IIUM Engineering Journal
Not a member yet
783 research outputs found
Sort by
Deep Learning-Based Skin Care Detection with Multi-method Explainability: Grad-CAM, Lime, and Occlusion Sensitivity
Full text linkSkin cancer is one of the most common malignancies worldwide, where early detection significantly improves treatment outcomes. While deep learning models show promise for automated skin lesion classification, their lack of interpretability limits clinical adoption. This study presents a comprehensive comparative analysis of three convolutional neural networks, ResNet-50, GoogLeNet, and SqueezeNet, for binary skin lesion classification (benign vs. malignant), integrating three explainable AI (XAI) methods (Grad-CAM, LIME, and Occlusion Sensitivity) to enhance clinical interpretability. We trained and evaluated these architectures on the Kaggle Skin Cancer dataset, which contains 2,637 dermoscopic images (1,440 benign, 1,197 malignant). Transfer learning employed ImageNet pre-trained weights with two-stage fine-tuning. Performance was assessed using accuracy, precision, recall, F1-score, specificity, and AUC-ROC metrics. ResNet-50 achieved the highest accuracy of 91.36% with an excellent AUC of 0.9721, demonstrating superior balanced performance. GoogLeNet achieved 88.94% accuracy with 73% fewer parameters, offering an optimal accuracy-efficiency trade-off. The proposed lightweight CNN, despite having the fewest parameters (1.2M), achieved 85.45% accuracy and a malignancy detection sensitivity of 92.7%, making it well-suited for screening applications. Training times ranged from 1.5 minutes (SqueezeNet) to 3 minutes 39 seconds (ResNet-50), demonstrating feasibility for resource-constrained settings. All XAI methods successfully generated clinically meaningful explanations, with models consistently focusing on lesion centers, color variations, and irregular borders. This study demonstrates that combining deep learning with XAI enables accurate and interpretable skin cancer detection. ResNet-50 is well-suited to well-resourced clinical settings, GoogLeNet offers balanced performance for resource-constrained deployments, and SqueezeNet enables mobile telemedicine applications with superior sensitivity.
ABSTRAK: Kanser kulit merupakan antara malignansi yang paling lazim di seluruh dunia, dan pengesanan awal terbukti dapat meningkatkan keberkesanan rawatan secara signifikan. Walaupun model pembelajaran mendalam menunjukkan potensi tinggi dalam pengelasan automatik lesi kulit, kekurangan kebolehinterpretasian telah mengehadkan penerimaan klinikal. Kajian ini membentangkan analisis perbandingan menyeluruh terhadap tiga rangkaian neural konvolusi, iaitu ResNet-50, GoogLeNet, dan SqueezeNet, bagi pengelasan binari lesi kulit (jinak vs. malignan), digabungkan dengan tiga kaedah kecerdasan buatan boleh jelas (XAI), iaitu Grad-CAM, LIME, dan Kepekaan Halangan, bagi menyokong interpretasi klinikal. Model dilatih dan dinilai menggunakan set data Kanser Kulit Kaggle yang mengandungi 2,637 imej dermoskopi, dengan menggunakan pembelajaran pindahan berasaskan pemberat pralatih ImageNet dan penalaan halus dua peringkat. Penilaian prestasi menggunakan metrik ketepatan, ketepatan ramalan, kepekaan, skor F1, pengkhususan, dan AUC-ROC menunjukkan bahawa ResNet-50 mencapai prestasi tertinggi dengan ketepatan 91.36% dan AUC 0.9721, manakala GoogLeNet menawarkan keseimbangan optimum antara ketepatan dan kecekapan dengan pengurangan parameter sebanyak 73%. SqueezeNet, walaupun paling ringan, mencapai kepekaan pengesanan malignan tertinggi sebanyak 92.7%, menjadikannya sesuai untuk aplikasi saringan dan teleperubatan mudah alih. Semua kaedah XAI berjaya menghasilkan penjelasan bermakna secara klinikal, dengan fokus konsisten pada pusat lesi, variasi warna, dan sempadan tidak sekata. Secara keseluruhan, kajian ini membuktikan bahawa penggabungan pembelajaran mendalam dan XAI membolehkan pengesanan kanser kulit yang tepat, boleh ditafsir, dan sesuai dalam pelbagai kekangan sumber klinikal
Evaluating Mechanical and Conductivity of Graphene-Silver Hybrid Inks on Copper Substrate at Elevated Temperature
