240106 research outputs found
Sort by
Ten questions concerning indoor dust
No full textDust is ubiquitous but heterogeneous and can be viewed through various lenses. Indoor dust commonly refers to particles in building interiors that have settled onto surfaces, albeit without a consensus scientific definition. Dust contains a myriad of chemicals and microbes in a complex mixture originating from multiple sources. Indoor dust thus serves as both an indicator of the constituents in an indoor environment and a source of exposure through ingestion, dermal contact, and/or inhalation of resuspended dust. The mass and composition of settled dust within a building varies by location. This variability and complexity of dust manifests in different physical and chemical properties on macro-, micro- and nanoscales, which in turn influences occupant exposure pathways and outcomes. For example, resuspension of allergens (e.g., by walking) may exacerbate asthma. Sampling via methods such as vacuuming or surface wiping allows for quantitative measurement of components to assess possible exposures. However, through collection or measurements, results may be biased, e.g., size. With large variations in dust composition and study methods, it can be difficult to compare studies. This heterogeneity impedes understanding the fate and transport of dust and, importantly, how dust impacts health. A standardized system to better define, sample, and characterize dust would help develop a more comprehensive understanding of dust across indoor environments. This paper aims to address outstanding questions regarding the chemical and biological characteristics of dust, how the characterization of dust is affected by sampling methods, and how design of indoor spaces affects the amount and qualities of accumulated dust.</p
Robust Correlated Equilibrium:Definition and computation
No full textWe study N-player finite games with costs perturbed due to time-varying disturbances in the underlying system and to that end, we propose the concept of Robust Correlated Equilibrium that generalizes the definition of Correlated Equilibrium. Conditions under which the Robust Correlated Equilibrium exists are specified, and a decentralized algorithm for learning strategies that are optimal in the sense of Robust Correlated Equilibrium is proposed. The primary contribution of the paper is the convergence analysis of the algorithm and to that end, we propose a modification of the celebrated Blackwell's Approachability theorem to games with costs that are not just time-average, as in the original Blackwell's Approachability Theorem, but also include the time-average of previous algorithm iterates. The designed algorithm is applied to a practical water distribution network with pumps being the controllers and their costs being perturbed by uncertain consumption due to the consumers. Simulation results show that each controller achieves no regret, and empirical distributions converge to the Robust Correlated Equilibrium.</p
Electromechanical Modeling of Localized Micro-Scale Piezoelectric Interaction at Impact Site
Get PDFThe non-conformal contact between a moving spherical ball and a fixed piezoelectric layer results in a micro-scale contact area. This research investigates the localized micro-scale piezoelectric interaction at the contact area on the piezoelectric layer’s electrical response under the impact excitation. The ball indentation causes time-variable 3D mechanical stress and electric displacement distributions around the impact site in the piezoelectric layer. In this study, an innovative semi-analytical model is developed that divides the piezoelectric layer into two zones. This technique enables 2D-axisymmetric analysis of non-axially symmetric patches, reducing computational complexity. The results indicate that a high-voltage zone appears beneath the contact area, which can significantly contribute to the piezoelectric’s electrical response. For instance, under a specific impact excitation and electrical boundary condition, the peak voltage across the piezoelectric thickness reaches 600 V while the output voltage is 1 V. Validation and comparison against experimental tests and the FEM method confirm the accuracy and efficiency of this model in the prediction of piezoelectric’s transient voltage signal and voltage peak applicable to energy harvesting and sensory applications. Finally, the developed model is applied in the practical optimization of an implantable energy harvester for biomedical applications
PosGNN:A Graph Neural Network Based Multimodal Data Fusion for Indoor Positioning in Industrial Non-Line-of-Sight Scenarios
