Procter & Gamble (United Kingdom)
Open Access Institutional Repository at Robert Gordon UniversityNot a member yet
9563 research outputs found
Sort by
Feasibility study of liquid metal-based thermal management for solid oxide electrolyzer cell (SOEC).
Get PDFSolid Oxide Electrolyzer Cells (SOECs) operate at elevated temperatures (700 to 1000 °C), leading to thermal management challenges due to significant temperature gradients that affect durability and performance. This study investigates the feasibility of using liquid metals including tin (Sn), sodium (Na), gallium (Ga), lead bismuth eutectic (LBE), and lithium (Li) as cooling fluids, in comparison to conventional air cooling. A 3D SOEC model coupled with electrochemical reactions is developed to assess maximum temperature, temperature distribution, and the Temperature Uniformity Index (TUI). Results show that liquid metals significantly reduce thermal gradients, with Ga and LBE achieving the lowest gradients of 3 K and 3.5 K, respectively, compared to air cooling at 7.5 K. Gallium is selected for further analysis to optimise cooling conditions. Increasing the Reynolds number from 900 to 8960 improves convective heat transfer (Nusselt number increases from 3.5 to 3.75) but raises pumping power by 323 percent. Lowering the middle cooling channel temperature from 1063 K to 1043 K further improves temperature uniformity and reduces peak temperatures. Among all strategies, parallel flow at 1043 K and 0.4 m/s achieves the best balance between thermal performance and energy consumption. Overall, gallium emerges as a highly effective cooling fluid for enhancing SOEC thermal management and long term reliability
Artificial intelligence-enabled transaction prediction.
Get PDFPredicting transaction behaviour and volume on e-commerce websites is difficult because transactional statistical methods don't do well with complex, unstructured data and cannot account for external factors like trends and promotions. Also, existing AI transactional predictive models suffer from historical data biases, interpretability concerns, and problems with human-crafted features. This research therefore proposes an AI machine learning transaction prediction system model that is more robust and transparent for predicting transaction behaviour when shopping online. This project designed a feedforward neural network (FNN) for binary classification to predict whether a transaction is made in the Santander customer transactions dataset. The input layer had 202 neurons, consisting of the data features, two hidden layers of 10 neurons each with ReLU activation, and a dropout rate of 0.2 to prevent overfitting. The output layer has 2 neurons with a sigmoid activation function to output separate probabilities, thus balancing complexity and enhancing class differentiation. The research made use of Bayesian optimization and tuned key hyperparameters, learning rate (0.001), batch size (32), drop rate (0.3), and optimiser (Adam), to balance regularization and convergence. The Adam algorithm combined with binary cross-entropy as the loss function was applied to a 70% training, 20% validation, and 10% split to improve stability and generalization. Our proposed approach demonstrated a high accuracy of 0.93, outperforming traditional models like decision tree (0.84), random forest (0.87), logistic regression (0.85), and gradient boosting (0.90). The Time-Weighted F1 Score of 0.52 and a Temporal ROC-AUC of 0.75 from time-dependent metrics are particularly important for real-world applications, as they reflect the model's capability to adapt to the evolving nature of transaction. The model development process is thoroughly documented, with a clearly defined architecture and an accessible implementation environment, thereby enhancing interpretability and helping to mitigate biases
ClusterSwarm: cluster-specific feature selection using binary particle swarm optimisation.
Get PDFFeature selection has become an important step in machine learning pipelines, contributing to model interpretability and accuracy. While the emphasis has been hugely on global feature selection techniques, these methods do not support feature attributions to the distinct groups within a dataset, since they assume that a single feature set is adequate to correctly undertake the classification task. Unlike unsupervised learning, moreover, feature selection techniques, whether global or local, have been well-developed for supervised learning. Due to the preceding reasons, this paper presents ClusterSwarm, a new approach towards cluster-based feature selection using Binary Particle Swarm Optimisation (BPSO) and the K-means algorithm, to identify cluster-specific feature sets. Evaluating using four publicly available datasets from the UCI repository, ClusterSwarm demonstrates superior performance to the standard K-means algorithm and agglomerative hierarchical clustering and performs similarly to Sparse K-means, a global feature selection technique. However, ClusterSwarm performs better than Sparse K-means in high-dimensional, multi-class and noisy contexts, while providing interpretability through feature attributions to each cluster. In comparison with CS Sparse K-means, a cluster-specific variant of Sparse K-means, ClusterSwarm produced better accuracies and more efficient feature selection, ignoring redundant features, unlike CS Sparse K-means. In addition to the four public datasets, we experimented with two synthetic datasets carefully curated to represent cases of noisy features and overlapping clusters. These datasets have been used to demonstrate the superiority of ClusterSwarm compared to Sparse K-means, CS Sparse K-means and the standard clustering techniques
Development of a novel platform for the optimal design, sizing and energy management of grid-integrated hybrid photovoltaic-hydrogen energy systems.
