Thermo-Mechanical Fatigue Using the Extended Space-Time Finite Element Method
Date
Authors
ORCID
Journal Title
Journal ISSN
Volume Title
Publisher
item.page.doi
Abstract
Thermomechanical high-cycle fatigue is a major failure mechanism for many engineering components in a diverse range of industries such as aerospace, automotive, and nuclear among others. Engineers trying to determine the fatigue life of a component typically rely on commercial fatigue analysis software which uses traditional fatigue criteria that are limited in their applicability. For instance, they are poor at handling multiaxial and variable amplitude loading. Furthermore, adding variable amplitude thermal loading into the mix makes using these traditional fatigue criteria even less appealing. In recent years, there have been attempts to establish methods for simulating high cycle fatigue based on finite element calculations rather than using it as a post-processing step. These include cohesive zone and continuum damage mechanics models. However, all of these methods employ cycle jumping strategies to cut down on the enormous computational time required. However, cycle jumping is not applicable for a random loading history or with random or out-of-phase temperature variation. Motivated by these current developments, this thesis proposes the use of the extended space-time finite element method (XTFEM) in combination with a two scale progressive fatigue damage model for the direct numerical simulation of thermomechanical high cycle fatigue. Instead of using the conventional explicit or implicit finite difference time integration methods, temporal approximations are introduced with FEM mesh and enriched based on the extended finite element method. After outlining the basic theory for XTFEM with thermomechanical coupling, the effectiveness of the computational framework is demonstrated in numerical examples including a coupled, thermomechanical fatigue simulation of a plate and hat stiffener model representative of a hypersonic aircraft’s structure.