Full text linkThis study investigates the effect of temperature on the damping behaviour, stiffness, and natural frequency of hybrid conductive ink (HCI) printed on a copper (Cu) substrate. The HCI comprises graphene nanoplatelets (GNPs), silver flakes (SF), and silver acetate (SA). The objective is to evaluate the HCI’s electrical conductivity and mechanical properties under varying thermal conditions. The HCI paste was formulated with a specified ratio of organic solvents, terpineol, and 1-butanol and cured in an oven at 260°C for 3 hours. The baseline droplet ratio was set at 1:1, and three Cu samples with varied HCl compositions were printed using a 60 µm mesh stencil process. The terpineol concentrations were changed while the 1-butanol droplet remained constant. The samples were then tested for electrical conductivity at room temperature using a two-point probe, in accordance with IEEE Std 118-1978, followed by mechanical behaviour testing using an ASTM E756-05 impact test. Furthermore, the samples were exposed to a range of temperature tests to evaluate mechanical and electrical conductivity under thermal stress. The results showed that the baseline composition exhibited minimal resistance and resistivity across the temperature range, with an average resistance of ? 0.2 ? and a resistivity of ? 0.8 ? · mm, respectively. The baseline composition also exhibited reduced stiffness and damping, with natural frequencies of 10.90, 1.40 kN/m, and 37.51 Hz across the samples. Therefore, the baseline composition exhibits relatively good electrical and mechanical properties for applications in flexible electronics.
ABSTRAK: Kajian ini meneliti kesan suhu terhadap sifat redaman, kekakuan, dan frekuensi semula jadi bagi dakwat konduktif hibrid (HCI) yang dicetak pada substrat kuprum (Cu). HCI ini terdiri daripada graphene platelet nano (GNP), serpihan perak (SF), dan perak asetat (SA). Objektif kajian ini adalah bagi menilai kekonduksian elektrik dan sifat mekanikal HCI di bawah keadaan haba berbeza. Pes HCI dibangunkan dengan nisbah tertentu pelarut organik, terpineol dan 1-butanol, dan dikeringkan dalam ketuhar pada suhu 260?C selama tiga jam. Nisbah titisan asas ditetapkan pada 1:1, dan tiga sampel Cu dengan komposisi HCI berbeza telah dicetak menggunakan proses stensil jaring 60 µm. Kepekatan terpinol diubah manakala titisan 1-butanol dikekalkan pada kadar yang sama. Sampel-sampel tersebut kemudiannya diuji bagi kekonduksian elektrik pada suhu bilik menggunakan probe Dua-Titik mengikut kod ujian piawaian IEEE Std 118-1978 dan diteruskan dengan ujian sifat mekanikal menggunakan ujian impak ASTM E756-05. Tambahan, sampel telah terdedah kepada pelbagai ujian suhu bagi menilai kekonduksian mekanikal dan elektrik di bawah tekanan haba. Dapatan kajian menunjukkan bahawa komposisi asas menunjukkan rintangan dan kerintangan minimum sepanjang julat suhu, dengan purata rintangan ? 0.2 ? dan kerintangan ? 0.8 ?/mm. Komposisi asas juga menunjukkan sifat mekanikal dengan kekakuan berkurangan, tingkah laku redaman, dan frekuensi semula jadi masing-masing sebanyak 10.90, 1.40 kN/m, dan 37.51 Hz merentasi sampel. Oleh itu, komposisi asas menunjukkan sifat elektrik dan mekanikal yang agak baik bagi aplikasi elektronik fleksibel
Optimizing Energy Efficiency in Three-Phase Induction Motors via GWO-Tuned PID Control
Full text linkInduction motors consume a significant share of industrial electricity, making their efficiency a crucial aspect of sustainable energy management. Traditional Proportional–Integral–Derivative (PID) controllers are commonly used to regulate motor performance; however, their fixed parameters often fail to maintain optimal control under varying load conditions. Although optimization methods, such as Genetic Algorithms and Particle Swarm Optimization, have been introduced to enhance PID tuning, they often encounter challenges, including premature convergence and limited adaptability. This creates a clear need for an optimization strategy that is both robust and dynamically responsive to ensure energy-efficient motor operation. To address this gap, this study introduces a Grey Wolf Optimization (GWO)-based PID tuning strategy that distinguishes itself from existing methods by achieving a superior adaptive balance between exploration and exploitation. This characteristic enables the controller to maintain stable, responsive performance even under fluctuating load conditions. Experimental results confirm that the proposed GWO-PID controller successfully reduces motor current from 285.21 A to 164.2 A and lowers power consumption from 150 kW to 84.5 kW, achieving a 42.4% reduction in current and a 44.7% improvement in energy efficiency. Additionally, electricity costs decrease by 43.5%, demonstrating strong economic potential. The novelty of this research lies in integrating GWO’s adaptive intelligence with PID control, yielding a more effective, reliable, and energy-efficient solution than existing optimization-based controllers for industrial induction motor systems.
