20005 research outputs found
Sort by
Capillary trapping in mixed-wet porous media: Implications for subsurface carbon dioxide sequestration
No full textSubsurface sequestration of carbon dioxide (CO2) is driving efforts to attain carbon neutrality. For the safe
and optimal operation of such complex applications, it is imperative to understand the physics of fluids
displacement. Direct numerical simulations are used to investigate the flooding of two immiscible fluids
having viscosity contrasts in mixed-wet porous media, which are ubiquitous in reservoirs characterized by
multifarious mineralogizes and complex physico-chemical histories. Three mixed-wet systems having different
wettability ranges and one mono-wet case are considered for investigation. Flooding by low viscosity fluid
caused fingering. Though the fingering patterns vary for different wettability distributions, the sample-scale
morphological metrics for all cases are closely comparable. The fingering profiles are preserved and later
subjected to flooding by high viscosity fluid. Entrapment of the defending phase due to capillarity for different
wettability systems are investigated. When the wettability range increases, the trapping efficiency is also
seen to increase linearly, suggesting that reservoirs with strong mixed-wet conditions present an attractive
option for CO2 sequestration. Pore-scale fluid displacements reveal that during viscous fingering the fluidfluid
interface initially developed in non-wet zones retract which contribute towards cooperative pore filling
in the surrounding wetting zones that influence the characteristic features of invading fluid’s flow morphology.
Additionally, various possibilities by which the defending phase gets trapped by flow bypassing are explored.
Trapping was prominent in zones having an affinity to the defending phase. The average trapped ganglia size
increases commensurately with degree of dispersion in wettability. The study also highlights shortcomings of
analyzing multiphase flows in mono-wet systems. Insights from this study can be used for improving pore
network models and training machine learning algorithms
Influence of multiaxial loading and temperature on the fatigue behaviour of 2D braided thick-walled composite structures
No full textWhile size effects in composite structures have been widely studied under quasi-static uniaxial loading, their influence under fatigue conditions, particularly in the presence of multiaxial stress states and elevated temperatures, remains insufficiently understood. This study investigates the fatigue behaviour of thick-walled ±45∘ braided glass fibre-reinforced polyurethane composite box structures under varying temperature and loading conditions. A combined experimental approach is adopted, coupling quasi-static and fatigue tests on large-scale structures with reference data from standardised coupon specimens. The influence of temperature (23–80 °C) and multiaxial shear–compression loading is systematically evaluated. The results demonstrate a significant temperature-dependent decrease in compressive strength and fatigue life, with a linear degradation trend that aligns closely between the box structure and coupon data. Under moderate multiaxial conditions, the fatigue life of box structures is not significantly impaired compared to uniaxial test coupon specimens. Complementary non-destructive testing using air-coupled ultrasound confirms these trends, demonstrating that guided-wave phase-velocity measurements capture the evolution of anisotropic damage and are therefore suitable for in situ structural health monitoring applications. Furthermore, these findings highlight that (i) the temperature-dependent fatigue behaviour of thick-walled composites can be predicted using small-scale coupon data and (ii) small shear components have a limited impact on fatigue life within the studied loading regime
Opinion dynamics with median aggregation
No full textUnderstanding the formation and evolution of opinions is of broad interdisciplinary interest. Many classical models for opinion formation focus on the impact of different notions of locality, e.g., locality due to network effects among agents or the role of the proximity of opinions. In practice, however, opinion formation is often governed by the interplay of local and global influences. In this paper, we study an asynchronous opinion dynamics in a social network. Each agent has a static intrinsic opinion as well as a public opinion that is updated asynchronously over time. Moreover, agents have access to a global aggregate (e.g., the outcome of a vote) of all public opinions. We focus on the popular median voting rule and show that pure Nash equilibria always exist. For every initial state of the dynamics, a pure equilibrium can be reached. The set of reachable equilibria forms a complete lattice, and extremal equilibria can be computed in polynomial time. Indeed, there are instances and initial states from which the number of reachable equilibria is exponentially large. The global median in these equilibria can be any of the initial opinions. We show that by uniformly increasing the influence of the aggregate median we can enforce that the median opinion is the same in every reachable equilibrium. We can compute the increase scheme that achieves this property in polynomial time. Furthermore, we show that finding the k most influential agents is NP-complete
Optimization of drag embedment anchors applying multi-objective evolutionary algorithm NSGA-II
No full textEstablishing renewables on a floating platform in the deep sea needs secure anchoring to the seabed, commonly achieved with drag embedment anchors (DEAs). The conventional design process relies heavily on empirical testing and is often time and resource-intensive, potentially leading to suboptimal designs. This research aims to overcome these limitations by applying an evolutionary optimization algorithm to existing analytical solutions for DEAs, identifying optimal anchor fluke and shank lengths. By leveraging an optimization strategy, we aim to enhance the design process while diminishing the dependency on exhaustive physical testing and high computational cost. We employ the Non-Dominated Sorting Genetic Algorithm II (NSGA-II) to optimize anchor shapes, with a focus on three key objectives: maximizing embedment depth and bearing capacity, and minimizing anchor volume. The methodology presents a Pareto front, encompassing all optimal solutions based on the formulated objectives, and demonstrates the efficiency of NSGA-II as a tool for optimizing anchor shapes
Data-driven pressure field prediction for ships in regular sea states
No full textMerchant shipping is responsible for more than 90% of the global trade and has a significant environmental impact, accounting for over 2% of global greenhouse gas emissions. Therefore, fuel-saving
