PROGRAM BOOK InECCE2023 : simplebooklet.com

Foreword Assalamualaikum wrt and welcome everybody,Distinguished keynote speakers and all participants.It is a glorious moment to extend my warm wishes onbehalf of the Faculty of Electrical and ElectronicsEngineering Technology (FTKEE), Universiti MalaysiaPahang Al-Sultan Abdullah (UMPSA). I want to conveymy heartfelt gratitude to the UMPSA management,particularly our Vice Chancellor, for supporting us inholding this event, namely "The 7th InternationalConference on Electrical, Control, and ComputerEngineering," better known as InECCE 2023, with theselected theme, "Embracing Innovation for theCommunity through Technological Transformation." We are also pleased to welcome our keynote speakers: Professor Emeritus David Al-Dabbas from Nottingham Trent University, England, United Kingdom; Dr. Shahrul YazidYahaya from Intel Technology Sdn. Bhd., Kulim, Kedah; and Professor Ir. Ts. Dr. KamarulHawari bin Ghazali from Universiti Malaysia Pahang Al-Sultan Abdullah, Pekan,Pahang. I would also like to express my heartfelt gratefulness as we are able to holdthis series of conferences in a physical mode after the necessity to hold it virtually in2021 due to the COVID-19 pandemic. There were 134 papers submitted; however, only107 relevant and high-quality papers have been accepted and will be presented inhybrid mode. This is also a great opportunity to share the new findings and innovationsin electrical, electronic, and computer engineering, particularly for the benefit of thecommunity, in alignment with the theme.On behalf of FTKEE, I am pleased to congratulate all the research scholars and paperpresenters from various backgrounds and countries who have made it to thisconference. These papers will enlighten all of us on the importance of a researchculture, especially in the area of electrical and electronics. It is hoped that all sharedknowledge here will be utilized for the betterment of humanity and nature for ourfuture generations.Last but not least, I am keen to congratulate the InECCE 2023 chairman and theorganizing committee for successfully holding this conference in the series. Despite allthe challenges faced, this conference has been successfully delivered and has served asa platform for knowledge sharing.Associate Professor Dr. Hamdan Bin DaniyalDeanFTKEE, UMPSA03

Page 4

Welcoming RemarkOn behalf of the organizing committee, it is my greatpleasure to welcome all participants to attend the 7th2023 International Conference on Electrical, Control andComputer Engineering (InECCE2023). This conference isorganized by Faculty of Electrical and ElectronicsEngineering Technology, Universiti Malaysia Pahang Al-Sultan Abdullah. It is a second time that the conferenceis held outside Pahang, Malaysia. The main aim of theconference is to provide an international platform forresearchers, academicians, engineers as well as industrialprofessionals from all over the world to share the findingsfrom latest research and developments in SustainableEnergy & Power Engineering, Instrumentation, Control &Computer Engineering and Applied Electronics &Computer Engineering research clusters. As stated by the conference theme, “Embracing Innovation for the CommunityThrough Technological Transformation”, it is hope that this conference will provide agood platform for the researcher to come together to contribute ideas and knowledgeto transform the innovation of technology for the benefit of the society or community.It is a unique opportunity for us to come together and to meet each other physicallysince the last conference is held virtually due to Covid19 pandemic. A total of 134papers have been submitted to this conference where the Microsoft ConferenceManagement Toolkits (CMT) was employed as a platform for paper submission, reviewsand camera-ready submission. After the reviewing process, 107 papers are accepted tobe presented in the conference. Among the 107 papers, 41 papers come from theresearchers from other universities including the universities outside Malaysia. Theaccepted paper will be presented in the conference before submitting to the SpringerLecture Notes in Electrical Engineering (LNEE) book series indexed by Scopus.I would like to take this opportunity to express my gratitude to conference committeemembers and reviewers for assisting me in the conducting the event and working hardto finish review in time to ensure the success of the event. I would like to thank ouracademic keynote speakers (Professor Emeritus David Al-Dabass and Professor Ir. Ts.Dr. Kamarul Hawari Ghazali) and industrial keynote speaker (Dr. Shahrul Yazid Yahaya)from Intel Corporation, Penang, Malaysia. Also, I would to acknowledge the greatsupport from my Dean of Faculty, Associate Professor Dr. Hamdan Daniyal who give mehuge opportunity to organize the Faculty main event in this year. I would like to extendmy high appreciation to the secretariats of the conference and all faculty staffs(academic & non-academic). Finally, lets enjoy and obtain benefit from the conference.Ir. Ts. Dr. Norizam Bin SulaimanGeneral ChairInECCE202304

Page 5

A Professor Emeritus of IntelligentSystems in the department ofComputer Science, TheNottingham Trent University. Hegraduated from Imperial College,worked for Redifon FlightSimulation for 6 years. Completeda Ph.D program in ParallelProcessing at StaffordshireUniversity. He is Fellow of the IET,IMA and BCS and editor-in-chiefof the International Journal ofSimulation: Systems, Science andTechnology; he currently serves asChair of the UK SimulationSociety.ProfessorEmeritus DavidAl-DabassBSc(Eng), ACGI, PhD, CEng,CMath, FIMA, FIET, FBCSSchool of Computing &InformaticsNottingham Trent UniversityKEYNOTE SPEAKER 1TITLE: Deep Learning Hybrid RecurrentAlgorithms using Data Mining forKnowledge Discovery SYNOPSIS:Starting with a model of the signaltrajectory to be mined, a recurrentdeep learning hybrid algorithm isderived to discover the knowledgeembedded within the data. Resultsshow good performance of thealgorithm in discovering the datamodel parameters online. Suggestionsfor future directions are given.05

Page 6

An accomplished R&D professionalin the semiconductor industry, heserves as the Department Managerfor Product Platform Developmentat Intel Corporation. With a Doctor ofEngineering in Engineering BusinessManagement, he possesses 28 yearsof diverse experience spanningmanufacturing, R&D, projectmanagement, and technologytransfer. A valued member ofuniversity advisory panels, hebridges industry expectations,academic capabilities, and graduatepotential. His expertise fuelsinnovation, research, and successfulteam leadership.Dr. ShahrulYazid YahayaDepartment Manager, System Test Hardware Design &Development, Sort TestTechnology Development/TMGIntel Technology Sdn. Bhd. KEYNOTE SPEAKER 2 TITLE: Challenges in the Deployment ofArtificial Intelligence SystemSYNOPSIS:Artificial intelligence (AI) technologyhas been steadily moving towardsbecoming the mainstream driver ofmany systems that affect how peoplelive and work. This covers manydifferent areas such as healthcare,transportation, retail, finance,manufacturing and many others. Thechallenges of developing AI system inthe laboratory mostly involvestechnical considerations such asarchitecture, workload balancing,performance, scalability tradeoffs, etc.The challenges of deploying suchsystem all the way to end usershowever, involve significant andoverwhelming effort to address. This keynote will introducechallenges, namely bias, privacy, ethicsand legal aspects, that must beconsidered and comprehended in thedeployment of any AI system. Thekeynote will mainly use autonomousdriving system as a case example. Thekeynote will suggest ways wheresocial leaders, lawmakers,technologists and industry leaders canformulate solutions together in ordernot to inhibit the growth of AItechnology.06

Page 7

KEYNOTE SPEAKER 3 TITLE: Early Warning System for Landslidesin Malaysia: A GNSS-Based MonitoringApproachSYNOPSIS:Landslides, escalating globally, pose apressing challenge in Malaysia due toclimate change impacts. To minimizedevastating consequences on life,property, and infrastructure, effectivecontrol measures are vital. Real-timemonitoring systems offer a proactiveapproach to identify potentiallandslide zones, averting disasters inadvance. Despite remote sensing,radar, and LiDAR being availabletechnologies, cost and accuracy limittheir broad use. Addressing theselimitations, a novel solution employsGNSS satellite technology forlandslide monitoring. Analyzing landmovement signals, it promptlydetects changes in position and slope,providing early warnings for potentiallandslides. Cost-effectiveness andprecision make the GNSS-basedmonitoring system suitable fordiverse locations. By deploying thistechnology in high-risk areas,Malaysia can promptly detect andmitigate potential landslides. Itsaffordability and accuracy make it avaluable asset for sustainabledevelopment in Malaysia and othersimilarly affected regions. Professor Ir. Ts.Dr.KamarulHawari GhazaliFaculty of Electrical andElectronics EngineeringTechnologyUniversiti Malaysia PahangAl-Sultan AbdullahA Professor in the Faculty ofElectrical and ElectronicEngineering Technology, UniversitiMalaysia Pahang Al-SultanAbdullah. His major research areasinclude Machine Vision System,Image Processing, SignalProcessing, Intelligent system,Vision Control, Thermal ImagingAnalysis. (in all relatedapplications – Electrical, Medical,Environment), Deep Learning forImage and Signal Classification.Currently a Professional Engineer(Ir.), Board of Engineer Malaysia(BEM).07

Page 8

Page 9

TimeProgram08:00 - 08:30Registration | Level 2, Royale Ballroom08:30 - 09:00Opening Remarks09:00 - 09:30Keynote Speech 1Prof. Emeritus Dr David Al-Dabbas, Nottingham Trent University, United Kingdom09:30 - 10:00Keynote Speech 2Dr Shahrul Yazid Yahaya, Intel Corporation, Malaysia.10:00 - 10:30GROUP PHOTO & COFFEE BREAK10:30 - 11:00Keynote Speech 3Prof. Ir. Ts. Dr. Kamarul Hawari Ghazali, Universiti Malaysia Pahang Al-Sultan AbdullahROOMParallel Sessions | Level 2ROYALE 1ROYALE 2ROYALE 311:00 - 13:00Session 1 (AppECE)Session 2(ICE)Session 3(SUPER)ID59 , ID4, ID14, ID24, ID26,ID28, ID30, ID39, ID42, ID58,ID114 ID86, ID3, ID10, ID6, ID12, ID31,ID22, ID32, ID46, ID95 ID61, ID2, ID7, ID29, ID52, ID72,ID85, ID106, ID107, ID35 13:00 - 14:00LUNCH TIME | Royale Coffee House14:00 - 16:00ROYALE 1ROYALE 2ROYALE 3Session 4(AppECE)Session 5(ICE)Session 6(SUPER)ID87, ID25, ID65, ID69, ID118,ID71, ID80, ID83, ID27, ID110,ID88ID77, ID124, ID76, ID40, ID13,ID44, ID63, ID75, ID23, ID20,ID115 ID62 , ID8, ID9, ID11, ID99, ID68,ID97, ID56, ID128, ID47, ID19 16:00 - 16:30COFFEE BREAK16:30 - 17:00Conference Awards | Level 2, Royale BallroomPARALLEL SESSIONS- PHYSICAL EVENT -SESSION INFORMATIONSession 1, 4: APPECE - Applied Electronic and Computer EngineeringSession 2, 5: ICE - Instrumentation & Control EngineeringSession 3, 6: SUPER - Sustainable Energy & Power Electronics09

Page 10

11:00 - 12:30MS TEAMSSession 7(AppECE)Session 8(ICE)Session 9(SUPER)ID104, ID109, ID111, ID119,ID125, ID133, ID135ID94 , ID116, ID127, ID129,ID131, ID132, ID122 ID18 , ID41, ID73, ID78, ID82,ID90, ID91 https://tinyurl.com/ParSess7https://tinyurl.com/ParSess8https://tinyurl.com/ParSess914:00 - 15:30MS TEAMSSession 10(AppECE)Session 11(ICE)Session 12(SUPER)ID36, ID37, ID38, ID45, ID117, ID103, ID123, ID101 , ID98, ID89, ID60,ID102, ID136, ID121 ID92, ID105, ID126, ID130,ID137, ID108, ID93, ID55https://tinyurl.com/ParSess10 https://tinyurl.com/ParSess11https://tinyurl.com/ParSess12PARALLEL SESSIONS- ONLINE EVENT -SESSION INFORMATIONSession 1, 4: APPECE - Applied Electronic and Computer EngineeringSession 2, 5: ICE - Instrumentation & Control EngineeringSession 3, 6: SUPER - Sustainable Energy & Power Electronics10

Page 11

Page 12

Understanding the influence of different dimensions on conductivity is fundamental in many fields,including electronics, material science, and electrochemical sensing. This study investigates the impacts ofvarying dimensions of electrodes on the conductivity performance of the porous paper-based humiditysensor. Hydrochloric acid (HCl) solution was chosen for its potential to alter the surface properties of thepaper to form porous paper. Interdigitated electrode structures with varying dimensions were fabricatedusing porous paper. It is found that a smaller electrode gap size provides better conductivity and lessresistance. A wider finger width provides more conductive pathways for electrons to travel, increasingconductivity. In comparison to a smaller electrode finger width, the wider electrode finger width transmitsmore current.Paper ID: TITLE: A b s t r a c t :AUTHOR(S): 59Porous Paper-based Device with Different Electrode Dimensions for HumiditySensor ApplicationsLoo Wei Yang, Gan Shin Pyng, Tan Jin Peng and Mastura Shafinaz ZainalAbidin*Paper ID: TITLE: AUTHOR(S): 4Surface anomaly detection using feature-based transfer learning for IoT-enabledsmart manufacturingMuhammad Ateeq, Matilda Isaac, Hadyan Hafizh, Bintao Hu, Ismail MohdKhairuddin, Mohd Amirul Abdullah and Anwar P.P. Abdul MajeedOwing to the advancement of computational technology, the employment of deep learning architecturefor defect detection in the manufacturing in-dustry has gained considerable attention. Traditional meansof defect detec-tion through manual visual inspection by operators are laborious as well as susceptible tomistakes. In the present study, a feature-based transfer learning approach is used to classify surfacedefects. The KolektorSDD database is used in the present study. Two pipelines were developed toinvestigate their efficacy in detecting the defects, namely the InceptionV3-SVM and VGG19-SVM pipelines,respectively. It was demonstrated from the study that the VGG19-SVM pipeline could provide desirableresults compared to the InceptionV3-SVM pipeline, suggesting that the VGG19 is a better feature extractorfor the evaluated surface defects. It could be concluded that the proposed pipeline is suitable for theclassification of surface defects.A b s t r a c t :PARALLEL SESSION 112

Page 13

The increasing demand and usage of quadcopter drone across fields and missions has led to activeinvolvement of quadcopter in research community. A proper controller is required for a quadcopter inorder to operate with a smooth and stable flight. Feature such as tracking has become one of therequirements to quadcopter. Such feature gives another challenge in designing a controller to en-sure astable and accurate quadcopter operation. Most system in industrial control application, includingquadcopter prefer conventional PID controller as the main or base control system. However, conventionalPID used fixed parameters for the system which could lead to undesired responses if the systemencountered any changes, and this could also lead to a long-time adjustment. Thus, fuzzy logic techniqueis integrated to system to form Fuzzy PID controller which the tuning process can be simplified andshorten. The proposed algorithm of Fuzzy PID is simulated with some trajectories and put to comparisonwith conventional PID. The result of conventional PID shows more tuning need to be conducted to im-prove the responses, while the Fuzzy PID controller provide better responses without further tuningapplied.Paper ID: TITLE: A b s t r a c t :AUTHOR(S): 14Fuzzy-PID Trajectory Tracking Control of a QuadcopterA'dilah Baharuddin and Mohd. Ariffanan Mohd. BasriPaper ID: TITLE: AUTHOR(S): 24A Comprehensive Review on Luminance Distribution for Visible LightCommunication (VLC) SystemsIzzah Hazirah binti Zainal, Zaiton binti Abdul Mutalip, Faezah bintiJasman, Wan Hafiza binti Wan HassanVisible light communications (VLC) are widely used in industries in or-der to transmit data signal fromtransmitter to receiver. This paper pro-vides comprehensive overview of the research conducted onluminance distribution for VLC. Luminance distribution can contribute significantly for visual performanceand visual comfort. The introduction to the basic ideas of VLC systems is followed by a review of themodulation strategies applied by re-searchers. The optimization of lighting schemes is also discussed,along with the challenges and limitations of various approaches. The benefits of using multiple lightsources are also included in this paper. The re-view concludes by summarizing the research gapsidentified and pro-posing future work in this area. Overall, the aim is to provide a comprehensiveunderstanding of the current state-of-art in luminance distribution of VLC systems with a focus onhighlighting areas for future improvement.A b s t r a c t :13

Page 14

The main element for reverse vending machine (RVM) is the waste classification technique. Theperformance of the RVM’s waste segregation can be enhanced by incorporating a good classificationtechnique. The objective of this project is to study the performance of CNN for image processing for thewaste segregation part of RVM. For this RVM, only polyethylene terephthalate (PET) bottles, aluminiumcans, and drink carton boxes are considered for recycling by using image classification based on transferlearning with Convolutional Neural Network (CNN) algorithms. The performance parameters that areevaluated are F1-score, computational time, and testing result for each neural network. In this paper,ResNet50 surpasses other neural networks due to the highest F1-score which is 0.9541 and the goodtesting performance although the computational time is longest among the other network which is240min 01s. The accuracy rate of ResNet50 achieved in this project is 0.9724.Paper ID: TITLE: A b s t r a c t :AUTHOR(S): 26Convolutional Neural Networks Performance Study for Image Processing ofWaste Segregation for Reverse Vending MachineTan Hor Yan, Zamani bin Md. Sani, and Sazuan Nazrah binti Mohd AzamPaper ID: TITLE: AUTHOR(S): 28Mechanical and Electrical Issues in Prefabricated Housing; Their Real Causesand Corrective StepsAhmad Bin Abd Jalil, Fadhilah Binti Md Fazil, Mohd Amir Shazwan BinHashim, Nurina Binti NawiHousing is most suitable to adopt prefabricated as it uses repetitious design, simultaneous construction, large scale andembed Mechanical & Electrical. Unlike develop countries which already far advanced, Malaysia still facing critical issuesincluding on fragmentation especially on M&E scope. Frag-mentation is defined as working in isolation, work inseparation and division that happens among different parties in the same project. Various research on prefabricatedhousing has been conducted focusing on supply chain, modularization, automation, design, payment and procurementbut lack fo-cus on fragmentation. This paper investigates the real causes of fragmenta-tion issues in Malaysianprefabricated housing that affected M&E and struc-tural to be miss-matched, their consequences and recommendsteps to over-come. The methodology used is mixed method with 118 questionnaires were analyzed using SPSS, thenfollowed with deep interviews with experienced professionals from M&E companies, prefabricated manufacturers,prefabri-cated installers, academicians, consultants, developers and main contractors. The result shows the causes offragmentation are M&E unfamiliar with pre-fabricated concept, late involvement of prefabricated companies and thepractice of segregation culture while the steps to overcome are by establish-ing close integration between M&E andproject team since beginning, each party must be appointed early and forming direct contract between prefabri-catedcompanies and developers. The contribution to the body of knowledge is by detailing the real causes that makefragmentation issue become critical, with descriptions on where the causes were actually started and also present-ingrecommendations that are practical which verified by all stakeholders who directly involve in Malaysia prefabricatedhousing.A b s t r a c t :14

