arXiv subjects

Andreas Windisch

Publications and source records attributed to Andreas Windisch.

At least 19 recordsLinked to original sources

arXiv2024

Vibrations and damping of the eigenmodes of viscoelastic nanospheres with thermal conductivity

In this paper, we derive a series of exact analytical closed expressions to calculate the complex eigenfrequencies and the displacement for the corresponding eigenmodes of a viscoelastic (nano-)sphere in the presence of linear damping. Where possible, we provide closed expressions for damping rates, including the contributions from viscosity, as well as thermal conductivity and solutions of the heat equation. We assume an isolated system, such that no energy/heat transfer to the environment is allowed. We find monotonic behavior of the damping as a function of frequency for breathing and torsional modes, however, for spheroidal modes we find non-monotonicity. Furthermore, we analytically analyze the thermodynamic limit for all mode types. We also investigate the frequency shift and find expected behavior, i.e. a reduced eigenfrequency with damping than without damping for breathing and torsional modes. For spheroidal modes, however, we find non-monotonic shifts, corresponding to the damping. For some eigenfrequencies we find anomalous frequency shifts.

cond-mat.mtrl-sci↗

arXiv2023

Evaluation of Data Augmentation and Loss Functions in Semantic Image Segmentation for Drilling Tool Wear Detection

Tool wear monitoring is crucial for quality control and cost reduction in manufacturing processes, of which drilling applications are one example. In this paper, we present a U-Net based semantic image segmentation pipeline, deployed on microscopy images of cutting inserts, for the purpose of wear detection. The wear area is differentiated in two different types, resulting in a multiclass classification problem. Joining the two wear types in one general wear class, on the other hand, allows the problem to be formulated as a binary classification task. Apart from the comparison of the binary and multiclass problem, also different loss functions, i. e., Cross Entropy, Focal Cross Entropy, and a loss based on the Intersection over Union (IoU), are investigated. Furthermore, models are trained on image tiles of different sizes, and augmentation techniques of varying intensities are deployed. We find, that the best performing models are binary models, trained on data with moderate augmentation and an IoU-based loss function.

cs.CV↗

arXiv2022

Automatic Detection of Interplanetary Coronal Mass Ejections in Solar Wind In Situ Data

Interplanetary coronal mass ejections (ICMEs) are one of the main drivers for space weather disturbances. In the past, different approaches have been used to automatically detect events in existing time series resulting from solar wind in situ observations. However, accurate and fast detection still remains a challenge when facing the large amount of data from different instruments. For the automatic detection of ICMEs we propose a pipeline using a method that has recently proven successful in medical image segmentation. Comparing it to an existing method, we find that while achieving similar results, our model outperforms the baseline regarding training time by a factor of approximately 20, thus making it more applicable for other datasets. The method has been tested on in situ data from the Wind spacecraft between 1997 and 2015 with a True Skill Statistic (TSS) of 0.64. Out of the 640 ICMEs, 466 were detected correctly by our algorithm, producing a total of 254 False Positives. Additionally, it produced reasonable results on datasets with fewer features and smaller training sets from Wind, STEREO-A and STEREO-B with True Skill Statistics of 0.56, 0.57 and 0.53, respectively. Our pipeline manages to find the start of an ICME with a mean absolute error (MAE) of around 2 hours and 56 minutes, and the end time with a MAE of 3 hours and 20 minutes. The relatively fast training allows straightforward tuning of hyperparameters and could therefore easily be used to detect other structures and phenomena in solar wind data, such as corotating interaction regions.

astro-ph.SR↗

arXiv2022

Vehicle management in a modular production context using Deep Q-Learning

We investigate the feasibility of deploying Deep-Q based deep reinforcement learning agents to job-shop scheduling problems in the context of modular production facilities, using discrete event simulations for the environment. These environments are comprised of a source and sink for the parts to be processed, as well as (several) workstations. The agents are trained to schedule automated guided vehicles to transport the parts back and forth between those stations in an optimal fashion. Starting from a very simplistic setup, we increase the complexity of the environment and compare the agents' performances with well established heuristic approaches, such as first-in-first-out based agents, cost tables and a nearest-neighbor approach. We furthermore seek particular configurations of the environments in which the heuristic approaches struggle, to investigate to what degree the Deep-Q agents are affected by these challenges. We find that Deep-Q based agents show comparable performance as the heuristic baselines. Furthermore, our findings suggest that the DRL agents exhibit an increased robustness to noise, as compared to the conventional approaches. Overall, we find that DRL agents constitute a valuable approach for this type of scheduling problems.

