Search Results (33)

This study presents a topology optimization approach to enhance the discharging performance of a latent heat thermal energy storage (LHTES) system using paraffin wax as the phase-change material (PCM) and a high-conductivity aluminium structure. Solidification is primarily governed by conduction, and the average heat transfer rate during this process is significantly lower than during melting; therefore, the optimization focused on the discharge phase. In a previous study, a novel LHTES device based on a Cartesian lattice was investigated experimentally and numerically. The validated numerical model from that study was adopted as the reference and used in a 2D topology optimization study based on the Solid Isotropic Material with Penalization (SIMP) method. The objective was to promote more uniform temperature distribution and reduce discharging time while maintaining the same aluminium volume fraction as in the reference device. Topology optimization produced a branched fin design, which was then extruded into a 3D model for comparison with the reference geometry. The optimized design resulted in improved temperature uniformity and a faster solidification process. Specifically, the time required to solidify 90% of the PCM was reduced by 12.3%, while the time to release 90% of the latent heat during the solidification process improved by 7.6%. Full article

Background: While timely blood transfusion is critical for restoring oxygen-carrying capacity after coronary artery bypass grafting (CABG), allogeneic blood product transfusions are independently associated with increased long-term mortality, necessitating a risk-stratified approach to balance oxygen delivery against immunological complications and infection risks. Methods: We retrospectively analyzed 3376 patients undergoing isolated CABG between 2005 and 2023 at a single tertiary center. Patients who died during their perioperative hospital stay within 30 days were excluded. Transfusion burden was assessed both as the absolute number of blood product units (packed red blood cells, platelet transfusion, fresh frozen plasma) and as a percentage of calculated patient blood volume. The primary outcome was all-cause mortality at 5 years. Flexible Cox regression with penalized smoothing splines, adjusted for EuroSCORE II, was used to model dose–response relationships. Results: From our cohort of 3376 patients, a total of 137 patients (4.05%) received >10 units of packed red blood cells (PRBC) perioperatively. These patients were older (median 71 vs. 68 years, p < 0.001), more often female (29% vs. 15%, p < 0.001), and had higher preoperative risk (EuroSCORE II: 2.53 vs. 1.41, p < 0.001). After 5 years, mortality was 42% in the massive transfusion group versus 10% in controls. Spline regression revealed an exponential increase in mortality with transfused units: 14 units yielded a 1.5-fold higher hazard of death (HR 1.46, 95% CI 1.31–1.64), rising to HR 2.71 (95% CI 2.12–3.47) at 30 units. When transfusion was indexed to blood volume, this relationship became linear and more tightly correlated with mortality, with lower maximum hazard ratios and narrower confidence intervals. Conclusions: Indexing transfusion burden to the percentage of patient blood volume replaced provides a more accurate and clinically actionable predictor of 5-year mortality after CABG than absolute unit counts. Our findings support a shift toward individualized, volume-based transfusion strategies to optimize patient outcomes and resource stewardship in a time of limited availability of blood products. Full article

Recycling plastic waste has emerged as one of the most pressing environmental challenges of the 21st century. One of the biggest challenges in polyethylene terephthalate (PET) recycling is the requirement to return bottles in their original, undeformed state. This necessitates storing large volumes of waste and takes up substantial space. Therefore, this paper seeks to address this issue and introduces a novel AIoT-based infrastructure that integrates the PET Bottle Identification Algorithm (PBIA), which can accurately recognize bottles regardless of color or condition and distinguish them from other waste. A detailed study of Azure Custom Vision services for PET bottle identification is conducted, evaluating its object recognition capabilities and overall performance within an intelligent waste management framework. A key contribution of this work is the development of the Algorithm for Citizens’ Trust Level by Recycling (ACTLR), which assigns trust levels to individuals based on their recycling behavior. This paper also details the development of a cost-effective prototype of the AIoT system, demonstrating its low-cost feasibility for real-world implementation, using the Asus Tinker Board as the primary hardware. The software application is designed to monitor the collection process across multiple recycling points, offering Microsoft Azure cloud-hosted data and insights. The experimental results demonstrate the feasibility of integrating this prototype on a large scale at minimal cost. Moreover, the algorithm integrates the allocation points for proper recycling and penalizes fraudulent activities. This innovation has the potential to streamline the recycling process, reduce logistical burdens, and significantly improve public participation by making it more convenient to store and return used plastic bottles. Full article