Page 15

This research focuses on the integration of Internet of Things (IoT) technol-ogy for monitoring andcontrolling the performance of wireless power trans-fer (WPT) systems. Overall, WPT allows for thetransmission of electrical energy without the use of physical connections, which provides simplicity andflexibility. The proposed framework incorporates IoT capabilities into the WPT system allowing for real-time monitoring and control. IoT devices deployed throughout the system collect data on parameterssuch as power transmission efficiency, temperature, voltage, and current. The experimental result showsthe transmitter and receiver for coil inductance by using 30 turns with 0.5 mm diameter helical and spiralcopper coils. Whereby, 145.00 µH inductance was entered to the transmitter and 165.00 µH for the receiverof the helical coil. Alternatively, the spiral coil inductance for the transmitter and receiver was 106.98 µHand 110.00 µH respectively. Following that, the output voltage for the WPT with load is 2.35 V at 10 mmdistance, which is higher than 0.08 V at 50 mm distance between coils. Next, it is noteworthy that theoutput voltage for the WPT without load exceeded the output voltage with load, which is 10.47 V at a10mm distance and 0.34 V at a 50 mm dis-tance. This research advances WPT technology by demonstratingthe con-venient of IoT-enabled monitoring and control. The proposed framework in-creases the overallperformance efficiency of WPT systems, making them ideal for use in industries such as automotive,consumer electronics, and healthcare.Paper ID: TITLE: A b s t r a c t :AUTHOR(S): 30Monitoring and Controlling Wireless Power Transfer System Performance via IoTLiew Hui Fang, Muhammad Izuan Fahmi Romli, Rosemizi Abd Rahim, NurIrwany Ahmad, Junaidah Binti Ali Mohd Jobran, M. Abdul Jabbar,Muhammad Khairul Bin JaafarPaper ID: TITLE: AUTHOR(S): 39Enhancing Driver Fatigue Detection Accuracy in On-Road Driving Systems usingan LSTM-DNN Hybrid Model with Modified Z-Score and Morlet WaveletRafiuddin Abdubrani ; Mahfuzah Mustafa ; Zarith Liyana ZahariDriver fatigue is a significant safety concern in transportation systems, with the potential to causeaccidents. Detecting and addressing driver fatigue in real time is crucial for improving road safety. Thisresearch paper introduces an innovative method for detecting driver fatigue using electroencephalogram(EEG) signals, enhanced by the Morlet mother wavelet and modified z-score feature. The Morlet wavelet isadapted to capture both temporal and frequency information from EEG signals associated with driverfatigue, while the modified z-score feature measures abnormal EEG activity. Three deep learning models,Long Short-Term Memory (LSTM), Deep Neural Network (DNN), and LSTM-DNN, are employed to classifythe data. The LSTM model captures long-term dependencies, the DNN model learns complexrelationships, and the hybrid LSTM-DNN model combines their strengths to improve classificationaccuracy. The proposed approach demonstrates its effectiveness through comprehensive experiments,achieving high accuracy, specificity, sensitivity, F1-score, and recall in driver fatigue detection. The LSTM-DNN hybrid model showed exceptional performance, achieving an accuracy of 99.99% in classifying EEGsignals. This showcases its remarkable precision in accurately categorizing the signals. Additionally, theLSTM-DNN model exhibited a specificity of 99.98% and a sensitivity of 100.00%, indicating its capability toclassify driver fatigue states accurately. Furthermore, the F1-score and recall for the LSTM-DNN model were99.99% and 100.00%, respectively.A b s t r a c t :15

Page 16

This research aims to provide a wireless and convenient charging solution, eliminating the need for multiple cables andadapters. By integrating RFID technology with Raspberry Pi, the system can identify and communicate with RFID tagsor chips embedded in devices, initiating the charging process wirelessly. However, most of the mobile phones andlaptop come with a very poor battery life. Such as, li-ion batteries used for mobile phones and laptops are consumableproducts. Due to the chemical properties of Lithium ions, the battery capacity will decrease by use over time, the user'senvironment and behavior. By leveraging RFID technology and the versatility of Raspberry Pi, the system aims to offeran efficient and convenient charging solution. RFID tags embedded in chargers and laptops enable automatic devicedetection and authentication on the charging station. The Raspberry Pi serves as the control unit, managing thecharging process and ensuring optimal power management. This system will limit the charging time to each user with5 minutes and the LCD display showing the actual time left. This system will cut off the current automatically which isin fully charged. The simulation and experiment proved that mobile phone and laptop charging output volt-age DC iswithin 5 VDC to 20 VDC, respectively. Moreover, output current estimated in the range 0.5 A to 5 A. Experimentalevaluations demonstrate the system's feasibility and effectiveness in enhancing user experience, reducing cable clutter,and streamlining the charging process. The proposed system has implications for diverse settings, catering to theincreasing demand for user-friendly wireless charging solutions.Paper ID: TITLE: A b s t r a c t :AUTHOR(S): 42Smart Charging System for Mobile Charges and Laptops Using RFID andRaspberry PiLiew Hui Fang, Muhammad Izuan Fahmi Romli, Rosemizi Abd Rahim, NurIrwany Ahmad, Junaidah Binti Ali Mohd Jobran, M. Abdul Jabbar, Lee JunRongPaper ID: TITLE: AUTHOR(S): 58IoT Smart Distribution Box (SDB) for Homestay Energy Management System byUsing Arduino and ESP32Muhammad Syurahbil Shamsudin ,Sim Sy Yi *, Mohd Abdul Talib MatYusoh Alvin John Lim Meng Siang , Azuwien Aida Binti BohariThis paper presents the design and implementation of a Smart Distribution Box (SDB) system to addressenergy waste behaviour by tenant of homestays during vacation. The system allows owners of homestaysto con-trol and monitor energy usage at their properties using a Blynk application and a GSM SIM900Acontrol system. Consumers can also limit their energy usage based on the amount they have "topped-up"to the system. The SDB includes a user-friendly Distribution Box (DB) that facilitates monitoring andcontrolling energy usage during stays. Additionally, a mock payment system for tenant has beendeveloped using messages sent to a GSM SIM 900A connected to an Arduino Uno and an ESP32 to link theBlynk apps for the owner to the system. The results show that the system can be used in standby mode,top-up mode, usage mode, low balance mode, and owner mode for the homestay. The system effectivelymanages energy usage between tenant and owners, makes payments for energy usage easy, and allowsfor upgrad-ing the DB system as technology evolves.A b s t r a c t :16

Page 17

Increasing demand for miniaturized antennas in wireless communication sys-tems has driven research onantenna miniaturization. This paper focuses on the miniaturization of a rectangular microstrip patchantenna, a commonly used antenna in wireless communication. To achieve miniaturization, a highpermit-tivity substrate, specifically an Epoxy-Barium Titanate nanocomposite, is em-ployed as areplacement for the conventional substrate, FR4. Different compo-sitions of epoxy filler (90:10%, 80:20%,70:30%, and 60:40%) substrates are investigated to determine the optimal design. The design andsimulation are performed using CST Studio Suite software at the Wi-Fi frequency of 2.4 GHz. Simulationare conducted to evaluate the antennas' performance in terms of S11, VSWR, efficiency, gain, anddirectivity. Among all the antennas with epoxy-barium titanate substrates, the antenna with a 20% epoxy-barium titan-ate composition exhibits the best performance at the Wi-Fi frequency of 2.4 GHz. Finally, theoptimized antenna design with the 20% and 30% epoxy-barium titanate substrate has potential to befabricated.Paper ID: TITLE: A b s t r a c t :AUTHOR(S): 114Modelling of Miniaturized Rectangular Microstrip Antenna Using Epoxy-BariumTitanate Nanocomposite Substrate for Wi-Fi ApplicationsNurulfadzilah Hasan, Nurul Hazlina Noordin, Mohamad Shaiful AbdulKarim, Nurhafizah Abu Talip Yusof, Noor Zirwatul Ahlam Binti Naharuddin, Mohd Ruzaimi Mat Rejab17

Page 18

Rotary-impact mechanisms generated high-impulse torques directly to the screws with relatively lowinput power. However, this type of tool also generates noise and vibration that may affect the operatorwho uses it. This paper develops a supervisory logic control based on a state-flow control algorithm on themodel to improve the shortfalls. Testing and evaluation are made through experiment and optimizationthrough simulation. The results showed a 19% vibration reduction from the motor’s reaction torque to theuser’s arms. The outcome may improve the user’s comfort, reducing the cause of hand-arm vibration andincreasing the workers’ allowable work time.Paper ID: TITLE: A b s t r a c t :AUTHOR(S): 86State Machine Logic Vibration Control Simulation of Rotary Impact DriverChi Hoe Leong and Rosmiwati Mohd-MokhtarPaper ID: TITLE: AUTHOR(S): 3Respiration Rate Monitoring using Optical-based Sensor integrated in Portableon Bed Device SystemMohd Anwar bin Zawawi, Nur Fatin Adawiah binti IsmailIn the past 3 years, coronavirus disease (COVID-19) which is caused by the SARS-CoV-2 virus has caused agreat challenge to human health condition. One of the major symptoms related to the COVID-19 isshortness of breath or breathing difficulty. People with severe COVID-19 may remain infectious beyond 10days. As the positive number of COVID-19 infectious patients in each country could increase higher thatthe capacity of available hospital beds, less critical cases will have to continue self-isolation at theirrespective home. The proposed respiration rate monitoring device is based on optical fiber sensorintegrated in a portable on bed device. The proposed device is able to provide continuous measurementof the respiration where the breath-ing rate that could be recorded is between 6 to 24 times per minutes.In summary, the implementation of portable on bed device with the optical fi-ber sensor can be appliedas an alternative method to provide continuous respiration rate measurement for COVID-19 patient whoare undergoing self-isolation period at their respective home.A b s t r a c t :PARALLEL SESSION 218

Page 19

The beampattern of the collaborative beamforming (CB) in wireless sensor network (WSN) suffers fromhigh maximum sidelobe level (SLL) due to the distribution of sensor nodes in random manner. The highSLL will cause high interference which is unreliable for wireless communication. Thus, this paper proposesa method of optimizing inter-element spacing of sensor nodes based on evolutionary algorithm (EA)optimizers which are Imperialist Com-petitive Algorithm (ICA), Backtracking Search Algorithm (BSA),Genetic Al-gorithm (GA) and Particle Swarm Optimization (PSO). The inter-element spacing betweensensor nodes is optimized in linear antenna array (LAA) configuration. The beampattern is optimized interms of peak SLL suppres-sion, control first null beam width (FNBW) and null placement in unintendeddirections. The algorithms are evaluated in different cases to fulfil the single and multiple fitness functions.The results show that all algorithms managed to control FNBW size in five cases, while reducing peak SLLup to 40% (ICA), 33% (PSO), 30% (GA) and 25% (BSA).Paper ID: TITLE: A b s t r a c t :AUTHOR(S): 10Beampattern Optimization of Collaborative Beamforming in Wireless SensorNetwork using Evolutionary Algorithms: a ComparisonNajla Ilyana Ab Majid, Nik Noordini Nik Abd Malik, Nor Aini Zakaria,Muhammad Zahid Zainul AbidinPaper ID: TITLE: AUTHOR(S): 6Improved Sparrow Search Algorithm for Test Redundancy ReductionMizanur Rahman, Kamal Z. Zamli, Md. Abdul KaderThe Sparrow Search Algorithm (SSA), a kind of innovative swarm intelligence algorithm, has been used ina variety of domains due to its special qualities, such as its robust global search capabilities, its limitednumber of adjustable parameters, and its clear structure. However, the SSA still has some weaknesses thatprevent its further development. These weaknesses include low population diversity, limited localsearchability, and a tendency to easily slip into local optima. Software testing is critical to fulfilling the userrequirement. the software is tested using test cases, and it is impossible to run all test cases in a regressiontest suite because the size of the test suite increases as the software changes over time, and this wouldtake a lot of time and effort. There are numerous approaches to minimize test suites, but none of themcan create a test suite with the ideal number of tests because the problem is NP-complete. In this context,in this paper, we proposed an improved sparrow search algorithm and we use the test redundancyreduction problem as a case study. Therefore, our algorithm has shown promising and superior resultscompared to standard SSA.A b s t r a c t :18

Page 20

Uncertainties and perturbations are the ‘enemies’ of a flying robot without a proper and reliable controller.The easiest controller to implement is the PID controller which its gains only need to be tuned properlyeither using Ziegler-Nichol's Method or manual tuning. Both are time consuming and single-acting PIDcontrollers cannot adapt in various situations with only single-tuning. Being in different situations requiresre-tuning, which is unfavorable during physical flight mode. In this paper, single-acting PID controller willbe transformed into a hybridized mode which includes the action of Radial Basis Function (RBF) NeuralNetwork. The objective is to help a quadcopter to survive various uncertainties and perturbations with aself-tuned algorithm. An RBF network is one kind of Artificial NN but with simpler network design andmore accurate local approximation. The performance of the proposed work is proved through simulationsusing MATLAB/Simulink. Different situations are presented to test the final system, which are the winddisturbance and trajectory tracking. Results are presented in this paper using visual simulation and ISEperformance index. After comparison between the proposed work and ZN-tuned PID controller is made,RBFPID controller wins the deal.Paper ID: TITLE: A b s t r a c t :AUTHOR(S): 12Radial Basis Function with PID for Quadcopter: Disturbed Trajectory TrackingNur Hayati Binti Sahrir, Mohd Ariffanan Bin Mohd BasriPaper ID: TITLE: AUTHOR(S): 31Kalman Filter Based Vehicle State of Charge Performance Analysis for DifferentBattery TypesHamzah Ahmad, Mohd Mawardi Saari, Mohd Syakirin Ramli, Nor AqilahOthmanThe Battery Management System (BMS) is critical to the battery's efficient and safe performance. Themonitoring of bat-tery parameters is one of the tasks performed by the BMS. The state of charge of abattery is a critical parameter that indi-cates the amount of charge it contains. Battery SOC estimate is atthe root of the battery management system and SOC has a direct impact on BMS decision-making andcontrol. The Kalman filter correction approach is used in this design, as well as the impact of charge anddischarge rates, temperature, and charge or discharge cycles on SOC estimates. The Kalman fil-tercorrection technique is presented based on this technology, with its application in the pure electric carbattery man-agement system. The findings reveal that the Kalman filter correction algorithm efficientlycorrects the Ah method error, increases estimate accuracy, and offers a more accurate SOC estimationtechnique for battery management systems. An ac-curate estimate of the battery's level of charge iscritical not only for alerting the user, but also for developing a control plan for maintaining the battery'scharacteristics within safe limits in order to extend its life.A b s t r a c t :20

Page 21

Coupled tank systems have been commonly employed across various industries to maintain accurateliquid levels in storage tanks. However, the dynamic nature and input-output reactions within TITO-coupled tank systems pose challenges for controller design. Possible enhancement towards accuratecontrolling of TITO systems through the application of the existing PID controller has been recognized. Adata-driven Adaptive Fuzzy-PID controller is designed in this article for improved control accuracy of TITOsystems. Safe Experimental Dynamics (SED) is employed as the data-driven tool to determine optimalFuzzy-PID parameters that minimize control tracking performance and errors. The proposed methodachieved effective Adaptive Fuzzy-PID parameter adjustments in the absence of the layout’smathematical modeling. An integrated approach between Fuzzy logic and PID control demonstratedsignificant potential for enhanced control performance over conventional PID controllers. Throughsimulation results, the proposed objective function has been demonstrated to yield superior controlaccuracy. The integration of data-driven techniques and utilization of the Adaptive Fuzzy-PID controlleroffers promising prospects in advancing the control performance of TITO-coupled tank systems.Paper ID: TITLE: A b s t r a c t :AUTHOR(S): 22Data-Driven Adaptive Fuzzy-PID Control of TITO Coupled Tank System withInput Delay: Design, Implementation, and Performance EvaluationMohd Riduwan Ghazali, Mohd Ashraf Ahmad, Wan Ismail Ibrahim,Mohamad Jamadil Akbar Jaafar, Suliana Ab. GhaniPaper ID: TITLE: AUTHOR(S): 32A ROS Based Mobile Robot Navigation with Imperfect Data AssociationHamzah Ahmad, Mohd Mawardi Saari, Mohd Syakirin Ramli, Nur AqilahOthmanThis paper attempts to design and analyze a mobile robot navigation system with imperfect dataassociation. In this paper, we address the issue related to navigation and control of a mobile robot whichenable it to deal with unexpected moving obstacles by sensor-based control. A probabilistic approach isproposed to deal with collision avoidance under certain uncertainty conditions. The mobile robotlocalization methodologies in common use at present will be introduced. A localization algorithm basedon Extended Kalman Filter (EKF) will be the technique to be investigate considering of environmentfeature extraction and map building. The technique can reduce the error in the calculation of the robot`sposition and orientation. The analysis is mainly take into account the observations about the mobile robotsurroundings from multiple sensors by EKF, which enables the robot to identify the surrounding objectsclearly and moves accordingly. The simulation and experimental results show that the proposednavigation method is effective.A b s t r a c t :21

Page 22

In this study, paper-based energy storage devices using various electrodes’ mate-rial through vacuumfiltration technique is studied. Paper substrate is becoming more common in supercapacitorsmanufacturing due to its properties of low cost and environmental friendliness. This study examined thephysical properties of fabricated devices using graphene and zinc oxide electrode, as active materials.Based on the result, it is shown that active materials that are coated on the paper substrate do notdeteriorate in terms of physical properties after vacuum filtration process in producing electrodes. Thisproves that vacuum filtration implements sufficient amount of active material for power storage into theelectrodes. This shows the feasibility for large scale production in future.Paper ID: TITLE: A b s t r a c t :AUTHOR(S): 46Paper-Based Device using Vacuum Filtration Technique with DifferentElectrode’s MaterialTan Jin Peng, Loo Wei Yang, Gan Shin Pyng and Mastura Shafinaz ZainalAbidinPaper ID: TITLE: AUTHOR(S): 95Lego Parts Recognition Based on Its Unique CharacteristicsNur Afifah Mohamad Yusob and Mohd Razali DaudThe recognition of Lego parts plays a crucial role in automated assembly and sorting systems. In thispaper, we propose a novel approach for Lego parts recognition based on their unique characteristics. EachLego part is characterized by the presence of circles on its surface, with the size of the circles generallycorresponding to the number of circles present. By analyzing the arrangement of circles, the shape of aLego part can be determined. Our approach focuses on two key aspects: the count of circles and theconnectivity of straight lines through the circle centers. By examining how straight lines intersect, we canaccurately identify the shape of a Lego part. To achieve Lego parts recognition, we employ computer visiontechniques and algorithmic analysis. We first detect and extract circles from images of Lego parts usingthe Hough transform. Next, we analyze the circle count and calculate the connectivity of straight linesbased on the circle centers. By considering the intersection patterns of these lines, we can classify Legoparts into various shapes, including squares, rectangles, triangles, and more complex configurations.Experimental results on a large dataset of Lego parts demonstrate the effectiveness of our approach. Weachieve a high accuracy rate in recognizing Lego part shapes, with minimal false positives and negatives.Our method offers a promising solution for automating Lego part recognition in assembly and sortingsystems, enabling improved efficiency and productivity.A b s t r a c t :22

