Procter & Gamble (United Kingdom)
Open Access Institutional Repository at Robert Gordon UniversityNot a member yet
9563 research outputs found
Sort by
Improved hyperparameter Bayesian optimization-bidirectional long short-term memory optimization for high-precision battery state of charge estimation.
No full textAt a time when new energy sources are constantly developing, mitigating the safety hazards of lithium batteries and prolonging their lifespan. In this paper, we take a ternary lithium-ion battery as an experimental object and carry out research based on the fusion method of deep learning and modeling for its high-precision state of charge (SOC) estimation requirements. This paper explores the construction of a battery dynamic model and hyperparameter optimization method based on a neural network. It also incorporates Kalman filter to investigate the noise correction strategy of a neural network model. Experimentally verified that the BO-BiLSTM-UKF fusion algorithm in this paper has a maximum error of only 0.113 %, which verifies the accuracy and strong robustness of the model. Its MAE and RMSE are reduced by 96.13 % and 95.73 % compared with the LSTM network model, which has better adaptability and estimation ability. In this paper, a network dynamic prediction fusion method based on the equivalent model is constructed and experimentally verified by different temperatures, complex working conditions and step-by-step simulation
A latency-efficient integration of channel attention for ConvNets.
Get PDFDesigning fast and accurate neural networks is becoming essential in various vision tasks. Recently, the use of attention mechanisms has increased, aimed at enhancing the vision task performance by selectively focusing on relevant parts of the input. In this paper, we concentrate on squeeze-and-excitation (SE)-based channel attention, considering the trade-off between latency and accuracy. We propose a variation of the SE module, called squeeze-and-excitation with layer normalization (SELN), in which layer normalization (LN) replaces the sigmoid activation function. This approach reduces the vanishing gradient problem while enhancing feature diversity and discriminability of channel attention. In addition, we propose a latency-efficient model named SELNeXt, where the LN typically used in the ConvNext block is replaced by SELN to minimize additional latency-impacting operations. Through classification simulations on ImageNet-1k, we show that the top-1 accuracy of the proposed SELNeXt outperforms other ConvNeXtbased models in terms of latency efficiency. SELNeXt also achieves better object detection and instance segmentation performance on COCO than Swin Transformer and ConvNeXt for small-sized models. Our results indicate that LN could be a considerable candidate for replacing the activation function in attention mechanisms. In addition, SELNeXt achieves a better accuracy-latency trade-off, making it favorable for real-time applications and edge computing
Assessment of surface water quality using chemometric tools: a case study of Jabi Lake, Abuja, Nigeria.
Get PDFWater pollution has become a growing threat to human society and natural ecosystems in recent decades. It increases the need to understand surface water quality assessment better using chemometric tools within aquatic systems. This study sampled the water quality of 21 parameters at multiple sampling points in Jabi Lake during wet and dry seasons (August–December 2021) using various statistical methods including cluster analysis, principal component analysis/factorial analysis, discriminant analysis, and box plot analysis. These samples were examined for physicochemical parameters employing standard techniques. The study revealed significant seasonal variations in water quality. During the wet season, key measurements included total dissolved solids (100.40 mg/l), dissolved oxygen (13.72 mg/l), and electrical conductivity (97.14 µs/cm). The dry season showed higher levels of most parameters, with total dissolved solids at 137.91 mg/l and electrical conductivity at 230.93 µs/cm. Statistical analysis identified strong correlations between various parameters, notably between phosphate and total hardness in the wet season (r = 0.978, α = 0.05) and between pH and temperature in the dry season (r = 0.995, α = 0.05). The study identified four principal components explaining 98.5–100% of the variance, representing various pollution sources including organic waste, domestic sewage, and natural factors. The findings indicated that dry season water samples were more polluted, with some parameters exceeding World Health Organisation standards, suggesting potential health risks. The research demonstrated the effectiveness of multivariate statistical techniques in analysing complex water quality data and provided valuable insights for water resource management, particularly regarding seasonal variations' impact on water quality
The role of fluorine substituents in the formation of the ferroelectric nematic phase.