Intricate structures with minimal weight and maximum stiffness are demanded in many practical engineering applications. Topology optimization is a method for designing these structures, and the rise of additive manufacturing technologies has opened the door to their production. In a recently published paper, a novel topology optimization algorithm, named the Updated Properties Model (UPM), was developed with the homogenization of strain level as an objective function and an updating Young modulus as the design variable. The UPM method optimizes mechanical structures without applying any constraints. However, including constraints such as volume, mass, and/or stress in topology optimization is prevalent. This paper uses the density-dependent Young modulus concept to incorporate the volume fraction in the UPM method. We address the critical problem of constraint-aware design without the complexity of constraint-handling formulations. We show the proposed methodology’s success and functionality by plotting the algorithm’s results in two- and three-dimensional benchmark structures. Key results present that adjusting algorithmic parameters can yield both binary (single-material) and graded-material solutions, offering flexibility for different applications. These findings suggest that the UPM can effectively replicate constraint-driven outcomes without explicitly enforcing constraints. The main novelty of this work lies in extending the constraint-free UPM framework to allow for controlled material distribution using a physically meaningful update rule. This extends the applicability of the UPM beyond previous efforts in the literature. We have also created a Julia package for our proposal. Full article

The advancement of topology optimization (TO) and additive manufacturing (AM) has significantly enhanced structural design flexibility and the potential for lightweight structures. However, challenges such as intermediate density, mesh dependency, checkerboard patterns, and local extrema in TO can lead to suboptimal performance. Moreover, existing AM technologies confront geometric constraints that limit their application. This study investigates minimum compliance as the objective function and volume as the constraint, employing the solid isotropic material with penalization method, density filtering, and the method of moving asymptotes. It examines how factors like mesh type, mesh size, volume fraction, material properties, initial density, filter radius, and penalty factor influence the TO results for a metallic gooseneck chain. The findings suggest that material properties primarily affect numerical variations along the TO path, with minimal impact on structural configuration. For both hexahedral and tetrahedral mesh types, a recommended mesh size is identified where the results show less than a 1% difference across varying mesh sizes. An initial density of 0.5 is advised, with a filter radius of approximately 2.3 to 2.5 times the average unit edge length for hexahedral meshes and 1.3 to 1.5 times for tetrahedral meshes. The suggested penalty factor ranges of 3–4 for hexahedral meshes and 2.5–3.5 for tetrahedral meshes. The optimal geometric reconstruction model achieves weight reductions of 23.46% and 22.22% compared to the original model while satisfying static loading requirements. This work contributes significantly to the integration of TO and AM in engineering, laying a robust foundation for future design endeavors. Full article

Gradient porous structures (GPS) offer significant mechanical and functional advantages over homogeneous counterparts. This paper proposes a computational design framework utilizing spatial Voronoi diagrams to create lightweight, stress-responsive spatial frames optimized for architectural double-curvature arched shell roofing components. The method integrates Solid Isotropic Material with Penalization (SIMP)-based topology optimization (TO) to establish initial stress-informed material distributions, adaptive Voronoi control point (CP) placement guided by localized stress data, and a bi-objective genetic algorithm (GA) optimizing maximum and average displacement. Following optimization, a weighted Lloyd relaxation (LR) refines Voronoi cells into spatial frameworks with varying densities corresponding to stress gradients. Finite Element Analysis (FEA) demonstrates that the optimized Voronoi-driven GPS achieves notable improvements, revealing up to 79.7% material volume reduction and significant improvement in structural efficiency, achieving a stiffness-to-weight ratio (SWR) exceeding 2200 in optimized configurations. Furthermore, optimized structures consistently maintain maximum von Mises (MVM) stresses below 20 MPa, well within the allowable yield strength of the Polyethylene Terephthalate Glycol (PETG) material (53 MPa). The developed framework effectively bridges structural performance, material efficiency, and aesthetic considerations, offering substantial potential for application in advanced, high-performance architectural systems. Full article