cs.LG↗

arXiv2021

A machine learning pipeline for autonomous numerical analytic continuation of Dyson-Schwinger equations

Dyson-Schwinger equations (DSEs) are a non-perturbative way to express n-point functions in quantum field theory. Working in Euclidean space and in Landau gauge, for example, one can study the quark propagator Dyson-Schwinger equation in the real and complex domain, given that a suitable and tractable truncation has been found. When aiming for solving these equations in the complex domain, that is, for complex external momenta, one has to deform the integration contour of the radial component in the complex plane of the loop momentum expressed in hyper-spherical coordinates. This has to be done in order to avoid poles and branch cuts in the integrand of the self-energy loop. Since the nature of Dyson-Schwinger equations is such, that they have to be solved in a self-consistent way, one cannot analyze the analytic properties of the integrand after every iteration step, as this would not be feasible. In these proceedings, we suggest a machine learning pipeline based on deep learning (DL) approaches to computer vision (CV), as well as deep reinforcement learning (DRL), that could solve this problem autonomously by detecting poles and branch cuts in the numerical integrand after every iteration step and by suggesting suitable integration contour deformations that avoid these obstructions. We sketch out a proof of principle for both of these tasks, that is, the pole and branch cut detection, as well as the contour deformation.

hep-ph↗

arXiv2021

Machine learning for predicting the Bz magnetic field component from upstream in situ observations of solar coronal mass ejections

Predicting the Bz magnetic field embedded within ICMEs, also known as the Bz problem, is a key challenge in space weather forecasting. We study the hypothesis that upstream in situ measurements of the sheath region and the first few hours of the magnetic obstacle provide sufficient information for predicting the downstream Bz component. To do so, we develop a predictive tool based on machine learning that is trained and tested on 348 ICMEs from Wind, STEREO-A, and STEREO-B measurements. We train the machine learning models to predict the minimum value of the Bz component and the maximum value of the total magnetic field Bz in the magnetic obstacle. To validate the tool, we let the ICMEs sweep over the spacecraft and assess how continually feeding in situ measurements into the tool improves the Bz prediction. Because the application of the tool in operations needs an automated detection of ICMEs, we implement an existing automated ICME detection algorithm and test its robustness for the time intervals under scrutiny. We find that the predictive tool can predict the minimum value of the Bz component in the magnetic obstacle with a mean absolute error of 3.12 nT and a Pearson correlation coefficient of 0.71 when the sheath region and the first 4 hours of the magnetic obstacle are observed. While the underlying hypothesis is unlikely to solve the Bz problem, the tool shows promise for ICMEs that have a recognizable magnetic flux rope signature. Transitioning the tool to operations could lead to improved space weather forecasting.

physics.space-ph↗

arXiv2019

Deep reinforcement learning for complex evaluation of one-loop diagrams in quantum field theory

In this paper we present a novel technique based on deep reinforcement learning that allows for numerical analytic continuation of integrals that are often encountered in one-loop diagrams in quantum field theory. In order to extract certain quantities of two-point functions, such as spectral densities, mass poles or multi-particle thresholds, it is necessary to perform an analytic continuation of the correlator in question. At one-loop level in Euclidean space, this results in the necessity to deform the integration contour of the loop integral in the complex plane of the square of the loop momentum, in order to avoid non-analyticities in the integration plane. Using a toy model for which an exact solution is known, we train a reinforcement learning agent to perform the required contour deformations. Our study shows great promise for an agent to be deployed in iterative numerical approaches used to compute non-perturbative 2-point functions, such as the quark propagator Dyson-Schwinger equation, or more generally, Fredholm equations of the second kind, in the complex domain.

