Measurement and Explicit Finite Element Modeling of Dynamic Crush Behavior of Carbon Fiber Reinforced Polymer Composites
Restricted (Penn State Only)
- Author:
- Haluza, Rudy T
- Graduate Program:
- Engineering Science and Mechanics
- Degree:
- Doctor of Philosophy
- Document Type:
- Dissertation
- Date of Defense:
- August 18, 2022
- Committee Members:
- Kevin Koudela, Co-Chair & Dissertation Advisor
Charles Bakis, Co-Chair & Dissertation Advisor
Jose Palacios, Outside Unit Member
J. Michael Pereira, Special Member
Robert Goldberg, Special Member
Reginald Hamilton, Outside Field Member
Albert Segall, Program Head/Chair - Keywords:
- Crush
composite
carbon
fiber
energy
CFRP
shear
model
simulation
crashworthiness
LS-DYNA
MAT213 - Abstract:
- Composite structures—typically used in aerospace vehicles due to their light weight, fatigue resistance, high specific strength and stiffness, and tailorable nature—also have shown promise as energy-absorbing structures, which is important for crashworthy vehicle design. The overall goals of the present investigation are to improve methods of crush testing and computational simulation of crushing of fiber reinforced polymer composites. The functionality, accuracy, and repeatability of a movable-boundary crash sled were verified, and an algorithm was implemented in a semi-automated program to process results from crash sled tests with impactor speeds of 3.81 and 7.62 m/s. Best practices for measuring shear strain in ASTM D5379 V-notched beam specimens using digital image correlation were developed and subsequently applied to the measurement of the out-of-plane shear behavior of IM7/8552 carbon/epoxy tape laminates. Computational simulations were run in LS-DYNA to verify the ability of a next-generation finite element material model (MAT213) to predict the orthotropic stress-strain behavior of unidirectional and multidirectional IM7/8552 tape laminates. The successful completion of the crash sled verification, shear testing, and verification models led to the final steps involving composite crush simulations. Two sets of simulations were performed involving simulations intended for calibration followed by predictive simulations. Simulations of flat-plate specimens (with buckling supports included in the test fixture) were iteratively run through trial-and-error to calibrate model parameters involving contacts, material damage, element erosion, friction, numerical filters, and the mesh. The results of calibration showed that MAT213 could produce a simulated force-displacement response within the scatter of experimentally measured curves. By incorporating multiple layers of shell elements and tiebreak contacts, the failure mode of the model was also shown to be comparable to the experimental tests. After the successful calibration of the flat-plate models, predictive simulations of C-channel specimens were completed using drop tower and crash sled test rigs and two pairs of impactor mass/velocity conditions. The drop tower simulations underpredicted the experimental stable crushing force by 33% for the higher-velocity condition and 23% for the lower-velocity condition (experimental data provided by University of Utah). The crash sled simulations underpredicted the experimental stable crushing force by 27% for the higher-velocity condition but fell within the experimental scatter for the lower-velocity condition. The more desirable results of the lower-velocity test condition may be related to the fact that the flat-plate crush simulations used for calibration were conducted at a similar velocity to the lower-velocity condition.