-
Numerical Investigation of the Local Thermo-Chemical State in a Thermo-Acoustically Unstable Dual Swirl Gas Turbine Model Combustor
Authors:
T. Jeremy P. Karpowski,
Federica Ferraro,
Matthias Steinhausen,
Sebastian Popp,
Christoph M. Arndt,
Christian Kraus,
Henning Bockhorn,
Wolfgang Meier,
Christian Hasse
Abstract:
In this work, the thermo-acoustic instabilities of a gas turbine model combustor, the so-called SFB606 combustor, are numerically investigated using Large Eddy Simulation (LES) combined with tabulated chemistry and Artificial Thickened Flame (ATF) approach. The main focus is a detailed analysis of the thermo-acoustic cycle and the accompanied equivalence ratio oscillations and their associated con…
▽ More
In this work, the thermo-acoustic instabilities of a gas turbine model combustor, the so-called SFB606 combustor, are numerically investigated using Large Eddy Simulation (LES) combined with tabulated chemistry and Artificial Thickened Flame (ATF) approach. The main focus is a detailed analysis of the thermo-acoustic cycle and the accompanied equivalence ratio oscillations and their associated convective time delay. In particular, the variations of the thermo-chemical state and flame characteristics over the thermo-acoustic cycle are investigated. For the operating point flame B ($P_{th}=25\,$kW), the burner exhibits thermo-acoustic instabilities with a dominant frequency of 392Hz, the acoustic eigenmode of the inner air inlet duct. These oscillations are accompanied by an equivalence ratio oscillation, which exhibits a convective time delay between the injection in the inner swirler and the flame zone. Two LES, one adiabatic and one accounting for heat losses at the walls by prescribing the wall temperatures from experimental data and Conjugated Heat Transfer (CHT) simulations, are conducted. Results with the enthalpy-dependent table are found to predict the time-averaged flow field in terms of velocity, major species, and temperature with higher accuracy than in the adiabatic case. Further, they indicate, that heat losses should be accounted for to correctly predict the flame position. Subsequently, the thermo-chemical state variations over the thermo-acoustic cycle for the enthalpy-dependant case are analyzed in detail and compared with experimental data in terms of phase-conditioned averaged profiles and conditional averages. An overall good prediction is observed. The results provide a detailed quantitative analysis of the thermo-acoustic feedback mechanism of this burner.
△ Less
Submitted 10 August, 2023;
originally announced August 2023.
-
Bayesian view on the training of invertible residual networks for solving linear inverse problems
Authors:
Clemens Arndt,
Sören Dittmer,
Nick Heilenkötter,
Meira Iske,
Tobias Kluth,
Judith Nickel
Abstract:
Learning-based methods for inverse problems, adapting to the data's inherent structure, have become ubiquitous in the last decade. Besides empirical investigations of their often remarkable performance, an increasing number of works addresses the issue of theoretical guarantees. Recently, [3] exploited invertible residual networks (iResNets) to learn provably convergent regularizations given reaso…
▽ More
Learning-based methods for inverse problems, adapting to the data's inherent structure, have become ubiquitous in the last decade. Besides empirical investigations of their often remarkable performance, an increasing number of works addresses the issue of theoretical guarantees. Recently, [3] exploited invertible residual networks (iResNets) to learn provably convergent regularizations given reasonable assumptions. They enforced these guarantees by approximating the linear forward operator with an iResNet. Supervised training on relevant samples introduces data dependency into the approach. An open question in this context is to which extent the data's inherent structure influences the training outcome, i.e., the learned reconstruction scheme. Here we address this delicate interplay of training design and data dependency from a Bayesian perspective and shed light on opportunities and limitations. We resolve these limitations by analyzing reconstruction-based training of the inverses of iResNets, where we show that this optimization strategy introduces a level of data-dependency that cannot be achieved by approximation training. We further provide and discuss a series of numerical experiments underpinning and extending the theoretical findings.
△ Less
Submitted 19 July, 2023;
originally announced July 2023.