With the explosion of information, the amount of text data has increased significantly, making text categorization a central area of research in natural language processing (NLP). Traditional machine learning methods are effective, but deep learning models excel in processing semantic information. Models such as CNN, RNN, LSTM, and GRU have emerged as powerful tools for text classification. Pre-trained models such as BERT and GPT have further advanced text categorization techniques. Contrastive learning has become a key research focus aimed at improving classification performance by learning the similarities and differences between samples using models. However, existing contrastive learning methods have notable shortcomings, primarily concerning insufficient data utilization. This study focuses on data enhancement techniques to expand the text data through symbol insertion, affirmative auxiliary verbs, double negation, and punctuation repetition, aiming to improve the generalization and robustness of the pre-trained model. Two data enhancement strategies, affirmative enhancement and negative transformation, are introduced to deepen the data’s meaning and increase the volume of training data. To address the introduction of false data, an instance weighting method is employed to penalize false negative samples, while complementary models generate sample weights to mitigate the impact of sampling bias. Finally, the effectiveness of the proposed method is demonstrated through several experiments. Full article

In the context of central solar receiver systems, the utilisation of S-CO₂ Brayton cycles as opposed to Rankine cycles confers a number of advantages, including enhanced efficiency, the requirement for less sophisticated turbomachinery, and a reduction in water consumption. A pivotal consideration in the design of such systems pertains to the thermal storage system. This work undertakes a comparative analysis of the performance of an S-CO₂ Brayton cycle utilising two distinct types of molten salts, namely solar salts and chloride salts (MgCl₂–KCl), as the heat transfer fluid on the thermal energy storage medium. The present study adopts an energetic and exergetic perspective with the objective of identifying areas of high irreversibility and proposing mechanisms to reduce them. The work is concluded with an analysis of the size of the different components. The overall energy efficiency is determined as 22.29 % and 23.76 % for solar and chloride salts, respectively. In the case of chloride salts, this efficiency is penalized by the higher losses in the solar receiver due to the higher operating temperature. The exergy analysis shows that using MgCl₂–KCl salts increases exergy destruction in the recuperators, lowering irreversibilities in other components. While the sizes of all components decrease when using chloride salts, the volume of the storage system increases. These results demonstrate that the incorporation of MgCl₂–KCl salts enhances the performance of S-CO₂ recompression cycles operating in conjunction with a central solar receiver. Full article

The lightweight and displacement-stable design of the mechanical support structure within the APTS (Acquisition, Pointing, and Tracking System) is crucial for enhancing the payload capacity of remote sensing, satellite communication, and laser systems, while still meeting specified functional requirements. This paper adopts the Solid Isotropic Material with Penalization (SIMP) method to investigate the structural topology optimization of the L-shaped bracket in the APTS, aiming to minimize structural compliance while using volume, key point displacement, and maximum stress as constraints. In the optimization model, differences in the topology of the L-shaped bracket structure are explored to minimize structural compliance, which was performed under volume, key point displacement, and stress constraints, and the results are compared with the initial reinforced structure. The innovative L-shaped bracket structure obtained through topology optimization uses significantly less material than the initial reinforced design, while still meeting the displacement and stress constraints. After smoothing, rounding, and finite element analysis, the displacement and stress performance of the optimized L-shaped bracket structure satisfies the set constraints. The method proposed in this paper offers an innovative solution for the lightweight design of mechanical support structures in APTS, with significant engineering application potential. Full article

