TY - GEN
T1 - Stochastic finite element analysis of composites
AU - Doles, Randall
AU - Cole, Courtney
AU - Sanei, Seyed Hamid Reza
N1 - Publisher Copyright:
© 2018 by DEStech Publications, Inc. All rights reserved.
PY - 2018
Y1 - 2018
N2 - Composite materials can have a large variation in material properties, especially in the transverse direction. This can cause difficulty when designing with these materials and may lead to overdesigning the material. Using finite element software, a model was created to simulate failure of composite materials in the transverse direction. Using micrograph images, a simulation of microstructures was generated based on fiber volume fraction and special positioning of fibers. Using a multi-level approach, the material properties were calculated in the micromechanical model. The comparison of the obtained elastic properties with rule of mixture revealed that the transverse properties are poorly predicted by rule of mixture. A mesomechanical model was then developed based on the material properties obtained from the micromechanics model. Using a progressive failure approach, each element could independently be disabled when the element fails, causing a simulated propagation of failed material. The results showed that failure follows a Weibull distribution consistent with experimental observation. The developed stochastic model will allow for infinite possibilities of failure replicating experimental findings.
AB - Composite materials can have a large variation in material properties, especially in the transverse direction. This can cause difficulty when designing with these materials and may lead to overdesigning the material. Using finite element software, a model was created to simulate failure of composite materials in the transverse direction. Using micrograph images, a simulation of microstructures was generated based on fiber volume fraction and special positioning of fibers. Using a multi-level approach, the material properties were calculated in the micromechanical model. The comparison of the obtained elastic properties with rule of mixture revealed that the transverse properties are poorly predicted by rule of mixture. A mesomechanical model was then developed based on the material properties obtained from the micromechanics model. Using a progressive failure approach, each element could independently be disabled when the element fails, causing a simulated propagation of failed material. The results showed that failure follows a Weibull distribution consistent with experimental observation. The developed stochastic model will allow for infinite possibilities of failure replicating experimental findings.
UR - http://www.scopus.com/inward/record.url?scp=85059433202&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85059433202&partnerID=8YFLogxK
M3 - Conference contribution
AN - SCOPUS:85059433202
T3 - 33rd Technical Conference of the American Society for Composites 2018
SP - 2426
EP - 2437
BT - 33rd Technical Conference of the American Society for Composites 2018
PB - DEStech Publications Inc.
T2 - 33rd Technical Conference of the American Society for Composites 2018
Y2 - 24 September 2018 through 27 September 2018
ER -