-
Invertible residual networks in the context of regularization theory for linear inverse problems
Authors:
Clemens Arndt,
Alexander Denker,
Sören Dittmer,
Nick Heilenkötter,
Meira Iske,
Tobias Kluth,
Peter Maass,
Judith Nickel
Abstract:
Learned inverse problem solvers exhibit remarkable performance in applications like image reconstruction tasks. These data-driven reconstruction methods often follow a two-step scheme. First, one trains the often neural network-based reconstruction scheme via a dataset. Second, one applies the scheme to new measurements to obtain reconstructions. We follow these steps but parameterize the reconstr…
▽ More
Learned inverse problem solvers exhibit remarkable performance in applications like image reconstruction tasks. These data-driven reconstruction methods often follow a two-step scheme. First, one trains the often neural network-based reconstruction scheme via a dataset. Second, one applies the scheme to new measurements to obtain reconstructions. We follow these steps but parameterize the reconstruction scheme with invertible residual networks (iResNets). We demonstrate that the invertibility enables investigating the influence of the training and architecture choices on the resulting reconstruction scheme. For example, assuming local approximation properties of the network, we show that these schemes become convergent regularizations. In addition, the investigations reveal a formal link to the linear regularization theory of linear inverse problems and provide a nonlinear spectral regularization for particular architecture classes. On the numerical side, we investigate the local approximation property of selected trained architectures and present a series of experiments on the MNIST dataset that underpin and extend our theoretical findings.
△ Less
Submitted 20 December, 2023; v1 submitted 2 June, 2023;
originally announced June 2023.
-
Reduced-Dimension Surrogate Modeling to Characterize the Damage Tolerance of Composite/Metal Structures
Authors:
Corey Arndt,
Cody Crusenberry,
Bozhi Heng,
Rochelle Butler,
Stephanie TerMaath
Abstract:
Complex engineering models are typically computationally demanding and defined by a high-dimensional parameter space challenging the comprehensive exploration of parameter effects and design optimization. To overcome this curse of dimensionality and to minimize computational resource requirements, this research demonstrates a user-friendly approach to formulating a reduced-dimension surrogate mode…
▽ More
Complex engineering models are typically computationally demanding and defined by a high-dimensional parameter space challenging the comprehensive exploration of parameter effects and design optimization. To overcome this curse of dimensionality and to minimize computational resource requirements, this research demonstrates a user-friendly approach to formulating a reduced-dimension surrogate model that represents a high-dimensional, high-fidelity source model. This approach was developed specifically for a non-expert using commercially available tools. In this approach, the complex physical behavior of the high-fidelity source model is separated into individual, interacting physical behaviors. A separate reduced-dimension surrogate model is created for each behavior and then all are summed to formulate the reduced-dimension surrogate model representing the source model. In addition to a substantial reduction in computational resources and comparable accuracy, this method also provides a characterization of each individual behavior providing additional insight into the source model behavior. The approach encompasses experimental testing, finite element analysis, surrogate modeling, and sensitivity analysis and is demonstrated by formulating a reduced-dimension surrogate model for the damage tolerance of an aluminum plate reinforced with a co-cured bonded E-glass/epoxy composite laminate under four-point bending.
△ Less
Submitted 29 February, 2024; v1 submitted 25 October, 2022;
originally announced October 2022.
-
Regularization Theory of the Analytic Deep Prior Approach
Authors:
Clemens Arndt
Abstract:
The analytic deep prior (ADP) approach was recently introduced for the theoretical analysis of deep image prior (DIP) methods with special network architectures. In this paper, we prove that ADP is in fact equivalent to classical variational Ivanov methods for solving ill-posed inverse problems. Besides, we propose a new variant which incorporates the strategy of early stop** into the ADP model.…
▽ More
The analytic deep prior (ADP) approach was recently introduced for the theoretical analysis of deep image prior (DIP) methods with special network architectures. In this paper, we prove that ADP is in fact equivalent to classical variational Ivanov methods for solving ill-posed inverse problems. Besides, we propose a new variant which incorporates the strategy of early stop** into the ADP model. For both variants, we show how classical regularization properties (existence, stability, convergence) can be obtained under common assumptions.
△ Less
Submitted 23 September, 2022; v1 submitted 13 May, 2022;
originally announced May 2022.
-
A Subspace-based Approach for Dimensionality Reduction and Important Variable Selection
Authors:
Di Bo,
Hoon Hwangbo,
Vinit Sharma,
Corey Arndt,
Stephanie C. TerMaath
Abstract:
An analysis of high-dimensional data can offer a detailed description of a system but is often challenged by the curse of dimensionality. General dimensionality reduction techniques can alleviate such difficulty by extracting a few important features, but they are limited due to the lack of interpretability and connectivity to actual decision making associated with each physical variable. Variable…
▽ More
An analysis of high-dimensional data can offer a detailed description of a system but is often challenged by the curse of dimensionality. General dimensionality reduction techniques can alleviate such difficulty by extracting a few important features, but they are limited due to the lack of interpretability and connectivity to actual decision making associated with each physical variable. Variable selection techniques, as an alternative, can maintain the interpretability, but they often involve a greedy search that is susceptible to failure in capturing important interactions or a metaheuristic search that requires extensive computations. This research proposes a new method that produces subspaces, reduced-dimensional physical spaces, based on a randomized search and leverages an ensemble of critical subspace-based models, achieving dimensionality reduction and variable selection. When applied to high-dimensional data collected from the failure prediction of a composite/metal hybrid structure exhibiting complex progressive damage failure under loading, the proposed method outperforms the existing and potential alternatives in prediction and important variable selection.
