Homogenization of Composites Focused on Time Domain Effective Viscoelasticity and Investigating Instabilities.

Tarkes Dora P. (Karlsruhe Institute of Technology (KIT), Germany)

Thursday 5th November, 2020 14:00-15:00 ZOOM (ID: 931 2780 4645)

Abstract

Topic: Applied Mathematics Seminar Tarkes Dora P.
Start Time : Nov 5, 2020 01:48 PM

Meeting Recording:
https://uofglasgow.zoom.us/rec/share/f8wbOBi8P9HffnE0nCduslHuM022XyhaqT016pA2hRKK5M5NISM41NP5msvzG7el.j3W4UgCymFyh-Faq

Access Passcode: See the e-mail with subject: Seminar recording password

Composites have ushered in materials science and engineering, allowing the design of materials with superior mechanical properties. Computing it’s effective behavior by mathematical homogenization is challenging in different field of application. The seminar would be in two folds covering viscoelastic homogenization in time domain with application focus on residual stress analysis in thermoset based composites and investigating instabilities in soft composites.

With an eye towards Industry 4.0, the virtual process chain plays a significant role in industries for process and part optimization. Time-domain viscoelastic homogenization problem is fundamental and vital for simulating residual stress in composite parts. The incremental variational based mean-field homogenization as proposed by Lahellec and Suquet (Lahellec and Suquet, 2007) involves the second-order statistical moments of internal strain. This method is extended and applied here to a viscoelastic composite with phases exhibiting Maxwellian behavior. Effective behavior and statistical moments of strain field are compared with full-field simulations. This is carried out for microstructures of unidirectional fiber-reinforced polymer (i.e., short, and long fiber) composite.

Meanwhile in the context of full-field homogenization, Bloch-Floquet analysis is used to investigate instabilities in non-linear composites with periodic arrangement of stiff inclusions. Experiments are performed using 3D printed composite samples for different microstructure. The instability patterns varied from a localized microscopic scale to a long-wave macroscopic scale by altering the initial geometry of the periodic microstructure. Inclusion spacing ratio and aspect ratio are the primary influencing factors. A qualitative agreement is observed between the numerically predicted trends and experimental results.

Add to your calendar

Download event information as iCalendar file (only this event)