The Integrated Multiscale Micromechanics Analysis Code (ImMAC) Software Suite developed by the NASA Glenn Research Center and the Ohio Aerospace Institute (OAI) consists of three components for the design and analysis of composite structures: (1) the Micromechanics Analysis Code with Generalized Method of Cells (MAC/GMC) performs rapid, standalone analysis of composite materials and laminates based on non-finite-element analysis (non-FEA) micromechanics methods (ref. 1); (2) the Finite Element Analysis--Micromechanics Analysis Code (FEAMAC) couples the efficient micromechanics capabilities of MAC/GMC with the ABAQUS finite element code (ABAQUS, Inc.) for multiscale analysis of composite structures; and (3) HyperMAC couples the MAC/GMC micromechanics capabilities with the HyperSizer stiffened structural optimization software (ref. 2).
FEAMAC has enabled multiscale stochastic progressive failure analysis of composite structures. It uses the generalized method of cells (GMC) micromechanics approach to model the local composite material behavior at the integration points within each finite element of a composite structure via the ABAQUS user-definable subroutines. GMC localizes to the level of the fiber and matrix constituent materials and thus enables the use of arbitrary nonlinear constitutive, damage, and life models (many of which are provided by MAC/GMC) for each monolithic constituent phase throughout the composite structure. This circumvents the need for the development and characterization of effective anisotropic constitutive models for the composite materials within the structure, which can be difficult in the presence of material nonlinearity. Furthermore, GMC provides access to the constituent-level stresses and strains throughout the structure, enabling the use of fiber- and matrix-scale failure and damage-evolution criteria. The well-documented computational efficiency of the GMC micromechanics approach (as compared with the finite element micromechanics approach, for example; refs. 3 and 4) permits the tractability of coupled structural FEA-micromechanics problems.
ABAQUS finite-element mesh of a dogbone specimen and MAC/GMC repeating unit cell (RUC) operating at each integration point.
The figures show an application of the developed modeling framework to the stochastic progressive failure analysis of a longitudinally reinforced SiC/Ti-21S composite1 dogbone specimen. A standard finite element mesh was employed, but the MAC/GMC micromechanics repeating unit cell was operative at each integration point in each element. The material input data thus became the material properties of the SiC fiber and Ti-21S matrix constituents. In this example, a deterministic viscoplastic model was employed to represent the Ti-21S matrix and a stochastic strength model was employed for the SiC fibers. The statistical strength properties for the fibers within each element were assigned randomly over the mesh geometry such that they sum to the vendor-supplied fiber strength histogram.
This process enables the failure initiation location, as well as the failure progression, to be random, rather than occurring at the highest stress riser. This is illustrated in the following figure, which plots the fiber damage at different times during a simulated tensile test on the composite specimen. Fiber failure initiated within the specimen gauge section and then arrested. Fiber failure then initiated at another location in the gauge section and also arrested. Finally, failure progressed through the specimen (indicating final failure) at the first failure initiation location.
Local fiber damage fraction as a function of time as fiber failure progresses within a longitudinal 33 wt% SiC/Ti-21S specimen with spatially distributed fiber strength statistics.
Deterministic simulations, and those that do not randomize the fiber failure properties over the specimen geometry, predict failure to initiate at the highest stress riser in the specimen, which is at the bottom of the transition region to the gauge section (above the predicted failure location shown). Correctly accounting for the stochastic nature of the failure initiation location is important because real structures fail in this manner. For example, in experimental tensile tests, SiC/Ti-21S specimens, like those modeled, repeatedly failed within the gauge section, not at the bottom of the transition region.
Find out more about this research:
ImMAC (Integrated multiscale Micromechanics Analysis Code) Software Suite:
http://www.grc.nasa.gov/WWW/LPB/mac/
HyperSizer product profiles:
http://www.hypersizer.com/Products/products.htm (external site)
Last updated: December 14, 2007
Responsible NASA Official:
Gynelle.C.Steele@nasa.gov
216-433-8258
Point of contact for NASA Glenn's Research & Technology reports:
Cynthia.L.Dreibelbis@nasa.gov
216-433-2912
SGT, Inc.
Web page curator:
Nancy.L.Obryan@nasa.gov
216-433-5793
Wyle Information Systems, LLC