Get PDFWith the World moving towards the Net-Zero energy transition, there is an increasing demand to scale-up the investment in renewable energy systems. To enable the increased integration of intermittent renewables such as wind and solar energies, while exploiting their full potential, adequate and sustainable energy storage systems are required. Hydrogen (H2) energy storage systems, given their eco-friendliness, fast response and versatility, are seen as potential solutions for mitigating the renewable energy intermittency. This thesis presents new strategies for optimising the design, sizing, and energy management of grid-integrated hybrid renewable-hydrogen energy systems. This thesis' key contribution to the knowledge lies in developing a novel precise dynamic system model that allows the accurate sizing and real-world dynamic simulation of hybrid Photovoltaic-Hydrogen (PV-H2) energy systems. This novel model captures the electrochemical dynamic behaviour of the individual system components in response to changes in operating conditions. Further contribution to the knowledge included developing real-word system-sizing optimisation models through integrating the developed novel model with both single-objective and multi-objective particle swarm optimisation algorithms. This integration allows optimising the hybrid system sizing while considering its real-world dynamic operation. The developed optimisation models respectively offer optimal system sizing for minimising the levelized cost of energy (LCOE) solely and for simultaneously minimising the LCOE along with the carbon footprint. A benchmark comparison versus the commercially available HOMER software is conducted to highlight the developed optimisation models' added privileges over HOMER. Finally, an innovative platform that encapsulates all the developed novel methodologies is presented to assist decision-makers in designing, optimally sizing, and simulating the real-world dynamic behaviour of grid-integrated hybrid PV-H2 energy systems prior implementation. This developed platform enables bridging critical gaps in existing software through modelling the system real-world dynamic behaviour, enabling the selection of customized decarbonisation levels, and allowing multi-objective optimisation for multi-prospect investment decisions. Key findings reveal that, compared to a generic system model that neglects the real-world dynamics of the system components, the developed novel precise dynamic model significantly allowed reducing the size of the H2 storage tank required by 72.5%, thus eliminating oversizing costs. The application of the developed single-objective optimisation model on the grid integrated case-study building has enabled achieving 31.54% reduction in the LCOE reaching 0.3697 £/kWh with a grid dependency of 53%, outperforming HOMER which resulted in a LCOE of 0.3976 £/kWh and a grid dependency of 62.78%. The multi-objective optimisation model has enabled further reduction in the grid dependency to 43%, while maintaining the LCOE at slightly higher reasonable value of 0.4117 £/kWh. Notably, it can be seen that for nearly the same LCOE as HOMER, the multi-objective optimisation model significantly lowered the grid dependency, thus providing a much greener solution at no additional costs
Determining the scope of the philosophy of computing education.
Get PDFThere are a number of different approaches to the investigation of teaching and learning within the subject of Computing Education. Many of the advances in pedagogy that have taken place over the past thirty years have been due to careful statistical analysis of empirical data, enhancing the reputation of the subject within the broader Computing discipline. Empirical, qualitative methodologies, of the kinds used extensively in the Social Sciences, have also appeared in the Computing Education literature, often investigating the socio-cultural aspects of the subject. More recently, there has been a proposal to develop a role for philosophical inquiry in Computing Education, which mirrors similar historical developments in Engineering Education. Rather than focus on the quantitative or qualitative analysis of the student experience, philosophical investigation instead relies on the use of conceptual analysis to investigate the detailed semantic content of ideas raised in the practice of computing education, careful analysis of the methodologies used to do such work, and a critique of the assumptions that underlie the subject. In this paper, we investigate ways in which an understanding of the Philosophy of Computing Education can assist research within the subject. We consider how it emerges from basic questions about nature of the subject, its scope, and how it can be applied fruitfully within the discipline
Unveiling the characteristics of ER70S-6 low carbon steel alloy produced by wire arc additive manufacturing at different travel speeds.
Get PDFWire Arc Additive Manufacturing (WAAM) produces metal components with crucial properties dependent on process parameters. Understanding the effects of these parameters on microstructure and mechanical properties is vital for optimizing WAAM. This study investigated the impact of varying travel speeds (TS) on the microstructure and mechanical properties of low carbon steel ER70S-6 alloy produced by WAAM process. The hypothesis centred on the impact of different TS values on heat input (HI) and cooling rates, and the subsequent effects on the resulting microstructure and mechanical properties of the deposited material. ER70S-6 alloy was deposited at three different TS: 120, 150, and 180mm/min. Microstructure and mechanical properties (microhardness, tensile strength, elongation) were evaluated for each TS condition. Distinct microstructures were observed in the deposited samples, influenced by cooling rates at different TS. Distinct microstructures emerged in different regions of the deposits due to varying cooling rates at different TS. Higher TS (180mm/min) significantly reduced pores and cracks while enhancing yield strength (YS) and ultimate tensile strength (UTS) up to 25.2 ± 0.77% elongation and 502.3 ± 3.17MPa UTS, respectively. However, UTS remained slightly lower (93%) than the catalogued value for ER70S-6 (540MPa), indicating a mild softening effect. TS significantly influenced the microstructure and mechanical properties of WAAM-produced ER70S-6 alloy. This study provides key insights into optimizing WAAM parameters for low carbon steel, paving the way for improved component production for diverse industrial applications
Machinability performance of single coated and multicoated carbide tools during turning Ti6Al4V alloy.
