lesion segmentation based on Interactive
Evolutionary Computation (IEC) is presented. In this method, a set of segmentation
parameters is interactively optimised to produce quality segmentation of skin lesion.
Extensive experimental evaluation is carried out on a public dataset of 200 ermoscopic
images. The performance of the proposed method is compared to several state-of-the-art
techniques in skin lesion segmentation. The results show that the proposed method is
superior in providing better segmentation performance.
54
PHR4: POWER SYSTEMS, HIGH VOLTAGE AND RENEWABLE ENERGY
R0069 An Asymmetrical Multilevel Inverter in Photovoltaic (PV) Application

Nur Iffah Amirah Yusop, Naziha Ahmad Azli and Norjulia Mohamad Nordin
In power electronics, one of the most significant recent advances in converting a DC source
voltage to an AC output voltage is by using a Multilevel Inverter (MLI). An MLI is popularly used
due to its advantages in improving the efficiency of a power conversion system. It typically
involves medium to high power applications including photovoltaic (PV) as the renewable
energy (RE) source. The performance of the power conversion system is determined by the
harmonics in the MLI output voltage. The MLI structure itself allows reduced harmonic
distortion in its output voltage waveform. This project focuses on an Asymmetrical Cascaded H-
bridge Multilevel Inverter (ACHMI) topology as the power converter in PV application. The MLI
uses Phase Disposition Pulse Width Modulation (PDPWM) as its modulation technique to
generate the output voltage. The operation of the ACHMI in PV application is simulated using
MATLAB/ Simulink. The results obtained are analyzed to evaluate its performance. It indicates
improved overall performance compared to the Symmetrical Cascaded H-bridge Multilevel
Inverter (SCHMI), which makes it a suitable candidate as an inverter topology that can be used
in PV application.
R0079 Effect of Silica Nanofiller in XLPE on Electrical Treeing

N.S.M. Nazar, N.S. Mansor, N.A. Muhamad, A. I. Mohamed, M. Mariatti and M. Kamarol
This paper presents the electrical treeing behavior in cross-linked polyethylene (XLPE)
nanocomposites with the addition of silica nanofiller. The concentration of silica nanofiller in
XLPE was varied from 0 wt%, 0.5 wt% and 1.0 wt%. The needle plate electrodes were used to
investigate the electrical treeing. The structure and the propagation length of electrical tree
growth were investigated after 30 minutes of tree inception voltage. The result of electrical
treeing was compared between pure XLPE and XLPE/Silica. The result shows the tree inception
voltage increases with the increasing of nanofiller concentration. The results also revealed that
the bush type tree was formed in all concentrations.
Electrical Treeing Characteristics of XLPE Material Containing Treated ZnO Nano-

R0082
Filler
N. S. Mansor, N. S. M Nazar, M. Fairus, D. Ishak, M. Mariatti, H. S. A. Halim, A. B. A. Ghani and
M. Kamarol
The investigation on the electrical trees characteristics in_unfilled cross-linked
polyethyline(XLPE)_and XLPE containing treated ZnO nanofillers has been carried out. The
characteristics of electrical trees in treated XLPE/ZnO nanocomposites were compared with
the unfilled one. The ZnO nanofiller was treated with Aminopropyltriethoxysilane (KH550).
Variation on the concentration of ZnO nanofiller in XLPE was carried out from 0.5wt% to
1.5wt%. The result reveals that the treated nanofillers in XLPE composite did not improve the
compare with the unfilled one. However, the addition of 0.5 wt% and 1 wt% of treated ZnO
nanofiller have improved 8.57% and 14.29% electrical tree growth, respectively in comparison
to the unfilled one .
55
Design and Simulation of SPWM and SVPWM based on Two Level Three Phase
R0093
Voltage Source Inverter, VSI for Grid – Connected PV System
Masri. S, Md Desa. M.K, Hariri. M.H.M
Grid – connected Photovoltaic (PV) system is one of the smart - grid system branch which offers
plenty of opportunity for further improvement. Apart from grid synchronizing procedure, pulse
width modulation (PWM) technique plays huge role in constructing the grid – connected PV
system to be more efficient and reliable. This paper presents the design and simulation of
Sinusoidal PWM (SPWM) and Space vector PWM (SVPWM) based on two level three – phase
voltage source inverter (VSI). The simulation is performed in MATLAB/Simulink platform.
Comparison between both PWM techniques in terms of the utilization of DC voltage and total
harmonic distortion (THD) were carried out. SVPWM demonstrates better performance as
compared to SPWM hence is qualified to use for grid – connected PV system.
Study of The Output Voltage of Asynchronous Raindrop for Piezoelectric Array