△ Less
Submitted 24 September, 2021; v1 submitted 3 June, 2021;
originally announced June 2021.
-
Photonic Crystal Fiber Metalens
Authors:
**gyi Yang,
Indra Ghimire,
Pin Chieh Wu,
Sudip Gurung,
Catherine Arndt,
Din ** Tsai,
Ho Wai Howard Lee
Abstract:
Optical fiber is a well-established efficient way to guide and manipulate light and allowing high bandwidth optical transmission for long distance communication with low attenuation. Although a dielectric optical waveguide is efficient for transmitting light, its functionality is limited by the dielectric properties of the materials of core and cladding.The light coming out from optical fiber is t…
▽ More
Optical fiber is a well-established efficient way to guide and manipulate light and allowing high bandwidth optical transmission for long distance communication with low attenuation. Although a dielectric optical waveguide is efficient for transmitting light, its functionality is limited by the dielectric properties of the materials of core and cladding.The light coming out from optical fiber is typically diverging, and the numerical aperture is determined by the refractive index of the fiber materials. Thus, the light intensity decreases significantly upon exiting the fiber. The use of metasurfaces provides the opportunity to tailor light properties for advanced light manipulation and to develop novel optical applications that are flat. By producing a specific phase profile using spatially-varied nano-antenna elements, metasurfaces can control the wavefront of the transmitted, reflected, and scattered light, and enable novel ultrathin optical components such as flat lenses. Here, we report an ultrathin optical metalens cascaded on the facet of optical fiber that enables light focusing in the telecommunication regime. In-fiber metalenses with focal lengths of 28 and 40 um and maximum efficiency of ~16.4% at a wavelength of 1550 nm are demonstrated. The integration of an ultrathin metalens and optical fiber will open the path to revolutionary in-fiber optical devices for practical applications in optical imaging, sensing, and fiber laser.
△ Less
Submitted 27 March, 2019; v1 submitted 12 June, 2018;
originally announced June 2018.
-
IPv6 Prefix Alteration: An Opportunity to Improve Online Privacy
Authors:
Dominik Herrmann,
Christine Arndt,
Hannes Federrath
Abstract:
This paper is focused on privacy issues related to the prefix part of IPv6 addresses. Long-lived prefixes may introduce additional tracking opportunities for communication partners and third parties. We outline a number of prefix alteration schemes that may be deployed to maintain the unlinkability of users' activities. While none of the schemes will solve all privacy problems on the Internet on t…
▽ More
This paper is focused on privacy issues related to the prefix part of IPv6 addresses. Long-lived prefixes may introduce additional tracking opportunities for communication partners and third parties. We outline a number of prefix alteration schemes that may be deployed to maintain the unlinkability of users' activities. While none of the schemes will solve all privacy problems on the Internet on their own, we argue that the development of practical prefix alteration techniques constitutes a worthwile avenue to pursue: They would allow Internet Service Providers to increase the attainable privacy level well above the status quo in today's IPv4 networks.
△ Less
Submitted 20 November, 2012;
originally announced November 2012.
-
Carbon Based Resistive Memory
Authors:
Franz Kreupl,
Rainer Bruchhaus,
Petra Majewski,
Jan B. Philipp,
Ralf Symanczyk,
Thomas Happ,
Christian Arndt,
Mirko Vogt,
Roy Zimmermann,
Axel Buerke,
Andrew P. Graham,
Michael Kund
Abstract:
We propose carbon as new resistive memory material for non-volatile memories and compare three allotropes of carbon, namely carbon nanotubes, graphene-like conductive carbon and insulating carbon for their possible application as resistance-change material in high density non-volatile memories. Repetitive high-speed switching and the potential for multi-level programming have been successfully d…
▽ More
We propose carbon as new resistive memory material for non-volatile memories and compare three allotropes of carbon, namely carbon nanotubes, graphene-like conductive carbon and insulating carbon for their possible application as resistance-change material in high density non-volatile memories. Repetitive high-speed switching and the potential for multi-level programming have been successfully demonstrated.
△ Less
Submitted 28 January, 2009;
originally announced January 2009.
