Bulletin of Electrical Engineering and Informatics
Not a member yet
2885 research outputs found
Sort by
Optimization of parabolic photovoltaic and solar thermal tracker model systems using heat transfer fluid
Get PDFRenewable energy sources such as photovoltaic (PV) and solar thermal (T) used in tropical Indonesia are still not optimal in converting electrical energy. This is caused by the movement of the sun which is not centered on the surface of the photovoltaic/thermal (PV/T). Based on this, PV/T technology was developed with a parabolic tracker system combined with T control with the aim of optimizing electrical energy conversion. The testing technique was carried out using an experimental method by comparing the installation of PV/T technology with a parabolic tracker system and a flat PV position. In the PV/T parabolic tracker, the supporting components use aluminum foil technology and heat transfer fluid (HTF) to reduce excess heat on the panel surface. The parabolic tracker system is implemented through a centralized system following the sun and aluminum foil and HTF technology function to keep the surface temperature stable. The results of PV performance testing at peak loads carried out at 10:00 am-3:00 pm, the installation of flat PV produced an average DC power of 71.68 Wp and the parabolic tracker system produced an increase in average DC power of 84.57 Wp, while with an increase in power of 17.98% during fluctuating weather conditions
Predicting the intention to adopt e-zakat payment services: a machine learning approach
Get PDFThe technology evolution in the zakat collection and payment services has brought about a profound transformation in the global processes of gathering and distributing charitable contributions. Despite witnessing a positive trend in annual zakat collection in Malaysia, it has yet to reach its optimal level. Therefore, predictions regarding performance and comparisons across multiple models for online zakat collection hold crucial significance in improving the overall collection rate. This paper, utilizing data from 230 zakat payers, presents an empirical assessment of various machine learning algorithms aimed at predicting zakat payer intentions when utilizing online platforms for zakat payments. Additionally, this paper presents the analysis of machine learning features importance to justify the effect of technology acceptance model (TAM) and theory of technology readiness (TR) attributes in the machine learning algorithms for predicting e-zakat payment service adoption intention. The findings show that many of the machine learning models are able to perform for highly accurate results, with most achieving over 80% accuracy. The most crucial attribute influencing these predictions was found to be the TAM. This study's methodology is designed to be easily replicable, allowing for further detailed exploration of both the influencing factors and the machine learning algorithms used
Enhancing realism in handwritten text images with generative adversarial networks
Get PDFImage synthesis is particularly important for applications that want to create realistic handwritten documents, which is why handwritten text generation is a critical area within its domain. Even with today's highly advanced technology, generating diverse and accurate representations of human handwriting is still a tough problem because of the variability in style. In this study, we tackle the problem of instability during the training phase of generative adversarial networks (GANs) for generating handwritten text images. Using the MNIST dataset, which includes 60,000 training and 10,000 test images of handwritten digits, we trained a GAN model to generate synthetic handwritten images. The methodology involves optimizing both the generator and discriminator using adversarial training, binary cross-entropy loss, as well as the optimizer Adam. A brand-new decaying learning rate schedule was introduced to speed up convergence. Performance was evaluated using the Fréchet inception distance (FID) metric. The results show that this model effectively generated high-quality synthetic images of handwritten digits, which resembled real data closely in the face of it all and also that there was a steady reduction in FID scores across epochs indicating improved performance
Integrating low-cost vision for autonomous tracking in assistive robots
No full textThis study presents the implementation of a real-time tracking system for the ARMOS TurtleBot, a robot designed for assistive applications in domestic environments. The system integrates two OmniVision 7670 (OV7670) camera modules positioned 7 cm apart to emulate human-like stereoscopic vision, enabling depth perception and three-dimensional object tracking. An embedded system platform 32-bit (ESP32) microcontroller captures and processes images from both cameras, calculates disparities, and transmits data to a Raspberry Pi via WebSockets. The Raspberry Pi, equipped with robot operating system (ROS), performs further analysis using open computer vision (OpenCV) and visualizes results in real-time with ROS visualization (RViz), allowing the robot to autonomously track moving objects such as humans or pets. Key optimizations, including image resolution reduction and data filtering, were implemented to enhance processing efficiency within the hardware constraints. The proposed approach demonstrates the feasibility of low-cost, real-time object tracking in assistive robotics, highlighting its potential for applications that require humanrobot interaction in dynamic indoor settings. This work contributes to the field by providing a practical solution for integrating stereoscopic vision and real-time decision-making capabilities into small-scale robots, promoting further research and development in affordable robotic assistance systems
A single-user electronic ticketing system using ERC-721 protocol for smart contracts
Get PDFSingle-user electronic ticketing systems face significant security challenges, including fraud and counterfeiting. While blockchain has been explored for electronic ticketing, existing solutions often remain centralized or focus solely on event-based scenarios, not single-user tickets such as flight, train, bus, big transport schemes, movie tickets, and vouchers. This paper presents a decentralized single-user ticketing system to address this gap by utilizing Ethereum's ERC-721 standard for smart contracts (SC). Transparency and privacy are ensured through asymmetric encryption. Digital signatures validate ticket authenticity, and an innovative ERC-721-based verification process is applied. Leveraging Ethereum's ERC-721 Protocols, digital signatures, and the interplanetary file system (IPFS) for decentralized metadata storage, this paper addresses centralization, security, traceability, and transparency concerns. The SC is integrated into a web application, and empirical analysis based on blockchain metrics demonstrates its performance. Results indicate that the system exhibits an efficient ticket transaction completion time of 19.64 seconds and a mean ticket verification time of 3.17 seconds. The outcome illustrates the efficiency of the system in mitigating fraud, counterfeiting, and security risks in single-user electronic ticketing systems
