http://bura.brunel.ac.uk/handle/2438/186 2026-05-05T07:35:12Z http://bura.brunel.ac.uk/handle/2438/33216 Title: Investigation of the use of power ultrasound to improve manufacturing processes in fluid phase Authors: Teyeb, Ahmed Abstract: Advanced manufacturing techniques play a crucial role in ensuring the efficiency and reliabil-ity of electric vehicle (EV) battery packs. In particular, the assembly of busbars and connect-ors—key components for current distribution—relies on joining dissimilar metals like copper and Aluminum. Aluminium-Copper (Al-Cu) alloys are often selected as best choice due to the combination of lightweight and effective conductivity. However, conventional joining methods of these metals often lead to the formation of brittle intermetallic compounds, compromising electrical performance and mechanical integrity. Despite the progressive interest in ultrasonic vibration–assisted laser welding for joining dissimilar metals in EV busbars and connectors, there remains a restriction in foundational understanding of how ultrasonic frequency, ampli-tude, and energy coupling interact with laser–material dynamics to control melt pool physics, intermetallic compound formation, and joint integrity. There is a rare integration of systematic transducer design and tuning with welding process optimization by existing studies, leaving the relationship between transducer characteristics and weld quality largely unexplored. Con-sequently, there is a critical knowledge gap in establishing quantitative, process–structure–property relationships that link ultrasonic transducer parameters to mechanical, electrical, and metallurgical performance of dissimilar-metal laser welds in EV applications. Therefore, this study addresses the use of ultrasound vibration-assisted laser welding as a manufacturing tech-nique, to join dissimilar metals such as Al-Cu, particularly for electric vehicles (EV) battery assembly. The research considers both theoretical and experimental approaches to the mecha-nism involved in formation of cavitation microbubbles and the effects of these agitated mi-crobubbles on the process-structure-property relationship arising from the integration of high-power ultrasonic vibration technology with laser welding. Owing to optimal cavitation bubble formation at ultrasonic frequencies between 20 kHz and 40 kHz – ranges widely adopted in industry for both performance and safety – the study presents the design and development of different transducer types operating at varying frequencies of 20, 28, and 40 kHz. Considera-tion of different experimental setups and varying sets of processing parameters, like transducer angle, transducer distance from metal plate, electrical impedance, and are put in place, to obtain optimal displacement and acceleration amplitudes. The initial section of the work focuses on comprehensive review of metal joining processes; the advancements and challenges associated with EV batteries in recent times and introduces power ultrasound in industrial applications. Also, it delves into the theoretical modelling of formation and collapse of microbubbles caused by the introduction of ultrasonic waves into the solidifying phase of the weld pool. In addition, experimental investigation of structural vibra-tion of lap and butt joints is discussed. Furthermore, the latter section elucidates the mechanical and microstructural analyses of ultra-sonic vibration-assisted laser welded joints. Results showed that the application of high-inten-sity ultrasound significantly disrupted epitaxial dendrite growth, refined grain structure, mini-mized plasma cloud formation, and altered the shape of intermetallic compounds from linear to spherical. The mechanical strength of vibration-assisted joints welded at 28 kHz and at weld-ing speed of 40 mm/s showed a significant increase of 24.5% against the non-vibration-assisted counterpart. The application of high-intensity ultrasound significantly improved weld quality by straightening the weld profile, reducing differential weld width-to-root by 14 - 62%, refining grain structure, and reducing defects such as spatter and plasma cloud formation. Also, SEM results showed that the ultrasonic-assisted laser welded joint was characterized by smaller in-termetallic formations and appeared mostly in the shape of spheres. Furthermore, it was ob-served that the volume density of the secondary phases within the grain boundaries reduced with the application of ultrasonic vibration. The pull test results indicated an approximate 10% increase in load capacity and a 27% increase in extension when vibration was introduced, com-pared to welds produced without vibration. This demonstrates that vibration-assisted welding enhances joint strength and ductility, contributing to improved mechanical performance and reliability of the weld. Overall, the study provides an innovative methodology to address critical issues in industrial and manufacturing processing of dissimilar metals, mitigating the detrimental effects of the formation and presence of hard, brittle intermetallic compounds associated with Al-Cu alloys. This method is highly effective in contemporary material processing of busbars and connectors of EV batteries. Description: This thesis was submitted for the award of Doctor of Philosophy and was awarded by Brunel University London 2026-01-01T00:00:00Z http://bura.brunel.ac.uk/handle/2438/33068 Title: Design and development of an AI enhanced channel coding technique with adaptive reconfigurable intelligent surface for terahertz 6G communication Authors: Ahmed AL-Joudi, Aya Khalid Abstract: This thesis presents a novel approach to design and implement a new channel coding method combined with a Adaptive RIS (ARIS) to enhance Terahertz (THz) communication in 6G networks. The research addresses the crucial requirements of 6G communication, including ultra-fast data transmission, minimal delay, extensive connectivity, and optimal energy usage. The innovative channel coding approaches, Polar Convolutional Serial Code (PCSC) and Polar Convolutional Parallel Code (PCPC), are specifically designed to enhance the reliability and data transfer rate of wireless communication systems operating at THz frequencies. Their performance is rigorously evaluated in congested network conditions, a common scenario in 6G applications, in conjunction with Non Orthogonal Multiple Access (NOMA) strategies. A key achievement in this research is the integration of ARIS into the commu-nication system, leading to the development of a ARIS Decision Making Algorithm (ARIS-DMA). This technology optimises signal strength and coverage by dynamically adjusting surface reflection and transmission properties based on the user’s location and network conditions. The ARIS-DMA effectively reduces power loss and latency, providing comprehensive coverage and a 70% signal power loss reduction, instilling confidence of users about the