Page 23

Recently, Flux Switching Generator has been widely used due to its ability of having simple constructionsyet high output performances. Hybrid Excitation Flux Switching Generator (HEFSG) are one of thesufficient machines that can reduce the armature winding on the excitation part. However, the impact ofthe HEFSG design parameters on overall electric generator performance cannot be underestimated. This isdue to the generator’s output voltage, power, and thus flux distribution is all determined by itsappropriate geometries. Consequently, the impact of various design factors, such as rotor span angle, rotorpole width, shaft radius, rotor outer radius, stator inner radius, stator pole width and stator back irondimension has been observed. The results for the preliminary design of HEFSG has been obtained with anoutput voltage of 384.87V with 4.26Nm of cogging torque.Paper ID: TITLE: A b s t r a c t :AUTHOR(S): 61A Preliminary Design of Hybrid Excitation Flux Switching Generator for SmallScale ApplicationsNur Afiqah Mostaman, Mahyuzie Jenal and Erwan SulaimanPaper ID: TITLE: AUTHOR(S): 2Maximizing Photovoltaic Cell Efficiency: Experimental Investigation on theImpact of Cleaning Methods on Power Output and Cost-effectivenessRuzlaini Ghoni, Mohd Aizat Bin Mohd Sulaiman, Ammar Husaini Hussian,Fuaad@Fuaat Mohamed Nawawi, Ahmad Firdaus Zali, Ahmad FaridRidhwan ZakariaThe performance of photovoltaic panels is greatly influenced by various environmental factors, affectingpower output, conversion efficiency, and energy expenses. Dust, being a significant factor, plays a crucialrole in this regard, with its properties varying based on location and including kind, size, shape, andmeteorology. Soil erosion poses a considerable challenge to the widespread adoption of solar photovoltaicsystems. To address these concerns, a study conducted at Epic Solar Sdn Bhd in Kemaman, Terengganu,Malaysia, investigated the impact of soiling and different dust-cleaning strategies on photovoltaic panels.The research explored the daily and monthly energy production variations based on cleaning agents andfrequency. Three panels underwent monthly cleaning with or without a clean-ing agent, while anothergroup was cleaned weekly without any agents. The findings revealed that soiling is non-uniform over timeand influenced by daily weather changes, with the highest levels occurring from July to November andthe least in May and June. The study concluded that cleaning the panels every two weeks is necessary tominimize losses caused by improper cleaning practices and emphasized that cleaning frequency out-weighs the choice of cleaning chemicals in achieving effective dust removal.A b s t r a c t :PARALLEL SESSION 323

Page 24

Induction motors are commonly employed for converting electrical power to mechanical power because they areuncomplicated, durable, sturdy, energy-efficient, and appropriate for use in challenging surroundings. An inductionmotor emits a magnetic field of low magnitude that can be harvested. In most cases, researchers focus on capturingmagnetic energy from electrical transmission lines, power plants, and electrical track paths. Though limited studieshave been conducted on magnetic energy harvesting from manufacturing machinery, such as an induction motor, theelectromagnetic transducer developed employs the clamped current-transformer technique. However, no investigationhas been carried out on the extraction of magnetic energy from an induction motor utilizing a clampless current-transformer method, which has similar capabilities for capturing magnetic field energy from an induction motor. Theproposed project aims to create an electromagnetic transducer to harvest magnetic energy from an induction motorwithout a clamp by developing a magnetic field energy harvester. The main goal is to design and build a currenttransformer to capture magnetic energy from an induction motor. Radiation from magnetic fields is used as an energy-harvesting element to provide voltage performance from the induction motor rod core. The design contrasts stainlesssteel and carbon steel for the core material. It also consists of cylindrical and square rods for the core shape and 300 or500 turns for the number of turns. Further, this design provides more flexibility since the transducer does not need tobe clamped to the incoming supply to the induction motor; instead, it can be placed anywhere near the engine wherethe strongest magnetic field exists. The findings demonstrate that a material Carbon Steel long cylindrical core designis the higher output power with 52.95µW for 500 turns.Paper ID: TITLE: A b s t r a c t :AUTHOR(S): 7Harvesting Magnetic Energy from Induction Motors: Design and Development ofan Energy HarvesterAmmar Husaini bin Hussian, Ruzlaini Ghoni, Mohd Tarmizi Ibrahim, YoucefMahboub, Afidatul Nadia Mok HatPaper ID: TITLE: AUTHOR(S): 29Analysis of Photovoltaic Module Degradation: An Experimental Investigation onthe Correlation Between Partial Shading, Hot Spots, and EVA DiscolorationMd. Imamul Islam, Mohd Shawal Bin Jadin, Ahmed Al Mansur, MohdSalmizan Bin Mohd Zain, Mohammad Asif Ul HaqThe performance of solar PV modules is impacted by several environmental stressors, including highambient temperatures, inadequate sunlight, shade, dust, soiling, cell damage, etc. To guarantee long-termclean and sustainable energy generation, it is crucial to understand performance degradation and PVmodule reliability. In this current research, an experimental investigation on the degradation analysis ofthree 80-watt monocrystalline PV modules that have been exposed for approximately ten years in atropical environment in Malaysia as well as a correlation between partial shading, hot spots, and EVAdiscoloration has been conducted. The use of visual inspection, I-V curve measurement, thermal imaging,and degradation estimation have all been used to conduct an extensive study throughout this work. Theoutcome demonstrates the rate of degradation and hot spots phenomena of each module because of thepartial shading impact. Hot spots development is to blame for the EVA discoloration seen on Panel 2 andPanel 1 cells. The degradation rates of Panel 1, Panel 2, and Panel 3 are 0.496%, 1.264%, and 0.189% peryear, respectively. The study addressed the fact that dust and dried algae serve as mechanisms of PVdegradation, and hot spots caused by partial shade may have an impact on cell discoloration.A b s t r a c t :24

Page 25

There have been several significant advances in the field of wind power. Wind turbines have been used forover 2,000 years and are still vital to the 21st century's global energy industry. The importance of advancedwind turbine control methods for inefficient energy collection is growing. The Pitch Control-ler adjusts theWind Turbine (WT) blade using sensors. Adjusting a WT's pitch angle dampens structur-al modes andreduces blade root moment. These researchers could generate new ideas by combining the TRIZapproach of the systematic invention with data mined from patent databases. Pitch control tech-nologydevelopment may benefit from the structured creativity made possible by the TRIZ framework. TRIZ'snovel approach to analysing patent portfolios, based on three key factors, may be fruitful. The threemetrics evaluate the development of patented technology over the previous two decades and offerrecommendations for the sector's future. Research and development in the scientific and technologicalfields covered by patents are essential sources of information for predicting and planning future techno-logical breakthroughs in various areasPaper ID: TITLE: A b s t r a c t :AUTHOR(S): 52Control of Low-Wind Energy-Generating Turbines using TRIZ Methodology: APortfolio of Patented Technological InnovationsMuhammad Saqib Iqbal and Zulhasni Abdul RahiimPaper ID: TITLE: AUTHOR(S): 72Design of V-Shape Magnets Sandwich Flux-Switching Permanent MagnetMachines with Modular Rotor TopologyIrfan Ali Soomro, Mahyuzie Bin Jenal, Erwan Sulaiman, Md Zarafi Ahmad,Nur Afiqah Binti Mostaman and Norsuhada Binti Zainal AbidinConventional flux-switching permanent magnet brushless machines (PMFSM) gained a lot of attractiondue to their high torque densities, simple and robust rotor structure, and the permanent magnets andcoils on the stator. The Vshape magnets sandwich PMFSM machine has been proposed to improvethetorque density of the machine in which two PM pieces in V-shape are sandwiched in one stator pole toenhance the PMs usage efficiency. 2D finite element analysis (2DFEA) method is employed to compare theperformance of V-shape magnets sandwich PMFSM with salient rotor topology with that of V-shapemagnets sandwich PMFSM with modular rotor, in terms of flux linkage, flux distribution, induced backEMF, and cogging torque. From the results it is shown that the salient rotor V-shape magnets sandwichPMFSM and V shape magnets sandwich PMFSM with modular rotor produces 0.03Wb and 0.02Wbrespectively.A b s t r a c t :25

Page 26

This paper examines losses distribution of devices in various H-Bridge Single-Phase grid-connectedtransformerless inverter systems. Topologies of the transformerless inverter with an ac-bypass circuit anddc-bypass circuit are investigated and verified by simulation results. To achieve maximum efficiency, theselected power semiconductor (IGBT and Diode) plays an important part in the inverter system design.Therefore, the power losses in the devices are also studied and analyzed in this paper. The power lossesconsist of power conduction losses and power switching losses. The estimation of the powersemiconductor losses is verified by PSIM 9.0 software with the thermal module. This estimation is basedon the insulated gate bipolar transistors (IGBTs) and diode manufacturer datasheets, which are includedin the thermal module. The calculation of device losses has been discussed in this paper. The parametersgiven in the thermal module devices datasheet contribute to the level of power semiconductor losses invarious photovoltaic H-Bridge transformerless grid-connected inverters. In this paper, the HGTG20N60A4DFairchild Semiconductor and PS21A79 Powerex IGBT antiparallel diode de-vice have been simulated andanalyzed. In conclusion, with the good thermal module in PSIM 9.1 software, the distribution losses indevices can be manageable. Therefore, regarding losses in the devices, the high performance of variousinverter circuits can be designed. This thermal module is calculated accurately with the experimental sothe user can identify which device is recommended before hardware is builtPaper ID: TITLE: A b s t r a c t :AUTHOR(S): 85Distribution of Semiconductor Device Losses in Photovoltaic TransformerlessGrid Connected Inverter TopologiesMaaspaliza AzriPaper ID: TITLE: AUTHOR(S): 106TCSC Optimization for Loss Minimization in Power System Using ComputationalIntelligence TechniquesN. Balasubramaniam, I. Musirin, N. A. M. Kamari, A.A. IbrahimMinimizing power loss in transmission systems is crucial for achieving energy efficiency, loweringtemperature rise and less monetary losses leading to a sustainable power system network in a utility.Flexible AC Transmission Systems (FACTs) have been vastly adopted in minimizing power loss andenhancing voltage profiles in power transmission systems. However, the effectiveness of FACTs devices inachieving these benefits relies heavily on their optimal placement and sizing within the transmissionsystem. Suboptimal solutions on FACTs devices' location and sizing result in under-compensation or over-compensation, both of which are undesirable outcomes. Therefore, robust optimization techniques arenecessary to attain optimal solutions. This study applies evolutionary programming (EP) and artificialimmune system (AIS) as computational intelligence techniques to examine the effects of thyristor-controlled static compensators (TCSC) for loss reduction in power systems. Transmission real power loss isto be minimized, and the voltage profile of the transmission power system is to be improved. This studyshows that the installation of TCSC substantially minimizes power system loss. The IEEE 30-Bus ReliabilityTest System (RTS) is used to validate the proposed application and compensation scheme The applicationof evolutionary programming and artificial immune system techniques provides valuable insights andsolutions to power loss reduction ultimately improving the performance of transmission power systems. Itwas discovered that both techniques are comparable in minimizing the transmission loss in the system.A b s t r a c t :26

Page 27

Congestion in the power system can result from progressing load in the power system. This phenomenonmay cause system instability which leads to failure in power delivery to the consumer. Thus, congestionmanagement needs to be performed in power system operation and planning. This initiative will require arobust optimization technique so that power failure can be avoided. This paper presents IntegratedAccelerated Mutation Evolutionary Programming for Congestion Management in Power Systems. In thisstudy, a new optimization technique is introduced termed Integrated Accelerated Mutation EP (IAMEP). IAMEP is utilized to identify the optimal sizing and locations for distributed generation installation as anoption to manage the congestion in the power system. A pre-developed voltage stability index, FVSI isutilized as the indicator for congested lines. Validation on the IEEE 30-Bus RTS demonstrates that theproposed technique managed to reduce the congestion in the power system. A comparative study withEP also reflects its superiority in managing the congestion phenomenon. The result would be beneficial topower system operators and planners for their transmission system management.Paper ID: TITLE: A b s t r a c t :AUTHOR(S): 107Congestion Management for Voltage Security Control in Power SystemNur Arina Rabuan, Ismail Musirin, Norbaiti Sidik, Nor Azwan MohamedKamari, Norziana Aminudin, Dalina Johari, A.V. Senthil KumarPaper ID: TITLE: AUTHOR(S): 35Analysis of Mismatch Power Loss and Degradation in Aged PV Modules andArrays: Insights from an Indoor Experimental InvestigationMd. Imamul Islam, Ahmed Al Mansur, Mohd Shawal Bin Jadin, Md. HasanMaruf, ASM ShihavuddinOnce PV modules are subjected to extreme internal and external stress over an extended period, theirperformance suffers. The PV system arrays met such variables that resulted in mismatched power losses inthe array while producing ener-gy at a large scale. The purpose of this research is to comprehend howirradiance and ageing contribute to mismatch loss, which occurs in PV arrays. In this study, three arrayconfigurations—series, parallel, and series-parallel—are taken into consideration while evaluating theoutput performance degradation of PV mod-ules at two distinct irradiances, 1000 W/m2 and 800 W/m2.The results of the ex-periment show that the deterioration of module power is caused by both irradi-anceand age. At 800 W/m2, the MML% for the series, parallel, and series-parallel topologies are 44.73%, 38.01%,and 42.2%, respectively. At 1000 W/m2, the MML% for the same configurations falls to 30.9%, 22.51%, and27.9%. Fur-thermore, for both irradiances, the parallel setup displays a lower MML than the other twoconfigurations. Additionally, Mono 1 (0.344/year) is the most deterio-rating panel out of the four, accordingto the examination of the deterioration rate of 10-year-old modules. These results emphasize the need toaddress irradiance and ageing in solar energy systems by stressing their negative impacts on powerdegradation and MML in PV modules and arrays.A b s t r a c t :27

Page 28

Filtration membrane for artificial kidney has been developed and introduced to the medical line as areplacement for hemodialysis and hemofiltration treatment. In this work, three different filter designsmade of Silicon Nitride with varying membrane thickness were studied and the mechanical behavior offiltration membrane which is displacement and stress are analyzed using ANSYS. During the simulation, itwas observed that the highest deflection occurred at the central region of the membrane. The relationshipbetween membrane thickness and displacement can be observed from the results. Thinner membraneswere found to generate higher displacement, as evidenced by Design 3, where a membrane thickness of15nm resulted in a recorded maximum deflection of 1.6um. Conversely, an increase in membranethickness led to a decrease in deflection. It was also found that, Design 1, featuring the thinnest membraneof 15 nm, recorded the highest stress of 2.42 N/m². The simulated results obtained from the simulation offiltration membrane can serve as a valuable reference for the future fabrication of filtration membranes forartificial kidneys.Paper ID: TITLE: A b s t r a c t :AUTHOR(S): 87Design and Analysis of Filtration Membrane for Artificial KidneyShazlina Johari, Nurul Izzatul Ain Ahmad Kamar, Bibi Nadia Taib, andMohd Hafiz IsmailPaper ID: TITLE: AUTHOR(S): 25Free Fatty Acid Detection in Heated Palm Cooking Oil using an Open PathOptical MethodW.S. Salleh, S. Nurulain, M.A.S. Aspar, M. R. Salim, H. ManapPalm cooking oil is widely used worldwide due to its affordability and excellent oxidative stability duringfrying. However, continuous usage of cooking oil can lead to chemical deterioration, resulting in theformation of harmful compounds for human health, specifically the presence of free fatty acids. In thisstudy, an open-path optical technique combined with spectroscopy was employed to identify these freefatty acids. With the open-path optical approach, a beam of incident light passes through the palm fryingoil sample at a specific angle before being collected by a spectrometer. The intensity of the emitted lightwas then evaluated using the Spectrasuite program. The research findings revealed that each heatedpalm oil sample exhibited a unique absorbance spectrum. The detection of free fatty acids wassuccessfully achieved at two specific wavelengths, namely 347.50 nm and 364.18 nm. Notably, thedetection sensitivity for heating durations of 0 to 3 hours was found to be superior at the wavelength of364.18 nm, as reported in the study.A b s t r a c t :PARALLEL SESSION 428

Page 29

Actions or behaviors are typically categorized as being normal or abnormal. A technique used to find rareor unusual patterns, events, or data points that significantly differ from a dataset's expected or typicalbehaviour is known as abnormal activity detection, sometimes referred to as Anomaly detection or outlierdetection. Walking, running, and crossing oil pipelines are seen as regular actions in the context of thisstudy, whereas excavating, drilling, and sawing on the pipeline are regarded as anomalous action. Adataset designed exclusively for oil pipeline Vandalism detection (anomaly) is not publicly available, a newdataset was created and used to train and test the algorithm. The two main activity/action recognitionprocesses are typically feature extraction and classification. Feature extraction entails collecting significantfeature vectors from a dataset (videos) that can capture the features of the behaviours, whereasclassification looks for the presence or absence of specific patterns to classify a video dataset as normal orabnormal. For feature extraction, Google and Xception Fine Tuned Networks are utilized, and acomparison is done to determine the best Network. Evaluation metrics such as elapsed time and accuracyserved as the benchmark for comparison/selection. Experimental results reveal that Xception Netoutperforms Google Network. Utilizing the feature vectors obtained from the Xception Network, LongShort-Term Memory (LSTM) is used to categorize the activity. To reduce crimes like oil pipeline vandalism,which has a significant impact on people's health, safety, the environment, and the economy, it is essentialto have an automated anomaly detection system.Paper ID: TITLE: A b s t r a c t :AUTHOR(S): 65Vandalism Detection in Videos using Convolutional Feature Extractor and LSTMClassifierYau Alhaji Samaila, Patrick Sebastian, Aliyu Nuhu Shuaibu, SulaimanAdejo Muhammmad and Isiaka ShuaibuPaper ID: TITLE: AUTHOR(S): 69Hevea Brasiliensis-Based Lightweight Cement Brick: Development andMicrowave Absorbing PropertiesShafaq Mardhiyana Mohamat Kasim, Hasnain Abdullah, Nazirah MohamatKasim, Mohamad Nasir Taib, Ahmad Puad Ismail, Nur Shafikah Rosli, AliOthman, and Basharudin Abdul HadiServices in telecommunications, particularly wireless communications, are in high demand as a result ofthe expansion of information technology through commu-nication media. However, while this progress isgood for the populace, it also ex-poses people to high levels of radio frequency radiation especially fromthe telecommunications tower that located close to residential which could be harmful to all humanraces. This study was demonstrated the effectiveness of Hevea Brasili-ensis-based biomass as absorbingmaterial in cement brick to eliminate radio fre-quency or more specifically microwave radiation. Theabsorption performance of anti-microwave Hevea Brasiliensis-based Non-autoclaved Aerated Cement(NAAC) brick was compared to commercial cement bricks in the frequency range of 1 to 12 GHz using thefree-space arch reflectivity measurement method. Hevea Brasiliensis-based NAAC brick which uses carbonpowder sawdust has the highest absorption performance compared to commercial cement brick inwhich show that it has potential as an anti-microwave material that can reach maximum absorption peak– 52.56 dB while for the commercial brick was only reaches maximum absorption peak at -12.74 dB. Inoverall, Hevea Brasiliensis-based NAAC brick performed up to -10 dB of absorption while the absorption forcommercial brick were below -10 dB.A b s t r a c t :29