ABSTRAK: Motor aruhan menggunakan sebahagian besar tenaga elektrik industri, menjadikan kecekapan operasi satu aspek penting dalam pengurusan tenaga mampan. Pengawal konvensional Kadar-Integral-Pembezaan (PID) lazim digunakan bagi mengawal prestasi motor; namun, parameter tetapnya sering gagal mengekalkan kawalan optimum di bawah keadaan beban berubah. Walaupun kaedah pengoptimuman seperti Algoritma Genetik dan Pengoptimuman Kawanan Partikel telah diperkenalkan bagi menambah baik talaan PID, pendekatan ini sering menghadapi masalah penumpuan awal dan keupayaan adaptasi yang terhad. Hal ini memerlukan satu strategi pengoptimuman yang lebih mantap dan responsif secara dinamik bagi memastikan operasi motor cekap tenaga. Bagi menangani masalah ini, kajian ini memperkenalkan satu strategi talaan PID berasaskan Pengoptimuman Serigala Kelabu (Grey Wolf Optimization, GWO) yang menonjol dari kaedah sedia ada melalui keseimbangan adaptifnya yang unggul pada penerokaan dan pengeksploitasian. Ciri ini membolehkan pengawal mengekalkan prestasi stabil dan responsif walaupun berkeadaan beban yang berubah. Dapatan kajian melalui eksperimen menunjukkan bahawa pengawal GWO–PID yang dicadangkan ini berjaya mengurangkan arus motor daripada 285.21 A kepada 164.2 A dan menurunkan penggunaan kuasa daripada 150 kW kepada 84.5 kW—mencapai pengurangan arus sebanyak 42.4% dan peningkatan kecekapan tenaga sebanyak 44.7%. Selain itu, kos elektrik turut menurun sebanyak 43.5%, sekaligus membuktikan potensi ekonomi yang kukuh. Keunikan kajian ini terletak pada integrasi kecerdasan adaptif GWO dengan kawalan PID, menawarkan penyelesaian lebih berkesan, boleh dipercayai, dan cekap tenaga berbanding pengawal berasas pengoptimuman lain bagi sistem motor aruhan industri
Editorial Message Vol. 27 No. 1 2026
Full text linkIt is with great pleasure that we present Volume 27, Number 1 of the IIUM Engineering Journal, an issue that reflects the breadth, depth, and evolving priorities of contemporary engineering research. The papers published in this volume collectively demonstrate how rigorous theoretical foundations, advanced computational techniques, and practical engineering solutions converge to address pressing technological, environmental, and societal challenges.This issue opens with contributions from Chemical and Biotechnology Engineering, where research on the growth kinetics of Pleurotus pulmonarius advances sustainable biotechnology practices through innovative methods to revive spawn culture. This work highlights how process optimization and biological engineering can contribute to food security and value-added agro-industrial applications.A substantial portion of this volume is devoted to Electrical, Computer, and Communications Engineering, underscoring the rapid progress in intelligent systems, data-driven modeling, and energy-aware technologies. Several papers explore the integration of artificial intelligence and machine learning into diverse domains, including graph neural networks for large-scale skyline query processing, transformer-based malware detection at the byte level, deep learning with explainability for skin care detection, and advanced sentiment prediction models. Complementing these are applied engineering studies on inertial sensor self-calibration for autonomous underwater vehicles, LoRaWAN-based IoT systems for electric fence monitoring, optimized PID control for energy-efficient induction motors, and vision-transformer-based groundwater-level forecasting for sustainable water-resource management. Together, these works illustrate how AI-enhanced engineering is reshaping decision-making, automation, and system reliability.In the Materials and Manufacturing Engineering section, the focus shifts to next-generation materials and devices that support