measures are becoming increasingly important in reducing the ecological footprint and increasing the fuel efficiency of maritime transport. Routing optimization systems, which require a rapid prediction of ambient-dependent fuel consumption, represent an essential pillar here, e.g. to reduce added resistances due to seaways and/or wind. The paper aims to predict the added resistance due to seaways. In contrast to conventional methods the goal is achieved by surrogate modelling of the entire pressure fields. To this end, an online/offline-procedure is applied to an exemplarily free-floating container vessel. The online approach to be trained consists of two building blocks, namely a convolutional autoencoder (CAE)-based order reduction step and a neural network-based (NN) regression step that links the reduced space of the autoencoder with three control parameters that describe the sea state (wave height/steepness, encounter angle and wave length). Training data is obtained from time-averaged values for simulating instantaneous ship motion and pressure fields. During the offline phase, the combined CAE/NN is trained to capture the time-averaged pressure fields for a variety of sea-state conditions. During ship operation (online phase), the surrogate model predicts the three-dimensional pressure fields in response to sea state conditions, projects the pressure fields onto the ship hull, and integrates the corresponding resistances to guide the route. The evaluation of the method shows promising results for the different building blocks and the concept could therefore represent an attractive approach for cost-effective surrogate modelling of complex multiphase flow fields
Geometric approach based on optimal toric packings for photonic crystals and metamaterials design
No full textWe introduce new classes of nanostructures—optimal toric packings of particles and their Voronoi tessellations—that exhibit remarkable properties for photonic and potentially for plasmonic/phononic crystals design. These structures connect the classical problem of optimal packings on tori with the development of optimized photonic architectures
Development of a print head for non-planar additive manufacturing of carbon fiber-reinforced polymers
No full textIn order to realize the potential of high-performance lightweight parts manufactured using the combination of Fused Filament Fabrication (FFF) and Carbon Fiber Reinforced Polymers (CFRP), two main challenges have been identified for a print head design intended for load-oriented non-planar printing. To address these objectives of enabling dynamic layer height variation and fiber cutting, all while avoiding collisions in the printing system, this work presents the methodological development of a new specialized print head. By analyzing the process and taking the existing non-planar printing system into account, the relevant optimization parameters are identified and accurately defined. The methodological development is aligned with the VDI standard 2221 and includes a formal requirement analysis, parameter definition, and functional structure. The solution space is discussed and partial solutions are compared in detail, before the integration and final design are presented. The print head is physically realized, functionally verified by printing CFRP parts, and shown to fulfill all set requirements. Minor necessary changes during the manufacturing and construction of the system are discussed. Finally, the insight, definition, process analysis, and final version are employed to present a further optimized design with a concrete methodological outlook on enabling the load-oriented non-planar FFF of CFRP with this new highly optimized design
Unveiling hidden intramolecular non-covalent interactions in a neutral serine, its zwitterion, cluster, and crystal by features of electron density
No full textWe investigate intramolecular non-covalent interactions (NCIs) in neutral serine, its zwitterion, molecular clusters, and crystal using electron density-based approaches, including QTAIM, RDG, IQA, and electronic pressure analysis. In addition to completed NCIs (hydrogen bonds with bond paths), we identify latent interactions—attractive, bond-path-free atomic pair interactions with negative interaction energies. These are classified into dynamic (vibration-induced and transient) and static (secondary, persistent but structurally passive) types. Analysis of the internal pressure in electronic continuum reveals that latent NCIs exhibit distinct signatures in the kinetic and exchange components, which evolve across the molecular, cluster, and crystalline states. Dynamic interactions are characterized by off-axis minima in the exchange part of the pressure, whereas static interactions lack such features. Upon crystallization, intramolecular latent NCIs may disappear due to electron density redistribution and the formation of intermolecular hydrogen bonds. These intermolecular contacts may also spatially constrain atoms, suppressing vibrational flexibility and effectively converting dynamic NCIs into static ones. The kinetic pressure highlights regions of electron localization, while the exchange pressure offers a physical criterion for distinguishing different types of NCIs. Our findings demonstrate the structural and stabilizing roles of latent interactions and establish electronic pressure as a sensitive and informative descriptor for their analysis
Deep learning for paranasal anomaly classification
No full textDie Anfälligkeit der Nasennebenhöhlen für allergische und nicht-allergische Infektionen erschwert die Diagnose, da Symptome von Schleimhautverdickungen bis zu polypoiden Massen variieren. Bisherige manuelle Analysen sind ressourcenintensiv und führen zu Ermüdung der Kliniker. Ein computergestütztes Diagnoseverfahren (CAD) kann diesen Prozess optimieren.
Während Deep Learning große Fortschritte in der medizinischen Bildanalyse gemacht hat, blieb die Erforschung paranasaler Anomalien bislang begrenzt. Diese Arbeit schlägt verschiedene Deep-Learning-Ansätze zur Klassifizierung von Kieferhöhlen-Opazitäten vor, darunter unüberwachtes Lernen, selbstüberwachtes Lernen und hybride Architekturen aus CNNs und Transformern.The susceptibility of the paranasal sinuses to allergic and non-allergic infections complicates diagnosis as symptoms vary from mucosal thickening to polypoid masses. Previous manual analysis is resource intensive and leads to clinician fatigue. Computer-aided diagnosis (CAD) can optimize this process.
While deep learning has made great strides in medical image analysis, the study of paranasal abnormalities has remained limited. This work proposes different deep learning approaches to classify maxillary sinus opacities, including unsupervised learning, self-supervised learning, and hybrid architectures of CNNs and transformers