Get PDFIn industrial environments, the wireless infrastructure is functional for offering services such as communication and positioning of industrial assets. However, the frequently occurring Non-Line-of-Sight (NLoS) conditions in industrial scenarios cause the wireless receiver to have positional information from a limited and varying number of wireless transmitters between consecutive time steps, leading to ambiguities in wireless infrastructure-based positioning. In this paper, we propose PosGNN, a novel data fusion solution based on the Graph Neural Network (GNN) approach that allows us to estimate the position of the User Equipment (UE) by fusing the positional information from the available wireless transmitters at each time step with the UE sensor technology. The performance of the proposed method is assessed using an experimental setup of Ultra-Wideband (UWB) technology as wireless infrastructure at 3.7 - 4.2GHz frequency band, the Inertial Measurement Unit (IMU) as UE-side sensor, and the Automated Guided Vehicle (AGV) as the target UE to be positioned. The experimental results demonstrate the exceptional performance of our approach over the conventional model-based approach, Extended Kalman Filter (EKF), and the data-driven approach, Deep Neural Network (DNN), achieving an average positioning error of less than 15 cm in harsh industrial environments.</p
Non-intrusive hourly occupancy detection in residential buildings using remotely readable water meter data: Validation and large-scale analysis
Get PDFOccupancy significantly influences building energy use; however, large-scale occupancy estimation, particularly in residential buildings, remains a challenge. While surveys and sensor data, particularly from remotely readable electricity meters, have been used for non-intrusive occupancy detection, water meter data have so far only been applied at daily resolution. This study introduces, validates, and applies a novel algorithm to infer hourly occupancy using data from remotely readable water meters. A new day definition, spanning noon to noon, is proposed to reflect typical daily rhythms better. Daily occupancy is inferred using water usage events and average consumption, akin to established methods. For days classified as occupied, hourly occupancy is estimated by associating water use with occupancy, followed by smoothing. Nighttime occupancy is inferred by linking the last evening and first morning occupancy. Validation was performed using one year of manually labelled data from ten households for daily occupancy (MCC = 0.982), and 147 days of unseen ground-truth data from nine households for hourly occupancy (MCC range: 0.267 to 0.795; building-weighted MCC = 0.594). Short absences were often misclassified as occupied, while longer absences (>3h) were reliably detected. Comparison with a state-of-the-art electricity-based algorithm showed superior performance of the proposed water-based method. Finally, the algorithm was applied to one year of data from 2,690 households, yielding plausible and interpretable occupancy patterns at both daily and hourly scales. These results demonstrate the method’s robustness and scalability, offering a promising approach for large-scale, non-intrusive occupancy monitoring.}<br/
Closing the data gap:leveraging pretrained neural networks for robotic surgical assessment on limited clinical data
Get PDFBackground: In robot-assisted surgery (RAS), surgical assessment is critical for ensuring competence and achieving optimal surgical outcomes. Artificial intelligence (AI)-based assessment offers an alternative to expert-based assessment but often requires large datasets, which are challenging to obtain. Transfer learning with pretrained algorithms may offer a potential solution and could reduce the need for clinical data. This study explores the use of transfer learning with preclinical porcine data to reduce the clinical data needed for action recognition (AC) and skills assessment (SA) in RAS. Methods: Abdominal, thoracic and urologic RAS procedures were video recorded. A convolutional neural network (CNN) with a Long Short-Term Memory (LSTM) layer, initially trained using preclinical data, was applied to the clinical dataset through three strategies; (1) direct application on the clinical dataset, (2) only training the LSTM and dense layers, and (3) retraining the entire network. For comparison, a baseline model was trained from scratch on clinical data. Results: Recordings from 15 procedures were included. The baseline clinical model achieved accuracies of 82.7% (AC) and 40.8% (SA). Direct application of the pretrained network resulted in accuracies of 84.8% (AC) and 51.6% (SA). Fine-tuning the LSTM and dense layers of the pretrained network yielded accuracies of 90.1% (AC) and 60.4 (SA), while retraining all layers achieved 90.5% (AC) and 57.6% (SA). Ablation analysis demonstrated higher accuracies with less data using transfer learning, 87.9% vs. 81.6%. Conclusions: Using pretrained preclinical AI models increases the accuracy of models trained on limited clinical data and reduces the need for clinical data. Public trial registry: www.clinicaltrials.gov (ID: NCT06612606).</p
Numerical investigation of sorption-enhanced ammonia synthesis for hydrogen storage: CFD modeling, experimental validation, and scale-up
Get PDFAmmonia (NH3) is increasingly recognized as a clean hydrogen (H2) carrier due to its high H2 content, carbon neutrality, and ease of storage and transport. Realizing its potential as an energy vector requires synthesis technologies that operate under mild conditions and are compatible with water electrolysis. This study develops a comprehensive computational fluid dynamics (CFD) model of sorption-enhanced NH3 synthesis with alternating catalyst and sorbent layers, validated against experimental data, to elucidate the in-situ sorption effects. The model is subsequently employed to scale up the process and to optimize the operation conditions of a pilot-scale reactor. Results indicate a 66.8% increase in H2 conversion with a six-layer configuration, while a 60-layer pilot reactor achieves over 90% conversion and 99.5% NH3 capture in single-pass operation. These findings demonstrate the scalability and potential of sorption-enhanced NH3 synthesis, providing a moderate and efficient pathway for H2 storage and transport.</p
