Carsten Könke - Academia.edu (original) (raw)

Papers by Carsten Könke

With the help of modern CAE-based simulation processes, it is possible to predict the dynamic beh... more With the help of modern CAE-based simulation processes, it is possible to predict the dynamic behavior of fatigue strength problems in order to improve products of many industries, e.g. the building, the machine construction or the automotive industry. Amongst others, it can be used to improve the acoustic design of automobiles in an early development stage. Nowadays, the acoustics of automobiles plays a crucial role in the process of vehicle development. Because of the advanced demand of comfort and due to statutory rules the manufacturers are faced with the challenge of optimizing their car's sound emissions. The optimization includes not only the reduction of noises. Lately with the trend to hybrid and electric cars, it has been shown that vehicles can become too quiet. Thus, the prediction of structural and acoustic properties based on FE-simulations is becoming increasingly important before any experimental prototype is examined. With the state of the art, qualitative comparisons between different implementations are possible. However, an accurate and reliable quantitative prediction is still a challenge. One aspect in the context of increasing the prediction quality of acoustic (or general oscillating) problemsespecially in power-trains of automobiles-is the more accurate implementation of damping in joint structures. While material damping occurs globally and homogenous in a structural system, the damping due to joints is a very local problem, since energy is especially dissipated in the vicinity of joints. This paper focusses on experimental and numerical studies performed on a single (extracted) screw connection. Starting with experimental studies that are used to identify the underlying physical model of the energy loss, the locally influencing parameters (e.g. the damping factor) should be identified. In contrast to similar research projects, the approach tends to a more local consideration within the joint interface. Tangential stiffness and energy loss within the interface are spatially distributed and interactions between the influencing parameters are regarded. As a result, the damping matrix is no longer proportional to mass or stiffness matrix, since it is composed of the global material damping and the local joint damping. With this new approach, the prediction quality can be increased, since the local distribution of the physical parameters within the joint interface corresponds much closer to the reality.

Computational Mechanics, Mar 1, 1995

This research presents a new simulation concept of damage evolution for metallic materials under ... more This research presents a new simulation concept of damage evolution for metallic materials under large displacements and deformations. The complete damage range is subdivided into both the micro-damage and the macro-damage range. The micro-damage phase is described by the Cocks/Ashby void-growth model for isotropic, ductile materials under isothermal conditions. After having reached a critical void-volume fraction, a macro-crack is introduced into the model. With such a concept the damage evolution from nucleation and growth of first micro-voids to initiation of macro-cracks and complete failure of the material can be simulated. Applying the Finite Element Method for the numerical formulation, at every incremental macro-crack step the Finite Element mesh is adapted such that the crack path remains independent of the initial mesh.

Computers & Structures, Nov 1, 2008

Multiscale approaches require the coupling of models on different spatial scales. In this paper, ... more Multiscale approaches require the coupling of models on different spatial scales. In this paper, a coupling using neural networks is proposed. Based on a set of mesoscale simulations of concrete a system of neural networks is trained to approximate the response. A macroscale constitutive model is obtained by homogenizing the mesoscale response. Special focus is put on the mesh sensitivity,

Engineering Fracture Mechanics, Oct 1, 2009

In polycrystal materials the intergranular decohesion is one important damage phenomena that lead... more In polycrystal materials the intergranular decohesion is one important damage phenomena that leads to microcrack initiation. The paper presents a mesoscale model, which is focused on the brittle intergranular damage process in metallic polycrystals. The model reproduces the crack initiation and propagation along cohesive grain boundaries between brittle grains. An advanced Voronoi algorithm is applied to generate polycrystal material structures based on arbitrary distribution functions of grain size. Therewith, the authors are more flexible to represent realistic grain size distributions. The polycrystal model is applied to analyze the crack initiation and propagation in statically loaded samples of aluminium on the mesoscale without the necessity of initial damage definition.

