Pusat Pengajian Kejuruteraaan Elektrik dan Elektronik - Monograf

Browse

Now showing 1 - 5 of 1182

Towards IR4.0 implementation for smart manufacturing and predictive maintenance
(2024-08)
Muhamad Fahmie bin Shahrulzahadie
This thesis explores the development and implementation of a smart manufacturing system aimed at enhancing predictive maintenance and operational efficiency. Using Industry 4.0 technologies, the system integrates Firebase Authentication for secure access, Google Sheets for real-time data storage, and a user friendly GUI for data visualization. The research involved collecting and analysing FFT vibration and temperature data from a milling machine using an Arduino Nano Wi-Fi Microprocessor, ADXL345 vibration sensor, and RTD Probe Temperature Sensor. The FFT analysis identified the machine's natural frequency at 25Hz and established upper limits for vibration magnitudes. The results from three samples collected under load conditions indicated that while the machine generally operated within acceptable parameters, Sample 3 revealed anomalies with one point on the X axis and five points on the Y-axis exceeding the upper limits. Additionally, the temperature peaked at around 80 degrees, suggesting potential issues such as misalignment, imbalance, or inadequate cooling and lubrication. These findings underscore the necessity of continuous monitoring and proactive maintenance. The study demonstrates that integrating real-time data acquisition and cloud-based storage in manufacturing processes can significantly improve the ability to prevent machine failures, thereby enhancing overall operational efficiency and machinery longevity. This research lays a solid foundation for the broader implementation of smart manufacturing systems, contributing to the advancement of predictive maintenance practices in the industry.
Universal suction cup for non uniform surface topology
(2024-07)
George, Ting Pek Jing
Universal suction cups are integral tools in various industries due to their ability to adhere to non-porous surface. However, their performance is often compromised on non-uniform surface or uneven surface which includes the surface of a mounted printed circuit board. This research aims to develop an innovative design for a universal suction cup and mechanical arm support that maintains strong adhesion towards non-uniform surface. The study involves designing a universal suction cup that caters the non-uniform surface topology primarily the mounted printed circuit board, designing a mechanical arm that enhances the performance of universal suction cup and test the maximum weight lifted by the universal suction cup. The universal suction cup is tested with increasing weight to identify the maximum weight lifted. Inclusion of mechanical arms that hold four universal suction cups, the result indicates it can lift mounted printed circuit board that showcase its ability to cater the non uniform surface thanks to the miniature size of the universal suction cup. The finding of the study suggests that the universal suction cup design can be widely applied in industries requiring reliable temporary adhesion which includes mounted printed circuit board, potentially reducing the need for specialized equipment. This research contributes to the advancement of adhesion technology and offers practical solutions for expanding the functionality of the universal suction cup.
Study on digital implementation space vector pulse width modulation with three phase inverter
(2024-07)
Shazriezal bin Saharuddin
Three-phase inverters frequently use pulse width modulation (PWM), a control technique that modifies pulse width to manage output voltage, improve system efficiency, and lower energy consumption. Sinusoidal Pulse Width Modulation (SPWM) is a widely used techniques in power electronics for regulating the output voltage inverters. However, improving SPWM to meet the efficiency and harmonic distortion standards of modern power systems presents challenge. SPWM often introduces harmonic components to the output waveform, resulting in higher losses. Due to its higher performance features, Space Vector Pulse Width Modulation (SVPWM) has increasingly replaced Sinusoidal Pulse Width Modulation (SPWM). Space Vector Pulse Width Modulation (SVPWM) in three-phase inverters is typically accomplished using a two-level technique due to its efficiency and ease of use. This technique maps necessary output voltage vectors to available inverter switching states directly, simplifying the construction of hardware and software. Compared to higher level approaches, two-level SVPWM improves real-time performance by lowering switching losses and representing different combinations of inverter switching states with a fixed set of voltage levels. Two-level SVPWM is the recommended option for three-phase inverters due to its user-friendly design, dependability, and efficient DC to AC power conversion, which guarantees a consistent and excellent output under a range of operating circumstances. This research work was aimed to show that SVPWM is better version for SPWM in three-phase inverters in DC bus utilization and Total Harmonic Distortion (THD).
Robotic glove rehabilitation system
(2024-08)
Ong, Qian Hui
Stroke patients often face challenges in restoring motor functions and independence. This final year project studies the integration of rehabilitation technology by developing a Robotic Glove Rehabilitation System for better post stroke patient recovery. The system proposed in this study is designed to provide targeted and personalized therapy that focuses on the rehabilitation of hand and finger movements. The robotic glove implements mirror therapy method which aims to optimize therapy effectiveness and accelerate the recovery process. To create a flexible and responsive rehabilitation environment, the Arduino MEGA controls the actuators and flex sensors in the rehabilitation glove. A PID controller is implemented to ensure the precision of the DC motor pump while also increasing the system’s accuracy. The lightweight soft actuators are fabricated using the Ecoflex silicone rubber to provide a comfortable and adaptable fit for post-stroke patients. The real-time data collection and analysis capabilities enable the healthcare professionals to monitor progress and make informed adjustments to rehabilitation programs and provide patients with timely feedback. The effectiveness of mirror therapy is based on the real-time feedback loop. It allows mirrored movements of the unaffected hand to promote neuroplasticity and improve motor functions in the affected hand.
Machine vision-based nondestructive inspection of wood surface cracks
(2024-07)
Nurul Syahirah binti Ismail
Wood inspection is a crucial process in industries such as construction, furniture manufacturing, and wood processing, where identifying defects like cracks, knots, and pinholes is vital to ensure the structural integrity and aesthetic appeal of products. Traditional manual inspection methods are labor-intensive, time consuming, and prone to human error, leading to inefficiencies and quality control issues. To address these limitations, recent advancements have introduced machine vision-based non-destructive inspection methods, utilizing high-resolution cameras and sophisticated computer vision algorithms for accurate defect detection. Despite these technological advancements, significant challenges persist, particularly in detecting small and complex defects due to limited and diverse datasets used for training deep learning models. The need for computational efficiency in practical industrial applications further complicates the deployment of these models. This research aims to enhance the accuracy, efficiency, and reliability of automated wood defect detection systems by leveraging the YOLO (You Only Look Once) algorithm, specifically YOLOv8, combined with data augmentation techniques. YOLOv8 is renowned for its speed and accuracy in real-time object detection, making it suitable for industrial applications requiring rapid and reliable defect detection. This study assesses the effectiveness of data augmentation techniques in improving the detection capabilities of YOLOv8 and evaluates its performance in terms of precision, recall, mean average precision (mAP), and F1-score. The research also includes a comparative analysis of YOLOv8 against traditional methods like SVM and older neural network models, highlighting its superior performance in diverse environments. Initial findings demonstrate that YOLOv8, enhanced with data augmentation, significantly improves the detection accuracy and generalization capability of the model, particularly for small and complex defects. The results indicate a substantial improvement in precision, recall, and mAP compared to conventional methods, validating the effectiveness of the proposed approach. These findings underscore the potential of advanced deep learning techniques in revolutionizing wood defect detection, leading to better quality control, reduced financial losses, and greater efficiency in the wood industry.

Browse

Recent Submissions