sustainability and high-performance applications. Contributions on eco-friendly tin-based perovskite solar cells and graphene–silver hybrid inks provide valuable insights into material characterization, performance optimization, and the development of greener energy and electronic solutions. The Mechanical and Aerospace Engineering contribution in this issue addresses surrogate model-based optimization of unmanned aerial vehicle wings, demonstrating how computational intelligence and response surface methodologies can accelerate design cycles while maintaining aerodynamic efficiency and structural integrity.Research in Mechatronics and Automation Engineering further enriches this volume, covering a wide spectrum of intelligent and autonomous systems. Papers in this section examine humanoid robot stability on uneven terrain, vehicle identification using YOLO-based deep learning, autonomous underwater vehicles for subaquatic exploration, FSM-based MBIST controller implementation, adaptive energy balance control for solar-powered hydroponics, and advanced medical image segmentation using attention-enhanced neural architectures. Collectively, these studies highlight the increasing role of autonomy, robustness, and sustainability in modern engineered systems.Taken as a whole, Vol. 27 No. 1 exemplifies the journal’s commitment to publishing high-quality, interdisciplinary engineering research that bridges theory and practice. The diversity of topics and methodologies presented here reflects not only the dynamic nature of engineering as a discipline but also its critical role in supporting sustainable development, digital transformation, and technological innovation.We extend our sincere appreciation to the authors for their valuable contributions, to the reviewers for their rigorous and constructive evaluations, and to the editorial team for their dedication in bringing this issue to publication. We hope that the articles in this volume will stimulate further research, inspire collaboration across disciplines, and serve as a meaningful reference for researchers, practitioners, and policymakers alike.
Prof. Ir. Ts. Dr. Teddy Surya GunawanExecutive EditorIIUM Engineering Journal
ISSN: 1511-788X E-ISSN: 2289-7860
Published by:IIUM Press,International Islamic University MalaysiaJalan Gombak, 53100 Kuala Lumpur, MalaysiaPhone (+603) 6421-5014, Fax: (+603) 6421-629
ORCA: AI-powered Autonomous Underwater Vehicle for Subaquatic Exploration
Full text linkThis paper presents research and development of an Autonomous Underwater Vehicle (AUV), named ORCA, to perform underwater missions independently without human intervention. ORCA plays a vital role in challenges that test robots' abilities in navigation, exploration, and interaction with the aquatic environment. Using advanced design tools, the AUV is meticulously designed in Inventor and manufactured via CNC machining, laser cutting, and 3D printing. We concentrate on the vehicle's design, manufacturing processes, control systems, PID controllers, and vision systems. Subsequently, the research and development effort is expanded to incorporate critical functionalities, including environmental perception, object detection, deep learning algorithms, and path-planning strategies. The culmination of this research has produced an AUV capable of autonomous underwater navigation, effective obstacle avoidance, efficient object detection, and precise payload manipulation using a gripper mechanism. ORCA has a comprehensive sensor suite comprising a BNO055 IMU, an MS5803-14BA pressure sensor for depth measurement, and a Logitech C525 camera for image processing. The system integration not only enhances the vehicle's operational capabilities but also represents a significant advancement in underwater robotics.