Computational science, engineering and technology series, May 22, 2009

Spatial and temporal multiscale simulations of damage processes for concret

Welding in The World, Jun 24, 2020

The possibilities and challenges of using fiber optical sensors to monitor the laser-assisted joi... more The possibilities and challenges of using fiber optical sensors to monitor the laser-assisted joining of metal-polymer joints have been described in this article. Fundamental investigation proves the basic suitability of the measuring method for this application and studies the effect of essential influencing variables of the joining process-e.g., the clamping force-on the resulting sensor signals. In addition, the strain state (because of the process temperature and shrinkage of the polymer) of the parts to be joined can be traced as a function of the joining partners, the process parameters, and the material thicknesses. It is shown that the fiber optical method is suitable for process monitoring directly in the joining zone of metal-polymer hybrids and providing a tool for detailed strain measurements in the joint zone during subsequent component testing.

The Annals of Thoracic Surgery, Aug 1, 2005

Background. The optimal closure technique of median sternotomy remains controversial. The objecti... more Background. The optimal closure technique of median sternotomy remains controversial. The objective of this study was to analyze the structural response of the separated sternum using computer-based numerical discretization techniques, such as finite element methods. Methods. Thoracic computer tomographic scans (2.5-mm slices) were segmented, analyzed by image processing techniques, and transferred into a three-dimensional finite element model. In a first approach a linear elastic material model was used; neglecting nonlinear and damage effects of the bones. The influence of muscles and tendons was disregarded. Nonlinear contact conditions were applied between the two sternal parts and between fixation wires and sternum. The structural response of this model was investigated under normal breathing and asymmetric leaning on one side of the chest. Displacement and stress response of the segmented sternum were compared regarding two different closure techniques (single loop, figure-of-eight). Results. The obtained results revealed that for the normal breathing load case the single loop technique is capable of clamping the sternum sufficiently, assuming that the wires are prestressed. For asymmetric loading conditions, such as leaning on one side of the chest, the figure-of-eight loop can substantially reduce the relative longitudinal displacement between the two parts compared with the single loop. Conclusions. The application of numerical simulation techniques using complex computer models enabled the determination of structural behavior of the chest regarding the influence of different closure techniques. They allowed easy and fast modifications and therefore, in contrast to a real physical model, in-depth parameter studies.

Computers and Concrete, Aug 25, 2011

ABSTRACT In this paper, a mesoscale model of concrete is presented, which considers particles, ma... more ABSTRACT In this paper, a mesoscale model of concrete is presented, which considers particles, matrix material and the interfacial transition zone (ITZ) as separate constituents. Particles are represented as ellipsoides, generated according to a prescribed grading curve and placed randomly into the specimen. In this context, an efficient separation procedure is used. The nonlinear behavior is simulated with a cohesive interface model for the ITZ and a combined damage/plasticity model for the matrix material. The mesoscale model is used to simulate a compression and a tensile test. Furthermore, the influence of the particle distribution on the loaddisplacement curve is investigated.

Computer Methods in Applied Mechanics and Engineering, Sep 1, 2007

... In this paper, the extended finite element method (XFEM) is used for a discrete crack simulat... more ... In this paper, the extended finite element method (XFEM) is used for a discrete crack simulation of concrete using an adaptive crack growth algorithm. Different criteria for predicting the direction of the extension of a cohesive crack are investigated in the context of the XFEM. ...

Springer eBooks, Oct 22, 2012

Today the numerical simulation of damage effects in heterogeneous materials is done by adaption o... more Today the numerical simulation of damage effects in heterogeneous materials is done by adaption of multiscale approaches. A consistent modeling in three dimensions with a high discretization resolution on each scale based on a hierarchical or concurrent multiscale model still has issues. The algorithms have to be optimized in regards to the computational efficiency and the distribution among available hardware resources often based on parallel hardware architectures. In the last 5 years high performance computing (HPC) as well as GPU computation techniques were established for investigation of scientific aims. Consequently, in this work substructuring methods for partitioning of FE meshed specimens were implemented, tested and adapted to the HPC computing framework using several hundred CPU nodes. An memory-efficient iterative and parallelized equation solver combined with a special preconditioning technique for solving the underlying equation system was modified and adapted to the consideration of combined CPU and GPU based computations.

