Loading...

Cyclic Behavior Modeling of Magnesium Alloy (AZ91) Under Thermo-Mechanical and Low Cycle Fatigue Loadings

Shamloo, Ali | 2013

797 Viewed
  1. Type of Document: M.Sc. Thesis
  2. Language: Farsi
  3. Document No: 44917 (08)
  4. University: Sharif University of Technology
  5. Department: Mechanical Engineering
  6. Advisor(s): Farrahi, Gholamhossein
  7. Abstract:
  8. Magnesium alloys have been used in the production of light weight components in automobile industries (for example: the ladder frame and the valve cover) due to a good combination of low density (1740 kg/m3), the tensile strength (160-365 MPa) and the elastic modulus (45 GPa). For the design and the optimization of these components under various loadings and different conditions, it is necessary to determine fatigue properties of the material by using thermo-mechanical and low cycle fatigue tests at different temperatures. In thermo-mechanical fatigue tests, the temperature varies in a specific range with the mechanical strain. The goal of this research is to investigate low cycle fatigue and thermo-mechanical fatigue behaviors of the AZ91 alloy. For this purpose, experimental results from fatigue tests were used to calibrate material parameters of constitutive models for the magnesium alloy. Four constitutive models, including the Chaboche’s model, the Chaboche’s model with a modified kinematic hardening rule, the Prandtl operator and the Skelton’s model, were used and their parameters were optimised. By considering some modifications to the Prandtl operator constitutive model and by using a more complete relation for the visco-plastic strain, the Prandtl operator constitutive model with a more complex visco-plasticity law, including isotropic and kinematic hardenings is also investigated. These models are used for the numerical analysis of the elastic-plastic behavior and stress-strain hysteresis curves of the material. Then, parameters were optimised by using the genetic algorithm and best fitted curves to experimental results were obtained. According to chosen error function, which is the average of the magnitude of the stress deviation from experimental data divided by the stress range, the modified Chaboche’s model with the error of 2.14% for the thermo-mechanical loading was selected as the most accurate model. The classic Prandtl operator model with the error of 2.61% for the thermo-mechanical loading, but with higher speed of computations (about 6 times faster) was also selected as another appropriate model
  9. Keywords:
  10. Thermomechanical Fatigue ; Low Cycle Fatigue ; Magnesium Alloy ; Chaboche's Constitutive Model ; Pranttle Number ; AZ91 Magnesium Alloy ; Hysteresis Curve

 Digital Object List

 Bookmark

No TOC