Can Real-Time Prehospital Medical Record Data Presented on A Screen Enhance Team Readiness in the Emergency Department?:A Pilot Study
No full textBackground Emergency Medical Services (EMS) continuously document treatment and patient condition information in the electronic prehospital medical record (ePMR) during care. Only selected information is communicated via telephone to the emergency department (ED) and the waiting ED team, potentially leading to loss of valuable information. Objectives To pilot-test whether implementing real-time, screen-based access to prehospital medical records before patient arrival enhances the ED team’s readiness. Methods Pilot study of implementing wall-mounted screens connected to the ePMR system in the ED trauma rooms in the North Denmark Regional Hospital. Three months before and four months after implementation, we measured the overall self-reported readiness of the ED team by questionnaires. The readiness rating was based on a visual analogue scale (VAS 0–15) and three sub-questions. Results We included 393 questionnaires (traumas N = 199, medical emergencies N = 194) corresponding to capture of 46% (141/307) of all events. For all questionnaires combined, overall readiness increased from a median of 7.1 (IQR 6.5–12.9) to 12.8 (IQR 9.7–14), p < 0.001. Stratified by event type, results persisted. Trauma: 7.1 (6.8–12.7) to 13.4 (9–14), p < 0.001; medical emergency: 7.1 (5.7–12.9) to 12.2 (9.7–13.9), p < 0.001. Conclusions Measured by questionnaires, we found that easy access to real-time EMS patient data, visualized on a screen in the trauma room before receiving patients with traumas or medical emergencies, significantly increases the overall self-reported readiness of the ED team members. Trial registration None.</p
Experimental investigations of internal macro-scale convection in the loose-fill wood fiber insulation layer of a full-scale wall element
No full textWith increasing restrictions on the energy efficiency of buildings, thicker insulation layers are installed in new and refurbished buildings to reduce heat losses. Previous studies have indicated that internal macro-scale convection cells can occur in thick porous insulation layers, decreasing the thermal performance of the envelope component. The focus of previous studies has been on horizontal insulation layers, most often composed of glass wool. Therefore, there is a lack of empirical data for loose-fill insulation and, in particular, bio-based materials, which have the potential of being more sustainable than conventional ones. The present investigation of this paper looks at the possibility of internal macro-scale convection inside loose-fill wood fiber insulation in a full-scale vertical wall element, with the modified Rayleigh number in the current investigation being between 20 and 45 and exhibiting internal convection in all cases. The experimental results show good agreement in terms of heat flux and temperature distribution with numerical simulations where the macro-scale convection is modelled explicitly. It also indicates that internal macro-scale convection can be modelled with existing building physics simulation tools, such as COMSOL. Finally, the internal macro-scale convection increases the effective U-value by up to 90 % for the highest temperature difference in steady-state conditions. This effect appears to diminish under dynamic boundary conditions, with a calculated effective U-value being within the uncertainty of the steady-state case with the lowest temperature difference, indicating that it might be less influential under real conditions.</p
Federated neural nonparametric point processes
No full textTemporal point processes (TPPs) are effective for modeling event occurrences over time but struggle with sparse and uncertain events in federated systems, where privacy is a major concern. To address this, we propose FedPP, a federated neural nonparametric point process model. FedPP integrates neural embeddings into sigmoidal Gaussian Cox processes (SGCPs) on the client side. SGCPs is a flexible and expressive class of TPPs, allowing FedPP to generate highly flexible intensity functions that capture client-specific event dynamics and uncertainties while efficiently summarizing historical records. For global aggregation, FedPP introduces a divergence-based mechanism to communicate the distributions of kernel hyperparameters in SGCPs between the server and clients, while keeping client-specific parameters local to ensure privacy and personalization. FedPP effectively captures event uncertainty and sparsity. Extensive experiments demonstrate its superior performance in federated settings, showing global aggregation with the KL divergence and the Wasserstein distance.</p