Springer eBooks, May 23, 2008

Meso-and microscale analysis are promising disciplines to cover the crack initiation as well as t... more Meso-and microscale analysis are promising disciplines to cover the crack initiation as well as the various crack propagation phenomena in engineering structures. On mesoscale it is observed that the crack propagation in polycrystalline metal materials occurs mainly along grain boundaries. Following this observation we present a two dimensional polycrystal meso model consisting of grains with an elastic orthotropic material law and cohesive interfaces along crystal boundaries, which is able to reproduce crack initiation and propagation in metallic materials. As an extension to classical Voronoi cell diagrams we apply an advanced algorithm to generate polycrystal material structures based on arbitrary distribution functions of grain size. Therewith we are more flexible to represent realistic grain size distributions. The polycrystal model is applied to analyze the crack initiation and propagation in statically loaded representative volume elements of aluminum on the mesoscale without the necessity of initial damage definition. Future research work is focused on the determination of constitutive relations for the cohesive interface law from mixed continuum atomistic simulations performed on a representative volume element on the microscale and homogenized to the mesoscale.

Kluwer Academic Publishers eBooks, Mar 15, 2006

ABSTRACT Using continuum damage theories the effects of discrete material defects are homogenized... more ABSTRACT Using continuum damage theories the effects of discrete material defects are homogenized to the macro-level. This results in an easy way of incorporating damage into existing continuum models. Nevertheless it remains necessary to recognize that the validity range of continuum damage mechanics terminates, if neighbouring voids start to coalesce, forming the first macro-cracks. From thereon, fracture mechanic concepts should substitute the continuum damage model. The introduced concept of coupling micro-mechanic damage measures with macro-crack models will be significantly enhanced by corresponding adaptive simulation concepts, guaranteeing the quality of the calculated results.

Engineering Computations, Feb 1, 1998

This paper presents a way to simulate damage for metal materials under large displacements and la... more This paper presents a way to simulate damage for metal materials under large displacements and large deformations. It divides the complete damage range into a micro-damage and a macro- damage phase. The micro-damage phase is described by the Cocks/ Ashby void-growth model for isotropic, ductile materials under isothermal conditions. After reaching the critical void-volume fraction, a macro crack is introduced into the model. Therewith it becomes possible to simulate the damage evolution from nucleation and growth of first micro-voids to initiation of macro cracks and total failure of the material. The Finite Element Method is applied for the numerical solution process. For every incremental crack step a special mesh has to be generated. Therewith the crack path remains independent of the initial FE-mesh. A special data structure allows for a most effective data manipulation for the incremental crack propagation simulation. The concept presented in this paper allows the numerical simulation of cracking processes in metal sheets. Tuning the calculated results with observations on existing steel bridges, statements on the (residual) lifetimes can be estimated.

Additional effort in the design and construction phase of reinforced concrete infrastructure can ... more Additional effort in the design and construction phase of reinforced concrete infrastructure can save money in the following phases of the life of these structures. This requires providing the design engineer with powerful nonlinear simulation tools, based on nonlinear structural analysis and Eurocode verifications. This "design for durability" will lead to products of high quality with reduced costs. The evaluation and rating of different designs under aspects of reliability and safety calls for appropriate tools to simulate and then classify the aging and damage evolution in structures. The same aspects are applicable in other phases (maintenance, special maintenance and repairs). This is very important in connection with the rehabilitation of existing civil infrastructure in Central and Eastern Europe. All of these aspects can only be covered with the nonlinear Finite Element Method as the most effective and general tool for numerical investigations with the support of the high-performance computers available today. This paper discusses the objectives of future research and possible research topics. Some relatively new topics include simulation tools that take into account the uncertainty of the input data, time dependent structural reliability, and the Internet.

Materials, Jul 16, 2023

This article is an open access article distributed under the terms and conditions of the Creative... more This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY

In this paper, a parameter identification procedure using Bayesian neural networks is proposed. B... more In this paper, a parameter identification procedure using Bayesian neural networks is proposed. Based on a training set of numerical simulations, where the material parameters are simulated in a predefined range using Latin Hypercube sampling, a Bayesian neural network, which has been extended to describe the noise of multiple outputs using a full covariance matrix, is trained to approximate the inverse relation from the experiment (displacements, forces etc.) to the material parameters. The method offers not only the possibility to determine the parameters itself, but also the accuracy of the estimate and the correlation between these parameters. As a result, a set of experiments can be designed to calibrate a numerical model.