hep-ph↗

arXiv2019

nEoS: Neutron Star Equation of State from hadron physics alone

We contribute a publicly available set of tables and code to provide Equations of State (EoS) for matter at neutron star densities. Our EoSes are constrained only by input from hadron physics and fundamental principles, without feedback from neutron star observations, and so without relying on General Relativity. They can therefore be used to test General Relativity itself, as well as modified gravity theories, with neutron star observables, without logical circularity. We have adapted state of the art results from NN chiral potentials for the low--density limit, pQCD results for the asymptotically high-density EoS, and use monotonicity and causality as the only restrictions for intermediate densities, for the EoS sets to remain as model-independent as is feasible today.

gr-qc↗

arXiv2017

Color superconductivity and charge neutrality in Yukawa theory

It is generally believed that when Cooper pairing occurs between two different species of fermions, their Fermi surfaces become locked together so that the resultant state remains 'neutral', with equal number densities of the two species, even when subjected to a chemical potential that couples to the difference in number densities. This belief is based on mean-field calculations in models with a zero-range interaction, where the anomalous self-energy is independent of energy and momentum. Following up on an early report of a deviation from neutrality in a Dyson-Schwinger calculation of color-flavor-locked quark matter, we investigate the neutrality of a two-species condensate using a Yukawa model which has a finite-range interaction. In a mean field calculation we obtain the full energy-momentum dependence of the self energy and find that the energy dependence leads to a population imbalance in the Cooper-paired phase when it is stressed by a species-dependent chemical potential. This gives some support to the suggestion that the color-flavor-locked phase of quark matter might not be an insulator.

nucl-th↗

arXiv2017

The quark propagator in the complex domain: A status report

In these proceedings I review the status of an ongoing project that aims at solving the quark propagator Dyson-Schwinger equation in the complex domain. The novel aspect of the approach is that the non-analyticities arising throughout the iteration of the equation are to be taken into account in an appropriate way through dynamic contour deformation. Because of the complexity of the approach, these studies are undertaken in a heavily truncated scenario that serves as a toy model for the development of the numerical techniques that are required to treat the system in a mathematically sound way.

hep-ph↗

arXiv2016

The analytic properties of the quark propagator from an effective infrared interaction model

In this paper I investigate the analytic properties of the quark propagator Dyson-Schwinger equation (DSE) in the Landau gauge. In the quark self-energy, the combined gluon propagator and quark-gluon vertex is modeled by an effective interaction (the so-called Maris-Tandy interaction), where the ultraviolet term is neglected. This renders the loop integrand of the quark self-energy analytic on the cut-plane -Pi < arg(x) < Pi of the square of the external momentum. Exploiting the simplicity of the truncation, I study solutions of the quark propagator in the domain x in [-5.1,0] GeV^2 X i[0,10.2] GeV^2. Because of a complex conjugation symmetry, this region fully covers the parabolic integration domain for Bethe-Salpeter equations (BSEs) for bound state masses of up to 4.5 GeV. Employing a novel numerical technique that is based on highly parallel computation on graphics processing units (GPUs), I extract more than 6500 poles in this region, which arise as the bare quark mass is varied over a wide range of closely spaced values. The poles are grouped in 23 individual 'trajectories', that capture the movement of the poles in the complex region as the bare mass is varied. The raw data of the pole locations and -residues is provided as supplemental material, which can be used to parametrize solutions of the complex quark propagator for a wide range of bare mass values and for large bound state masses. This study is a first step towards an extension of previous work on the analytic continuation of perturbative one-loop integrals, with the long-term goal of establishing a framework that allows for the numerical extraction of the analytic properties of the quark propagator with a truncation that extends beyond the rainbow by making adequate adjustments in the contour of the radial integration of the quark self-energy.

hep-ph↗

arXiv2016

Three-point vertex functions in Yang-Mills Theory and QCD in Landau gauge

Solutions for the three-gluon and quark-gluon vertices from Dyson-Schwinger equations and the three-particle irreducible formalism are discussed. Dynamical quarks (``unquenching'') change the three-gluon vertex via the quark-triangle diagrams which themselves include fully dressed quark-gluon vertex functions. On the other hand, the quark-swordfish diagram is, at least with the model used for the two-quark-two-gluon vertex employed here, of minor importance. For the leading tensor structure of the three-gluon vertex the "unquenching" effect can be summarized for the nonperturbative part as a shift of the related dressing function towards the infrared.