In this work, we conducted a numerical study on the cavitation flow around a circular cylinder with $Re=200$ and $\sigma =1,$ through the implementation of a porous coating. The primary objective addressed the effectiveness of utilizing a porous surface to control cavitation. We analyzed the cavitation dynamics around the cylinder and the hydrodynamic performance at different permeability levels of the porous surfaces ($K={10}^{-12}-{10}^{-10}$). The flow was governed by the density-based homogeneous mixture model, and the volume penalization method was used to deal with the porous layer. A high-order compact numerical method was adopted for the simulation of the cavitating flow through solving the preconditioned multiphase equations. The hydrodynamic findings demonstrated that the fluctuations in the lift coefficient decreased when the porous layer was applied. However, it is not possible to precisely express an opinion about drag because the drag coefficient may vary, either increasing or decreasing, depending on the permeability within a constant thickness of the porous layer. The results revealed that the application of a porous layer led to the effective suppression of cavitation vortex shedding. In addition, a reduction of the shedding frequency was obtained, which was accompanied by thinner and elongated vortices in the wake region of the cylinder. With the proper porous layer, the inception of the cavitation on the cylinder was suppressed, and the amplitude of pressure pulsations due to the cavitation shedding mechanism was mitigated. Full article

Deep learning models have gained prominence in human activity recognition using ambient sensors, particularly for telemonitoring older adults’ daily activities in real-world scenarios. However, collecting large volumes of annotated sensor data presents a formidable challenge, given the time-consuming and costly nature of traditional manual annotation methods, especially for extensive projects. In response to this challenge, we propose a novel AttCLHAR model rooted in the self-supervised learning framework SimCLR and augmented with a self-attention mechanism. This model is designed for human activity recognition utilizing ambient sensor data, tailored explicitly for scenarios with limited or no annotations. AttCLHAR encompasses unsupervised pre-training and fine-tuning phases, sharing a common encoder module with two convolutional layers and a long short-term memory (LSTM) layer. The output is further connected to a self-attention layer, allowing the model to selectively focus on different input sequence segments. The incorporation of sharpness-aware minimization (SAM) aims to enhance model generalization by penalizing loss sharpness. The pre-training phase focuses on learning representative features from abundant unlabeled data, capturing both spatial and temporal dependencies in the sensor data. It facilitates the extraction of informative features for subsequent fine-tuning tasks. We extensively evaluated the AttCLHAR model using three CASAS smart home datasets (Aruba-1, Aruba-2, and Milan). We compared its performance against the SimCLR framework, SimCLR with SAM, and SimCLR with the self-attention layer. The experimental results demonstrate the superior performance of our approach, especially in semi-supervised and transfer learning scenarios. It outperforms existing models, marking a significant advancement in using self-supervised learning to extract valuable insights from unlabeled ambient sensor data in real-world environments. Full article

During the digital transformation of construction projects, the significant volume of project data raise a multitude of data responsibility issues. Project stakeholders, often motivated by financial interests and other considerations, frequently engage in data fraud, namely the alienation of project digital responsibility (APDR), which ultimately hinders the benefits released by the digital transformation of projects. However, the causes of APDR are still unclear. This study aims to bridge this knowledge gap by empirically investigating the factors influencing APDR and delineating their pathways. A model outlining the mechanism of APDR formation, rooted in fraud risk factor theory (FRFT) and information asymmetry theory (IAT), is proposed. To collect data from 276 Chinese construction project practitioners, a questionnaire was meticulously designed. Confirmatory factor analysis (CFA) was subsequently applied to assess the validity of the proposed model. Finally, the proposed model consisting of six variables was examined using structural equation modeling (SEM). The results showed that opportunity (OPP), motivation (MOT), and information asymmetry (INF) had a positive effect on APDR, while exposure probability (EXP), penalty strength (PEN), and ethics (ETH) had a negative effect on APDR. Through revealing the formation mechanism of APDR, the findings are beneficial for understanding why stakeholders adopt APDR at the risk of being penalized. This study aims at deepening the systematic understanding of APDR and enriches the relevant theories on project digital responsibility (PDR). Such knowledge would also contribute to project managers proposing effective interventions to inhibit APDR and promote PDR. Full article

