Mechanical Engineering
http://hdl.handle.net/10576/3095
2024-03-30T00:27:38ZIntegrated Machine Learning Approaches for Comprehensive Bearing Health Monitoring and Fault Classification Using Multi-Sensory Data
http://hdl.handle.net/10576/51455
Integrated Machine Learning Approaches for Comprehensive Bearing Health Monitoring and Fault Classification Using Multi-Sensory Data
Alhams, Amir Ali Abdulrahman
Modern industries heavily rely on machines equipped with rolling-element (RE) bearings. However, these machines face substantial risks due to potential bearing faults, where even minor defects can lead to catastrophic failures. Shockingly, statistics reveal that up to 40-51%of induction motor failures can be attributed to bearing damage. Early fault detection through Condition Monitoring is therefore crucial. Over the last two decades, various machine learning (ML) techniques have been explored to detect defects in rolling element bearings. This research project focuses on optimizing ball-bearing fault detection through diverse ML techniques, supported by comprehensive experimental work.
The experimental work entails multiple steps, including the preparation of a varied set of bearings and the enhancement of equipment with advanced sensors. These experiments resulted in a rich dataset comprising vibrations, currents, and sound – vital for ML model training. The experiments spanned two different machines, incorporating variations in speed and applied forces. Analysis of the experimental data unveiled the significant influence of defect shapes on bearing responses. Notably, at lower speeds, defect shapes prominently affected vibrations, with rectangular defects displaying logarithmic growth while circular defects exhibited exponential behavior. With increasing speed, the behavior of bearings with different defect shapes tended to converge. Current emerged as a robust choice for fault detection, maintaining consistent behavior regardless of defect size and shape, while sound closely resembled vibrations, with slight variations in the case of circular defective bearings.
The second part of this research is dedicated to developing ML fault detection models. Three models, utilizing ensemble techniques, were created. These models, employing Decision Trees (DT), Random Forest (RF), and XGBoost, achieved prediction accuracies of 82%, 91%, and 92%, respectively. A feature importance analysis identified CRSF and SF as dominant parameters. Furthermore, a sound-to-vibration transformation ML model was introduced. This model, built on a 1D Operational U-Net (Op-UNet) framework, is capable of synthesizing realistic vibration signals from sound measurements across different working conditions, fault types, and severities achieving a striking minimum accuracy of 97%. The models and datasets presented in this research signify a significant advancement in bearing health condition monitoring.
2024-01-01T00:00:00ZDesign, Development, and Testing of Torque and Angle Measurement System for a Traction Machine
http://hdl.handle.net/10576/48561
Design, Development, and Testing of Torque and Angle Measurement System for a Traction Machine
Farhan, Mohammad
This thesis presents the development of a traction measuring system designed to evaluate the interaction between shoes and playing surfaces to prevent lower extremity injuries. The interaction between shoes and playing surfaces has been identified as a key risk factor for lower extremity injuries. The traction system was designed and built with an advanced mechatronics interface which is going to be used as an add-on to an existing manually operated EXETER S2T2 traction machine utilized by Aspetar. The components of the traction machine such as the electronics interface, operating software, and connecting parts were developed in the workshop. The mechanical interactions such as rotational torque, and angular displacement, between the shoes and the natural grass as well as artificial grass of playing surfaces were measured. The values of torque were calculated at various applied loads within 90o angles for inward and outward rotations, and it was plotted against the angular displacement, and speed of rotation. The peak torque obtained for the Nike Tiempo shoe has the highest traction of 32.44 Nm for outward rotation on natural grass (NG) and the lowest of 25.89 Nm for inward rotation on artificial grass (AG) which are considerably different for various loading and speed conditions. The result analysis shows the improved efficiency and ability of the mechatronic add-on over the existing manually operated traction measuring machine. A numerical model was developed using the modified Bouc-Wen mode parameters for the calculation of torque with various conditions. The numerical results show that the rotational stiffness can be represented by sub-systems and each with their own properties. As the loading condition or speed changes, the model's degree of freedom also changes which leads to a reduction in sudden change in stiffness. The traction measuring system has the potential to contribute significantly to the understanding of lower extremity injury mechanics and to help design footwear and playing surfaces that minimize injury risk. The results of this study will be useful for sports medicine practitioners, footwear designers, and sports organizations looking to reduce the incidence of lower extremity injuries in athletes.
2023-06-01T00:00:00ZA HYBRID MAGNETORHEOLOGICAL ELASTOMER FOR SEMI-ACTIVE ISOLATION OF LONGITUDINAL AND TORSIONAL VIBRATIONS
http://hdl.handle.net/10576/45066
A HYBRID MAGNETORHEOLOGICAL ELASTOMER FOR SEMI-ACTIVE ISOLATION OF LONGITUDINAL AND TORSIONAL VIBRATIONS
ALI, ABDELRAHMAN
Magnetorheological elastomers (MREs) are materials that exhibit a change in their mechanical properties, such as stiffness and damping, in response to an applied magnetic field. In general, MREs consist of an elastomeric matrix and homogeneously dispersed magnetic particles, which allow for adjustable viscoelastic properties when a magnetic field is applied. This property, known as magnetorheological (MR) effect, is a result of the alignment of magnetic particles in response to the field and enables the control of the material's mechanical properties and makes it useful for semi-active vibration isolation. Despite the inherent viscoelastic-property change of MREs, their damping capabilities and MR effect are adversely affected by the slow response time and suspension of the particles that are dispersed within the material. To address this, a hybrid MRE is developed by encapsulating MR fluid inside the elastomer, enhancing its performance in longitudinal and torsional vibration isolation. A semi-active base isolator is fabricated using the hybrid MRE as the elastomeric element, showing effectiveness in attenuating vibrations and exhibiting superior MR effect and damping characteristics compared to conventional MREs. The study concludes that the hybrid isolator can enhance the performance and MR effect while simultaneously reducing magnetic field requirements. Additionally, a parametric model that can predict the nonlinear and hysteretic behavior of the hybrid MRE is proposed. These findings reveal the potential of hybrid MREs for developing smart isolation systems in future research.
2023-06-01T00:00:00ZINVESTIGATING FATIGUE LIFE IN BOLTED FLANGE CONNECTION IN WIND TURBINE TOWERS
http://hdl.handle.net/10576/45068
INVESTIGATING FATIGUE LIFE IN BOLTED FLANGE CONNECTION IN WIND TURBINE TOWERS
SALAMEH, ABDULLAH M.
A comprehensive assessment of fatigue was performed on an L-flanged bolted connection under four different wind speeds. The wind turbines were subjected to average wind speeds of 5, 10, 15, and 20 m/s. The primary objective was to investigate how the fatigue life was influenced by increasing the number and size of bolts while also developing a systematic approach for analyzing the fatigue life of bolted flange connections more broadly. The study determined that increasing the size or number of bolts can notably improve the fatigue life of bolted flange connections. Additionally, the curves derived from the assessment data demonstrated a steeper slope for a greater number of bolts, indicating that the percentage increase of adding bolts is not consistent for each additional bolt. Instead, the percentage increment rises exponentially when increasing the number of bolts. However, selecting the most suitable design improvement strategy depends on the specific circumstances. For example, increasing the number of bolts may not always be possible due to spatial limitations. In the majority of cases, the study observed that increasing the number of bolts resulted in significant improvements in fatigue life, regardless of the size of the bolts used. This noteworthy finding can be particularly advantageous when assessing the cost-effectiveness of possible solutions for enhancing the durability of bolted flange connections.
2023-06-01T00:00:00Z