Page 30

Thresholding is a type of segmentation that involves dividing the pixels into separate groups based on their intensitylevel according to one or more threshold values. Thresholding is a popular image segmentation technique forconverting gray-level images to binary images. This paper will demonstrate a Multilevel Thresholding (MTH) for imagesegmentation based on Artificial Bee Colony (ABC) algorithm. This paper shows the designing process for an imagesegmentation module using multilevel thresholding. The multilevel thresholding algorithm divides pixels into discretezones, that segment the objects in the image, and it is the greatest solution for segmenting real-world pictures. Thegrey level histogram is used to determine the threshold point. It shows that each image has its own set of optimumthreshold values. From there, we can get the optimum threshold value for multilevel thresholding based on theArtificial Bee Colony (ABC) algorithm. To get the optimum threshold value, Otsu’s method has been used. After applyingOtsu’s Method, the algorithm is then analyzed by comparing the performance of multilevel thresholding using well-known benchmarks, Peak Signal-to-Noise Ratio (PSNR) and Structural Similarity Index (SSIM). The result shows that theABC algorithm performs almost the same as the other three methods, which are Harmony Search (HS), DifferentialEvolution (DE) and Particle Swarm Optimization (PSO). This can be seen from the results of PSNR and SSIM which showsmall margins of difference between the four algorithms. It can be concluded that the ABC algorithm is proven toperform well in image segmentation. From the collected data, there are significant changes in the PSNR and SSIMvalues due to the various multilevel thresholding techniques. As the level rises, so does the quality of the segmentedimage.Paper ID: TITLE: A b s t r a c t :AUTHOR(S): 118Enhancing Image Segmentation: Multilevel Thresholding using Artificial BeeColony AlgorithmNor Farizan Binti Zakaria , Muhammad Nazmi bin Mohamad Rosly, MohdHerwan Sulaiman, Rohana Abul Karim and Nurul Wahidah ArshadPaper ID: TITLE: AUTHOR(S): 71Resting state EEG for Personality Traits ClassificationUmay Kulsoom, Dr. M. Naufal B.M. Saad, Dr. Syed Saad Azher AliPersonality classification plays a vital role in providing a deeper understanding of human behavior andindividual differences. The pandemic has brought about significant changes and challenges that haveunderscored the importance of personality classification. Personality classification facilitatesunderstanding individuals’ cognitive and emotional responses to stressors, and their potentialimplications for mental health, tailored support, workforce adaptation, communication, and public healthmessaging. Power spectral features extracted from EEG and self-reported assessments (NEO-FFI scores)are used as inputs to the Support Vector Machine (SVM) classifier to determine the feasibility of classifyingpersonality from resting state EEG.A b s t r a c t :30

Page 31

Internet of Things (IoT)-based sensors have transformed the measurement of water quality parameterssuch as a pH, dissolved oxygen (DO), electrical conductivity (EC), total dissolved solids (TDS), ortemperature sensor. A recent study using Saccharomyces boulardii as a model microorganism at theUniversiti Malaysia Pahang (UMP) biotechnology facility demonstrates this. After an hour of bacterialinoculation, the DO sensor was able to detect the presence of bacteria in the water, with the maximal DOreading increasing from 4 ppm to 8 ppm. However, after five hours, the DO level returned to normal due tothe yeast's acclimation process. When the concentration of Saccharomyces boulardii was increased from50 to 100 percent, the study found that the DO readings fluctuated. The 50% sample required up to 4hours to normalize from 7.824 mg/l to 4.804 mg/l, whereas the 100% sample required 6 hours to normalizefrom 6.288 mg/l to 4.36 mg/l. This was due to the yeast depleting more oxygen from the culture medium,which harmed the yeast in the long run. The study concluded that IoT-based sensors can accuratelymeasure water quality parameters and detect the presence of microbes in water. The study also revealedthat the concentration of Saccharomyces boulardii can impact the DO reading, with higherconcentrations taking longer time to normalize. This emphasizes the significance of employing IoT-basedsensors to accurately measure water quality parameters.Paper ID: TITLE: A b s t r a c t :AUTHOR(S): 80IoT-Enabled Water Quality Sensor: Detecting Concentration of SaccharomycesBoulardii Bacteria to Enhance Water SafetyMuhammad Aqil Hafizzan Nordin, Mohd Faizal Jamlos and AbdelmoneimA. BakhitPaper ID: TITLE: AUTHOR(S): 83Deep Learning-Based Yield Prediction for the Die Bonding SemiconductorManufacturing ProcessMuhammad Ali Akbar, Ahmad Jazlan, Azhar Mohd Ibrahim and ArfahAhmadIn the semiconductor manufacturing industry, consistently achieving a high yield is the primary target tomeet customer demands and ensure continuous profitability. The ability to predict the yield of a particularmanufacturing process at either the Front of Line and End of Line facilities is therefore essential in order toanalyze Return of Investments (ROI), predictive maintenance and condition monitoring. Howeverachieving high quality predictions with good accuracy is challenging due to the various uncertainties inthe manufacturing process such as unexpected machine downtime and stoppage for maintenance. In thispaper we propose a method using Deep Learning Long Short Term Memory Recurrent Neural Networks(LSTM-RNN) to perform day ahead forecasting of the yield from the Die Bond process at a particularsemiconductor manufacturing facility. The method was implemented using MATLAB software, and theresults demonstrate that the proposed approach achieves accurate yield forecasts with less than 8% error.Further improvements can be made by utilizing hourly data instead of daily dataA b s t r a c t :31

Page 32

This paper applies a readout circuit to improve measurement output when reading the grouped resistivevalue in a matrix array format. The circuit is designed to address the main challenge faced in utilizingresistive pressure sensor arrays for foot plantar applications. The proposed approach, called the NodalArray Approach (NAA), modifies the Wheatstone Bridge Circuit using nodal analysis technique andKirchhoff's Current Law. By solving simulta-neous equations derived from the voltage readings of thereadout circuit, the NAA accurately calculates the resistance values of the sensors. The readout circuitconnection is of low complexity, utilizing resistive elements as the major components of the readingtechnique with only three iterations in-volved for each voltage node. Hardware results demonstrate thatthe NAA achieves high accuracy in obtaining a sensor's resistance value, while adher-ing to severallimitations to avoid miscalculation (with an average calcula-tion error of less than 5%).Paper ID: TITLE: A b s t r a c t :AUTHOR(S): 27Development of Readout Circuit using Nodal Array Approach (NAA) for FiveResistive Sensors ArrangementFairuz Rizal Bin Mohamad Rashidi, Muhammad Farid Akmal BinFahrurrazi and Airul Sharizli Bin AbdullahPaper ID: TITLE: AUTHOR(S): 110Investigation of Sensor Probe Performance on Soil Moisture ConditioningMonitoring SystemNik Adham Faris, Muhammad Arif Osman, Aisyah Illani Sulaiman, Roshahliza M. Ramli and Nadzirah Mohd MokhtarThe Internet of Things (IoT) is continuously advancing with improved inter-net technology, providing newopportunities for innovation such as in agri-culture that can benefit from IoT devices. This paper focuseson investigat-ing the performance of a sensor probe in an environmental conditioning monitoring system.The project aims to develop an IoT-based moisture sen-sor capable of collecting and storing data on thecloud using a cloud server. This study explores optimal conditions for accurate data collection and ana-lyzes the sensor's performance in different soil conditions, including dry, waterlogged, and normal. Awireless sensor network integrated with the moisture sensor is designed and implemented for real-timedata collection through the IoT infrastructure. Then, the acquired information is stored on a cloudplatform. By varying soil conditions, the sensor's performance is thor-oughly examined, providing insightsinto its accuracy and reliability. This knowledge is valuable for effective strategies in agriculture and waterre-source management, improving decision-making and resource allocation. The research contributes tothe advancement of environmental conditioning monitoring systems, informing the development ofenhanced sensor technol-ogies, and promoting more efficient and sustainable agricultural practices.A b s t r a c t :32

Page 33

The extraction of the region of interest (ROI) in hand vein images plays a crucial role in their detection.Accurately extracting the vein area faces chal-lenges such as variations in hand poses, lighting conditions,orientation, ap-pearance, and noisy background. Although numerous techniques have been proposed forROI extraction of hand veins, their capabilities are often limited to a specific hand pose and location. Toaddresses this limitation, this paper presents a deep learning- approach using Faster R-CNN for adaptiveROI vein extraction. The proposed system is evaluated using two hand vein data-bases: self-acquisitionand SUAS, encompassing diverse hand poses. To as-sess the performance of the proposed technique, it iscompared to two exist-ing ROI vein extraction techniques. The comparative results demonstrate that theproposed technique achieves impressive performance in accurately locating the ROIs for various handposes and locations. By employing a deep learning-based approach and evaluating its effectiveness ondifferent hand vein databases, this appear offers a promising solution for adaptive ROI vein extraction,overcoming the limitations of existing techniques.Paper ID: TITLE: A b s t r a c t :AUTHOR(S): 88Hand Vein ROI Extraction using Faster R-CNNMarlina Yakno, Junita Mohamad Saleh, Mohd Zamri Ibrahim, SyamimiMardiah Shaharum, Rohana Abdul Karim33

Page 34

In order to predict the blood glucose values of diabetic patients, this study uses two AutoregressiveIntegrated Moving Average Model (ARIMA) models—the self optimized ARIMA model and the ARIMAmodel based on Bayesian optimization to analyze historical data from the continuous blood glucosemonitoring system (CGM) and the equipment calibration value as the training data set to foretell apatient's blood sugar level in the future in order to prevent hypo- and hyperglycemic episodes. CGM datafrom 8 patients obtained by Suzhou Municipal Hospital in Jiangsu Province, China, was used to validatethe data in this paper on the two models. Obtain and compare the minimum Mean Square Error (MSE)values of the prediction results of the two models at 15 minutes, 30 minutes, and 45 minutes. In order toimprove the accuracy of the patient's blood glucose level prediction, the model using the smallest MSEvalue among the two ARIMA models was used as a method for predicting the patient's blood glucose levelprediction.Paper ID: TITLE: A b s t r a c t :AUTHOR(S): 77The modified ARIMA predicting algorithm apply on Glucose Values Prediction.Bian QingXiang, Azizan As’array, Cong XiangGu, Khairil Anas bin Md Rezali,and Raja Mohd Kamil bin Raja AhmadPaper ID: TITLE: AUTHOR(S): 124Experimental Evaluation of Sensor Readings in Invasive Electrical CapacitanceTomography for Conducting Pipe Applications – Initial StudyHaziq Syakir Hamzah, Yasmin Abdul Wahab, Ain Eazriena Che Man,Nurhafizah Abu Talip @ Yusof, and Mohd Mawardi SaariElectrical Capacitance Tomography (ECT) is a promising imaging technique for monitoring gas-liquidflows in industrial pipelines. In this study, we de-veloped an ECT system with eight electrodes to detectthe presence of static gas inside oil, specifically for oil-gas regime. We optimized the sensor design using aDesign of Experiment (DOE) method and developed a signal genera-tor and signal conditioning system tocollect data offline. Our experimental approach involved the use of an invasive approach with theplacement of sensors within the pipe to improve the accuracy of the measurements. We conducted aseries of experiments with different oil-gas regimes to evaluate the performance of the ECT system. Theresults demonstrate that the inva-sive approach significantly enhances the sensor readings, leading to im-proved imaging accuracy and reliability for detecting static gas inside oil. This study contributes to theunderstanding of the sensor reading perfor-mance in invasive ECT for oil-gas regime and provides valuableinsights for optimizing the measurement process. The findings have implications for the development ofmore robust and accurate ECT systems in industrial settings.A b s t r a c t :PARALLEL SESSION 534

Page 35

Condition based monitoring (CBM) has emerged as a promising technique for assessing the condition andperformance of various mechanical systems, including lift motors. This paper focuses on the application ofvibration signal analysis for SHM of lift motors, enabling the early detection of potential faults andensuring reliable and safe operation. Lift motors are subjected to various mechanical stresses, vibrations,and operational loads during their service life. These factors can lead to wear, fatigue, and other structuralis-sues that may compromise the motor's performance and reliability. Vibration analysis has proven to bean effective non-intrusive method for monitoring the dynamic behavior of lift motors and detecting earlysigns of degradation. The acquired vibration signals are processed and analyzed using various techniques,such as Fourier analysis, time-frequency analysis and Auto-regressive Integrated Moving Average (ARIMA)model. The ARIMA model is capable of capturing the underlying patterns and trends in time series data,making it suitable for forecasting and anomaly detection in vibration signals. This paper also compares themeasured vibration signatures with reference baselines or established thresholds for possible early faultdetection and diagnosis.Paper ID: TITLE: A b s t r a c t :AUTHOR(S): 76Predictive Analysis of a Lift Motor using Autoregressive Integrated MovingAverage (ARIMA) Model for Vibration-based Condition MonitoringSharafiz Rahim, Adnan Dehghani, Khairil Anas Md Rezali, Abdul Murad,Siti Nor Azila Khalid, Azahar bin Mohd, Zamil Hisyam, Mohamad Fikri BinMohamad YunusPaper ID: TITLE: AUTHOR(S): 40Application of Granitic Residual Soil and Palm Oil Fuel Ash as an AbsorbingMaterial to Develop High-Strength Anti-Microwave Brick WallsNur Hashira Narudin, Hasnain Abdullah, Khairunnisa Ab Razak, MohdNasir Taib, Basharudin Abdul HadiElectromagnetic pollution from modern technologies has resulted in electromag-netic radiation. Wirelesscommunication, power transmission, and communica-tions gadgets utilized in our daily lives, such as mobile phones,tablets, and laptop computers, expose people to electromagnetic pollution. The pervasiveness of these technologieshas raised concerns regarding the safety of human radio fre-quency radiation exposure. The effectiveness of radiation-absorbing materials (RAM) in absorbing microwave energy has led to an increase in attention in re-cent years.Substantial research has been conducted to produce new radiation-absorbing materials with great absorbingperformance. In this research, anti-microwave brick walls were developed. As we know, brick is a fundamental buildingunit and the most often used construction material. However, the exist-ing brick is incapable of absorbing microwaveenergy. This study aims to exam-ine the absorption performance of solid anti-microwave brick walls with various rawmaterial composition ratios incorporating granitic residual soil and Palm Oil Fuel Ash (POFA) as absorbing materials.Agricultural waste is high in lignin content, with about half of the components containing carbon which is a goodelectromagnetic wave absorbent. The absorption performance of the anti-microwave bricks was tested using the NavalResearch Laboratory (NRL) Arch free-space method, and the findings were studied over a frequency range of 1 to 12GHz.According to the results of the investigation, solid anti-microwave bricks containing 15% landslide and 35% POFAachieved the best absorption perfor-mance, with a maximum absorption of -39.19dB with great compressive strength,13.615MPa.A b s t r a c t :35

Page 36

Giant Magnetoresistance (GMR) sensors have become increasingly popular for Non-Destructive Testing(NDT) applications, especially in eddy current inspec-tion. GMR sensors have high sensitivity and low noisecharacteristics, and they have been used for detecting surface cracks in conductive materials and monitor-ing the structural health of steel bridge cables. According to the GMR sensor ap-plication, it also can beused for detecting corrosion on pipelines compared to conventional eddy current systems that are onlyapplicable to measure the coating thickness and depth of defect on pipeline. In this work, the designing ofa GMR sensor probe by using coil as the excitation signal transmitter has been done for measuring thedepth of defect based on a calibration block on two types of mate-rial within brass and mild steelmaterial. This system will measure the lower depth is 0.5mm and the higher depth 5mm. The turn of coilfor the excitation sig-nal transmitter is set on 600turns where the diameter of the coil is 0.17mm. Based onthe Brass steel calibration block, a sudden increase occurs at a defect depth of 5mm which is as much as2.443Vp compared to the previous one which is 2.427Vp (4.5mm) and 2.421Vp (4mm). In addition, a suddenincrease also oc-curred for mild steel, which is in the defect depth of 3.5mm, which is 2.458Vp comparedto the previous 2.441Vp (3mm) and 2.431Vp (2.5mm). Finally, the re-sult shows that when the depth ofdefect is increased then the value of signal Voltage peak (Vp) also increases.Paper ID: TITLE: A b s t r a c t :AUTHOR(S): 13Designing GMR Sensor Probe using Transmitter Coil for Artificial CrackIdentification on Brass and Mild Steel Calibration BlockKharudin Bin Ali, Nazry Abdul Rahman, Damhuji Bin Rifai, Zulfikri Salleh.Ahmad Anwar Zikri OthmanPaper ID: TITLE: AUTHOR(S): 44Performance evaluation of Smoothed Functional Algorithm based methods forSigmoid-PID Control optimization in MIMO Twin-Rotor SystemsMok, RenHao; Ahmad, Mohd AshrafThis paper explores the tuning of the Sigmoid Proportional-Integral-Derivative (SPID) controller usingvariations of the Smoothed Functional Algorithm (SFA) for the underactuated Multiple-Input Multiple-Output (MIMO) twin-rotor system. The SPID controller, incorporating a sigmoid function, extends theapplicability of traditional PID controllers to complex, non-linear systems. However, SPID tuning presentschallenges due to the added control parameters and the inherent non-linearity of the sigmoid function. Toeffectively tune the SPID con-troller, SFA is recommended, which stochastically optimizes the parameterspace without requiring an explicit mathematical model. However, the standard SFA suffers from unstableconvergence issues, necessitating modified approaches such as the Norm-Limited SFA (NLSFA) andMemory-Based SFA (MSFA). NLSFA constrains gradient approximation within boundaries, preventingexcessively large approximations that lead to divergence but at the cost of an additional optimizationparameter. The MSFA introduces a memory function to consider optimal solutions from previousiterations, promoting continuous convergence. The effectiveness of these SFA variations in optimizingSPID controllers for a MIMO twin-rotor system is compared, offering insights into the control andoptimization of complex non-linear systems.A b s t r a c t :36

Page 37

Crowd collectiveness paradigm is a promising field for a practical monitoring and controlling the mass ofpilgrims during the Hajj period. This technique describes the degree of the individuals union on acollective motion. In this paper, a coherent motions merging is applied to detect and investigate thecollectiveness of the pilgrims in three states of lane formation during the Hajj; laminar, stop-and-go andturbulent flow. The experiment has been conducted on simulated agents-series dataset of Hajj crowd.Experimental results show that the collectiveness degree has changed and it has been affected by thecoherent motion nature of the crowd based on these three states.Paper ID: TITLE: A b s t r a c t :AUTHOR(S): 63Monitoring Case Study for Motion Crowd State Transitions in MinaSami A. M. Saleh, A Halim Bin Kadarman, Waheed Ghanem, Sanaa Ghaleb,Antar Abdul-Qawy, Zainal Abidin Arsat, Solehuddin ShuibPaper ID: TITLE: AUTHOR(S): 75Pick and Place Robot Arm Using PLC Modeling of LLD and PNZulfakar Aspar, Nurul Huda Abd Rahman and Mohd Zhafri bin BaharudinRobot Arm is extensively being used in the industry for various applications. With advances inmicrocontrollers and various other digital controllers, this re-search is going to investigate if it is stillfeasible to develop a robot arm applica-tion by using a Programmable Logic Controller (PLC). Since PLCmodeling has become more complex, the project is limited to the functions of the robot arm for pick andplace activities which are limited to two axes. A flowchart is good for controller planning while the actualimplementation is done by using a Ladder Logic Diagram (LLD). This project is going to discuss in detailthe development of the robot structure, pneumatic control and PLC modeling using LLD. This project isalso going to compare the robot arm with another pick and place robot arm which has similar functions.The pick and place robot arm was verified by running a simple Printed Circuit Board (PCB) as a load in anactual operation. For PLC modeling improvement, Petri Net was used to develop the PLC model followedby automatically generate the equivalent LLD model. By modeling at a higher level of abstraction, it iseasier to develop, improve and maintain the LLD model. In the future, the pick and place robot arm canalso be modified to fulfill a specific mission by modifying the PLC model.A b s t r a c t :37