R0135
Energy Harvester
Chin-Hoong Teoh, Zuraini Dahari and I.M Izrin
This paper proposed the array piezoelectric configurations on asynchronous drop timing for
piezoelectric raindrop energy harvester (PREH). The effect of different types of configurations of
piezoelectric on the output RMS voltage are studied. The result revealed that the voltage
increased as the number of piezoelectric connected in series increased. It is found that six
piezoelectric generated the highest RMS voltage, 255.2 mV. In addition, 3S-2P configuration has
the highest RMS voltage generated compared to the other configurations for combination array
circuit.
Performance Evaluation Of 6-Gbps Hybrid Dwdm/Multibeam Free-Space Optical

R0063
Network In An Unusual Haze
Samer A. Al-Gailani, Mohd Rizal B. Ashad, Osayd M. Kharraz and Redhwan Q. Shaddad
To prevail against atmospheric attenuation considering unusual haze in tropical areas and to
scale the growing demand for high capacity communication bandwidth and scalability, a
prominent technique is used known as dense hybrid wavelength division multiplexing
(DWDM)/multibeam free-space optics (FSO). In this study, attenuation arising due to haze
scattering which is given as a function of link distance, wavelength and visibility is estimated
from visibility data obtained from Malaysian meteorological department. Maximum attenuation
of 20 dB/km has been observed due to unusual haze in Malaysia. This attenuation of 20 dB/km is
applied on a proposed hybrid four channel of 1.5-Gb/s DWDM/multibeam FSO network with
four wavelengths having standard downlink channel spacing of 100 GHz (~0.8 nm). The hybrid
DWDM/multibeam FSO method has enhanced the performance of an FSO link in terms of the
data rate, link distance, received power and scalability. The proposed approach facilitated access
data to four end users at a data rate of 1.5 Gb/s each, over a link distance of 1,645 m at BER of
10-9 under 20 dB/km attenuation.
56
Design and analysis of compact perfect metamaterial absorber for dual band
R0067
applications
Mohammad Jakir Hossain, Mohammad Rashed Iqbal Faruque, Md. Jubaer Alam, Eistiak Ahamed
and Mohammad Tariqul Islam
A new multi-ring hexagonal with meandered line design structure based on simple outline is
proposed for dual band application in microwave ranges. The finite-difference time-domain
(FDTD) method based CST simulator was approved to examine the perfect metamaterial
absorbance (PMA). The structure revealed the resonance frequency within the C-band and X-
band of the microwave spectra. At resonance frequencies, the absorbance’s pinnacles are
99.18% at 7.96 GHz and 99.50% at 11.62 GHz, correspondingly. Dual band operating frequencies
have been obtained by design configurations. The dimension of the offered structure is 10.50
mm × 10 mm × 1.6 mm that comprises all geometrical parameters to suitable the structure
inside the area of substrate material. The results of the offered PMA exhibited dual band
absorber response over the frequency ranges from 5 to 14 GHz. Therefore, the offered structure
enables numerous application areas, for example, satellite communications, radar, security,
protection as well as stealth innovation.
Optimal Power Allocation with Time Power Switching Relaying Protocol in

