WCFA'13 & PUM05
Programme of the meeting
The program can be downloaded here. It consists of these lectures:
Lecturer: Robert Basan
Timing: October 23, 10:10-12:20
Timing: October 23, 8:00-9:40
The main aim of the lecture is to present and discuss the recent developments of the Fictitious Notch Rounding (FNR) approach as originary proposed by Neuber. The theoretical basis of the method are summarised, describing the fundamental steps for its application. Taking advantage of some recent analytical solutions valid for blunt V-notches and V-notches with end holes under Mode 1, 2 and 3 loading, the FNR approach is applied providing the multiaxiality factor, s, as a function of the notch opening angle and of the actual notch radius. Some open problems regarding the application of the FNR approch to mixed mode loading conditions are also discussed.
In the second part of the presentation a criterion based on the averaged strain energy density over a control volume is presented. The approach combines together the idea of the Neuber's material microstructural support length, applied here as a real volume, having a material dependent critical radius surrounding the notch tip, and the Beltrami's total strain energy density, thought as the critical parameter when averaged over the control volume. Some recent applications of the approach to uniaxial and multiaxial fatigue data from welded and notched components are summarised showing some advantages of the approach with respect to the stress based criteria.
Lecturer: Marina Franulović
Timing: October 22, 15:00-17:20
Genetic algorithm (GA), as a very powerful evolutionary method, is recommended to be used for the determination of material parameters in complex material models because of its capability to provide very good approximation of the solution in systems with large number of unknown variables. It is also recommended to be used in problem solutions that require manipulation of very large number of data, such as the case of material fatigue investigations that take into account large number of phenomena, which occur within the structure. Practical use of GA for parameter identification is presented for Chaboche’s material model that takes into account isotropic and kinematic hardening/softening of the material in low-cycle fatigue regime. The procedure of GA that consists of definition of the objective function, definition and implementation of the genetic representation and definition and implementation of the genetic operators is presented for a given problem, as well as a possibility to automate parameter identification procedure and simulation of material behavior.
Lecturer: Patrik Huter
Timing: October 21, 12:40-14:50 - Session B
Lecturer: Hynek Lauschmann
Timing: October 21, 18:10-19:00 - Session A
The conventional quantitative fractography of fatigue fractures is based on striations and/or beach lines. As an alternative, applications of image textural analysis were developed, called Textural fractography. SEM images are characterized by a set of numerical characteristics - image features, estimated by decomposition methods, structural analysis or random field models. The relation between image features and crack growth rate (CGR) is expressed by means of a multivariate statistical model or a neural network. For cases of different variable cycle loadings, reference crack growth rate (RCGR) was proposed which is related to common features of all fatigue fractures, and may be estimated directly from images of fracture surfaces. Physical interpretation was suggested on the basis of cycle-by-cycle crack growth description. Applications are presented to sets of specimens from aluminium alloys and steels loaded by constant cycle, periodic and random blocks.
Lecturer: Jan Kohout
Timing: October 22, 10:30-12:15
S-N curve is in this time of deep studies of fatigue mechanisms quite out of the interest of materials scientists, but for design engineers it is still very useful tool. Its assessment need not be trivial task if certain reliability is asked. Even the choice of dependent and independent variable in regression need not be fully self-evident for very large and along S-N curve changing dispersion and for the asymptote(s) of S-N curve perpendicular to stress axis.
The contribution tries to choose suitable regression function containing regression parameters with clear and definite meaning. It presents very rough but for basic use sufficient presentation of reliability of S-N curve assessment, which can be performed and understood also by non-specialists in mathematical statistics and without necessity of special software.
Lecturer: Marko Nagode
Timing: October 21, 16:00-18:05 - Session B
The aim of the lecture is to present the damage operator approach to thermo-mechanical fatigue modelling. Low cycle fatigue assuming arbitrary mechanical and thermal loads is considered. The approach implemented into the LMS Virtual.Lab consists of the steps to follow:
A turbine housing of a turbocharger and exhaust downpipe are considered for validation purposes.
Lecturer: Jan Papuga
Timing: October 22, 8:00-10:00
Lecturer: Jan Papuga
Timing: October 21, 16:10-18:05 - Session A
Lecturer: Milan Ruzicka
Timing: October 21, 12:40-15:30 - Session A
Timetables of past volumes
firstname.lastname@example.org, last update: September 13, 2013";