Get PDFThis paper presents the machinability performance of uncoated, single-coated, and multicoated carbide tools during turning of Grade 5 (Ti6Al4V) Titanium alloy, which is challenging to machine due to its distinctive material properties. Coated tools with single-coated Titanium Aluminium Nitride (TiAlN) and multi-coated layer of Titanium Aluminium Nitride with Aluminium Chromium Nitride (TiAlN + AlCrN) coated inserts were utilized to assess surface roughness (Ra), tool wear rate (R), and chip morphologies under various cutting conditions using dry machining. Analysis of the used tools revealed that coated tools exhibited improved tool life and surface quality compared to uncoated tools across all cutting conditions. Multi-coated tools of TiAlN + AlCrN demonstrated a tool life increase of up to 15% compared to uncoated and single-coated tools, with surface roughness improvements ranging from 30 to 45% depending on cutting speed. Chip morphology analysis indicated an increase in the chip reduction coefficient with higher cutting speeds for all tool types. Coated tools exhibited the lowest chip-reduction coefficient due to the presence of TiAlN and AlCrN coatings, which control the tool chip contact length. Conversely, uncoated chip morphology resulted in larger chip thickness values compared to coated tools, particularly at cutting speeds above 100 m/min, attributed to poor heat dissipation and chemical reactions at the tool chip interface. Energy dispersive X-ray scanning electron microscopy (SEM/EDXS) analysis of worn uncoated inserts revealed a higher tendency towards Titanium adhesion compared to coated tools. The proposed multi-layer coatings of (TiAlN + AlCrN) used for dry machining proved highly beneficial for achieving economic machining objectives and may reduce the need for lubrication when processing Ti6Al4V alloys
Innovative building techniques: the time performance of panelised and modular construction systems in a home development research project.
Get PDFModern methods of construction (MMC) have been embraced by the construction sector of many states, largely due to assertions that some MMC processes can be economically effective and efficient. MMC have been received as opportunities by governments to stimulate economies and meet demands for new homes, and contractors to enter new markets, and generate profits whilst improving the health and safety conditions of work. Although many studies have commented on the implications of using specific MMC, arguing the degrees of their superior time performance, no evidence has been presented on the time performance of multiple MMC in a live house construction experiment, which was also a research project managed by one contractor on one construction site. This paper contributes to the fulfilment of this research gap and presents an analysis of the time performance of five MMC, namely steel-based and timber-frame modular construction, light-gauge modular and panelised light-gauge steel frame construction, and aerated concrete panelised construction. The research method includes an investigation of the residential building design, interviews with relevant construction personnel, and analysis of project data. The actual time performance of these MMC, unlike their planned time performance, was inferior to traditional construction. The planned time performance of panelised construction was inferior to some modular methods. Inadequate site logistics and design issues were the main barriers to achieving the planned time efficiencies. These research findings identify areas of improvement in the time performance of the aforementioned MMC, enlighten governments, employers, contractors, and end users, and offer valuable research data
Exploring the development and implementation of practice-based interprofessional education for student pharmacists in Scotland: a case study.
Get PDFInterprofessional collaborative practice (IPCP) is considered essential to address the increasingly complex needs of patients. Subsequently, developing a workforce with the right competencies is a requisite to ensure safe, effective and efficient person-centred care within healthcare systems. This in turn has increased focus on interprofessional education (IPE) as a necessary step in preparing a "collaborative practice-ready" workforce. The overarching aim of this research programme was to explore the development and implementation of practice-based IPE in the experiential learning (EL) curriculum of Master of Pharmacy (MPharm) programmes in Scotland. Ethical approval was granted by the School of Pharmacy and Life Sciences Ethics Review Committee at Robert Gordon University
Avoiding (offshore) renewable and critical mineral-based resource curse in Africa: the role of a continent-wide legal regime.
Get PDFAfrica has significant offshore renewable energy and critical minerals, crucial for transitioning to a carbon-neutral economy. Africa holds 30% of the world's critical mineral reserves, with countries like Congo, South Africa, and Nigeria having significant quantities. Offshore wind, hydropower, and solar energy potential is vast. However, the Global North is increasingly acquiring these resources, referred to as the 'New Colonialism'. This lack of a robust regulatory framework risks a renewed Resource Curse in the context of offshore energy and critical minerals transition
Data sources:
Access Repository Dashboard
Do you manage Open Access Institutional Repository at Robert Gordon University? Access insider analytics, issue reports and manage access to outputs from your repository in the CORE Repository Dashboard!