Get PDFThe synthesis and characterization of a group of molecules based on the skeletal structures of 4-[(4-nitrophenoxy)carbonyl]phenyl 4-methoxybenzoate and (4-nitrophenyl) 2-methoxy-4-(4-methoxybenzoyl)oxybenzoate is reported. Fluorine substituents are added to these structures and their liquid crystalline behaviour is characterised. All eight compounds reported exhibit the ferroelectric nematic phase, NF, and in four of them the antiferroelectric phase labelled NX is observed. These include rare examples of enantiotropic NF and NX phases, and several examples of the direct NF-isotropic transition. The addition of a fluorine substituent reduces the nematic-isotropic transition temperature, and this is attributed to the reduction in both shape anisotropy and the ability of the molecules to form antiparallel dimers. The effect of the fluorine substituents on the NF and NX phase is less regular and interpreted in terms of the changes to molecular shape and electron distribution. The stability of the NX phase follows that of the NF phase in the series without a lateral methoxy group. The addition of a lateral methoxy group reduces the stability of the NF phase and completely suppresses the NX phase
Embedding sustainability in engineering education: empowering students with knowledge and skills for a sustainable future.
Get PDFIntegrating sustainability into engineering disciplines is essential for preparing students to tackle the complex environmental, social, and economic issues of the 21st century. This paper examines the significance of incorporating sustainability into engineering curricula, with the goal of equipping students with a thorough understanding of how their specialized knowledge can foster a sustainable future. By embedding sustainability principles in engineering education, students are inspired to critically consider the lasting effects of their work and to devise innovative solutions that align with the United Nations Sustainable Development Goals (SDGs). The paper outlines effective strategies for the integration of sustainability within engineering programs, emphasizing course design, teaching methods, and assessment practices that encourage interdisciplinary collaboration and real-world problem-solving. Through case studies and practical examples, it demonstrates how aligning engineering disciplines with sustainability objectives can nurture crucial skills such as systems thinking, ethical decision-making, and technical problem-solving. Additionally, it underscores the need to raise awareness of sustainability issues across engineering fields, empowering students to understand the interconnections among natural systems, technological advances, and societal needs. In conclusion, the paper promotes a transformative approach to engineering education—one that enables students to leverage their expertise to create sustainable innovations, support environmental stewardship, and drive meaningful change. By equipping future engineering leaders with sustainability knowledge and competencies, we can ensure they are prepared to confront the pressing challenges of our global society and contribute to building a more sustainable future for generations to come
Sulfur-linked cyanoterphenyl-based liquid crystal dimers and the twist-bend nematic phase.
Get PDFThe synthesis and characterisation of two series of cyanoterphenyl-based liquid crystal dimers containing sulfur links between the spacer and mesogenic units, the 34-{ω-[(4′-cyano-[1,1′-biphenyl]-4-yl)thio]alkyl}-[11,21:24,31-terphenyl]-14-carbonitriles (CBSnCT), and the 34-({ω-[(4′-cyano-[1,1′-biphenyl]-4-yl)thio]alkyl}oxy)-[11,21:24,31-terphenyl]-14-carbonitriles (CBSnOCT) are described. The odd members of both series show twist-bend nematic and nematic phases, whereas the even members exhibit only the nematic phase. This is consistent with the widely held view that molecular curvature is a prerequisite for the observation of the twist-bend nematic phase. The nematic–isotropic and twist-bend nematic–nematic transition temperatures are higher for the dimers containing cyanoterphenyl groups than for the corresponding cyanobiphenyl-based dimers. This change is more pronounced for the nematic–isotropic transition temperatures and is attributed to the enhanced interaction strength parameter associated with the cyanoterphenyl fragment whereas the molecular shapes, as governed by the spacer, are rather similar. The behaviour of CBS2CT appears somewhat anomalous and exhibits a higher value of the twist-bend nematic–nematic transition temperature than expected, and this is attributed to the presence of highly bent molecular conformations
Collaborative and social learning in virtual environments.