Page 38

Corrosion is one of the most common problems associated with steel structures. The occurrence ofcorrosion may lead to metal loss, at which point might threaten the integrity of a steel structure.Therefore, the employment of magnetic flux leak-age (MFL) and eddy current testing (ECT) is beneficial inproviding detection of metal loss due to corrosion. Thus, a differential magnetic probe using both meth-ods is developed. The probe consists of two fluxgate sensors and an excitation coil. Then, a line scanmeasurement is conducted on a 6-mm mild steel sample with metal loss defects. From the result of theline scan measurement of the MFL signals, the presence and depth of the defects could be identified.Meanwhile, on-ly the defect presence can be identified from the ECT signals, although only re-stricted tohigher frequencies detectionPaper ID: TITLE: A b s t r a c t :AUTHOR(S): 23Development of a Differential Magnetic Probe to Evaluate Metal Loss due toCorrosionMohd Aufa Hadi Putera Zaini, Mohd Mawardi Saari, Ummi Sabihah MohdYusdi, Nurul A’in Nadzri and Zulkifly AzizPaper ID: TITLE: AUTHOR(S): 20Magnetic Field Sensor Based on Corroded Multi-mode Fiber StructureNurainie Husin, Chew Sue Ping, Latifah Sarah Bt. Supian, Anis ShahidaNiza Bt. MokhtarMagnetic field detection is an important area in magnetic field sensing as it has various usages such asmeasuring earth’s magnetic field strength, detecting mag-netic anomalies of different characteristics aswell as detecting submarines for the military. This paper proposed magnetic field optical sensor based onmagnet-ic fluid (MF) using the singlemode-multimode-singlemode (SMS) structure. The SMS structurewas corroded with hydrochloric acid (HCl) to enhance its sensitiv-ity in detecting magnetic field strength.The short segment of MMF in the SMS sensor acts as sensing element and has occurrence of multimodeinterference (MMI). The sensor was proven can detect magnetic field strength from 1.5 mT to 7.4 mT.. Theachieved magnetic field strength sensitivities are -795.2 pm/mT and -327.9 pm/mT, corresponding to thelength of MMF for 39mm and 45mm, respec-tively as the sensors are corroded for 900s. The linear fittingcoefficient achieved for 39mm and 45mm of MMF are 0.8884 and 0.9814.A b s t r a c t :38

Page 39

This study aims to develop an advanced categorization mechanism using e-nose and optical sensortechnologies to effectively evaluate the quality of palm oil. The objective is to enhance the precision andeffectiveness of palm oil classification by integrating odor and color concentration assessment. Byutilizing an optical sensor that relies on color concentration, the primary focus is on categorizing thequality of palm oil based on its chromatic characteristics. This approach ensures an impartial and effectiveevaluation of the oil. Additionally, the study evaluates palm oil quality by employing e-nose and opticaltechniques to consider both aroma and chromatic properties. The electronic nose captures the olfactoryprofile, while the optical sensor measures chromatic concentration. By combining these measurements, acomprehensive assessment of palm oil quality is achieved, encompassing sensory and visual attributes.This research significantly contributes to improving quality control and assurance procedures in the palmoil sector through the development of an intelligent classification system that integrates e-nose andoptical sensor data. The proposed system offers an enhanced and unbiased method for grading palm oil,promoting sustainable production techniques, and meeting the increasing demand for premium palm oilcommodities.Paper ID: TITLE: A b s t r a c t :AUTHOR(S): 115Classification of Palm Oil Quality using Case-Based Reasoning Base on Odor andOptical DataMujahid bin Mohamad, Muhammad Sharfi bin Najib, Razali bin Muda,Saiful NIzam bin Tajuddin and Mohammad Fakhireen Aminudin39

Page 40

. Battery lifetime and performance are critical concerns for electric vehicles (EV’s)and energy storagesystems (ESS). However, factors such as environmental variations and manufacturing defects often resultin charge imbalance among battery cells, leading to reduced energy capacity and power performance ofthe entire battery pack. To address this issue, cell equalization becomes necessary. In recent years, theadvancements in deep reinforcement learning (DRL) have made it a viable tool for battery managementsystems in electric vehicles. In this research paper, a new method is introduced for active cell balancing ofa battery pack consisting of four series connected lithium-ion batteries. The approach utilizes deepreinforcement learning (DRL) within a MATLAB simulation. A deep Q-learning (DQL) algorithm is used forthe training of DRL agent, and a dc-dc Zeta converter is used to transfer the excess charge from theovercharged cell to the under-charged cell in the battery pack. Additionally, the proposed equalizationtopology focuses on selecting the specific cells that requires balancing, aim-ing to enhance the speed ofthe equalization process. The simulation results demonstrated that SoC convergence among four Li-ioncells (with an SoC difference of as little as 0.5 %) occurs within 500 seconds using the pro-posed novelintegration.Paper ID: TITLE: A b s t r a c t :AUTHOR(S): 62A Novel Active Cell Balancing Approach Based on Reinforcement Learning forSoC Balancing Of Four Lithium-ion Battery Cells.Neha khan and shreasthPaper ID: TITLE: AUTHOR(S): 8Performance Comparison of Conventional and V-Shape Magnets SandwichFlux-Switching Permanent Magnet Machines with Modular Rotor TopologyIrfan Ali Soomro, Mahyuzie Bin Jenal, Erwan Sulaiman, Md Zarafi Ahmad, ,and Nur Afiqah Binti MostamanConventional flux-switching permanent magnet brushless machines (PMFSM) gained a lot of attractiondue to their high torque densities, simple and robust ro-tor structure, and the permanent magnets andcoils on the stator. The sandwich PMFSM and V-shaped magnets PMFSM machine has been proposed toim-prove the torque density of the machine in which two PM pieces are sandwiched in one stator pole toenhance the PMs usage efficiency. 2D finite element analysis (2DFEA) method is employed to compare theperformance of conventional and V-shape magnets sandwich PMFSM with modular rotor topology, interms of flux linkage, flux distribution, induced back EMF, cogging torque and average torque. From theresults it is shown that both motors have almost produced same flux linkage and Sandwich PMFSMgenerates slightly high torque than V-shape magnets sandwich PMFSM with modular rotor.A b s t r a c t :PARALLEL SESSION 640

Page 41

This paper presents a study on the impact of Distributed Generation (DG) types and penetration levels onloss control in transmission systems. To achieve this, an Artificial Immune System (AIS) algorithm isutilized in conjunction with MATLAB simulations on the IEEE 30-Bus Reliability Test System (RTS).Penetration level is defined as the number of DGs installed within the system, and four types of DGs areconsidered based on their real and reactive power delivery/consumption characteristics. The simulationresults are divided into two parts: the first part focuses on randomizing DG location and sizing to obtainthe minimum total system loss, while the second part investigates the effect of DG penetration level ontotal system loss reduction with fixed DG capacity. The study demonstrates that installing four Type-1 DGsleads to the highest total system loss reduction. Moreover, higher DG penetration levels result in greatertotal system loss reduction with DGs of the same capacity. These findings offer insights for optimizing DGdeployment strategies in transmission systems, thus enhancing their efficiency and reliability.Paper ID: TITLE: A b s t r a c t :AUTHOR(S): 9Assessing the Effects of DG Types and Penetration Levels on TransmissionSystem Loss Control through Computational Intelligence-Based TechniqueMohd Helmi MansorPaper ID: TITLE: AUTHOR(S): 11Optimal Placement of Renewable Distributed Generation and Capacitor Bank toMinimize losses with Particle Swarm OptimizationSN Syed Nasir, T Sohail, JJ Jamian and R AyopPower loss is an important aspect of the power system that must be kept as low as possible. The optimalsiting and sizing of distributed generation (DG) and shunt capacitor at distribution networks for thepurpose of minimizing real power loss is attracting a lot of attention from electric power utilities thesedays. DG is expected to play an important role in the power system's residential, commercial, andindustrial sectors. Traditional electricity sources can be replaced with DG, which can also be used toimprove the current electrical system. Capacitor bank and DG combined will improve systemperformance even further. This paper presents a method based on an analytical approach for optimalallocation (sizing and siting) of DG and capacitor bank to reduce overall real power losses in thedistribution network subject to equality and inequality constraints. It is common practice to reduce powerlosses and im-prove the voltage profile of a distribution system by placing DG and capacitors in the bestpossible locations. For this research, the IEEE-33 and IEEE-69 bus systems are used. A backward-forwardsweep load flow analysis will be performed using MATLAB software to investigate power losses and volt-age magnitude. Particle swarm optimization was used to optimize the placement of the DG andcapacitors in this project to achieve the lowest possible power losses. The research demonstrates that theproposed method effectively reduces real power losses in the distribution network. The optimizedallocation of DG and capacitors resulted in improved voltage profiles and minimized power losses.A b s t r a c t :41

Page 42

This paper presents a performance analysis of Genetic Algorithm (GA) and Par-ticle Swarm Optimization(PSO) based Proportional-Integral (PI) controllers for regulating the output voltage of a Cascaded H-Bridge Multilevel Inverter (CHMI) during load variations. Proportional-Integral (PI) controller is the mostcommon controller used to solve this issue due to its simple structure, high stabil-ity, minimal steady stateerror and ease of implementation. However, the perfor-mance of this controller is sensitive to parametervariation and has limited per-formance for nonlinear systems such as CHMI. The study aims to evaluateand compare the effectiveness of GA and PSO in achieving voltage regulation in the CHMI system usingMATLAB Simulink. GA and PSO are applied to optimize the parameters of the PI controller to maintain thedesired output voltage despite load variations. The simulation experiments are conducted under loadvariations to assess the performance of the controllers. The simulation results demonstrate that both GAand PSO based PI controllers effectively regulate the output voltage of the CHMI system during loadvariations. However, a comprehensive analysis reveals that the PSO algorithm outperforms the GAalgorithm in terms of voltage regulation accuracy and response time.Paper ID: TITLE: A b s t r a c t :AUTHOR(S): 99Performance Analysis of GA and PSO Based PI Controller for Cascaded H-bridgeMultilevel Inverter Output Voltage RegulationNur Huda Binti Ramlan, Habri bin Marzuki, Suliana Ab. Ghani, NizaruddinM. Nasir, Muhammad Afiq Iqmal NorazmanPaper ID: TITLE: AUTHOR(S): 68The Impact of Cleaning Bird Drops for PV Power Increment: An ExperimentalStudy in Dhaka, BangladeshAhmed Al Mansur, Sabbir Hasan Tohid, Md. Mostafizur Rahman, Md.Shahin Alom, Md. Sabbir Alam, Shaquar Islam Leyon, Chowdhury ShajjadHaider, Md. Imamul Islam, Mohd Shawal Jadin, and Ratil H AshiqueNon-uniform shading causes significant Power reduction in rooftop Photovoltaic (PV) systems. Bird dropsare one of the key factors which cause non-uniform shading on the PV module. Regular cleaning of dustyPV modules can enhance the output power significantly. In this work, a water-based cleaning method isapplied on a 2×2, 40W PV array with bird-dropping conditions to investigate the output powerenhancement. The experimental test is done for seven different cases of the interconnection of the PVmodules for both clean and unclean conditions. The experimental results show that the output power isincreased significantly after cleaning the bird drops. The maximum output power, 35.2 W is achieved forcase 6, while the percentage of power enhancement (PPE) is made at 6.21%. The highest PPE is achievedat 16.69% for case 7.A b s t r a c t :42

Page 43

The sustainability of power systems is a vital need for modern societies. The increasing frequency of powershutdowns triggered by severe weather events, which are worsened by the effects of climate change, hasintensified research efforts aimed at enhancing the resilience of power systems. Reme-dial action needs tobe planned for improving the power system’s resilience. The installation of distributed generation (DG) isone of the suitable efforts to alleviate this phenomenon. This paper presents enhancing power systemresilience through evolutionary programming for high-impact low probability (HILP) events. Validation onIEEE 30-Bus Reliability Test System (RTS), solved us-ing Evolutionary Programming (EP) under extremeweather demonstrates its capability in improving the power system resilience. In this study, the EPtechnique is used to find the optimal location and sizing of DG for the pur-pose of improving the powersystem’s resiliency in the case of HILP events. The results demonstrate that this algorithm effectivelyquantifies the system’s resilience under extreme weather events. The results could be beneficial to powersystem operators and planners.Paper ID: TITLE: A b s t r a c t :AUTHOR(S): 97Enhancing Power System Resilience through Evolutionary Programming forHigh Impact Low probability EventsFathiah Zakaria, Ismail Musirin, Nor Azwan Mohamed Kamari, NorzianaAminuddin, Dalina Johari, Sharifah Azwa Shaaya, Abdullah Akram Bajwa,A. V. Senthil KumarPaper ID: TITLE: AUTHOR(S): 56Development of Flood Early Warning System by Monitoring Pagoh River WaterLevel and Rainfall DistributionAtiqah Amiera Kamarudin, Muhammad Rusydi Muhammad Razif, OmarAbu Hassan, Muslim Abdullah Zaik and Nurul Hasyimah Mohd MustaphaFloods occur on a yearly basis in Peninsular Malaysia, mainly during the monsoon seasons, which begin inOctober for the second inter monsoon and end in November and December for the early Northeastmonsoon. This project is being carried out to provide locals living near the Pagoh River with an earlywarning system. The goal of this project is to monitor the river’s water level, measure the intensity ofrainfall around the river, and collect live stream footage of the river before, during, and after the flood tohelp citizens keep informed about the river’s current status. Flooding frequently causes both physical andfinancial damage. By utilizing a centralized Internet of Things (IoT) application, Blynk, the output fromthese three components can give actual results regarding the current state of river water levels and thedanger of impending floods. This project will be able to wirelessly alert individuals about the river’s waterlevel and the current status in the river near them.A b s t r a c t :43

Page 44

This paper presents voltage control in a DC-DC buck converter for the output tracking problem of DC-DCconverters with varying in input voltage. Switching dc-dc converter systems are non-linear and time-varying in nature. Therefore, achieve stable output voltage, a super twisting sliding mode controlleralgorithm (STWA) is proposed. The STWA is proposed to assure the sliding surface con-verge in finite timewith reduction in chattering effect as discontinuous of conven-tional SMC causes high frequencyoscillation in control input and sliding surface. The performance of the proposed controller is compared toconventional SMC in order to see the effectiveness of the controller in reducing chattering effects andimprove steady-state error. The simulation results have shown that STWA able to improve chatteringeffect improve the transient performance and reduce stead-state errorPaper ID: TITLE: A b s t r a c t :AUTHOR(S): 128DC-DC Buck Converter Control using Super Twisting Sliding Mode ControlMaziyah Mat-Noh, Ng Yu Jun, M.S BakarPaper ID: TITLE: AUTHOR(S): 47Virtual Power Plant Management Using PID ControllerAisyah Bukhari, Siti Hajar Yusoff, Muhammad Sharir Fathullah MohdYunus, Siti Nadiah Mohd Sapihie, Nur Syazana Izzati RazaliVirtual Power Plant (VPP) is a reliable system for energy production and the at-tractiveness comes fromthe fact that VPP can control the energy production in order to fulfill the demand of consumers. However,the delay in communication and the varying output of the Distributed Energy Resources (DER) can causein-stability to the system and the VPP will then operate in a suboptimal condition. Therefore, this projectaims to design a VPP system that uses a PID controller to achieve stable signal that can deliver the powerefficiently with minimal delay for a solar and wind farm with load. From observation, the tuning of the PIDcon-troller's Kp, Ki and Kd results in a signal with reduce rise time and overshoot, eliminate steady stateerror, improve transient response and increase stability of the system. The simulation observes theperformance of PID with reference volt-age of 25V and 30V. The output voltage is constant throughout thesimulation proving the stability of the circuit while eliminating errors.A b s t r a c t :44

Page 45

The purpose of optimal coordinating Overcurrent Relays (OCR) is to establish a precise time interval,known as the coordinated time interval, CTI between the primary and backup relays' operations. In orderto reduce all primary relays' tripping time and ensure power system dependability the value of TDS and PSmust be optimized. A nature-inspired meta-heuristic algorithm called Barnacles Mating Optimizer (BMO)is implemented to solve this optimal OCR coordination problem. BMO is tested to the IEEE 8-bus testsystem with normal inverse characteristic curve (IDMT) as in the IEC guideline. The findings are thencompared to the other well-known optimization algorithms, such as GA and PSO to evaluate the BMO'sfeasibility and efficiency performance of the method. The study showed that the proposed methodimproves OCR coordination’s OF in an IEEE-8 bus system better than other selected methods testedPaper ID: TITLE: A b s t r a c t :AUTHOR(S): 19Optimal Coordination of Overcurrent Relays using Barnacles Mating Optimizer(BMO)Noor Zaihah Jamal ; Muhammad Yusuf Shamsuddin45

Page 46

Page 47

This project presents the development of an aerial monitoring system, known as a UMP FPGA-Cube,designed for environmental surveillance ap-plications. The UMP FPGA-Cube incorporates an FPGA DE10-Lite Devel-opment Board, an Arduino Uno R3 Microcontroller, and various sensors in-cluding temperatureand humidity, GPS, barometric, and accelerometer sen-sors. The system aims to collect real-timetelemetry data from below 1km al-titude and transmit it wirelessly to a ground station for analysis. Byintegrat-ing FPGA technology, Arduino control, and a selection of sensors, the UMP FPGA-Cube enablesaccurate and reliable data acquisition. The wireless communication system ensures timely transmission oftelemetry data, en-hancing efficiency in environmental monitoring. The analysis of data col-lected fromdifferent locations and altitudes provides valuable insights into environmental conditions, contributing toinformed decision-making and ef-fective environmental management. This research offers a cost-effectiveand versatile solution for environmental surveillance, showcasing the potential of the UMP FPGA-Cubesystem in gathering and analyzing telemetry data for a wide range of environmental applications.Paper ID: TITLE: A b s t r a c t :AUTHOR(S): 104UMP FPGA Cube: An Ariel Monitoring System Design for EnvironmentalSurveillance using FPGANurul Hazlina Noordin, Amir Farhan Bin Mohd Rasidi, Rosdiyana Samad,Mohamad Shaiful Abdul KarimPaper ID: TITLE: AUTHOR(S): 109Empowering Traffic Management: Anomaly Detection in Vehicle Traffic Flowusing XGBoost and Isolation Forest AlgorithmsQamil Zhafri bin Ahmad Nizam, Mohd Zamri Ibrahim, Norasyikin Fadilah,Md Rizal Othman and Ahmad Afif bin Mohd FaudziAnomaly detection in vehicle traffic flow plays a crucial role in ensuring efficient transportation systemsand maintaining public safety. However, traditional methods for anomaly detection present certainlimitations. For instance, older techniques often rely on statistical-based approaches, such as usingstandard deviation and assuming a normal data distribution, to identify anomalies based on statisticalattributes. While these methods have paved the way for more advanced approaches, such as XGBoost andIsolation Forest, which capture complex patterns and relationships in the data, providing improvedaccuracy and flexibility in anomaly detection. This paper proposes a method for anomaly detection invehicle traffic flow using XGBoost and Isolation Forest algorithms. XGBoost is a powerful gradient boostingframework that effectively captures complex patterns in the data, while Isolation Forest is an unsupervisedlearning algorithm that isolates anomalies based on their unique characteristics. The approach involvespreprocessing the traffic data, extracting relevant features, and training the models using XGBoost andIsolation Forest. Experimental results on real-world traffic datasets demonstrate the effectiveness of theproposed method, achieving a high accuracy using a threshold of 75% for XGBoost and 40% for IsolationForest in detecting anomalies. This approach has the potential to enhance traffic management systemsand improve overall traffic flow efficiency.A b s t r a c t :PARALLEL SESSION 747