progress in the field. In addition, the thesis investigates the application of Deep Learning (DL) methods for decoding PCPC. It suggests a Deep Q Network (DQN) based Deep Q Network ARISDMA (DQN-ARISDMA) to improve beamforming and increase spectral efficiency. The findings exhibit significant enhancements in data transmission speeds, utilisation of the frequency spectrum, and the ability of the system to respond promptly, all of which are vital for time-sensitive applications in 6G networks. The outcomes of this study contribute significantly to the development of communication systems that can meet the rigorous standards of future 6G networks while also being scalable, energy-efficient, and reliable. This advancement creates opportunities for progress in areas such as smart cities, autonomous vehicles, and augmented/virtual reality experiences, demonstrating the practical implications of our research. Description: This thesis was submitted for the award of Doctor of Philosophy and was awarded by Brunel University London 2026-01-01T00:00:00Z http://bura.brunel.ac.uk/handle/2438/32827 Title: Development of YOLO-based object detection and tracking systems for airport ground safety and simulation-based collision analysis Authors: Bingol, Emre Can Abstract: Airport ground incidents remain one of the major safety concerns in aviation, causing operational disruption and substantial economic losses. Although conventional surveillance and sensing can support surface awareness, cost, deployment constraints, and limited interpretability at close range motivate low-cost, vision-based alternatives. Key research gaps include (i) apron-specific, part-level, openly available labelled datasets, (ii) systematic and fair benchmarking of modern deep learning architectures under consistent evaluation conditions, and (iii) integrated risk analysis that progresses from perception to actionable early warning. This thesis addresses these gaps by developing and validating a scalable early-warning framework based on Computer Vision (CV) and Deep Learning (DL), designed for integration with existing Closed-Circuit Television (CCTV) infrastructures. The thesis follows a multi-stage methodology. First, a new detection and segmentation dataset was constructed with five aircraft classes (airplane, wing, nose, tail, and fuselage) to support part-level perception for apron safety. Using this dataset, twelve modern object detection and segmentation architectures were trained and evaluated under consistent experimental settings to establish a benchmarking baseline. YOLOv8-Seg (You Only Look Once, version 8-Segmentation) emerged as the most suitable backbone for the intended operational constraints. Second, to improve robust detection and segmentation of aircraft and their components, YOLOv8-Seg was systematically optimised through a six-step, ablation-driven pipeline, spanning parameter tuning, loss-function refinement, data augmentation, and inference-efficiency improvements. Third, a Multi-Object Tracking (MOT) dataset was created and annotated in the MOTChallenge format to benchmark leading trackers under identical evaluation settings. Finally, two complementary safety layers were developed: (i) a reactive module that issues immediate alerts using image-plane geometric proximity derived from segmentation outputs, and (ii) a proactive module that forecasts short-horizon conflicts by extrapolating past motion and evaluating future overlap using Intersection over Union (IoU). Experimental results show that the optimised YOLOv8-Seg model significantly improves segmentation performance by +8.04 points in mAP@0.5:0.95 (mean Average Precision averaged over IoU thresholds from 0.50 to 0.95) and +4.74 points in mAP@0.5. On the tracking benchmark with airplane-only ground truth, DeepSORT achieved the strongest overall performance, reaching 92.77% Multi-Object Tracking Accuracy (MOTA) with 93.27% recall. The framework was validated across multiple representative scenarios using both simulated and real-world video, supporting the feasibility of a low-cost, CCTV-compatible approach for enhancing apron safety through integrated perception, tracking, and early warning. Description: This thesis was submitted for the award of Doctor of Philosophy and was awarded by Brunel University London 2025-01-01T00:00:00Z http://bura.brunel.ac.uk/handle/2438/32754 Title: Analysis of backhaul networks for developing countries to support next generation communication systems Authors: Al-Zubaidi, Inas Abstract: In its most basic form, the target of fifth generation (5G) is to provide reliable and continued connectivity for the user despite obstacles and challenges. These obstacles and challenges vary depending on the scenario, whether it’s an urban or rural area in developed or developing countries. This thesis focuses on the 5G backhaul for standalone (SA) network and the im-pacts of backhaul technologies on the Quality of Service (QoS) and user experiences, in particular, end to end delay (E2E) and capacity planning requirements. In par-ticular, the aim is to facilitate the work of providers, developers, and investors when planning to introduce 5G technology to developing countries. This work looks into employing simulation-based approach to consider bandwidth aspects when design-ing/ upgrading current/ future cellular systems in developing countries. It presents a scheme to maximize the use of bandwidth considering both capacity and delay aspects and helps to identify major parameters that influence system design for different 5G use cases and scenarios. The result proves that the method to determine the required link capacity is by observing the traffic delay and users access statistics as well as by increasing the capacity incrementally by changing the factor for each link in the network, un-til optimal capacity is achieved. It also indicates that within the ”broadband in the crowd” scenario for 5G services and applications, the necessary bandwidth for last-mile network connections can vary depending on the service type. Specifically, bandwidth requirements can be lessened for ultra-low latency services and applica-tions, with even greater reductions possible for those that do not require such low latency. These adjustments are observed when the backbone link is operating at its full capacity. The finding shows that for developing countries, and by considering the cost of the bit per second, user down link/ uplink, and convenience of user terminals as more critical considerations for the adoption of one of the technologies for backhauling 5G traffic, a satellite and hybrid topology based on the existing networks, financial considerations will play an important role in determining the backhaul network topology with optimizing for the specific requirements. Description: This thesis was submitted for the award of Doctor of Philosophy and was awarded by Brunel University London 2024-01-01T00:00:00Z