78146 research outputs found
Sort by
A novel contrastive loss for zero-day network intrusion detection
Get PDFMachine learning has achieved state-of-the-art results in network intrusion detection; however, its performance significantly degrades when confronted by a new attack class—a zero-day attack. In simple terms, classical machine learning-based approaches are adept at identifying attack classes on whichthey have been previously trained, but struggle with those not included in their training data. One approach to addressing this shortcoming is to utilise anomaly detectors which train exclusively on benign data with the goal of generalising to all attack classes—both known and zero-day. However, this comes at the expense of a prohibitively high false positive rate. This work proposes a novel contrastive loss function which is able to maintain the advantages of other contrastive learning-based approaches (robustness to imbalanced data) but can also generalise to zero-day attacks. Unlike anomaly detectors, this model learns the distributions of benign traffic using both benign and known malign samples, i.e. other well-known attack classes (not including the zero-day class), and consequently, achieves significant performance improvements. The proposed approach is experimentally verified on the Lycos2017 dataset where it achieves an AUROC improvement of .000065 and .060883 over previous models in known and zero-day attack detection, respectively. Finally, the proposed method is extended to open-set recognition achieving OpenAUC improvements of .170883 over existing approaches
Seismic enhancement of masonry-infilled substandard reinforced concrete frames using lightweight steel exoskeleton
Get PDFA significant portion of existing reinforced concrete (RC) structures in seismically active regions was constructed prior to the adoption of modern seismic design standards, leaving them highly susceptible to earthquake-induced damage. The vulnerabilities of these structures, often exacerbated by material degradation, have been starkly revealed in recent seismic events. This study addresses the urgent need for effective retrofitting solutions by evaluating the seismic performance of deficient masonry-infilled RC frames retrofitted with a novel lightweight steel exoskeleton system—Resisto 5.9 Tube—designed to enhance structural resilience. Three full-scale RC frame specimens, replicating typical deficiencies of older construction practices, were subjected to quasi-static cyclic loading up to near-collapse conditions, with interstory drifts ranging from 0.05% to 2.50%. The test series included: a bare frame (BF), an infilled frame with unreinforced hollow clay masonry units (IF), and a retrofitted infilled frame (RIF) incorporating the steel exoskeleton. Results reveal that masonry infill substantially increases lateral load capacity—by factors of 2.42 (IF) and 3.59 (RIF) compared to BF. However, IF exhibited a brittle failure mode, with significantly reduced displacement capacity. In contrast, the exoskeleton-enhanced RIF demonstrated a 147% increase in load capacity relative to IF, extended peak force occurrence to 0.70% drift, and achieved improved cyclic stability. While initial stiffness remained comparable between IF and RIF (within 4% difference), energy dissipation in RIF at 1.50% drift was threefold that of IF. Further, the exoskeleton system markedly improved the performance of the infill wall, extending the defined limit and damage states at ultimate by up to 150% and 344%, respectively. These enhancements facilitated sustained infill–frame interaction under large drifts, a behavior often neglected in conventional seismic design. The findings position the Resisto 5.9 Tube as a cost-effective and scalable retrofitting solution, offering a paradigm shift in how infill contributions are considered in seismic response assessments. This work establishes a foundation for advanced analytical modeling and practical implementation in earthquake-prone regions
Advanced optimisation software framework for floating offshore wind farm logistics and operations
Get PDFOffshore wind farms are being built farther from shore to maximise energy production, but this creates significant logistical challenges for maintaining these remote turbines. Service Operation Vessels (SOVs) and Crew Transfer Vessels (CTVs) play a vital role in keeping turbines operational. However, their routes need to be carefully planned to reduce travel distances and fuel consumption while meeting tight maintenance schedules. This study introduces a Python-based optimisation framework designed to streamline both day-to-day and campaign-style maintenance operations. By integrating geospatial analysis, clustering algorithms, and multi-day scheduling, the framework generates efficient vessel routes considering different turbine locations, weather windows, technician capacity, and vessel availability. When full daily servicing is unfeasible, the framework prioritises tasks and creates multi-day schedules to ensure efficient use of resources. Clustering techniques further streamline the process by grouping nearby turbines for maintenance. A real-world case study demonstrated a 36 % reduction in fuel consumption compared to conventional methods, underscoring the framework's potential to lower operational costs and enhance sustainability. In addition to this efficiency gains, the solution mitigates risks by ensuring timely maintenance, thereby supporting the offshore wind sector's capacity to meet escalating energy demands reliably