Page 48

The aim for this paper is to develop an enhanced horse stable security monitoring using deep learning.The YOLO techniques are developed using normalized standard initialization and trained with a dataset ofsample photos from horse stables, encompassing both humans and horses. The trained weights arevalidated using a test video from a horse stable at Tanjung Lumpur. The implementation of the findings iscarried out using the Python language. Consequently, this study investigates the YOLO techniques andtheir architecture, analyzing the best approaches for the proposed system. The proposed approachachieves a high precision of 80% for optimum video detection and 86% for real-time detection. Theperformance analysis identifies YOLOV4 with a threshold value of 0.7 and a larger dataset as the mosteffective system to implement. Overall, this research delves into the investigation of YOLO techniques andtheir architecture, contributing to the improvement of security monitoring in horse stables. By employingdeep learning and advanced object detection methodologies, the performance and reliability of securitymonitoring systems in equestrian environments are enhanced.Paper ID: TITLE: A b s t r a c t :AUTHOR(S): 111Enhanced Horse Stable Security Monitoring using Deep Learning: InvestigatingYOLO Techniques and ArchitectureNurul Alea Ashifah, Syamimi Mardiah Shaharum, AAM Faudzi, MarlinaYakno, WSW Samsudin and Paiza Md DomPaper ID: TITLE: AUTHOR(S): 119Optimizing Image Segmentation: A Multilevel Thresholding Based OnDifferential EvolutionNor Farizan Binti Zakaria, Luqman Hakim Bin Amirol Husainy, MohdHerwan Sulaiman, Rohana Abul Karim and Nurul Wahidah ArshadThresholding is a sort of segmentation that divides pixels into discrete groups depending on their intensity levelaccording to one or more threshold values, as we all know. Thresholding is a popular image segmentation technique forconverting grey-level images to binary images,. This project is a Multilevel Thresholding Algorithm for imagesegmentation based on Differential Evolution. We have design an image segmentation module using MultilevelThresholding (MTH). The Multilevel Thresholding (MTH) technique separates pixels into discrete zones that respect theimage’s objects and is the best option for segmenting real-world images. The histogram of an image determines thethreshold point. It demonstrates that each image has a unique set of optimal threshold values. From that, we get theoptimum threshold value in Multilevel Thresholding (MTH) based on Differential Evolution. We apply the Otsu methodto identify the best optimum threshold value. After using Otsu’s techniques, we examine the system by comparingMultilevel Thresholding (MTH) performance with different optimization, which is called benchmark function, PeakSignal-to-Noise Ratio (PSNR) and Structural Similarity Index (SSIM). The outcome demonstrates that the DifferentialEvolution algorithm almost matches the performance of the Harmony Search (HS), Artificial Bee Colony (ABC), andParticle Swarm Optimization (PSO) approaches. This is evident from the PSNR and SSIM values, which reveal littledifferences across the four techniques. It can be concluded that the Differential Evolution algorithm performs well andachieves nearly the same PSNR and SSIM value as other image segmentation techniques. Due to the numerousmultilevel thresholding techniques, the PSNR and SSIM values from the obtained data show substantial variations. Thequality of the segmented image increases with level.A b s t r a c t :48

Page 49

This paper presents the design and analysis of an H-shaped patch antenna for wireless communication at5 GHz. The antenna design process considers important factors such as geometrical parameters, substrateselection, and feeding technique. Copper is used for the ground plane and patch element, while Roger RT5880 is chosen as the substrate material. Computer software technology software is used to design theantenna together with simulated results for return loss, voltage standing wave ratio, radiation pattern, andgain. Through parametric analysis, optimal values for various dimensions are determined to achieve adesirable return loss value. The results demonstrate a well-matched impedance with a return loss value of-67.55 dB at 5 GHz, indicating efficient signal propagation. Overall, the proposed antenna design offerspromising performance for wireless communication applications at 5 GHz.Paper ID: TITLE: A b s t r a c t :AUTHOR(S): 125Design of 5 GHz H-Shaped Patch AntennaNurfarhana Mustafa, Muhammad Khairul Ikhwan Tahir, Nur Sofia IdayuDidik Aprianto, Nurul Hazlina Noordin, and Mohamad Shaiful Abdul KarimPaper ID: TITLE: AUTHOR(S): 133An Analysis of Driver Drowsiness Detection Using Electromyography (EMG)Facial MusclesFaradila Naim, Ashvien Kumar Subramaniam, Mahfuzah Mustafa, NorizamSulaimanDrowsiness during driving can lead to fatal vehicle crashes and deaths. Facial responses when drowsy is auseful signal to detect driver drowsiness. Most driver drowsiness studies that uses facial responses arevision, EEG or EOG based inputs. There is lack of studies using facial EMG. This project aims to analyse theeffectiveness of using electromyography (EMG) signals from fa-cial muscles (Masseter and OrbicularisOrris) to detect driver drowsiness. 12 participants drove for 1 hour in a simulated driving were taken assamples. 7 time domain features were extracted from the raw EMG and kNN classifier was used as thesignal processing model to detect driver drowsiness. The highest accuracy achieved is 85.71% with 70:30training and test data ratio, k values (kNN) of 2 and 4, seven-time domain features, for both Masseter andOrbicularis Orris muscles. The study concludes that it is possible to de-tect driver drowsiness with highaccuracy using EMG signals from facial musclesA b s t r a c t :49

Page 50

Heart rate, a critical health indicator, is typically measured using contact methods such aselectrocardiography (ECG). However, there is a growing interest in the development of non-contactmonitors, which offer considerable advantages. These include eliminating skin irritation, reducingdiscomfort, and minimizing the risk of infection, making them perfectly suited for long-term, continuousmonitoring. Importantly, when patients are unaware of being measured, it can prevent intentional orsubconscious alterations in their heart rate, thereby enhancing the reliability and accuracy of the results.This study aimed to examine the comparative efficacy and precision of two heart rate measurementdevices the mmWave high radio frequency radar and oximeter, at varying distances and conditions,including situations with and without obstruction. The devices were tested for accuracy and consistencyin readings at distances of 30 cm, 60 cm, 90 cm, and 120 cm. The findings demonstrated that both devicesprovided highly accurate and consistent readings when there were no obstructions, regardless of thedistance. The waveforms produced by the devices were comparable, suggesting minimal deviation and astrong correlation between the two sets of results. Even in the presence of obstructions, the devicesmaintained their high degree of accuracy, with only minor variations in readings. At 60 cm, the margin oferror was found to be just plus or minus 4 bpm, indicating a remarkable tolerance to obstacles. Whencompared at greater distances, i.e., at 90 cm and 120 cm, with obstructions, the error margin slightlyincreased but remained within an acceptable range. This underlined the devices' robustness and theirability to deliver reliable results under less-than-ideal conditions. In conclusion, the mmWave radardemonstrated high levels of precision and consistency across various distances and conditions to estimatethe heart rate value.Paper ID: TITLE: A b s t r a c t :AUTHOR(S): 135Non-Contact Heart Rate Estimation using mmWave High Radio FrequencyRadarAl-Hasanol Gumanti Sudirman, Mohd Zamri Ibrahim, Ikhwan HafizMuhamad, Rosdiyana Samad and Wan Nur Azhani W. Samsudin50

Page 51

This study investigated the usability of optimization for anomaly detection of unmanned aerial vehicles(UAV). This research detects anomalies via the particle swarm optimization (PSO) method, focusing on themotor and blade faults. The vibration of fault data was measured via acceleration. Combining the PSOmethod with the monitoring-based fitness function identified the exact place where the fault hadhappened. The vibration velocity increased two times from the usual velocity when the fault was detected.The fitness function was developed via three stages, i.e., frame setting, tolerance checking, and computingthe PSO standard to differentiate among faulty, turning, and usual data peaks. This study achieved a highdetection accuracy of 76% using simulation programs of mission planner, ardupilot, and flight gearPaper ID: TITLE: A b s t r a c t :AUTHOR(S): 94Detecting Anomalies in Unmanned Aerial Vehicles via the Optimization MethodFatimah Dg Jamil, Mohammad Fadhil Abas, Vasantha Raj Rajaram,Norhafidzah Mohd Saad, Mohd Firdaus AbasPaper ID: TITLE: AUTHOR(S): 116Investigation of Odor from Surface Water Intensity Based on Pressure VariationsUsing an Intelligent Classification ApproachMohammad Danial Izzuddin bin Razali , Muhammad Sharfi Bin Najib,Mujahid bin Mohamad, and Suhaimi bin Mohd DaudThis article presents an investigation of odor intensity in surface wa- ter based on pressure variations usingan intelligent classification approach. The study aims to develop a reliable method for assessing odorlevels in water bod- ies, which can have significant implications for environmental monitoring andmanagement. A dataset of pressure variations and corresponding odor intensity levels was collected andused to train a k-nearest neighbors (k-NN) classifier. The performance of the classifier was evaluated, andreal-time simulations were con- ducted to demonstrate the applicability of the proposed method. Theresults show promising accuracy in odor intensity classification, suggesting the potential of the proposedapproach in practical odor monitoring systems.A b s t r a c t :PARALLEL SESSION 851

Page 52

This project aims to identify the initial covariance value of a ROS-based mobile robot, specifically the Turtlebot3 Burger.The basic navigation of the robot re-quires a significant amount of data and resources to process the output path. Toaddress this challenge, the Kalman Filter algorithm is implemented in this robot, as it is widely used for mobile robotnavigation and system integration. One crucial parameter for implementing the Kalman Filter is the covariance matrix,which needs to be determined. Understanding the specifications of the robot is essential for programming andoperating it effectively. The system model of this robot is developed based on the kinematic model of a two-wheeledmobile robot. To exe-cute this project, an experimental setup consisting of a laptop and a robot, serving as the ROSMaster and ROS Slave respectively, is required. Furthermore, the project aims to comprehend the function andefficiency of the robot's performance, including the LiDAR sensor, Inertial Measurement Unit (IMU) sensor, andOdometry sensor. These sensors are mounted on the robot to achieve accurate localization. An indoor experiment wasconducted to determine the covariance value. Different sources of sensor information are fused into a singlerepresentational format called sensor fusion. By using an extended Kalman filter (EKF), data from Odometry and IMUsensors were combined to estimate the position and orientation of the mobile robot. The identified covariance valuewill serve as the initial covariance matrix for the implementation of the Kalman Filter-based system using this robot.The experimental results indicate that the proposed method is suitable and practical for real-world applicationsPaper ID: TITLE: A b s t r a c t :AUTHOR(S): 127Enhancing Navigation Accuracy of Turtlebot3 Burger Mobile Robot throughInitial Covariance Matrix DeterminationMuhammad Haniff Gusrial, Muhammad Luqman Hakim Abdullah, NurAqilah Othman and Hamzah AhmadPaper ID: TITLE: AUTHOR(S): 129Development of Portable Solar Fertigation SystemNajwa Ayuni Abdullah, Muhammad Shakirin Shapee, Muhammad ArifOsman and Roshahliza M RamliPortable solar fertilization devices offer numerous benefits in agricultural systems, including cost reduction andincreased efficiency. The proposed development utilizes solar energy to power fertigation systems, making them idealfor gardens, farms, and agricultural settings. The fertilization process involves using a liquid drop system that ensuresplants efficiently absorb the available nutrients. A 12 V solar panel converts solar energy into electrical power, which isstored in a battery for later use. By utilizing timers, the fertigation system can be activated at specific times each day,further streamlining operations. The integration of the Internet of Things (IoT) concept has revolutionized farmingpractices. Real-time monitoring and control of the pump, facilitated through the IoT, enhance water usage efficiencyand enable convenient farming operations. A portable and eco-friendly water pump, powered by a solar panel, can becontrolled remotely using a mobile application that also provides environmental monitoring. Temperature andhumidity sensors measure air conditions, while a soil moisture sensor detects changes in moisture levels, alerting thesystem to the need for fertilization. A rain sensor module detects raindrops and measures intensity, ensuring that plantsare not overwatered during the rainy season. The system incorporates Blynk software, which displays temperature andhumidity conditions and enables remote control of the fertilization system via Wi-Fi communication. Using this efficientfertilizer system can lead to a steady income for farmers. Compared to traditional agriculture, labour costs aresignificantly reduced, resulting in lower operational costs and the ability to sell crops at more affordable prices.A b s t r a c t :52

Page 53

Olive oil, a vegetable oil with numerous health benefits, possesses distinctive nutritional characteristicsclosely linked to its chemical composition. The formation of these characteristics primarily takes placeduring the oil's production, storage, and handling of raw materials. It is worth noting that olive oil alsocontains acidity, which can present challenges. Consumption of olive oil can potentially contribute todisorders associated with saturated fat. Hence, the primary objective of this project is to examine the levelof free fatty acids (FFA) in heated olive oil and identify the optimal absorption wavelength. Spectroscopytechniques will be employed to analyse free fatty acid in olive oil. UV/Vis spectroscopy allows theinvestigation of a sample's electronic structure and facilitates the identification of compounds present.Four different olive oils are tested at varying heating times. By measuring the FFA due to heating, theimpact of different heating conditions on the oil can be assessed. Prolonged heating leads to a greaterdegree of oxidation. Through this analysis, consumers will gain a better understanding and be morecautious when using olive oil.Paper ID: TITLE: A b s t r a c t :AUTHOR(S): 131Analysis of Heated Olive Cooking Oil Using A Spectroscopy TechniqueN.F.M Fadzila, W.S. Salleh, S. Nurulain, M.A.S. Aspar, H. ManapPaper ID: TITLE: AUTHOR(S): 132Formulation Of Fitness Function To Estimate PH Value Of Adjacent Block Via PHValue And Water FlowNurul Najihah Mohd Radzi, Mohammad Fadhil Abas, Muhammad SyukriBin Ahmad, Norhafidzah Mohd Saad, Mohd Hisyam Mohd AriffWater quality is measured by several factors, such as the potential of hydrogen (pH), the concentration ofdissolved oxygen (DO), bacteria levels, salinity, or turbidity. This project focuses on the pH of water as itgives more impact on determining the quality of water. It is noticed that the speed of water flow doesmanipulate the value of pH water. A large set of data which comprises five locations, four of the locationspH and water speed are used to determine the fifth location pH (known as unsampled pH). To estimatethe un-sampled pH, a fitness function has been formulated using Multi-Layer Neural Network by GeneticAlgorithm (MLNN-GA) and compares the results in terms of accuracy between the estimation of pHwithout water speed and pH with water speed. Both estimated results will be compared with the actualpH value. The results of the estimated data pH with speed is 94.27% compared to the estimated datawithout speed is 93.83%. The result showed that speed is one of the factors that can be used to increaseaccuracy in estimating the pH value.A b s t r a c t :53

Page 54

This paper presents the development of an IoT-based smart fridge sys-tem aimed at improving theefficiency of food thawing for working mothers and young professionals. Traditional thawing methods areknown to be time-consum-ing and unreliable, compromising food safety and quality. The project seeks toprovide a safe, convenient, and cost-effective solution for preparing healthy meals. The objectives includeinvestigating and developing a customized thawing model for the smart fridge, designing the systemusing Arduino programming, and conducting performance analysis. The research incorporates Salvadoriand Masheroni's thawing time prediction method based on different food types. The system isimplemented using the Arduino Uno board and the Blynk application, with real food samples used fortesting. Experimental results show a close corre-lation between predicted and actual thawing times forslab beef, but significant deviations are observed for cylinder fish samples. In conclusion, this paper con-tributes to enhancing the defrosting process, prioritizing food safety, and offering a convenient solution forindividuals with busy schedules. By harnessing IoT technology, the proposed smart fridge system improvesconvenience and effi-ciency in food thawing.Paper ID: TITLE: A b s t r a c t :AUTHOR(S): 122Empowering Food Thawing with IoT: Design and Development of a Smart FridgeSystemChai Kuan Jie, Syamimi Mardiah Shaharum, Mohd Ali N. Z54

Page 55

Wireless power transfer (WPT) technology eliminates the need for cables and cords, providing aconvenient and efficient method for transferring electrical power. The transmitter sends outelectromagnetic waves that are converted into electrical energy by the receiver, with the powertransferred through an air gap. This study aims to enhance the performance of WPT by developing ahybrid coupling topology that combines Inductive Power Transfer (IPT) and Capacitive Power Transfer(CPT) to create a high-efficiency system. The hybrid topology is simulated using MATLAB@Simulink inboth Series-Series (SS) and Parallel-Parallel (PP) compensation. The simulations vary in frequency and dutycycles and different load resistances at the receiving end to obtain precise results. The study evaluates theoutput power and efficiency to minimize any losses in the system and maximize efficiency and powertransfer. The results provide insights into how the hybrid coupling topology can be optimized for WPT,leading to more efficient and effective wireless power transfer systems. Such systems have a wide range ofapplications in various industries, including transportation, healthcare, and consumer electronics. With thegrowing demand for wireless power transfer, more efficient systems can lead to significant advancementsin these industries.Paper ID: TITLE: A b s t r a c t :AUTHOR(S): 18Breaking Free from Cords: Enhancing Wireless Power Transfer with a HybridCoupling TopologyAnisya Sakinah Abdol Samat, Mohd. Shafie Bakar, Mohd. Shawal Jadin,Omar AlimanPaper ID: TITLE: AUTHOR(S): 41Maximum Power Point Tracking (MPPT) based Particle Swarm Optimization(PSO) for Hydrokinetic Energy HarnessingWan Ismail IbrahimThis paper presents a design and modeling of the Particle Swarm Optimization (PSO)-based maximumpower point tracking (MPPT) algorithm specifically tailored for variable-speed fixed-pitch vertical axishydrokinetic turbines. The proposed algorithm can maximize electrical power without requiringadditional sensors and prior knowledge of the water turbine characteristics. Unlike the conventional MPPTalgorithm, the PSO-based MPPT algorithm exhibits minimal oscillations at the maximum power once thetrue peak is located. The PSO MPPT algorithm is characterized by its simplicity, flexibility, accuracy, andefficiency in tracking the maximum power under different water velocities. The simulation results byMATLAB/ Simulink indicated that the PSO MPPT can achieve 83 % efficiency in terms of output power andreduce the oscillation during dynamic steady-state.A b s t r a c t :PARALLEL SESSION 955