Analysis methods to determine the optimal combination of volume and intensity of objectively measured physical activity (PA) with prospective outcomes are limited. Participants in UK Biobank were recruited in the UK between 2006 and 2010. We linked the questionnaire and accelerometer with all-cause mortality data from the NHS Information Centre and NHS Central Register up to April 2021. We developed a novel method, extending the penalized spline model of Augustin et al. to a smooth additive Cox model for survival data, and estimated the prospective relationship between intensity distribution and all-cause mortality, adjusting for the overall volume of PA. We followed 84,166 men and women (aged 40–69) for an average of 6.4 years (range 5.3–7.9), with an observed mortality rate of 22.2 deaths per 1000. Survival rates differed by PA volume quartile, with poorer outcomes for the lowest PA volumes. Participants with more sedentary to light intensity PA (<100 milligravities (mg)) and/or less vigorous intensity PA (>250 mg) than average for a given volume of PA, had higher mortality rates than vice versa. Approximate hazard ratios were 0.83 (95% credible interval [CI]: 0.79, 0.88) for an average-risk profile compared to a high-risk profile and 0.80 (95% CI: 0.74, 0.87) for a low-risk profile compared to an average-risk profile. A high- versus low-risk profile has the equivalent of 15 min more slow walking, but 10 min less moderate walking. At low PA volumes, increasing overall volume suggests the most benefit in reducing all-cause mortality risk. However, at higher overall volumes, substituting lighter with more vigorous intensity activity suggests greater benefit. Full article

The iron and steel industry is a large emitter of CO₂ globally. This is especially true for the iron and steel industries in China, Japan, and Korea due to their production volumes and the prevalence of carbon-based steel production. With few low-carbon and commercially available alternatives, the iron and steel industry is truly a hard-to-abate sector. Each of the countries of interest have committed to a net-zero future involving the mitigation of emissions from steel production. However, few studies have investigated the means by which to achieve decarbonization beyond the inclusion of price signalling policies (e.g., carbon tax or emission trading schemes). Here, we use E3ME-FTT:Steel to simulate technology diffusion in the ISI under several policy environments and we investigate the likely impacts on the wider economy. The results show that penalizing carbon intensive processes can incentivize a transition towards scrap recycling, but it is relatively unsuccessful in aiding the uptake of low carbon primary steelmaking. A combination of support and penalizing policies can achieve deep decarbonisation (>80% emission reduction compared with the baseline). Mitigating the emissions in the iron and steel industry can lead to economic benefits in terms of GDP (China: +0.8%; Japan: +1.3%; Korea: +0.1%), and employment (Japan: +0.7%; Korea: +0.3%) with China, where job losses in the coal sector would negate job gains elsewhere, as the exception. Full article

We compared the image quality and quantification parameters through bayesian penalized likelihood reconstruction algorithm (Q.Clear) and ordered subset expectation maximization (OSEM) algorithm for 2-[¹⁸F]FDG-PET/CT scans performed for response monitoring in patients with metastatic breast cancer in prospective setting. We included 37 metastatic breast cancer patients diagnosed and monitored with 2-[¹⁸F]FDG-PET/CT at Odense University Hospital (Denmark). A total of 100 scans were analyzed blinded toward Q.Clear and OSEM reconstruction algorithms regarding image quality parameters (noise, sharpness, contrast, diagnostic confidence, artefacts, and blotchy appearance) using a five-point scale. The hottest lesion was selected in scans with measurable disease, considering the same volume of interest in both reconstruction methods. SUL_peak (g/mL) and SUV_max (g/mL) were compared for the same hottest lesion. There was no significant difference regarding noise, diagnostic confidence, and artefacts within reconstruction methods; Q.Clear had significantly better sharpness (p < 0.001) and contrast (p = 0.001) than the OSEM reconstruction, while the OSEM reconstruction had significantly less blotchy appearance compared with Q.Clear reconstruction (p < 0.001). Quantitative analysis on 75/100 scans indicated that Q.Clear reconstruction had significantly higher SUL_peak (5.33 ± 2.8 vs. 4.85 ± 2.5, p < 0.001) and SUV_max (8.27 ± 4.8 vs. 6.90 ± 3.8, p < 0.001) compared with OSEM reconstruction. In conclusion, Q.Clear reconstruction revealed better sharpness, better contrast, higher SUV_max, and higher SUL_peak, while OSEM reconstruction had less blotchy appearance. Full article