Solvent driven pore engineering in coffee-derived activated hydrochar : implications for post-combustion CO2 capture
Get PDFEngineering the pore structure of biomass-derived activated carbons is critical for optimizing their performance in adsorption-based applications. This study demonstrates for the first time that washing hydrochars in solvents of different polarity before activation is a simple yet powerful strategy to tailor pore size distribution. Hydrochar is produced from spent coffee grounds via hydrothermal carbonization, followed by washing in various solvents and activation in KOH. This results in carbons with a very large surface area (∼2700 m2/g), and washing is demonstrated to significantly increase product yield. Furthermore, washing in non-polar or mixed-polarity solvents removes long-chain carboxylic acids and esters from the hydrochar, promoting the development of narrow micropores while suppressing mesopore formation. To illustrate the impact of this structural control of porous carbons, post-combustion CO2 capture is investigated as a case study. Narrower pore size distribution enhances CO2 uptake, significantly improving capacity from 2.8 mmol/g for unwashed samples to 3.8 mmol/g for acetone-washed samples. Interestingly, moderate pore size (9–12 Å) is shown to be optimal for CO2:N2 selectivity, while smaller pores result in lower selectivity due to stronger interactions between N2 and the pore walls. These findings highlight the potential role of solvent washing in directing pore architecture of hydrochars for adsorption-based carbon capture technologies and beyond
Monitoring bridge vibrations via spaceborne SAR micro‐Doppler
No full textThe advent of Synthetic Aperture Radar (SAR) imaging has presented the possibility of remote monitoring of civil infrastructure on a large scale. Although well established for observing slow and long-term phenomena, its application to vibration-based structural health monitoring (SHM) remains relatively unexplored in the current literature. This study demonstrates the use of micro-Doppler SAR (MDSAR) using data from spaceborne platforms for measuring structural vibrations of a real bridge, specifically the line of sight velocity time histories of the deck. These measurements are compared to synchronous ground truth data to validate the method and assess its accuracy. Experimental results show that MDSAR measures vibration with an error in velocity on the order of 1 mm/s and successfully identifies the bridge’s dominant frequencies from two separate SAR acquisitions at different times. Spectral correlation with ground truth data reaches values up to 0.88. Frequency estimation errors are essentially controlled by the resolution of the spectrum, which in turn is limited by the acquisition time. In this work, a frequency resolution of 0.06 Hz is achieved for an acquisition duration of 16 s. Given these results, it is expected that MDSAR could be suitable for monitoring natural frequencies and performing modal recognition for bridges. Further improvements in the technology and in the analysis algorithm could potentially enable the accurate measurement of mode shape components
Strengthening school-university collaborations
No full textChemistry is an experimental science that for many learners only comes alive in the laboratory. But specialized equipment is increasingly out of reach of school budgets. Strengthening school–university collaborations can help to bridge the gap
Evaluating uncertainty in global wave storm characteristics using CMIP6-derived wave climate simulations with SWAN and WAVEWATCH III models