ABSTRAK: Kajian ini membentangkan penyelidikan dan pembangunan Kenderaan dalam Air Berautonomi (AUV) yang dikenali sebagai ORCA, dibangunkan bagi tujuan misi bawah air secara autonomi tanpa sebarang campur tangan manusia. ORCA memainkan peranan penting dalam menghadapi cabaran menguji kebolehan robot dalam aspek navigasi, penerokaan, serta interaksi dengan persekitaran akuatik. Dengan memanfaatkan alat reka bentuk yang canggih, AUV ini direka bentuk dengan teliti menggunakan perisian Inventor dan dihasilkan melalui teknik pemesinan CNC, pemotongan laser, dan percetakan 3D. Penekanan utama dalam penyelidikan ini adalah pada reka bentuk kenderaan, proses pembuatan, sistem kawalan, pengawal PID, serta sistem penglihatan. Selain itu, penyelidikan dan pembangunan ini turut diperluaskan bagi menyertakan fungsi-fungsi kritikal seperti persepsi persekitaran, pengesanan objek, algoritma pembelajaran mendalam, dan strategi perancangan laluan. Dapatan kajian ini telah menghasilkan AUV yang mampu melaksanakan navigasi bawah air secara autonomi, menghindari halangan dengan berkesan, mengesan objek dengan cekap, serta mengendali beban dengan tepat menggunakan mekanisme pemegang. ORCA dilengkapi dengan suit pengesan yang komprehensif, termasuk pengesan IMU BNO055, pengesan tekanan MS5803-14BA bagi pengukuran kedalaman, serta kamera Logitech C525 bagi pemprosesan visual. Integrasi sistem ini bukan sahaja meningkatkan keupayaan operasi kenderaan tetapi juga mewakili satu langkah penting dalam perkembangan robotik bawah air
Development of Electric Fence Fault Sensing and Monitoring System with LoRaWAN IoT
Full text linkThe persistent challenge of Human-Elephant Conflict (HEC) in regions like Malaysia necessitates robust and efficient mitigation strategies. Electric fences are effective but often face maintenance inefficiencies due to delayed fault detection. This study presents a smart electric fence monitoring system designed for real-time fault diagnosis and localisation. The system employs IoT-enabled in-place sensor nodes comprising 10 kV voltage sensors, short-circuit detection sensors, and 3-axis gyroscope sensors. Sensor data is transmitted via a LoRaWAN network, selected for its long-range, low-power characteristics, which are well-suited to rural, low-bandwidth environments where electric fences are typically deployed. Field validation of a 50 m, 10 kV, 6 A electric fence segment achieved 100% voltage and short-circuit detection rates and 99.91% gyroscope tilt accuracy. Reliable data transmission was maintained up to 1.3 km, with an RSSI of -110 dBm, in campus environments with concrete obstructions. Supplementary testing at the same positions using antennas at increased height yielded an RSSI of -79 dBm, with a 41 dB link margin, highlighting the potential for range scaling in future work. The system has so far been validated on a short 50 m electric fence segment with a practical LoRaWAN range of 1.3 km under campus conditions, indicating the need for further optimisation and field-scale trials. The system provides a practical solution for the Department of Wildlife and National Parks, Peninsular Malaysia (PERHILITAN), with direct application to their Sistem Pagar Elektrik Gajah (SPEG) for HEC mitigation. Beyond this application, the approach demonstrates potential for future use of IoT-based Structural Health Monitoring (SHM) concepts in resource-constrained rural infrastructures.
ABSTRAK: Cabaran berterusan Konflik Manusia-Gajah (HEC) di kawasan seperti Malaysia memerlukan strategi mitigasi yang kukuh dan berkesan. Pagar elektrik berkesan dalam menangani konflik ini, tetapi sering berhadapan masalah penyelenggaraan akibat kelewatan pengesanan kerosakan pada pagar. Kajian ini membentangkan satu sistem pemantauan pagar elektrik pintar yang direka bagi mengdiagnosis penentuan lokasi kerosakan secara masa nyata. Sistem ini menggunakan nod pengesan dengan keupayaan internet benda terdiri daripada pengesan voltan 10 kV, pengesan pengesanan litar pintas, dan pengesan giroskop. Data pengesan dihantar melalui rangkaian LoRaWAN, yang dipilih atas dasar rangkaian jarak jauh dan penggunaan tenaga minimal. Ciri-ciri ini dianggap bersesuaian bagi persekitaran luar bandar di mana pagar elektrik biasanya dipasang, kerana kawasan ini seringkali tidak mendapat rangkaian telekomunikasi komersial. Ujian lapangan pada segmen pagar elektrik sepanjang 50 m, 10 kV 6 A telah berjaya mengesan voltan dan litar pintas