Date

5-1-2018

Document Type

Thesis

Degree

Master of Science

Department

Mechanical Engineering

First Adviser

Vermaak, Natasha

Abstract

This work investigates, both experimentally and computationally the cyclic behavior of stainless steel 316 under uniaxial loading at room temperature. Elastoplastic investigations of SS 316 are important in the development of an understanding of the possible behavior and their contribution to the material performance under cyclic loading. Cyclic plasticity can occur in an SS 316 component or structure depending on the loading conditions. Therefore, it is vital that the cyclic behavior of SS 316 is recognized and understood. In particular SS316 cylindrical rods specimens were tested under uniaxial cyclic loading. The experimental results show that ratchetting behavior regimes exist under the conditions presented. In order to simulate the experiments, an elastoplastic material model based on the Chaboche model is utilized in the commercial finite element (FE)-software ABAQUS. The Chaboche constitutive model utilized for cyclic loading, which includes nonlinear kinematic and isotropic hardening is discussed in detail. The kinematic and isotropic hardening parameters for the Chaboche model are also identified. The kinematic hardening parameters are calibrated using experimental data from the first half-cycle of loading, and the isotropic hardening parameters are defined by using cyclic experimental data from a test with symmetric strain (up to 1%). A 2D axisymmetric model is created in ABAQUS, where the same geometry, boundary conditions and loading cases are applied as those recorded experimentally. A mesh sensitivity study is also carried out. The error between simulation and experimental results is calculated and sources for the error are discussed. One strategy to decrease the numerical error is applied and evaluated. This work provides evidence that the Chaboche model can predict the cyclic behavior of stainless steel 316. However, there remain significant questions about the accuracy of the model parameters identified as they lead to errors in the predicted plastic strain at large cycle numbers. It is concluded that an improved method for calibrating the parameters or a more complex constitutive model is needed to better predict the cyclic behavior of stainless steel 316.

Share

COinS