Damage and Optimization Models for Analysis and Design of Discontinuous Fiber Composite Structures

1
Damage and Optimization Models for Analysis and
Design of Discontinuous Fiber Composite Structures

• Ba Nghiep Nguyen
• Acknowledgements PNNLs Computational Science
  Engineering Initiative
• Korolev Vladimir, Brian Tucker (contributors)

NSF/DOE/APC Workshop Future of Modeling in
Composites Molding Processes (Design
Optimization Session), June 9-10, 2004,
Arlington, Virginia
2
Damage and Optimization Models for Analysis and
Design of Discontinuous Fiber Composite
Structures
PNNL has developed

• A multiscale mechanistic approach to damage
  based on
• micromechanical and continuum damage mechanics
  descriptions
• An optimization approach using the optimal
  control theory accounting for the composite
  microstructure
• An experimental procedure for acquiring
  acoustic emission signals to identify damage

3
Damage and Optimization Models for Analysis and
Design of Discontinuous Fiber Composite
Structures
Optimal Control Theory Approach to Short-Fiber
Composites
Fiber volume fractions Fiber aspect ratios
Fiber orientation parameter
4
Current State of the Art in Design and
Optimization of Discontinuous Fiber Composites

• Elastic analysis-based design
• Micromechanical models rely on material database
  (fiber volume fraction, aspect ratio, orientation
  distribution, etc.) to predict effective
  properties
• Process modeling to predict fiber orientation
• Control of process and microstructural parameters
  to improve composite stiffness
• Elastic finite element analysis of the as-formed
  composite structure
• Nonlinear analysis based design
• Phenomenological models rely on material database
  and testing of specimens
• Nonlinear micromechanical models derived from the
  self-consistent and Mori-Tanaka frameworks (e.g.
  elastic-plastic, damage, creep)
• PNNL damage models using a multiscale mechanistic
  approach
• ORNL micromechanical models
• Formal optimization methods
• Only at the beginning
• Duvaut et al. (2000). Optimization of Fiber
  Reinforced Composites, Composite Structures, 48,
  83-89
• PNNL optimization model using the optimal control
  theory

5
Vision on Future Directions

• Design optimization methods should be reliable
  to effectively assist processing manufacturing
  of composite components and parts
• Development of new process and constitutive
  models accounting for the constituents
  characteristics and properties, and their
  interaction with each other
• Interface between process modeling and structural
  modeling to create and design a composite part
  through simulations
• Processing manufacturing can rely on efficient
  design optimization methods rather than on
  trial-and-error approaches
• Reduce the number of experimental tests and trial
  moldings

Manufacturing
Structural modeling
Process modeling
6
Perceived Gaps

• Where we are now
• Micromechanical models predict elastic properties
  and some nonlinear responses
• Process models provide qualitative predictions of
  fiber orientations in injection molding
• Phenomenological constitutive models exist in
  commercial FE codes for structural analyses
• Limited interface between process and structural
  modeling
• Analysis and design are still based on intensive
  material database obtained through experiments
• Initiation of multiscale mechanistic models based
  on micromechanics and continuum mechanics

• Where we should be
• Accurate micromechanical models accounting for
  concentrated fiber volume fractions
• Accurate process models for short- and long-fiber
  thermoplastic injection molding
• Constitutive physics-based models for predicting
  durability and time-dependent behavior
• Interface between process and structural modeling
  for linear and nonlinear analyses
• Optimization methods accounting for process,
  design and loading variables and constraints
• Analysis and design should rely on reliable
  physics-based models to assist processing
  manufacturing

7
Research Thrusts

• Micromechanics
• Process micromechanics Effects of fiber content,
  length on the rheology and fiber orientation
• Micromechanics of materials Homogenization
  accounts for interaction between constituents and
  defects
• Continuum mechanics Need of constitutive models
  for
• Fatigue
• Time dependent behaviors (creep, relaxation,..)
• Impact
• Moisture
• Optimization models accounting for nonlinear
  behaviors
• Minimization of damage
• Improvement of durability (fatigue, creep)
• Multi-scale modeling
• From a microstructural to a continuum model