dengan ketepatan 100%, manakala pengesan giroskop pula mampu mengesan kecondongan pada kadar 99.91%. Penghantaran data daripada pengesan pula berjaya mencapai jarak maximum 1.3 km dengan RSSI -110 dBm dalam persekitaran kampus dengan penghalang konkrit. Ujian sampingan penghantaran data yang dilaksanakan menggunakan antena berkedudukan tinggi pada lokasi dan jarak yang sama pula, menunjukkan RSSI -79 dBm dengan margin pautan 41 dB. Ini menunjukkan potensi peningkatan jarak penghantaran data untuk kajian seterusnya. Pasa masa ini, sistem ini telah disahkan kepenggunaanya bagi segmen pagar elektrik pendek dengan kepanjangan 50 m, manakala penghantaran data secara berkesan dihadkan pada 1.3 km dalam persekitaran kampus. Had ini menunjukkan, wujud keperluan bagi kerja-kerja penambah baikan bagi meningkatkan kadar keberkesanan melalui ujian berskala lapangan pada masa depan. Sistem ini menyediakan penyelesaian kejuruteraan yang praktikal untuk Jabatan Perlindungan Hidupan Liar dan Taman Negara Semenanjung Malaysia (PERHILITAN), dengan aplikasi langsung pada Sistem Pagar Elektrik Gajah (SPEG) bagi mitigasi HEC. Kajian ini juga menunjukkan potensi bagi penggunaan konsep Pemantauan Kesihatan Struktur (SHM) berasaskan IoT pada masa hadapan dalam infrastruktur luar bandar
Development of a Model for Malware Detection and Classification at the Byte Level Based on Transformer
Full text linkMalware threats have been a critical concern in cybersecurity, particularly due to the increasing complexity and constantly evolving variants that are difficult to detect using conventional signature-based or static rule-based methods. This research focused on developing a Transformer-based model at the byte level to detect and classify malware effectively and adaptively, thereby streamlining the analytical process without requiring specialized feature tokenization. The primary objective was to design and evaluate a Transformer model that captures universal and adaptive malware patterns directly from raw byte representations, enabling cross-platform applicability. A quantitative experimental approach was employed using three public datasets: Malware Detection PE-Based Analysis, MC-dataset-binary, and Malware.zip. Data processing involved byte embedding, dilated 1D convolution, multi-head self-attention, and attention pooling. Model optimization was conducted using AdamW with a combined scheduler, Stochastic Weight Averaging (SWA), random byte masking augmentation, and Mixup Embedding. Experimental results showed that the byte-level Transformer model achieved high classification accuracy across the three datasets, namely 99%, 92%, and 94%, respectively. These results demonstrate that a byte-level Transformer can effectively capture universal malware patterns in binary data, offering a flexible and highly accurate approach to developing resilient defenses against modern cyber threats.
ABSTRAK: Ancaman perisian perosak (malware) merupakan isu kritikal dalam keselamatan siber, terutama apabila disebabkan oleh peningkatan kerumitan dan variasi yang sentiasa berevolusi, sekaligus menyukarkan pengesanan menggunakan kaedah konvensional berasaskan tandatangan atau peraturan statik. Kajian ini memfokuskan kepada pembangunan model berasaskan Transformasi pada peringkat bait (byte-level) bagi mengesan dan mengklasifikasikan perisian perosak secara berkesan dan adaptif, tanpa memerlukan pengekstrakan atau penandaan ciri khusus. Objektif utama kajian adalah bagi mereka bentuk dan menilai keberkesanan model Transformasi yang mampu menangkap corak perisian perosak bersifat universal dan adaptif secara langsung daripada representasi bait mentah, seterusnya membolehkan kebolehgunaan merentas platform. Pendekatan eksperimen kuantitatif digunakan dengan melibatkan tiga set data awam, iaitu Analisis Berdasarkan Pengesanan Perisian Perosak PE, Set Data Binari MC, dan Malware.zip. Pemprosesan data merangkumi pembenaman bait, konvolusi 1D terlarut (dilated), perhatian kendiri berbilang kepala, serta pengumpulan berasaskan perhatian. Pengoptimuman model dilaksanakan menggunakan AdamW dengan penjadual gabungan, Purata Pemberat Stokastik (SWA), penambahan data melalui penyamaran bait rawak, dan Penggabungan Embedding. Dapatan kajian melalui eksperimen menunjukkan bahawa model Transformasi peringkat bait mencapai ketepatan pengelasan tertinggi, iaitu masing-masing 99%, 92%, dan 94% bagi ketiga-tiga set data. Dapatan ini membuktikan bahawa Transformasi peringkat bait berupaya menangkap corak perisian perosak universal daripada data binari, sekaligus menawarkan pendekatan fleksibel dan berketepatan tinggi bagi membangunkan pertahanan yang lebih berdaya tahan terhadap ancaman siber moden