Get PDFThis study evaluates the performance and uncertainties of two third-generation spectral wave models, SWAN and WW3, in simulating global wave storm characteristics, including mean and maximum storm wave heights, storm duration, and storm power. The models were forced with CMIP6-derived EC-Earth3 wind and sea ice data for 1984–2014 and validated against ERA5 reanalysis and in-situ buoy observations. Results show that SWAN model consistently underestimates storm wave height, particularly in tropical and high-energy regions, whereas WW3 aligns more closely with ERA5 but tends to overestimate storm wave heights and storm power in the Southern Ocean. Both models reproduce storm durations reasonably well, although WW3 exhibits fewer significant biases and narrower confidence intervals, reflecting higher reliability. Storm power analysis reveals SWAN's systematic underestimation and WW3's better overall performance, albeit with localized overestimations in extreme-energy basins. Comparisons with buoy data confirm WW3's improved accuracy in estimating storm durations and power, though challenges remain in replicating extreme wave heights. These findings underscore the inter-model uncertainty associated with different wave models and emphasize the need for refined wave model physics and regional calibration to improve the reliability of global wave-storm projections and better inform coastal planning and climate adaptation
Optimal screening procedures for items with a random number of defects
Get PDFIn this paper, we study optimal screening procedures for items with random number of defects. Each defect that causes an item’s failure during this screening procedure is repaired/removed. Upon observing the numbers of removed defects, a decision is made whether to discard an item or to justify its future field operation. It is shown that these decisions depend on the distribution of the number of defects in an item. Three discrete distributions are considered: negative binomial, Poisson and binomial. It is shown, e.g., for the negative binomial case, that screening out of items with any number of removed defects improves the quality of remaining items. On the other hand, for the Poisson distribution of defects, there is no need to screen out items, as the distribution of the number of remaining defects does not depend on the number of the removed defects. The optimal screening policies to minimize the corresponding expected cost functions for each case are analyzed. The numerical illustrations of the obtained results are provided. Through the numerical examples, it is shown that the optimal screening policy significantly differs depending on the distribution of the number of defects in an item as well as the involved costs
OMICmAge quantifies biological age by integrating multi-omics with electronic medical records
Get PDFBiological aging reflects complex cellular and biochemical processes that can be measured across multiple omic layers. Using routine clinical laboratory data from ~31,000 participants in the Mass General Brigham Biobank, we developed EMRAge, a biomarker of mortality risk that can be broadly recapitulated across electronic medical records. Here we show that EMRAge can be modeled using elastic net regression with DNA methylation and multi-omics to generate DNAmEMRAge and OMICmAge, respectively. Both biomarkers are strongly associated with incident and prevalent chronic diseases and mortality, performing comparably or better than current biomarkers across discovery (Massachusetts General Brigham Aging Biobank Cohort, n = 3,451) and validation cohorts (TruDiagnostic, n = 14,213; Generation Scotland, n = 18,672). Importantly, OMICmAge leverages epigenetic biomarker proxies to integrate proteomic, metabolomic and clinical domains while remaining quantifiable from DNA methylation alone. This framework establishes an accessible, scalable measure of biological aging with potential to reveal molecular interconnections that shape healthspan and disease risk. [Abstract copyright: © 2026. The Author(s).
Targeting non-canonical NF-κB signalling in CYLD cutaneous syndrome by selective inhibition of IκB kinase alpha
No full textCYLD cutaneous syndrome (CCS) skin tumours develop from puberty onwards, can number in the hundreds and progressively grow over time. CCS patients lack medical therapies and require repeated surgery to control tumour burden. CYLD loss of heterozygosity (LOH) drives tumour growth, and CCS tumours have previously been shown to demonstrate increased canonical NF-κB and Wnt signalling. Here, we demonstrate evidence of non-canonical NF-κB signalling in CCS tumour keratinocytes, with increased p100 to p52 processing and RelB protein expression compared to normal skin. Utilizing complementary transcriptomics and proteomics on patient derived CCS tumour cell fractions, we identify IκB kinase alpha (IKKα) as a candidate target in the non-canonical NF-κB signalling pathway. A novel, highly selective, IKKα inhibitor (SU1644) used in patient derived CCS tumour spheroid cultures demonstrated that IKKα inhibition reduced tumour spheroid viability. These data provide the pre-clinical rationale for the assessment of topical IKKα inhibitors as a novel preventative treatment for CCS
Access Repository Dashboard
Do you manage University of Strathclyde Institutional Repository? Access insider analytics, issue reports and manage access to outputs from your repository in the CORE Repository Dashboard!