Get PDFThe Higher Education Institution (HEI) sector is changing, driven by digital advancements and social expectations. Educators must reflect on how, what and why they teach ensuring it is meaningful and relevant. Experiential learning (Kolb, 2014) has been heavily prioritised in UK HEI's, where integrating the 'real world' into the classroom is positive, enhancing students skills development and employability. However, as the workplace becomes more virtual, accelerated by the pandemic, there is a need to consider collaborative and social learning. Collaborative Online International Learning (COIL) attracts attention as an innovative pedagogy that empowers students to engage meaningfully in virtual, cross-cultural learning spaces, preparing them for the interconnected world of work (Swartz, et al., 2020)
Enhancing entrepreneurial skills through creativity, diversity and collaborative online international learning.
Get PDFThis paper evaluates staff and students' perspectives of an entrepreneurial Collaborative Online International Learning (COIL) project, which examined students' creative thinking and practice whilst working across three different countries, institutions and educational systems, with the total student body representing multiple nationalities. The diversity of the students' backgrounds and experiences is at the heart of the paper, which seeks to identify the links between the experiential learning undertaken, the diversity of a group and its subsequent creative outputs. The paper aims to evaluate the extent to which international, online, interdisciplinary group work fosters creative learning environments, specifically evaluating the effect of an entrepreneurial task, and the impact of differing educational levels, subject fields and location. The ability to work in a heterogeneous team comprising different backgrounds, locations and practices is a significant challenge for international collaboration online. With increasing literature focusing on authentic assessments, students' ability to articulate how their learning transfers to the world of work, and the increased importance of students cultivating transferable employability and entrepreneurial skills (including communication, problem-solving and critical thinking), this paper evaluates the extent to which these growing concerns are addressed by COIL projects, using a series of in-depth interviews with the staff and students from all three institutions. The study found that COIL projects can offer a valuable addition to curricula of further and higher education institutions with the benefits of internationalisation, cross-cultural communication and teamwork being highlighted within the findings. The key reflections identified that structured communication for staff and students alike is critical to the success of the COIL, without which the students' ability to create is diminished. Upon evaluation it became clear that whilst COILs are inherently of value in a broader sense, the interdisciplinarity of the students' subject knowledge itself is a critical factor in determining the success of the project
European justice: the Scottish contribution.
Get PDFOf the eleven UK members of the Court of Justice from 1973 until 2020, three have been Scottish jurists, namely AJ Mackenzie Stuart, David Edward and Ian Forrester. Each of them has played an important role at the Court and they all contributed to the shaping of EU Law while there. In particular, Scottish judges' names can be associated with important judicial statements in the fields of free movement of goods and EU Citizenship. In so doing, the three Scottish judges can be said to have made a unique and enduring contribution to the moulding of EU legal principles. Can it be asserted that this contribution was facilitated by the Scots law background and connections of these three judges? That may be not easily proven but by looking at some of the factors which shaped the three Scottish judges and by examining some individual judgments with which they are connected, a picture emerges of an especially robust level of influence exerted by the Scottish judges at the EU Court, an influence which endures there even after UK membership has ended. Perennial sovereignty v supranationalism frictions have long been at the heart of European integration. Scots lawyers and judges willingly entered that fray and sought resolutions from within. None of this is a co-incidence; from the time of the Act of Union, Scots lawyers have adapted to and used their unique heritage of mixed legal traditions and their intellectual rigour and values of the Scottish Enlightenment in order to take their place outwith their jurisdictional boundaries
Assuring privacy of AI-powered community driven Android code vulnerability detection.
No full textThe challenge of training AI models is heightened by the limited availability of data, particularly when public datasets are insufficient. While obtaining data from private sources may seem like a viable solution, privacy concerns often prevent data sharing. Therefore, it is essential to establish a system that effectively balances privacy concerns with the need for data. In our previous work, we introduced "Defendroid", which focuses on real-time Android code vulnerability detection using a blockchain federated neural network with explainable artificial intelligence. In this study, the Defendroid approach is enhanced by incorporating variable differential privacy techniques to ensure the privacy of the model training process. The proposed method significantly improves privacy, achieving a privacy budget between 1 and 1.5, while maintaining Defendroid's baseline accuracy of 96% and an F1-Score of 0.96. As a result, this research thoroughly addresses concerns about the privacy of source code, filling a critical gap. This advancement not only showcases the effectiveness of the new approach but also its capability to address the significant challenges of privacy and data scarcity in AI-driven, community-focused Android code vulnerability detection
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!