Performance Investigation and Efficiency Enhancement of Eco-Friendly Tin-Based CH3NH3SnI3 Perovskite Solar Cell via SCAPS-1D
Full text linkHalide perovskite materials, particularly lead-based CH3NH3PbI3, have garnered significant attention in the PV industry for their exceptional efficiency in solar cell applications. However, due to the toxicity of lead, research interest has shifted toward Sn-based alternatives. This study explores a lead-free Sn-based perovskite solar cell (PSC) with the structure ITO/TiO2/CH3NH3SnI3/CBTS/Ni, where CH3NH3SnI3 (MASnI3) serves as the absorber material, TiO? as the electron transport layer (ETL), Cu2BaSnS4 (CBTS) as the hole transporting layer (HTL). Device performance is analyzed using the SCAPS-1D simulation software. The impact of key performance-determining parameters, including the thickness, doping density, and defect density of the absorber, ETL, and HTL, has been accounted for. The proposed PSC architecture, optimized for key performance-determining parameters, achieves a power conversion efficiency PCE (?) of 27.28%, an open-circuit voltage (VOC) of 1.0283 V, a fill factor (FF) of 83.62%, and a short-circuit current density (JSC) of 31.72 mA/cm2. This study examines the influence of interface defect density, shunt and series resistances, back-contact metal work function, and operating temperature on the performance of PSCs. Furthermore, the analysis includes current density-voltage (J-V) and quantum efficiency (QE) characteristics to provide a comprehensive evaluation of the effectiveness of the proposed PSC.
ABSTRAK: Bahan perovskit halida, khususnya CH?NH?PbI? berasaskan plumbum, telah menarik perhatian besar dalam industri fotovolta (PV) berikutan kecekapan tinggi dalam aplikasi sel suria; namun, isu ketoksikan plumbum telah mengalih tumpuan penyelidikan kepada alternatif berasaskan timah (Sn). Kajian ini meneroka sel suria perovskit (PSC) bebas plumbum berasaskan Sn dengan seni bina ITO/TiO?/CH?NH?SnI?/CBTS/Ni, di mana CH?NH?SnI? (MASnI?) bertindak sebagai bahan penyerap, TiO? sebagai lapisan pengangkut elektron (ETL), dan Cu?BaSnS? (CBTS) sebagai lapisan pengangkut lubang (HTL). Prestasi peranti dianalisa menggunakan perisian simulasi SCAPS-1D dengan mengambil kira parameter penentu prestasi utama, termasuk ketebalan, ketumpatan pendopan, dan ketumpatan kecacatan bagi lapisan penyerap, ETL dan HTL. Seni bina PSC yang dicadangkan, selepas pengoptimuman parameter, mencapai kecekapan penukaran kuasa (PCE) sebanyak 27.28%, voltan litar terbuka (V_OC) 1.0283 V, faktor pengisian (FF) 83.62%, dan ketumpatan arus litar pintas (J_SC) 31.72 mA/cm². Kajian ini turut menilai pengaruh ketumpatan kecacatan antara muka, rintangan siri dan pirau, fungsi kerja logam sentuhan belakang, serta suhu operasi terhadap prestasi PSC. Di samping itu, analisis ciri ketumpatan arus–voltan (J–V) dan kecekapan kuantum (QE) disertakan bagi memberikan penilaian menyeluruh terhadap keberkesanan sel suria perovskit yang dicadangkan
FSM-Based MBIST Controller Implementation using the March AZ2 Test Algorithm
Full text linkAn efficient on-chip memory testing requires the use of a Memory Built-In Self-Test (MBIST) that applies a low-complexity test algorithm that offers excellent fault coverage, to ensure high test quality at a minimal cost. The March AZ2 algorithm, with 14N complexity, was previously established to balance the fault coverage and test complexity. It was previously implemented in an MBIST controller through an automatic generation process using an Electronic Design Automation (EDA) tool, which limits its customizability and optimization potential. This paper presents the implementation of an MBIST controller that employs the March AZ2 test algorithm and is based on a Finite-State Machine (FSM) architecture. The developed MBIST controller was implemented on a Field-Programmable Gate Array (FPGA) to validate its functionality and fault coverage. A comparison with an equivalent MBIST controller automatically generated using Siemens EDA Tessent MemoryBIST tool demonstrates that the proposed FSM-Based MBIST controller achieves 84% lower circuit area and 32% power consumption, while offering similar fault coverage.
