Change search
CiteExportLink to record
Permanent link

Direct link
Cite
Citation style
  • apa
  • ieee
  • modern-language-association-8th-edition
  • vancouver
  • Other style
More styles
Language
  • de-DE
  • en-GB
  • en-US
  • fi-FI
  • nn-NO
  • nn-NB
  • sv-SE
  • Other locale
More languages
Output format
  • html
  • text
  • asciidoc
  • rtf
Sensitivity analysis and uncertainty propagation through new lifetime prediction model of welded metal assemblies
KTH, School of Engineering Sciences (SCI), Solid Mechanics (Dept.), Solid Mechanics (Div.).
2018 (English)Independent thesis Advanced level (degree of Master (Two Years)), 20 credits / 30 HE creditsStudent thesis
Abstract [en]

Fatigue failures are critical in armored vehicle design but the lack of knowledge on fatigue phenomena forces the use of large safety factors which induces weight and costs increase. This master's thesis has been driven by the will of improving the simulation of low cycle fatigue phenomenon of metal parts and welded assemblies in order to reduce safety margins. Two distinct but complementary paths are investigated: improvement of the fatigue simulation model and awareness of fatigue phenomenon dispersion. Firstly, the implementation of a new lifetime prediction model is proposed based on the ONERA fatigue criterion that has been developed for both low cycle and high cycle fatigue phenomena. The method introduced in this report is composed of the identification of material behavior laws, the simulation of experimental tests by finite element analysis and the calibration of fatigue criterion parameters. Low cycle fatigue experimental tests (uniaxial traction-compression tests) data for base metal and welded samples are used for the identifications. The material behavior follows Chaboche's theory and integrates two nonlinear kinematic hardenings and one isotropic hardening. However, the use of isotropic hardening is nearly impossible in practice due to computation cost and it has been decided to work with stabilized cycle laws. The finite element analyses are controlled by tests stresses. The ONERA fatigue criterion is only partially identified because the full identification requires additional experimental tests that have not been performed yet. Despite those simplifications, the lifetime prediction value is accurate and better than the ones given by previously used model. The process still has to be validated on other geometries. During the implementation of the whole process, elastic limit and fatigue limit appeared to be the most influential variables and the refinement of their values will increase the model's accuracy. Secondly, it is well known that fatigue phenomenon shows high dispersion. One way to take into account this aspect is to assign a stochastic aspect to lifetime prediction model variables and propagate it through the model. However, the combination with finite element analysis is impossible because of the computation time cost of each simulation forcing the use of metamodelling methods. PhimecaSoft offers to gather in a user-friendly interface a broad range of methods to execute uncertainties propagations in structural design environment. A coupling with in-house fatigue model has been set up to study the influence of model variables on the lifetime. The definition of the probability laws of the input variables is crucial. In the framework of this thesis it has been necessary to truncate the laws to keep plausible dispersions of the input variables and realistic values of the lifetime. Different functionalities of PhimecaSoft have been investigated such as: the use of Sobol's coefficients for global sensitivity analysis or the generation of metamodels. Reliability analysis is also part of PhimecaSoft and can be done whether by simulations methods or approximation methods. The use of reliability index is encouraged in the future to move from safety factor's based design toward reliability based design optimization. This second part of the thesis confirms that the elastic limit and fatigue limit are the most influential variables of the problem.

Place, publisher, year, edition, pages
2018. , p. 39
National Category
Applied Mechanics
Identifiers
URN: urn:nbn:se:kth:diva-232536OAI: oai:DiVA.org:kth-232536DiVA, id: diva2:1237795
External cooperation
Nexter
Subject / course
Solid Mechanics
Supervisors
Examiners
Available from: 2018-08-10 Created: 2018-08-10 Last updated: 2018-08-10Bibliographically approved

Open Access in DiVA

No full text in DiVA

By organisation
Solid Mechanics (Div.)
Applied Mechanics

Search outside of DiVA

GoogleGoogle Scholar

urn-nbn

Altmetric score

urn-nbn
Total: 289 hits
CiteExportLink to record
Permanent link

Direct link
Cite
Citation style
  • apa
  • ieee
  • modern-language-association-8th-edition
  • vancouver
  • Other style
More styles
Language
  • de-DE
  • en-GB
  • en-US
  • fi-FI
  • nn-NO
  • nn-NB
  • sv-SE
  • Other locale
More languages
Output format
  • html
  • text
  • asciidoc
  • rtf