Compatibility of transformer construction materials with mineral, natural ester, and synthetic ester insulating oils
Get PDFThis paper presents the experimental findings on the compatibility of kraft paper strips and gasket materials (fluoroelastomer (FE) and nitrile butadiene rubber (NBR)) with mineral (MO), natural ester (NE), and synthetic ester (SE) insulating oils. First, three insulating oil samples were prepared, and kraft paper strips, pressboards, and FE and NBR gasket materials were immersed in the oils. Metal catalysts were added into the insulating oil samples to simulate the actual conditions of oil-immersed transformers. The samples were thermally aged at 130 °C for 400 h. The results show that the tensile strength of the kraft paper immersed in NE increased by 1.82%, while the tensile strength of the kraft paper immersed in MO and SE decreased by 6.23 and 0.80%, respectively. The Shore A hardness of FE thermally aged in MO and SE decreased by 2.64 and 11.16%, respectively. In contrast, the FE thermally aged in NE became slightly harder, with a percentage degradation of +1.62%. On the other hand, the NBR thermally aged in MO, NE, and SE drastically decreased by 94.30, 86.70, and 93.67%, respectively. Hence, it is concluded that NBR is incompatible with the insulating oils tested in this study. In contrast, FE is most compatible with NE, followed by MO and SE
A machine and DL approach for classifying customer sentiments from online shopping reviews in Bangla text
Get PDFDue to the widespread availability of the internet all across the world, people prefer shopping online rather than going to a shop. There are various online marketplaces available in Bangladesh, like Daraz, Pickaboo, Rokomari, Othoba, Bikroy, Food Panda, and Robi Shop. With the increasing quantity of customers on online shopping platforms, the number of product reviews also increases with it. Data is classified utilizing machine learning (ML), deep learning (DL), transfer learning, and other data mining algorithms to facilitate the customer’s comprehension of the primary subject of the review before making a purchase. Natural language processing techniques are employed to categorize data in any given language for such issues. There are no Bengali shopping review datasets available on online sites. So, we manually collected a dataset of 2,600 reviews. In this paper, reviews are classified into 5 categories (satisfied, very satisfied, not satisfied, fairly satisfied, and satisfied but delivery problem). DL (long short-term memory (LSTM) and convolutional neural network (CNN)) and ML (support vector machine (SVM), random forest (RF), gradient boosting (GB), and extreme gradient boosting (XGBoost)) model have been applied. Among the DL models, CNN has the best accuracy (91.27%), and the RF classifier provides the highest accuracy (84.39%) out of all the ML models
Surface-mount device design cycle time reduction using hybrid predictive modeling and optimization algorithm
Get PDFThis study develops a hybrid predictive and optimization model for surface-mount device (SMD) design, addressing the extended design cycle times in the semiconductor industry caused by high computational demands. Challenges are tackled effectively through integration of convolutional neural network (CNN) for high-accuracy predictions and simulated annealing (SA) algorithm for optimization of SMD physical parameters. CNN model that trained on Monte Carlo simulation (MCS) data, achieved a predictive accuracy of 99.91% in forecasting SMD design errors. Concurrently, SA algorithm refined design parameters and substantially reducing error rates to nearly zero after 800 iterations. Our results indicate that combining predictive modeling with an optimization algorithm significantly enhances SMD design efficiency, providing a robust tool for mitigating time-to-market risks in semiconductor manufacturing
A systematic literature review on the use of artificial intelligence for cybercrime rate forecasting
Get PDFCybercrime has a significant impact on the quality of life and economy of individuals, businesses and countries, and the speed of the increase has made it a pressing issue in today's digital age. This systematic review aims to identify the artificial intelligence models recently developed to forecast the rate of cybercrime and to help authorities and police forces define strategies in the fight against cybercrime. The PRISMA methodology was used with 229 articles retrieved from Scopus, IEEE and Web of Science, of which 30 met the eligibility criteria. The results showed that the traditional machine learning methods random forest, support vector machine (SVM) and logistic regression (LR) excel in their use to forecast cybercrimes by achieving more accurate results among the different methods tested. It was concluded that machine learning methods are, so far, effective in forecasting the rate of cybercrime, with accuracy ratios of up to 99.9%. However, the potential for future research lies in creating new forecasting models such as autoregressive integrated moving average long short term memory (ARIMA-LSTM) proposed in this study to improve the performance and accuracy of cybercrime forecasting
Transformers for aerial images semantic segmentation of natural disaster-impacted areas in natural disaster assessment
Get PDFAerial image segmentation of natural disaster-impacted areas and detailed and automatic natural disaster assessment are the main focus of this study. Detecting and recognizing objects on aerial images of areas impacted by natural disasters and assessing natural disaster-impacted areas are still difficult problems. To solve these problems, this study utilizes four of the latest transformer-based semantic segmentation network models, bidirectional encoder representation from image transformers (BEIT), dense prediction transformer (DPT), OneFormer, and SegFormer, and proposes a detailed and automatic natural disaster assessment of the segmented image. The SegFormer model achieved the first-best result, and the OneFormer model achieved the second-best result. The SegFormer model outperformed OneFormer by 1.58% higher for the mean accuracy value and 4.28% for the mean intersection over union (mIoU) value. All receiver operating characteristics (ROC) curves have mean area under curve (AUC) values above 0.9, which means that the SegFormer model performs well in generating semantic segmentation images. The fuzzy c-means (FCM) clustering algorithm performed well and could automatically cluster the natural disaster assessments into four categories. This study has produced semantic segmentation of aerial images of areas impacted by natural disasters and natural disaster assessments, which can be used in natural disaster management systems