R0081
Cooperative Networks
F. K. Ojo and M. F. M. Salleh
In this paper, we study the optimal power allocation for energy efficiency maximization in
cooperative energy harvesting enabled relay networks with time power switching relaying
(TPSR) protocol. In particular, we develop channel capacity and total power consumption
expressions in terms of power splitting (PS) and time switching (TS) coefficients using TPSR
protocol in the considered network. We formulate an optimization problem and then apply a
power control technique to solve our formulated problem. Subsequently, we extend our study
to a multiple-relay scheme, where the best relay is selected from the available multiple relays.
We then propose a Max-Min relay selection scheme. Our simulation results show that the
energy efficiency maximization using TPSR protocol outperforms the energy efficiency
maximization using both the PS-based and TS-based relaying protocols.
A Modified Probability of Success Transmission Approach for MAC Protocol in

R0089
Wireless Networks
D.O Akande and M.F.M Salleh
We present a modified approach of obtaining the probability of successful packet transmission
for MAC layer protocol. Over a decade, several works have proposed different ways of obtaining
the probability of successful transmission to compute the throughput for both the legacy 802.11
DCF and cooperative-based MAC protocols. Although, most work in the MAC layer perspective
has employed the traditional probability of successful transmission, we exploit the successful
exchange of control and data packet transmission between nodes to obtain the probability of
successful transmission. We then compare the performance of our modified probability of
success transmission and the traditional through simulation study. The result obtained shows
that the network throughput of the cooperative-based MAC protocol with modified probability
of successful transmission outperforms the traditional by a performance gain of 3.06% at the
lowest number of nodes in the network.
57
Embedded Ethernet and Controller Area Network(CAN) in Real Time Control

R0036
Communication System
Ching Chia Leong and Mohamad Khairi Ishak
Embedded Ethernet and Controller Area Network (CAN) protocol can be used in control network
to achieve hard real-time communication. For embedded Ethernet protocol, Carrier Sense
Multiple Access with Collision Detection (CSMA/CD) is the media access control (MAC) used to
control data transmission between nodes in network. Back-off algorithm in CSMA/CD is used to
handle packet collisions and retransmission. For CAN protocol that developed for automotive
application, it has priority arbitration to handle collisions and retransmission. In this paper,
embedded Ethernet network models and CAN network models are developed and simulated in
MATLAB Simulink software. Several back-off algorithms, which are Binary Exponential Backoff
(BEB), Linear Back-off Algorithm, Exponential-Linear back-off Algorithm and Logarithm Back-off
Algorithm are proposed and implemented into Embedded Ethernet network model to evaluate
the performance. Both embedded Ethernet and CAN network models are extended to 3, 10 and
15 nodes to evaluate performance at different network condition. The performance criteria
evaluated and discussed are average delay and jitter of packets. The results show that in network
with high number of nodes, Linear Back-off Algorithm and Exponential-Linear back-off Algorithm
shows improvement in packets delay, 61% and jitter, 83% compared to standard algorithm, BEB.
For CAN network, the packet jitter is relatively low, 0.293ms.
R0022 4x1 Array Antenna with Harmonic Suppression Capability

Nur Aisyah Amir, Shipun Anuar Hamzah, Shaharil Mohd Shah, Khairun Nidzam Ramli, Mohamad
Md Som, Mohd Shamian Zainal, Fauziahanim Che Seman, Mazlina Esa and Nik Noordini Nik Abd
Malik
A 4 × 1 rectangular patch array antenna with electromagnetic bandgap (EBG) structure and filters
is proposed in this work. The antenna operates at 2.45 GHz which is suitable to be used in the
rectenna system and has the ability to suppress the higher order harmonics. To combine
harmonic suppression and filtering functions within the antenna, two filters on the feed line and
EBG structure on the ground plane are implemented into the array antenna configuration. The
filter is designed such that unwanted signals are removed to achieve higher order harmonic
rejection mode. The EBG structure which is located on the ground plane, act as a Defected
Ground Structure (DGS). The structure provides a low pass representative with improved
harmonic suppression performance in the upper frequency of the stop band. For demonstration,
a 4 × 1 rectangular patch array antenna operating at 2.45 GHz is designed, fabricated, and
measured. Reasonable agreements between simulated and measured results are observed. The
prototype exhibited excellent suppression at 5.56 GHz and 7.2 GHz in the presence of the EBG
structure and filters. The return loss also increases to -29.99 dB at the operating frequency with
increases in gain.
58
ACKNOWLEDGEMENTS
The organizing committee acknowledges the efforts of all those

