Refine
Document Type
- Conference Proceeding (28)
- Article (11)
- Doctoral Thesis (1)
Institute
Keywords
- Angewandte Mathematik (33)
- Strukturmechanik (32)
- Architektur <Informatik> (4)
- Computerunterstütztes Verfahren (4)
- CAD (3)
- Finite-Elemente-Methode (2)
- API (1)
- Angewandte Informatik (1)
- Bruchmechanik (1)
- Computer Science Models in Engineering; Multiscale and Multiphysical Models; Scientific Computing (1)
The modeling of crack propagation in plain and reinforced concrete structures is still a field for many researchers. If a macroscopic description of the cohesive cracking process of concrete is applied, generally the Fictitious Crack Model is utilized, where a force transmission over micro cracks is assumed. In the most applications of this concept the cohesive model represents the relation between the normal crack opening and the normal stress, which is mostly defined as an exponential softening function, independently from the shear stresses in tangential direction. The cohesive forces are then calculated only from the normal stresses. By Carol et al. 1997 an improved model was developed using a coupled relation between the normal and shear damage based on an elasto-plastic constitutive formulation. This model is based on a hyperbolic yield surface depending on the normal and the shear stresses and on the tensile and shear strength. This model also represents the effect of shear traction induced crack opening. Due to the elasto-plastic formulation, where the inelastic crack opening is represented by plastic strains, this model is limited for applications with monotonic loading. In order to enable the application for cases with un- and reloading the existing model is extended in this study using a combined plastic-damage formulation, which enables the modeling of crack opening and crack closure. Furthermore the corresponding algorithmic implementation using a return mapping approach is presented and the model is verified by means of several numerical examples. Finally an investigation concerning the identification of the model parameters by means of neural networks is presented. In this analysis an inverse approximation of the model parameters is performed by using a given set of points of the load displacement curves as input values and the model parameters as output terms. It will be shown, that the elasto-plastic model parameters could be identified well with this approach, but require a huge number of simulations.
ESTIMATING UNCERTAINTIES FROM INACCURATE MEASUREMENT DATA USING MAXIMUM ENTROPY DISTRIBUTIONS
(2010)
Modern engineering design often considers uncertainties in geometrical and material parameters and in the loading conditions. Based on initial assumptions on the stochastic properties as mean values, standard deviations and the distribution functions of these uncertain parameters a probabilistic analysis is carried out. In many application fields probabilities of the exceedance of failure criteria are computed. The out-coming failure probability is strongly dependent on the initial assumptions on the random variable properties. Measurements are always more or less inaccurate data due to varying environmental conditions during the measurement procedure. Furthermore the estimation of stochastic properties from a limited number of realisation also causes uncertainties in these quantities. Thus the assumption of exactly known stochastic properties by neglecting these uncertainties may not lead to very useful probabilistic measures in a design process. In this paper we assume the stochastic properties of a random variable as uncertain quantities caused by so-called epistemic uncertainties. Instead of predefined distribution types we use the maximum entropy distribution which enables the description of a wide range of distribution functions based on the first four stochastic moments. These moments are taken again as random variables to model the epistemic scatter in the stochastic assumptions. The main point of this paper is the discussion on the estimation of these uncertain stochastic properties based on inaccurate measurements. We investigate the bootstrap algorithm for its applicability to quantify the uncertainties in the stochastic properties considering imprecise measurement data. Based on the obtained estimates we apply standard stochastic analysis on a simple example to demonstrate the difference and the necessity of the proposed approach.
Iso-parametric finite elements with linear shape functions show in general a too stiff element behavior, called locking. By the investigation of structural parts under bending loading the so-called shear locking appears, because these elements can not reproduce pure bending modes. Many studies dealt with the locking problem and a number of methods to avoid the undesirable effects have been developed. Two well known methods are the >Assumed Natural Strain< (ANS) method and the >Enhanced Assumed Strain< (EAS) method. In this study the EAS method is applied to a four-node plane element with four EAS-parameters. The paper will describe the well-known linear formulation, its extension to nonlinear materials and the modeling of material uncertainties with random fields. For nonlinear material behavior the EAS parameters can not be determined directly. Here the problem is solved by using an internal iteration at the element level, which is much more efficient and stable than the determination via a global iteration. To verify the deterministic element behavior the results of common test examples are presented for linear and nonlinear materials. The modeling of material uncertainties is done by point-discretized random fields. To show the applicability of the element for stochastic finite element calculations Latin Hypercube Sampling was applied to investigate the stochastic hardening behavior of a cantilever beam with nonlinear material. The enhanced linear element can be applied as an alternative to higher-order finite elements where more nodes are necessary. The presented element formulation can be used in a similar manner to improve stochastic linear solid elements.
This paper presents the combination of two different parallelization environments, OpenMP and MPI, in one numerical simulation tool. The computation of the system matrices and vectors is parallelized with OpenMP and the solution of the system of equations is done with the MPIbased solver MUMPS. The efficiency of both algorithms is shown on several linear and nonlinear examples using the Finite Element Method and a meshless discretization technique.