Experimental investigation and FE-based modeling of temperature dependent nonlinear deformation and damage of short fiber reinforced thermoplastics for automotive applications

Efficient and reliable numerical simulation methods play an important role in the design of industrial components using short fiber reinforced thermoplastics (SFRT). The main goal of this dissertation is to develop a Finite Element (FE) based simulation method capable of capturing the complex mechanical behaviors and predicting the crack initiation of SFRT under a variety of loading conditions, while ensuring that the required computation and parameter calibration efforts are acceptable for industrial applications.
To establish a comprehensive material database for modeling purposes, this research extensively characterizes PBT-GF30, a commonly used SFRT material in the automotive industry. The experimental investigation considers a range of loading conditions, including short-term and long-term quasi-static loading, as well as cyclic loading. Furthermore, influential factors such as temperature, strain rate, injection flow direction, thermal aging, and micro-damage are comprehensively examined, providing essential insights into the mechanical properties of SFRT.
Mechanical responses of the investigated SFRT are modeled utilizing a hybrid approach combining micro-mechanical and macro-mechanical modeling strategies. This approach employs the Mori-Tanaka mean-field homogenization method to determine the effective linear elastic properties of SFRT, while the macroscopic plastic deformation is described using a macro-mechanical anisotropic viscoplastic model. Within the Continuum Damage Mechanics (CDM) framework, the effects of the matrix micro-damage on the macroscopic behaviors of SFRT are taken into account. A nonlinear damage evolution law, correlated with the material deformation history, accurately captures the different processes of damage accumulation observed in quasi-static and low cycle fatigue (LCF) tests.
The proposed material model is implemented in the ANSYS software, employing a hybrid time scheme to ensure compatibility and efficiency for industrial applications. In this approach, the standard implicit time scheme is employed for iterative equilibrium solving at each time step, while an explicit scheme is utilized to evaluate and update damage variables. To enhance software compatibility, standard material models available in the commercial FE software are utilized instead of user-defined subroutines. Moreover, to improve computational efficiency, a clustering technique is employed to group similar fiber orientations and damage variables.
Finally, the validity of the simulation method, incorporating calibrated material parameters, is demonstrated through accurate predictions of the mechanical responses observed in both testing specimens and a real industrial component subjected to various loading conditions.