who have contributed their valuable time and efforts as reviewers
in ensuring high quality technical papers for ROVISP2018 &
RCEEE2018. In addition to this, to the International Advisory
Committee, thank you for the continuous support to this
conference.
Deepest appreciation is due to all staff of the School of Electrical

and Electronic Engineering, Universiti Sains Malaysia (USM). We
would like to express our gratitude to Springer for their technical
supports, as well as AUN/SEED-Net , IET and Tekmark Group for
their financial supports.
Finally, the organizing committee would like to express our utmost

thanks to all the participants, attendees and exhibitors during
ROVISP2018 & RCEEE2018.
59
ASEAN University Network/Southeast
Asia Engineering Education
Development Network (AUN/SEED-
Net) has been established to promote
human resource development in
engineering for sustainable socio-
economic development of the ASEAN
region. Due to the need for sustainable
development after the economic
downturn in ASEAN region in 1997,
AUN/SEED-Net was officially
established as a sub-network under the
auspices of ASEAN University Network
(AUN) in 2001 from leaders’ initiatives
derived at earlier meetings related to
ASEAN – Japan Summit.
Full operation of AUN/SEED-Net Project started in 2003 with main

support from the Japanese Government through Japan International
Cooperation Agency (JICA), Member Institutions and Member
Governments. The network of AUN/SEED-Net is composed of 26
Member Institutions from 10 ASEAN countries selected from leading
institutions in engineering in respective countries. Project activities
have been implemented by the Member Institutions and AUN/SEED-
Net Secretariat with support from 14 leading Japanese Supporting
Universities and JICA Project Office for AUN/SEED-Net.
For more than a decade and a half, the AUN/SEED-Net project has
made significant contribution to the development of human resource
in ASEAN region. Their effort to promote higher education in the field
of engineering has produced the great outcomes. Each of the member
institutions offers great support to establish a region-wide system for
advanced institution and research by ASEAN member institutions in
collaboration with Japanese Supporting Universities.
60
61
Established in 1994 by Dato’ Ralph Khor and other partners, Tekmark Group of
Companies started with a clear vision of becoming a reputable technology business,
offering customers only the most advanced and quality Test and Measurement
solutions.
With over two decades of deep electronic measurement industry experience,
Tekmark has grown from strength to strength, from partnering with the best
measurement manufacturers, to design and operating our very own accredited
service and calibration laboratory facilities.
Today, Tekmark has more than 10 offices across the ASEAN region, with regional
head office situated in Singapore and Kuala Lumpur to deliver unparalleled service
and attention to our valued customers, wherever they may be. Among other
complimentary agencies, Tekmark Group takes pride in being Keysight and IXIA’s
Authorised Technology Partner for Malaysia, Singapore, Myanmar and the
Philippines.
The Group also provides Asset Managed Services, Calibration and Repair, System
Integration for Medical Industry and Industry 4.0 Solution (Smart Factory, Industrial
Automation, Cyber Security and Internet of Things Enablement).
Tekmark Group At A Glance
For more information, please call Samantha/Benjamin at 03-9057 8999 or email to 62

[email protected] or visit www.tekmarkgroup.com

RCEEE2018 Programme Book

Uploaded by

Copyright:

RCEEE2018 Programme Book

Uploaded by

Document Information

Original Description:

Copyright

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

RCEEE2018 Programme Book

Uploaded by

Copyright:

The 10th International Conference on Robotics, Vision, Signal Processing & Power Applications

14th & 15th August 2018 Wembley St-Giles Hotel, Penang

May God reward you with bountiful goodness to all.

