2063 research outputs found
Sort by
Efficient Localization via Soft Information With Generic Sensing Measurements
Full text linkAccurate location awareness is essential for various context-based applications. This calls for efficient methodologies to collect, communicate and process position-dependent measurements, especially in situations with limited computational resources. The soft information (SI) approach has recently shown significant improvements in accuracy over conventional localization methods. By developing efficient SI-based techniques, it is possible to achieve higher precision also in case of stringent computational constraints. This paper proposes new SI-based localization techniques that utilize belief condensation and maximum entropy methods to reduce both communication burden and computational complexity. In addition, the techniques presented enable the use of generic sensing measurements, including those taking discrete and categorical values. Through two case studies involving time and angle measurements, we demonstrate how the proposed approach can significantly improve localization accuracy and computational efficiency.PID2022-137063NB-I0
Discontinuity-Induced Dynamics in the Conductance-Based Adaptive Exponential Integrate-and-Fire Model
Full text linkIn this article, we present a computational study of the Conductance-Based Adaptive Exponential (CAdEx) integrate-and-fire neuronal model, focusing on its multiple timescale nature, and on how it shapes its main dynamical regimes. In particular, we show that the spiking and so-called delayed bursting regimes of the model are triggered by discontinuity-induced bifurcations that are directly related to the multiple-timescale aspect of the model, and are mediated by canard solutions. By means of a numerical bifurcation analysis of the model, using the software package coco, we can precisely describe the mechanisms behind these dynamical scenarios. Spike-increment transitions are revealed. These transitions are accompanied by a fold and a period-doubling bifurcation, and are organised in parameter space along an isola periodic solutions with resets. Finally, we also unveil the presence of a homoclinic bifurcation terminating a canard explosion which, together with the presence of resets, organises the delayed bursting regime of the model
Charge Transport at Atomic Scales in 1D-Semiconductors: A Quantum Statistical Model Allowing Rigorous Numerical Studies
Full text linkThere has been a recent surge of interest in understanding charge transport at atomic scales. The motivations are myriad, including understanding the conductance properties of peptides measured experimentally. In this study, we propose a model of quantum statistical mechanics which aims to investigate the transport properties of 1D-semiconductor at nanoscales. The model is a two-band Hamiltonian in which electrons are assumed to be quasi-free. It allows us to investigate the behaviour of current and quantum fluctuations under the influence of numerous parameters, showing the response with respect to varying voltage, temperature and length. We compute the current observable at each site and demonstrate the local behaviour generating the current.PID2023-146683OB-100 funded by MICIU/AEI
/10.13039/501100011033 and by ERDF, E
A GENERIC-guided active learning SPH method for viscoelastic fluids using Gaussian process regression
Full text linkWhen applying machine learning methods to learn viscoelastic constitutive relations, the poly-
mer history dependence in viscoelastic fluids and the generalization ability of machine learn-
ing models are challenging. In this paper, guided by the general equation for nonequilib-
rium reversible-irreversible coupling (GENERIC) framework, a novel GENERIC-guided active
learning smoothed particle hydrodynamics (G2ALSPH) method is proposed to obtain effective
constitutive relations for reliable simulations of viscoelastic flows. By utilizing the GENERIC
framework, the target viscoelastic constitutive relation is reduced to a simple functional relation
between the eigenvalues of the conformation tensor and the eigenvalues of its thermodynam-
ically conjugated tensorial variable, which incorporates the flow-history-dependent memory
effect. Based on data and Gaussian process regression (GPR), a new active learning strategy
is developed to obtain the simplified constitutive relation, in which the generalization ability is
ensured by actively acquiring more data points when needed. Moreover, a novel relative uncer-
tainty is devised to establish an accuracy evaluation tool for the GPR prediction results, which
reduces the number of required training data points while maintaining accuracy. Furthermore,
the SPH method combined with the latest techniques serves as an effective macroscopic numer-
ical method. Eventually, the Poiseuille flows and the flows around a periodic array of cylinders
at different Weissenberg numbers are simulated to validate the effectiveness and accuracy of
the G2ALSPH method. The Oldroyd-B model is used as the ground truth constitutive relation
to provide the required data for GPR, hence bringing analytical solutions for comparison. The
excellent performance demonstrates that the G2ALSPH method has a promising prospect of
applications for data-driven simulations of viscoelastic fluids
Augmenting MRI scan data with real-time predictions of glioblastoma brain tumor evolution using faster exponential time integrators
No full textIntegrated damage detection and time-series data augmentation for floating offshore mooring systems via variational semi-supervised learning
No full textThe dynamics and stability of the semi-submersible offshore platforms are significantly impacted by the degradation of the mooring system. Identifying structural integrity issues in mooring systems through a data-driven approach is challenging due to the infrequency of damage events and the difficulties in recording them. To address these challenges, this study proposes the Time-Series Variational Semi-Supervised Learning (TSVSSL) framework, which effectively bridges the gap between supervised and unsupervised learning by leveraging unlabelled data for damage detection. The proposed framework features a distinctive training procedure in which the encoder-decoder and classifier components are trained concurrently. This process produces a well-clustered latent representation that enhances damage detection and supports class-specific artificial data generation. A numerical study using simulated responses of a 5 MW semi-submersible FOWT under varying metocean conditions demonstrated that the proposed framework outperformed existing deep learning methods in damage detection, achieving superior accuracy, precision, recall, and F1 score. Further, a rejection sampling technique is also introduced to effectively generates artificial data that closely aligns with actual time series displacement response. The novelty of the proposed framework lies in its dual focus on damage detection and artificial data generation marking a significant advancement in the data-driven assessment of mooring systems
