Date of Completion

12-24-2019

Embargo Period

12-24-2019

Keywords

Composite Processing

Major Advisor

Dianyun Zhang

Associate Advisor

Tai-Hsi Fan

Associate Advisor

Julian Norato

Associate Advisor

Leslie Shor

Associate Advisor

Jiong Tang

Field of Study

Mechanical Engineering

Degree

Doctor of Philosophy

Open Access

Open Access

Abstract

Fiber-reinforced composites have been widely employed in structural applications due to their high stiffness-to-weight ratio and easily-tailored mechanical properties. However, manufacturing of these lightweight materials involves an infusion and curing process, and flow-induced defects and residual stresses can easily occur, which can compromise the mechanical performance of the composite structures. The purpose of this work is to develop an integrated flow-curing processing model to accurately predict the resin flow front and the residual stresses in the Liquid Composite Molding (LCM) process. In the infusion model, the fiber fabrics are treated as a homogeneous and porous material, and the resin flow movement is governed by the Navier-Stokes equations. The resin flow can be solved using the Volume of Fluid (VOF) method. Due to the pressure equilibrium, the resin flow movement and the compaction of the fiber preform are two-way coupled, and the coupling between the flow and compaction models can be captured in the proposed model. After the infusion simulation, the residual stresses are predicted through a curing model. Since the thermal and chemical strains are major contributing factors to the residual stresses, the temperature and cure progression inside the composite are first predicted through a thermal-chemical analysis. Based on the predicted temperature and cure progression, the residual stress development can be captured through a thermo-viscoelastic model. Based on the elastic moduli of the fiber and viscoelastic properties of the resin, the effective relaxation moduli of the composite can be predicted through the correspondence principle and micromechanics models. The effective composite properties are then included in a 3D anisotropic viscoelastic constitutive law, and the differential form of the constitutive law is developed for numerical implementation to improve the computational efficiency. The accuracy of the processing model is assessed by comparing the simulation results against experiments through a set of benchmark examples. The proposed coupled flow-curing processing model is physics-based and experimentally-validated, which can be employed to understand the variability in composite manufacturing and identify the root causes of processing-induced defects. The integrated model shows great promise as a modeling toolkit to guide the design of optimal manufacturing procedures with minimized defects.

Download

Share

COinS