“When sustainability is viewed as being a matter of survival for your business, I

PROF. MOHD RIZAL MOHD ARSHAD

It is my great pleasure to welcome you to the 10th International Conference on

“Empowering Research and Innovation towards Sustainability” is the conference

ASSOC. PROF. Dr. MOHD FADZLI SALLEH

CONFERENCE ORGANIZING COMMITTEE

SECRETARIAT FINANCE & FUND

TECHNICAL & LOGISTIC &

MULTIMEDIA & EVENT PUBLICITY & WEBSITE

OPENING CEREMONY SCHEDULE

8:00 – 8:20 Arrival of VIPs/guests/delegates

Opening Remarks by ROVISP/RCEEE Conference

Welcome Speech by SEEE Dean

9:00 – 9:10 AUN/SEED-Net Representative

9:10 – 9:30 Officiating Ceremony & Souvenir Presentation

9:30 – 10:00 Keynote 1

10:00 – 10:30 Opening Ceremony ends & Coffee Break

Wembley St-Giles Hotel, Penang, Malaysia

Keynote Speech 1 – Prof. Dr. Satoshi Matsumoto

10.00 – 10.30 Coffee Break

Keynote Speech 2 – Prof. Dr. Junzo Watada

11.00 – 11.30 Keynote Speech 3 – Prof. Dr. Naoki Uchiyama

B1 – DEC/EDAS1 B2 – VIS1 B3 – PHR1 B4 – RF1

20.00 – GALA DINNER [RCEEE PARTICIPANT & COMMITTEE ONLY]

Wembley St-Giles Hotel, Penang, Malaysia

8.30 – 9.00 Keynote Speech 4 – Prof. Dr. Uda Hashim

Keynote Speech 5 –Patrick Klotz

9.30 – 10.00 Coffee Break

12.00 – AUN/SEED-Net Introduction and Closing Ceremony

B1 – AI2 B2 – VIS4 B3 – PHR4 B4 –RF3 RCEEE2018

Sustainable Energy System in Collaboration with

Electric power demand is increasing year by year as the world population

In the lecture, we will introduce the current situation on development of high

Professor Dr. Junzo Watada

Risk and VaR’s Role in Power System’s

Risk can be defined as Variance. Most often, optimization problem targets

Professor Dr. Uda Hashim

Nanotechnology in Medical Diagnostics

Embrace Artificial Intelligence – Don’t Fear

Self-driving cars, Chat-bots, Robo-servants in restaurants…

Time Technical Areas

Prof. Hla Myo Assoc. Prof.

11.30 – 11.45 C0020 R0084 C0003 C0012

Assoc. Prof. Dr.

14.00 – 14.15 R0034 R0096 R0019 R0043

Chairperson Dr. Teoh Dr. Jiashen Prof. Fadzil Ain

Time Technical Areas

Analysis on Band Diagram Design and Physical Properties of p-GaN/n-ZnO Light

C0028 Implementation of an analog-to-digital stochastic conversion circuit in FPGA

C0027 Reactive Command Shaping for Vibration Control

A Hardware Model for the Control and Datapath of a Convolutional Neural

R0070 Internet of Things: Security Challenges and Its Future Direction

VIS1: VISION, IMAGE AND SIGNAL PROCESSING

An Improved Initialization Based Histogram of K-Mean Clustering Algorithm for

C0031 Parallelization of CCSDS Hyperspectral Image Compression using OpenMP

R0072 Intelligent Mobile Dictionary and Thesaurus

A Literature Review on Data Conversion Methods on EEG for Convolution Neural

PHR1: POWER SYSTEMS, HIGH VOLTAGE ENGINEERING AND RENEWABLE

C0003 Investigation of Grid Faults on Performance of Grid Tied Photovoltaic System

R0005 Load Management Benefits of Electrical Power Grids Performance: A Review