Boundary layers for the upper-convected Beris–Edwards model of nematic liquid crystals
No full textIn this paper, we derive and analyze a set of Prandtl-type equations for the
boundary layers in nematic liquid crystals. We focus on a two-dimensional
model where the hydrodynamics are governed by the Beris–Edwards equations
with a shape parameter ξ= 1, specifically emphasizing the upper convected
derivative in the order-tensor equation. We introduce a novel decomposition of
the order tensor, which, combined with an Ansatz inspired by Prandtl’s theory,
leads to a set of limiting equations as the Reynolds, Ericksen, and Deborah
numbers approach infinity. We explore two distinct regimes of the dimen-
sionless parameters in the Beris–Edwards equations. The first regime results
in a partial decoupling in the limiting equations, where the velocity field is
unaffected by the order tensor, though the order tensor is influenced by the
flow. In the second regime, we derive a fully coupled system. Our analytical
investigation of the derived models reveals that, in the decoupled case, the
limiting equations admit analytic-type solutions, while in the coupled setting,
the equations allow for shear-flow type solutions
A Least-Squares-Based Neural Network (LS-Net) for Solving Linear Parametric PDEs
Full text linkDeveloping efficient methods for solving parametric partial differential equations is crucial for addressing inverse problems. This work introduces a Least-Squares-based Neural Network (LS-Net) method for solving linear parametric PDEs. It utilizes a separated representation form for the parametric PDE solution via a deep neural network and a least-squares solver. In this approach, the output of the deep neural network consists of a vector-valued function, interpreted as basis functions for the parametric solution space, and the least-squares solver determines the optimal solution within the constructed solution space for each given parameter. The LS-Net method requires a quadratic loss function for the least-squares solver to find optimal solutions given the set of basis functions. In this study, we consider loss functions derived from the Deep Fourier Residual and Physics-Informed Neural Networks approaches. We also provide theoretical results similar to the Universal Approximation Theorem, stating that there exists a sufficiently large neural network that can theoretically approximate solutions of parametric PDEs with the desired accuracy. We illustrate the LS-net method by solving one- and two-dimensional problems. Numerical results clearly demonstrate the method's ability to approximate parametric solutions
Dynamic counteraction of Maxwell-Wagner polarization with frequency-dispersive pseudo-inductive effect in V2O5 nanorods: role of oxygen vacancy and detrapped carriers towards RF decoupler
Full text linkA frequency-dispersive impedance analysis of polycrystalline V2O5 nanorods is performed under an external DC field to investigate the intricate interplay between dielectric polarization and delocalized carrier transport. Rietveld refinement of powder XRD patterns confirms the phase purity and provides essential crystallographic information, while SEM and HRTEM analyses offer comprehensive insights into the morphology. The elemental composition, chemical bonding, stoichiometry, and surface characteristics are examined using XPS spectra. Evidence of oxygen vacancies from both Rietveld refinement and XPS analysis motivates a DFT-based comparative study of the electronic properties of pristine V2O5 and oxygen-vacancy-incorporated V2O5 supercell. Beyond the conventional frequency-dispersive dielectric, impedance, and admittance spectra, a key observation is the capacitance switching from positive to negative values at higher frequencies (~MHz range) under varying external DC bias. Detailed fitting of the experimental data reveals a bi-relaxation mechanism: one component arises from Maxwell-Wagner interfacial polarization between grain cores and boundaries, dominant at lower frequencies, while the second mechanism involves the hopping of delocalized carriers, resulting in a high-frequency degenerative pseudo-inductive response. Notably, both of the characteristic frequencies decrease with increasing bias. The counteractive interference between these mechanisms manifests in the asymmetric Breit-Wigner-Fano (BWF) profile observed in the dielectric susceptance spectra. Finally, the equivalent circuit parameters and overall frequency-dispersive characteristics suggest the potential application of V2O5-based metal-insulator-metal (MIM) devices as radio-frequency decouplers.JDC2022-049793-I/MCIN/AEI/10.13039/501100011033
RES, QHS-2023-2-003
Learning the Graph Structure of Regular Vine-Copulas from Dependence Lists
No full textRegular vine copulas (R-vines) provide a comprehensive framework for modeling high-
dimensional dependencies using a hierarchy of trees and conditional pair-copulas.
While the graphical structure of R-vines is traditionally derived from data, this
work introduces a novel approach by utilizing a (conditional) pairwise dependence
list. Our primary goal is to construct R-vine graphs that include the maximum
possible number of dependence relationships specied in such lists. To tackle this
optimization challenge, characterized by exponential growth in the search space and
the structural constraints of R-vines, we propose two distinct methodologies: A 0-1
linear programming formulation and a Genetic Algorithm (GA). Additionally, the
Randomized Constructive Technique (RCT) is employed to generate initial population
of the GA, serving as a baseline for our comparison. Experimental results reveal the
superior performance of the GA over the RCT in terms of success rate, incorporating
more relationships than RCT into the constructed R-vine graphs and achieving near-
optimal or optimal graph structures.IT1504-22
PID2022-137442NB-I00
PID2023-149195NB-I0