Page 56

Battery State of Charge (SOC) estimation is of utmost importance for the ef-ficient operation of battery systems.Artificial Neural Networks (ANN) have emerged as powerful tools for accurately estimating SOC. This study explores theapplication of ANN for SOC estimation in batteries using a comprehensive dataset obtained by simulating the usage ofa lithium poly-mer cell model, specifically the ePLB C020, in an electric car resembling the Nissan Leaf. The datasetencompasses various battery parameters, including voltage, current, temperature, and other relevant variables. Theresearch fo-cuses on developing an optimized ANN architecture, training methodology, and evaluation metrics toensure precise SOC estimation. The proposed ANN architecture consists of input, hidden, and output layers withcareful-ly optimized neuron numbers and activation functions. Through an iterative training process employingbackpropagation and gradient-based optimiza-tion algorithms, the weights and biases of the ANN are adjusted toenhance its performance. Evaluation metrics such as Mean Squared Error and corre-lation coefficient are employed toassess the accuracy and reliability of the SOC estimations. The experimental findings underscore the effectiveness of theANN model in achieving accurate SOC estimation for the simulated electric car battery. This study highlights thepotential practical applica-tions of ANN in battery management systems, enabling reliable SOC esti-mation forimproved control and optimization strategies. Additionally, a comparison between the proposed ANN model andExtreme Learning Ma-chines (ELM) reveals superior performance of the ANN in SOC estimation for the simulatedelectric car battery.Paper ID: TITLE: A b s t r a c t :AUTHOR(S): 73State of Charge Estimation by Using Artificial Neural Networks for LithiumPolymer Battery of Electric VehicleMohd Izzat Mohd Zalam, Mohd Herwan Sulaiman, Zuriani Mustaffa, AddieIrawan HashimPaper ID: TITLE: AUTHOR(S): 78Optimal Distributed Generation (DG) Allocation for Transmission LossesMinimization using Arithmetic Optimization Algorithms (AOA)Abdulah, Nor Rul Hasma; Mustafa, Mahfuzah; Samad, RosdiyanaThis research paper presents research on the optimal placement and sizing of distributed generation (DG)units in power systems. The integration of DG units plays a key role in addressing the growing demand forrenewable energy sources and improving grid stability. The objective of this study is to minimizetransmission losses using Arithmetic Optimization Algorithms (AOA). Extensive experiments and analyzesare conducted with a focus on testing AOA performance on both weak and secure bus scenarios. Theresults demonstrate the effectiveness of AOA in achieving optimal DG place-ment, resulting in reducedtransmission loss and improved system performance. Findings from this research provide valuable insightsfor power system planners and operators, assist decision-making processes in optimizing DG integration,and increase the overall efficiency of power systems.A b s t r a c t :56

Page 57

To maintain the long-term reliability of photovoltaic modules while maximizing the power output,possible faults in the photovoltaic modules need to be diagnosed early. Aerial thermal image inspection iscommonly used to detect and locate the hotspots of the photovoltaic modules. However, detecting ahotspot from this image can be severely affected by noises and thus can wrongly locate the hotspot dueto thermal reflection from the surrounding. One of the solutions is by examining both visual and thermalimages of the photovoltaic modules. This paper presents multi-spectral image matching of thephotovoltaic modules. First, absolute structure map (SMi) and directional structure map (DSMi) areproposed. A histogram of the oriented gradient is then used to describe each interest point’s local regionbased on the SMi and DSMi. Next, the Gabor wavelet filter is applied to the SMi, whereas the average filteris applied to the DSMi to construct the histogram bins. Finally, the normalized feature vectors areconjoined. Experiments were conducted to evaluate the proposed structure map feature descriptor’sperformance. The results showed that this method could provide Precision and recall up to 0.82 and 0.97,respectively.Paper ID: TITLE: A b s t r a c t :AUTHOR(S): 82Local Feature Descriptor Based on Directional Structure Map for Improving theHotspot Detection in the Multispectral Aerial Image of a Large-Scale PV SystemTan Li Ven, Mohd Shawal Bin Jadin, Muhammad Khusairi Osman, Mohd.Shafie Bakar, Md. Imamul Islam Ahmed Al Mansur and Mohammad Asif UlHaqPaper ID: TITLE: AUTHOR(S): 90Economic and Emission Dispatch Solution using Evolutionary Mating AlgorithmAhmad Shahier Abdul Aziz, Mohd Herwan Sulaiman, Zuriani MustaffaThis paper presents the Evolutionary Mating Algorithm (EMA) as a novel evolutionary algorithm foraddressing economic emission load dispatch (EELD) problems. The optimization of power systems withrespect to eco-nomic and emission considerations is of utmost importance in contempo-rary powersystem engineering. Emphasizing cost and emission reduction is essential for efficient power systemoperation. In this study, the economic and emission dispatch problem is tackled using the EvolutionaryMating Algorithm (EMA). The performance of the EMA algorithm is evaluated on a 10 and 40-unitgenerator test system. Comparative analyses are conducted with other algorithms, namely the CuckooSearch Algorithm (CSA), Flower Pollination Algorithm (FPA), and Barnacles Mating Optimizer (BMO). Theresults indicate the effectiveness of the Evolutionary Mating Algorithm in solving economic emissiondispatch problems, thereby demonstrating the efficacy of the proposed EMA approach for addressingEELD problems.A b s t r a c t :57

Page 58

Electricity consumption is a significant indicator of modern society's development and advancement. It isinfluenced by factors such as population growth, urbanisation, and economic activity. However, predictingelectricity consumption is a tough task due to the complexity and fluctuations of the energy market. Inthis paper, Support Vector Regression (SVR) was proposed in developing a predictive model for Malaysianelectricity consumption. SVR was chosen as our proposed method as it can handle nonlinear and high-dimensional data using kernel functions. Data used for this study were retrieved from various sourcesincluding macrotrends.net, the World Bank's Climate Knowledge Portal, and the World Bank's indicatordatabase. The dataset consists of relevant variables such as temperature, population density, andeconomic growth to anticipate future electricity demand. Results from this study showed that the SVRmodel outperforms other methods in terms of accuracy and error metrics. Additional components,hyperparameter fine-tuning, ensemble approach research, and long-term forecasting are all advocated forfurther improvement.Paper ID: TITLE: A b s t r a c t :AUTHOR(S): 91Predictive Model for Electricity Consumption in Malaysia using Support VectorRegressionMUHAMMAD AIMANDZIKRI MOHD NIZAM ; Sahimel Azwal Sulaiman ; NorAzuana Ramli58

Page 59

Power consumption, area, and timing affect the efficiency of the processor. There is an increasing demandfor higher standards and improved performance in chip technology. The pursuit of better performancehas become a prevailing objective among individuals and industries. As technology continues to advance,the stand-ards for chip performance will undoubtedly rise, motivating researchers and manufacturers tostrive for even greater achievements. The aim of this project is to study and analyse the performancerequirements in terms of area, power con-sumption, and speed of a 16-point Radix-4 Fast FourierTransform (FFT) pro-cessor. Furthermore, this project aims to create an efficient FFT processor chip. Thesimulation was carried out using Mentor Graphics QuestaSim, and the syn-thesis was done using MentorGraphics Oasys-RTL. The project focused on the front-end design part, implementing the design with130nm CMOS process tech-nology. The research methodology of the study implemented a modified flop-based design to facilitate timing optimization. The design has a low power con-sumption of 620.105mW, acell area of 1374521µm², and high overall perfor-mance. The design exhibits high speed and a smaller cellarea. The performance of the proposed design has improved compared to the original design.Paper ID: TITLE: A b s t r a c t :AUTHOR(S): 36The Design and Analysis of Fast Fourier Transform Processor using VLSITechniqueKuan Pei Xian, Fahmi SamsuriPaper ID: TITLE: AUTHOR(S): 37Face Detection Through Conceal with Deep LearningEswaran Rasentheran, Fahmi SamsuriThe presented research paper purposes a method to detect concealed faces using the pre-trained model,Convolutional Neural Network (CNN) and using Tensor-Flow as the deep learning method. Concealed facedetection has emerged as a critical field of research and application due to its relevance in numerousdomains. With the growing concerns regarding security, law enforcement, public health, and privacy, theneed to accurately detect and identify individuals whose faces are concealed has become paramount. Inthis paper there will be 2 approaches used to detect concealed faces utilizing the above stated model andtechnique.A b s t r a c t :PARALLEL SESSION 1059

Page 60

This technical report summarizes the whole developing process of this project. The project is aboutdeveloping a web application to detect early diabetes mellitus. The problem statement for this project isthe lack of convenience in detecting dia-betes. The only way to diagnose diabetes is through blood testbut blood test is not recommended to conduct frequently. There is also lack of application to help todetect diabetes. To solve the problem, we have found the most accurate ma-chine learning algorithmwhich is Random Forest to detect diabetes. Random Forest algorithm will be the prediction model andwill be deployed into the web application. The web application is developed by using Visual Studio Codewith the assistance of Streamlit framework. To develop this web application, there will be two parts, whichare prediction model development and user interface devel-opment. Prediction model developmentinvolves choosing the most suitable ma-chine learning algorithm to be the prediction model. Userinterface development consists of the UI of the web application. There is only one result discussed in thisreport.Paper ID: TITLE: A b s t r a c t :AUTHOR(S): 38Design and Analysis of an Early Diabetes Mellitus Detection Using OpenCVLoh Shu Yi, Fahmi SamsuriPaper ID: TITLE: AUTHOR(S): 45Epileptic Seizure Classification in EEG Signals Using KNN and SVMFathin Naadiah Mohd Razif, Mahfuzah Mustafa, Rosdiyana Samad, andNor Rul Hasma AbdullahEpilepsy is a neurological disorder that affects millions of people worldwide, and accurate classification ofepilepsy based on seizure type and epilepsy syndrome is crucial for effective treatment. However,distinguishing between different types of epilepsy can be challenging due to the complexity of EEGsignals. This study investigated the effectiveness of using eight key features extracted from EEG signals inaccurately classifying epilepsy using KNN and SVM algorithms, achieving an accuracy of 100% for bothalgorithms. The study's findings provide a promising approach to accurately classify epilepsy, which canpotentially improve the accuracy of epilepsy classification and develop more effective treatmentstrategies for epilepsy patients.A b s t r a c t :60

Page 61

Because of its simplicity, flexibility, and robustness, the Artificial Bee Colony (ABC) algorithm, a swarmintelligence-based optimisation method, has been widely applied in a variety of fields. However, itsapplication in hyperparameter optimisation for machine learning classifiers deserves exploration. Theeffectiveness of ABC and its modified version, JA-ABC5, for hyperparameter optimisation across variousclassifiers, including Support Vector Machine (SVM) and K-Nearest Neighbour (KNN), is studied in this re-search. The Wisconsin dataset is used to evaluate the performance of these classifiers, and thehyperparameters are optimised using the JA-ABC5 algorithm. The performance of JA-ABC5 is comparedto that of grid search, standard ABC, Bayesian optimisation, and random search. The results show that JA-ABC5 performs well in terms of SVM, which is accuracy, specificity, and sensitivity, while its performance inKNN is comparable. This re-search contributes to a better understanding of machine learning modeloptimisation, with the potential to improve the performance of these models in a variety of applications.Paper ID: TITLE: A b s t r a c t :AUTHOR(S): 117Evaluating JA-ABC5 Hyperparameter Optimisation with ClassifiersRavindran Nadarajan, Noorazliza Sulaiman, Junita Mohamad-SalehPaper ID: TITLE: AUTHOR(S): 103Enhancing Squat Safety and Performance with Computer Vision and DeepLearning ModelMuhamad Aqil Hilman Hazlan, Ikhwan Hafiz Muhamad and Mohd ZamriIbrahimExercise is good for one's health and fitness, however, it can also be ineffective and potentially dangerousif done improperly by the user. Exercise mistakes are made when users don't use Correct form or pose.Poor squat posture for example can damage the knee health for a long period. Thus, maintaining ahealthy squat posture is crucial for a person to workout effectively. This project introduces the use ofcomputer vision to develop a model using the MediaPipe Pose, that recognizes and classifies the bestsquat posture and provides recommendations on how users can improve the form. The data is collectedfrom exercise videos of correct squat posture by a professional coach. The developed algorithmsuccessfully classifies correct posture with overall accuracy of 85%.A b s t r a c t :61

Page 62

A signature is a handwritten representation of a person's name, nickname, or other mark affixed to papers to verify theiridentity and intent. A hand-written signature identifies the work as well as its creator. Every person's signature isdifferent, make it crucial to recognize a person's handwritten signature. Besides, doing human verification might beimperfect and uncertain at times. In some cases, signatures on papers are occluded by document texts or rubberstamp, which making them difficult to verify. The same might be said for antique documents, which the signatures maynot always be visible. Others, when a client's signature has slight alterations in the pattern, some financial transactionrequests cannot be validated precisely. Handwritten signature verification is a challenging task that has been widelystudied in recent decades to solve all the problems. Despite the progress that has been made, it remains as an activearea of research, with new techniques and approaches being developed all the time. Researchers are experimentingwith a variety of methods for distinguishing between genuine and forged signatures. Motivation to these problems, anovel method is proposed for reliably recognizing of signatures in documents and performing identification checkingusing CNN-based deep learning algorithm. The proposed algorithm's validity is determined by the matching sign withthe online signature database. An autoencoder is utilized to create random distortions in genuine photos from thedatabase, which were then given to the classifier during training to create false signatures. The proposed algorithm isbased on Siamese Network, which works on two inputs from VGG-16 with the same weight and same structure, andpro-duce two features. The classification results of the proposed algorithm are about 99% accurate.Paper ID: TITLE: A b s t r a c t :AUTHOR(S): 123Signature Verification using CNN Deep Learning-Based Approach (FYP Student)Wan Nur Azhani W. Samsudin, Mohd Zamri Ibrahim, Muhammad HaziqZainul Asri, Wan Syahirah W. Samsudin, Suraya Abu Bakar62

Page 63

Social distancing is a non-medical practice that helps slow down the trans-mission of viruses which issuggested by the World Health Organization (WHO). However, as every country has battled the spread ofthe virus for almost three years, social distancing practice now seems ignored by public people due tosome reason such as they are in a rush. This study aims to analyze human detection using deep learningmethods in various positions and to develop social distancing detection using the proposed method. Thus,human detection and social distancing detection using a deep learning algo-rithm which is You Only LookOnce (YOLO) version 3 is developed. This method uses custom datasets and the Euclidean distanceformula to compute the distance between two people for the social distancing detector. The out-putdistance is measured in the real world (centimetres). As a result, the cur-rent datasets for each positionsuch as front view, back view, side view, and the crowd gives the result of human detection at 94.44%,91.67%, 97.50%, and 88.89% respectively. Hence, the highest accuracy of human detection goes to the sideview position with a percentage accuracy of 97.50%. Next, the distance between the two people iscalculated correctly with the accepta-ble range of ±0.3cm.Paper ID: TITLE: A b s t r a c t :AUTHOR(S): 101Analysis of AI-Powered Human Detection Method for Social DistancingMonitoringNur Aina Syafinaz Mohd. Atfan, Rosdiyana Samad, Nor Rul HasmaAbdullah, Mahfuzah Mustafa1, Nurul Hazlina Noordin, Dwi PebriantiPaper ID: TITLE: AUTHOR(S): 98Child Left in the Car Detection: Image Enhancement for Day and NightRohana Abdul Karim, Hong Zhuang Shen, Marlina Yakno, Yasmin AbdulWahab, Mohd Zamri IbrahimDespite the government's awareness campaign on the safety of children in non-moving vehicles, the casesof children being trapped and suffocated in unattended cars keep rising. Children are often being ignoredby their par-ents with the engine off. Furthermore, there are also children being ignored due tounstoppable calls from work. Next, existing systems with only 1 de-tection used have a limit eitherdetecting face or signal detection. Most of the software tools have a limitation in detecting human face ina low light situation. Other than that, the obstacle like hand will decrease the accuracy of face detection.This project aims to develop a complete and adequate face detection system for detecting the presence ofchildren in the car by detect-ing human physical features with temperature and sensor technology. Theobjective is to enhance the visualization of images for human identification and to measure theperformance of the features selections for detection sys-tem. There are 100 sample images of child facesbeing collected, and three filters are being compared for image enhancement: fastNIMe-anDenoisingColored, histogram equalizer and median filter. During normal daylight,fastNIMeanDenoisingColored achieves the highest percentage of accuracy of face and hand detectionwith 90%, followed by without filter and Median Filter with 90% accuracy. During 100 to 150 value ofdimmer images, Histogram Equalizer achieves the highest accuracy percentage of face and handdetection with 85%, followed by FastNlMeanDesnoisingCol-ored and without filter with 81% accuracy.A b s t r a c t :PARALLEL SESSION 1163

Page 64

Diabetic retinopathy (DR) is a prevalent health issue associated with long-term diabetes, often leading tovarious complications. One characteristic manifestation of DR is the development of exudates in theretinal region. Fundus image analysis serves as a common method for diagnosing DR, rely-ing on theexpertise of ophthalmologists. Recently, computer-assisted diag-nostic tools have enhanced the speed ofdiagnosis, provided diverse perspec-tives, and evaluated treatment outcomes. Consequently, numerousstudies have focused on identifying diabetic retinopathy lesions in fundus images. This research aims toextract exudate features using the superpixel and Ga-bor methods in fundus retinal images. Acombination of image processing with the superpixel algorithm is compared with the Gabor featureextraction method. The findings reveal that the superpixel method outperforms the Ga-bor method inaccurately extracting exudate characteristics. The achieved re-sults indicate a 90% accuracy for thesuperpixel method and 78.3% accuracy for the Gabor method. These outcomes underscore the superiorperformance of the superpixel method in exudate extraction from fundus images. This study contributesto advancing the field of diabetic retinopathy analysis and highlights the potential of the superpixelmethod in improving diagnostic accuracy and efficiency.Paper ID: TITLE: A b s t r a c t :AUTHOR(S): 89Comparative Analysis of Superpixel and Gabor Methods for Exudate FeatureExtraction in Diabetic Retinopathy Fundus ImagesNur Munirah Suhaimi, Rosdiyana Samad, Nor Rul Hasma Abdullah,Mahfuzah Mustafa, Mohd. Zamri Ibrahim, Dwi PebriantiPaper ID: TITLE: AUTHOR(S): 60Passenger's Demographic Analytical System in an Artificial Intelligence ofThings (AIoT) Edge Device for Public BussesSyafiq Fauzi KamarulzamanBusses has become crucial resources in metropolitan area where public transport were widely utilizedefficiently. Such resources can be overused or underused at a certain period of time, making managementof the resources are not optimized on a certain period of time. Artificial Intelligence (AI) can be embeddedas one of the optimization approaches on a bus passengers analytic. This proposed project utilizes the AIto detect and analyze the demographic of the bus users in real-time environment. In this project, weutilized an AI edge device to capture facial characteristics of the passengers for demographic analysis,where the results of the analytic are presented in a dashboard based on time the passengers board thebusses. This aim of this proposed system is to optimize the bus resources and increase the efficiency andreliability of the public busses management.A b s t r a c t :64

