Multiscale modeling of viscoelastic behavior of unidirectional composite laminates and deployable structures

Research output: Contribution to journalArticlepeer-review

54 Scopus citations

Abstract

Due to the inherent viscoelasticity of constituent matrix and the possibility of long-term storage, space deployable structures made of composites are likely to exhibit relaxation in the stored strain energy, which may degrade their deployment performance. This paper presents a bottom-up finite element based multiscale computational strategy that bridges the experimentally measurable properties of constituent fibers and matrix to numerical predictions of viscoelastic behavior of composite laminates and general shell structures. A user-friendly RVE analysis plug-in tool is developed in Abaqus/CAE to rapidly estimate the effective orthotropic viscoelastic properties of unidirectional composites by taking as input the microstructure geometry as well as the known properties of fibers and matrix. Some benchmark problems were solved, and the accuracy and efficiency of the proposed plug-in tool were verified. Next, the strategy is shown to be applicable to model the viscoelastic behavior of macroscale composite laminates and deployable shell structures, by utilizing built-in functions in Abaqus to define the stacking sequence and accordingly update the material properties. In particular, the proposed multiscale strategy was employed to simulate the influence of modulus relaxation on the deployment dynamics of a composite tape-spring hinge, and good agreement was achieved as compared to reported experimental results.

Original languageEnglish
Article number110754
JournalMaterials and Design
Volume219
DOIs
StatePublished - Jul 2022

Keywords

  • Composite laminates
  • Deployable structures
  • Multiscale modeling
  • Viscoelastic relaxation

Fingerprint

Dive into the research topics of 'Multiscale modeling of viscoelastic behavior of unidirectional composite laminates and deployable structures'. Together they form a unique fingerprint.

Cite this