ABSTRAK: Pengujian memori atas cip yang cekap memerlukan penggunaan Ujian-Kendiri Terbina-Dalam Memori (MBIST) menggunakan algoritma ujian berkompleksiti rendah dan menawarkan liputan kesalahan yang tinggi, bagi memastikan kualiti ujian baik pada kos minimum. Algoritma March AZ2, dengan kekompleksan 14N, telah diperkenalkan sebelum ini bagi mengimbangi antara liputan kesalahan dan kompleksiti ujian. Ia telah dilaksanakan dalam pengawal MBIST melalui proses penjanaan automatik menggunakan Automasi Reka Bentuk Elektronik (EDA), yang menghadkan tahap kebolehsuaian dan potensi pengoptimuman. Kajian ini membentangkan pelaksanaan pengawal MBIST yang menggunakan algoritma ujian March AZ2, berasaskan seni bina Mesin Keadaan-Terhingga (FSM). Pengawal MBIST yang dibangunkan telah dilaksanakan pada Litar Terpadu-Serba Guna (FPGA) bagi mengesahkan fungsi dan liputan kesalahan. Perbandingan dengan pengawal MBIST yang dijana secara automatik menggunakan Siemens EDA Tessent MemoryBIST membuktikan bahawa pengawal MBIST berasaskan FSM yang dicadangkan mengurangkan keluasan litar sebanyak 84% dan penggunaan kuasa 32% lebih rendah, sambil mengekalkan liputan kesalahan yang sama
Vehicle Identification and Classification Using YOLO Algorithm
Full text linkVehicle identification and classification are among the challenging activities for the management and control of a large number of different vehicles moving in the inner city. Among many identification and classification systems, the YOLO algorithm stands out for its ability to analyze at high speed and with high accuracy. The algorithm is continually evolving, with notable versions including YOLOv8. This research presents a method for identifying and classifying vehicles using the YOLOv8 algorithm. The assessment of the proposed method's effectiveness was conducted using two COCO datasets (328,000 images) and a real-world dataset from Ho Chi Minh City (HCMC) with more than 1,000 images. The findings indicate that the proposed method can be applied to identify and classify vehicles with an accuracy of 93%-98%. Comparative results with prior studies also demonstrate the superiority of the YOLOv8 algorithm.
ABSTRAK: Pengecaman dan pengelasan kenderaan adalah salah satu aktiviti mencabar dalam pengurusan dan kawalan sejumlah besar kenderaan berbeza yang bergerak di bandar. Di antara kebanyakan sistem pengecaman dan pengelasan, algoritma YOLO menonjol kerana keupayaannya menganalisa pada kelajuan berketepatan tinggi. Algoritma ini terus dibangunkan dengan penambahbaikan yang banyak dan versi yang ketara ialah YOLOv8. Penyelidikan ini membentangkan kaedah mengenal pasti dan mengelaskan kenderaan menggunakan algoritma YOLOv8. Penemuan penilaian keberkesanan kaedah yang dicadangkan telah dijalankan menggunakan dua set data COCO dengan 328,000 imej dan set data sebenar di Ho Chi Minh City (HCMC) lebih daripada 1,000 imej. Dapatan kajian menunjukkan kaedah ini diaplikasikan dalam mengenal pasti dan pengelasan kenderaan berketepatan 93% hingga 98%. Hasil perbandingan dengan kajian lepas juga menunjukkan keunggulan algoritma YOLOv8