Page 65

Sitting is a basic action and resting position in which the body weight is supported primarily by lowerparts of the body that are in contact with the ground or a horizontal surface such as a chair seat. Poorsitting posture can damage the spine health for a long period. Thus, maintaining a healthy sitting postureis crucial for a person who needs to sit for a long time. By detecting the sitting posture of a person able torepair and warn the bad sitting posture. This project proposed the use of computer vision to develop ahuman skeleton model using MediaPipe Pose by plotting the landmarks on the joint point throughout thebody that is called the keypoint. The pipeline's posture estimation component predicts the location of all33-person keypoints with three degrees of freedom. These keypoints will be used to calculate body angleand classifies the best sitting posture. This project able to produce overall precision of 92.5% for straightsitting posture and recall result that achieve 84.091% in real-time data and image. The overall accuracy forthis project is 82%.Paper ID: TITLE: A b s t r a c t :AUTHOR(S): 102Harnessing Computer Vision and Deep Learning Model for Optimal SittingPosture DetectionMuhamad Mirza Azfar Damanhuri, Ikhwan Hafiz Muhamad and MohdZamri IbrahimPaper ID: TITLE: AUTHOR(S): 136Region of Interest Detection in Thermal Image of DC Motor Using ImageProcessing for Average Temperature CalculationAhmad Afif Mohd Faudzi, Tuan Nur Atikah Tuan Shukri, Mohd ZamriIbrahim, Syamimi Mardiah Shaharum and Mohd Azri HizamiAnomalies in temperature, such as excessive heat, often indicate potential equipment breakdown ormalfunction. Therefore, it is crucial to implement preventive maintenance programs that closely monitorthese variations. In mo-tor thermal modeling, calculations involving the convection thermal coeffi-cientare necessary. The coefficient relies on the average temperature of the motor surface, traditionallymeasured using a thermocouple. However, using the predefined shapes provided by Testo Software mayinclude irrelevant are-as, which can affect the accuracy of the average temperature calculation. To addressthis limitation, this paper proposes an image-processing technique to segment only the relevant area ofinterest. By utilizing this technique, the av-erage temperature can be calculated more accurately.Furthermore, a compara-tive analysis will be conducted to compare the average temperature obtainedthrough the image-processing technique with that obtained using Testo Soft-ware. This analysis aims todemonstrate the effectiveness of the proposed technique in achieving more precise temperaturemeasurements in the context of motor thermal analysis.A b s t r a c t :65

Page 66

Bell’s palsy is one the main causes of facial nerve paralysis and it is a condition in which the muscles onone side of the face become suddenly weak. The weakening causes the lower half of the face to droop.One-sided smile is common in Bell’s palsy cases and the affected eye refuses to close. This can cause theeye to become extremely dry and inflamed, often leading to exposure keratitis. By studying the differencesof eyes blinking between normal and Bell’s palsy patient, the aim of this study is to design a computer-assisted diagnosis tool in evaluating of the facial nerve function. This study aims to extract the significantfeatures by using the Viola-Jones algorithm. A combination of image processing methods with the Viola-Jones algorithm achieved the promising result where the developed tool is successfully detected thedifferences between the normal and Bell’s palsy patient during the eyes movement. This study contributesas a great aid tool to clinicians or medical professionals for an efficient facial nerve evaluation.Paper ID: TITLE: A b s t r a c t :AUTHOR(S): 121Eye Blinking Assessment of Bell’s Palsy PatientWan Syahirah W. Samsudin, Ellina Farzana Zarini, Wan Nur Azhani W.Samsudin, Syamimi Mardiah Shaharum , Kenneth Sundaraj and MohdZaki Ahmad66

Page 67

In an application with a high temperature and restricted volume constraints, the synchronous reluctance(SynRM) machine with passive rotor is a prevalent solu-tion. To attain a high torque density, the rotorstructure exhibits complex geome-try of either flux barrier or segmented rotor to achieve a high saliencyratio. This study evaluates the consequence of having the flux barrier and segmented rotor structure onthe stress and deformation of a small SynRM with a rotor diameter measuring 25.24 mm. The analysis isconducted using FE where torsional force and centrifugal for were affected to the rotor structure at atorque of 0.2 Nm and rotational speed of 10,000 rpm. The results show that the centrifugal force is morecritical as it generates higher stress for both rotor designs. The flux-barrier rotor experiences lesscentrifugal-induced stress at 5.46 MPa compared to 10.92 MPa of the segmented rotor. However, bothrotors will not suffer from perma-nent deformation. A hypothetical dimension scale-up indicates that theflux barri-er has a higher stress of 547.56 MPa and deformation due to centrifugal force and exceeds thematerial yield strength at 10 times its current dimension.Paper ID: TITLE: A b s t r a c t :AUTHOR(S): 92Comparison of Stress and Deformation due to Electromagnetic Torque in SynRMwith Flux-Barrier and Segmented RotorC. Yik How, M. S. Mat Jahak and M. A. H. RasidPaper ID: TITLE: AUTHOR(S): 105Optimal Planning of Photovoltaic Distributed Generation Considering Time-Varying LoadsAhmad Syahmi Aiman Mohd Zuri, Norhafidzah Mohd Saad, Nur SyazanaMohd Sayuti, Mohammad Fadhil Abas, Suliana Ab Ghani, Norazila Jaalam,Abid AliThe integration of photovoltaic distributed generation (PVDG) into power systems has gained significantattention due to its potential for renewable energy generation and the reduction of greenhouse gasemissions. However, the intermittent nature of solar power and the presence of time-varying loads posechallenges to the optimal planning and utilization of PV systems. This research focuses on addressing theoptimal planning of PVDG considering time-varying loads. The backward/forward sweep power flow(BFSPF) with mix-integer optimization by genetic algorithm (MIOGA) methods are used to optimally sizeand locate PVDGs in the radial distribution network (RDN) while considering the dynamic nature of loadsover time. There are three time-varying load cases: residential, commercial, and industrial. In MATLAB, theapproach is evaluated using a conventional 33-bus RDN. With the installation of PVDG, the simulationresults show a reduction in total power loss and an improvement in voltage magnitudes for the network.According to the findings, multi-PVDG installation in the residential, commercial, and industrial loadmodels can minimize power losses by up to 58.96%, 54.49%, and 56.92%, respectively. Aside from loweringlosses, installing PVDG also helps to enhance the voltage profile of the radial distribution network. Thefindings highlight the importance of considering load fluctuation to achieve optimal integration of PVDGinto power distribution networks, ultimately contributing to the transition towards a sustainable energyfuture.A b s t r a c t :PARALLEL SESSION 1267

Page 68

This paper presents the development of an Electric Vehicle (EV) Charging Pillar integrated with Internet ofThings (IoT) technology and Smart Energy Meter features. The system aims to address the challenges of EVcharging by providing efficient energy consumption, real-time monitoring, and a convenient userexperience. The EV Charging Pillar incorporates IoT capabilities, allowing for seamless communicationbetween the charging infrastructure and EVs. This enables real-time data exchange, effective control, andmonitoring of the charging process. The Smart Energy Meter feature enhances efficiency through accurateenergy measurement, billing, and load balancing. The hardware components include charging sockets,communication modules, and energy meters, interconnected through a central control unit. The softwarecomponent comprises an IoT plat-form facilitating data exchange, charging control, and user interactionthrough a mobile application or web interface. The developed system offers benefits to both EV ownersand infrastructure providers, such as convenient access to charging stations, monitoring capabilities, andoptimized energy consumption. Infrastructure providers can benefit from centralized monitoring,proactive maintenance, and integration with the grid. The proposed EV Charging Pillar with IoT and SmartEnergy Meter features provides a comprehensive solution to enhance the accessibility, efficiency, andreliability of EV charging infrastructure, promoting sustainable transportation systems.Paper ID: TITLE: A b s t r a c t :AUTHOR(S): 126Development of EV Charging Pillar with IoT (Smart Energy Meter) FeaturesMuhammad Ikram bin Mohd RashidPaper ID: TITLE: AUTHOR(S): 130Prediction of Solar Power Generation using Random Forest Regression ModelNur Zahirah binti Mohd Ali, Muhammad Zulfadhli bin Mohd Azhar,Syamimi Mardiah Shaharum, Wan Syahirah W. SamsudinRenewable energy sources include sunshine, wind, flowing water, internal heat, and biomass. Solar energyis a significant source of electricity generation due to its accessibility. Knowing the real generation andconsumption of power is the first step of making a good electrical system. To save resources and reducecosts, power utilities are required to balance between produced power and customers’ consumption.Prediction is essential for the future operation of smart grids. To predict the generation, input featuresmust be evaluated based on historical data on ISolarCloud. Supervised machine learning algorithm is usedto create a predictive model. In this project, Random Forest (RF) Regression model have been chosen topredict the power generation from solar energy. By finding the best-fit algorithm, more investigationwould be taken place for improvement in future work such as Correlation Coefficient (R), Mean AbsoluteError (MAE), Root Mean Square Error (RMSE) and Root Relative Squared Error (RRSE). The best techniqueensures excellent precision in energy forecasting with a very low error rate and the results.A b s t r a c t :68

Page 69

Several factors cause the output degradation of the photovoltaic (PV) module. The main affectingelements are the higher PV module temperature, the shaded cell, the shortened or conducting bypassdiode, and the soiled and degraded PV array. However, one key factor in making photovoltaic installationsa profita-ble investment is regular and effective inspections in order to detect occurred defects.Unmanned aerial vehicles (UAV) are increasingly used in various in-spection fields in Large Scale Solar(LSS). Nowadays, infrared thermography (IRT) technology is widely used for hotspot detection. Comparedwith manual inspection, the use of unmanned aerial vehicles (UAVs) can improve work efficiency greatlyin large-scale PV plants. The IRT image processing of PV modules is important for hotspot detection.Without the segmentation of PV modules, the hotspot location cannot be determined. In this paper, weproposed a method to combine mask images with IRT images to acquire segmentation. MATLAB imageprocessing and computer vision using color threshold are used. This paper shows experimental results forthirty photovoltaic (PV) mod-ules. From thirty photovoltaic (PV) modules, there are five photovoltaic (PV)modules that cannot be segmented very well. The color and temperature of the IR image cannot besimply to segment. The hotspot cell may occur due to re-flection from the sunlight to the photovoltaic.Quantitative evaluation is used to assess our quality method. The average quality of the output mask is83.3%, which indicates the method performs well in segmentation.Paper ID: TITLE: A b s t r a c t :AUTHOR(S): 137Fusion Colour Model for Photovoltaic (PV) SegmentationAzura @ Nurul Shuhada Binti Daud, Rohana Binti Abd Karim, MohdShawal Bin JadinPaper ID: TITLE: AUTHOR(S): 108Solar-Powered IoT-Integrated Air Quality System with ESP-NOW for Real-TimeOutdoor MonitoringLiphia Law Li Wen, Norasyikin Fadilah, Mohd Zamri Ibrahim, Ikhwan HafizMuhamad and Rohana Abdul KarimAir quality in Malaysia is worsening due to increased vehicle usage and wildfire in Indonesia. Theseactivities contribute to higher pollutant levels which are harmful to human health. Therefore, there is aneed for a portable, self-sufficient air quality monitoring system that is easy to install, maintain, transport,and does not rely on a fixed power source. Additionally, a system that seamlessly integrates with otherdevices and networks is required to provide real-time data for timely analysis and access by authorities.This paper presents a solar-powered system that integrates with the Internet of Things (IoT) and measuresvarious air quality parameters, including temperature, humidity, particulate matter 2.5, carbon monoxide,and nitrogen oxide. Real-time data transmission to a remote server enables authorities to analyze andaccess data promptly. The system also incorporates an alert system for surpassing measurementthresholds, enabling swift actions. By facilitating effective monitoring and addressing of air qualityconcerns, this system supports authorities, researchers, and stakeholders. Moreover, its long-range wirelesscapability using ESP-NOW technology allows communication up to 10m for indoors with obstacles, 20mfor indoors without obstacles, and 40m for outdoors.A b s t r a c t :69

Page 70

Dye-sensitized solar cells (DSSCs) have emerged as a promising technology for converting solar energyinto electrical energy due to their cost-effectiveness and environmentally friendly nature. This paperpresents a fabrication process for DSSCs utilizing copper (I) iodide (CuI) as a key component, which offersan attractive alternative to conventional materials. The DSSC was fabricated with four different sizes,producing current from 27.8µV to 1.1µV. The intervention is the use of eco-friendly copper (I) iodide orcuprous iodide (CuI), a p-type semiconductor material, with the incorporation of a organic ligand, calledTetramethylethylenediamine (Tmed) in the preparation for solid state dye sensitized solar cells (DSSC).Paper ID: TITLE: A b s t r a c t :AUTHOR(S): 93Fabrication of Dye-Sensitized Solar Cells (DSSC) using Copper (I) Iodide: ASustainable Approach to Solar Energy ConversionAli Imran Bin Rozli Sham, Muhamad Firdaus Naim Bin Rozilani@Azman, Ayib Rosdi Bin Zainun, Izan Izwan Bin Misnon, Lin Jin Kiong, MohdHisham bin Arif, Norazian Binti Subari, Noor Zirwatul Ahlam BintiNaharuddinPaper ID: TITLE: AUTHOR(S): 55Reliability Assessments of Distributed Generation Penetration Level on PowerSystem NetworksAhmad Zairi Mohd Zain and Mohd Ikhwan Muhammad RidzuanReliability assessment represents a crucial role in assessing the stability and performance of power systemnetworks. With the integration of distributed generation (DG), the reliability dynamics of these networkshave undergone significant transformations. However, most of the existing research in this area haspredominantly focused on voltage and power loss-based DG placement strategies, neglecting theimportance of reliability-based placement methods. Therefore, this paper aims to bridge this gap byinvestigating the impact of placing DGs with different penetration levels on the effectiveness of networkreliability. To achieve this objective, the study utilizes the IEEE 9-bus and 14-bus systems as test cases forsimulating network reliability. The widely adopted Monte Carlo Simulation (MCS) approach is employedwithin comprehensively evaluate both customer-related and system reliability performance. Thesimulation outcomes yield important reliability indices, including System Average Interruption FrequencyIndex (SAIFI), System Average Interruption Duration Index (SAIDI), and Customer Average InterruptionDuration Index (CAIDI). The results demonstrate the varying impacts of DG placement and penetrationlevels on network reliability, unveiling the intricate relationship between these factors and overall systemreliability.A b s t r a c t :70

Page 71

ORGANIZING & TECHNICAL COMMITEESIr. Ts. Dr. Mohd Zamri bin IbrahimDr. Wan Syahirah binti W SamsudinPuan Norazian binti SubariPuan Nor Farizan binti ZakariaDr. Suliana binti Ab GhaniPuan Nor Fadzillah binti ZulkipliPuan Nornashua Farhani AbdulGhaffarTs. Dr. Zainah binti Md. ZainIr. Ts. Dr. Norizam bin SulaimanDr. Mahfuzah binti MustafaDr. Mohammed Nazmus ShakibP.M. Dr. Waheb Abdul Jabbar ShaifAbdullahTs. Dr. Mohd Shawal bin JadinDr. Mohd Syakirin bin RamliP.M. Dr. Mohd Mawardi bin SaariP.M. Dr. Mohd Herwan bin SulaimanDr. Amir Izzani bin MohamedTs. Dr. Saifudin bin RazaliDr. Mohd Amir Shahlan Mohd AsparTs. Dr. Raja Mohd Taufika Raja IsmailP.M. Dr. Abu Zaharin bin AhmadPROGRAM MANAGER & REGISTRATIONPUBLICATION CHAIR@EDITORSPUBLICATIONSTs. Nidzamuddeen bin IshakP.M. Ts. Dr. Hadi bin ManapEn. Omar bin AlimanEn. Azri bin IdrisEn. Mohd Maliki bin Md SaadEn. Mohd Nizam bin Md IsaEn. Ahmad Saifuddin bin AbdulMananIr. Dr. Ayib Rosdi bin Zainun Dr. Rosdiyana binti SamadPuan Rosyati binti HamidPuan Faradila binti NaimTs. Dr. Nurhafizah binti Abu Talip@ YusofDr. Norazila binti JaalamDr. Norhafidzah binti Mohd SaadEn. Mohd Falfazli bin Mat JusofPuan NurulFadzilah binti HasanEn. Ikhwan Hafiz bin MuhamadPuan Nurul Wahidah bintiArshadLOGISTIC & LOCAL ARRANGEMENTPROMOTION & SPONSORSHIPWEBSITE & MULTIMEDIAP a t r o nYBHG. Profesor Dato’ Ts. Dr. Yuserrie bin ZainuddinA d v i s o rProfesor Madya Dr. Hamdan bin DaniyalC h a i rIr. Ts. Dr. Norizam bin SulaimanS e c r e t a r yTs. Dr. Noorazliza binti Sulaiman & Puan Wahida binti HussinF i n a n c eEn. Mohd Redzuan bin Ahmad & Puan Kamisah binti Kamaruddin71

Page 72

Abdul Nasir Abd Ghafarabdnasir@ump.edu.myAbu Zaharin bin Ahmadzaharin@ump.edu.myAddie Irawanaddieirawan@ump.edu.myAhmad Afif Mohd Fauziafif@ump.edu.myAhmad Nor Kasruddin Nasirkasruddin@ump.edu.myAhmad Zaki Haji Shukorzaki@utem.edu.myAirul Sharizli Abdullahsharizli@ump.edu.myAmir Shahlan Asparamirs@ump.edu.myAmir Izzani Mohamedamirizzani@ump.edu.myAyib Rosdi Zainunayib@ump.edu.myBakri Hassanbakri@ump.edu.myDavid Al-Dabassdavid.aldabass@btinternet.comMohammed Nazmus Shakibnazmus@eee.green.edu.bdHadi bin Manaphadi@ump.edu.myDwi Prebianti dwipebrianti@iium.edu.myFahmi Samsuri fahmi@ump.edu.myHamzah Ahmadhamzah@ump.edu.myIkhwan Muhamadikhwanh@ump.edu.myIzzeldin Ibrahim Mohamedizzeldin@ump.edu.myKharudin Alikharudin@uctati.edu.myKrismadinata Krismadinatakrisma@ft.unp.ac.idMahfuzah Mustafamahfuzah@ump.edu.myMarlina Yaknomarlinayakno@ump.edu.myMaziyah Mat Nohmaziyah@ump.edu.myMohamad Shaiful Abdul Karimmshaiful@ump.edu.myMohammad Abasmfadhil@ump.edu.myLIST OF REVIEWERS72

Page 73

Mohd Razali Daudmrazali@ump.edu.myMohd Anwar Zawawimohdanwar@ump.edu.myMohd Ashraf Ahmadmashraf@ump.edu.myMohd Azri Abdul Azizazriaziz@uitm.edu.myMohd Falfazli Mat Jusofmfalfazli@ump.edu.myMohd Herwan Sulaimanherwan@ump.edu.myMohd Ikhwan Muhammad Ridzuanikhwanr@ump.edu.myMohd Ismifaizul Mohd Ismailismifaizul.ismail@mimos.myMohd Mawardi Saarimmawardi@ump.edu.myMohd Riduan Ghazaliriduwan@ump.edu.myMohd Shawal Jadinmohdshawal@ump.edu.myMohd Syakirin Ramlisyakirin@ump.edu.myMohd Zamri Ibrahimzamri@ump.edu.myMuhamad Zahim Sujodzahim@ump.edu.myMuhammad Sharfi bin Najibsharfi@ump.edu.myNoor Zaihah Jamalzaihah@ump.edu.myNoor Zirwatul Ahlam Naharuddinzirwatul@ump.edu.myNoorazliza Sulaimanazliza@ump.edu.myNor Farizan Zakarianorfarizan@ump.edu.myNor Hana Mamatnorhana@uctati.edu.myNor Rul Hasma Abdulahhasma@ump.edu.myNorazian Subariaziansubari@ump.edu.myNorazila Jaalamzila@ump.edu.myNorizam Bin Sulaimannorizam@ump.edu.myNurhafizah Abu Talip @ Yusofhafizahs@ump.edu.myNurulfadzilah Hasannurulfadzilah@ump.edu.myLIST OF REVIEWERS73

PROGRAM BOOK InECCE2023

Page 1

Page 2

Page 3