hep-ph↗

arXiv2016

Dynamics of semi-superfluid fluxtubes in color-flavor locked quark matter

At very high densities, as for example in the core of a neutron star, matter may appear in the color-flavor locked (CFL) phase, which is a superfluid. This phase features topologically stable vortex solutions, which arise in a spinning superfluid as localized configurations carrying quanta of angular momentum. Despite the topological stability of these vortices they are not the lowest energy state of the system at neutron star densities and decay into triplets of semi-superfluid fluxtubes. In these proceedings we report on the progress of our numerical study in the Ginzburg-Landau approach, where we investigate lattices of semi-superfluid fluxtubes. The fluxtubes are obtained through controlled decay of global vortex configurations in the presence of a gauge field. Understanding the dynamics of semi-superfluid string configurations is important in the context of angular momentum transfer from a quark matter core of a neutron star beyond the core boundary, since not vortex-, but fluxtube pinning seems to be the relevant mechanism in this scenario.

nucl-th↗

arXiv2016

Vortex structure in superfluid color-flavor locked quark matter

The core region of a neutron star may feature quark matter in the color-flavor- locked (CFL) phase. The CFL condensate breaks the baryon number symmetry, such that the phenomenon of superfluidity arises. If the core of the star is rotating, vortices will form in the superfluid, carrying the quanta of angular momentum. In a previous study we have solved the question of stability of these vortices, where we found numerical proof of a conjectured instability, according to which superfluid vortices will decay into an arrangement of so-called semi-superfluid fluxtubes. Here we report first results of an extension of our framework that allows us to study multi-vortex dynamics. This will in turn enable us to investigate the structure of semi-superfluid string lattices, which could be relevant to study pinning phenomena at the boundary of the core.

nucl-th↗

arXiv2016

Stability of superfluid vortices in dense quark matter

Superfluid vortices in the color-flavor-locked (CFL) phase of dense quark matter are known to be energetically disfavored relative to well-separated triplets of "semi-superfluid" color flux tubes. However, the short-range interaction (metastable versus unstable) has not been established. In this paper we perform numerical calculations using the effective theory of the condensate field, mapping the regions in the parameter space of coupling constants where the vortices are metastable versus unstable. For the case of zero gauge coupling we analytically identify a candidate for the unstable mode, and show that it agrees well with the results of the numerical calculations. We find that in the region of the parameter space that seems likely to correspond to real-world CFL quark matter the vortices are unstable, indicating that if such matter exists in neutron star cores it is very likely to contain semi-superfluid color flux tubes rather than superfluid vortices.

nucl-th↗

arXiv2015

Unquenching the three-gluon vertex: A status report

We discuss unquenching of the three-gluon vertex via its Dyson-Schwinger equation. We review the role of Furry's theorem and present first results for the quark triangle diagrams using non-perturbatively calculated dressing functions for the quark propagator and the quark-gluon vertex.

hep-ph↗

arXiv2014

On propagators and three-point functions in Landau gauge QCD and QCD-like theories

Recent progress in our studies of propagators and three-point functions in Landau gauge for QCD and QCD-like theories is presented. Special emphasis is put on the properties of the three-gluon vertex and the quark-gluon vertex. The effect of unquenching is investigated. Furthermore, an exploratory study for a large number of light flavours is described, from where clear evidence for the qualitative behaviour of propagators in the so-called conformal window can be extracted.

hep-ph↗

arXiv2013

How to determine the branch points of correlation functions in Euclidean space

Two-point correlators represented by either perturbative or non-perturbative integral equations in Euclidean space are considered. In general, it is difficult to determine the analytic structure of arbitrary correlators analytically. When relying on numerical methods to evaluate the analytic structure, exact predictions of, e.g., branch point locations (i.e., the multi-particle threshold) provide a useful check. These branch point locations can be derived by Cutkosky's cut rules. However, originally they were derived in Minkowski space for propagators with real masses and are thus not directly applicable in Euclidean space and for propagators of a more general form. Following similar considerations that led Karplus et al., Landau and Cutkosky more than 50 years ago to the mass summation formula that became known as Cutkosky's cut rules, we show how the position of branch points can be derived analytically in Euclidean space from propagators of very general form.

hep-ph↗