Presentation
Modeling of mechanical failure phenomena is important both in materials research and for our fundamental understanding of nature. On a microscopic scale, molecular dynamics (MD) simulations can be used to directly model the failure of materials. However, this can be time and resource-consuming to use on large sample spaces. Machine learning methods can learn the mapping between material structures and physical phenomena.
In this case, we focus on the relationship between the mechanical yield stress of α-quartz crystals with a porous layer under shear and tensile stress. We use simplex noise to create structures that can generate geometries that appear similar to natural terrains and surfaces. It is widely used for generating scenery in video games and animations. We further suggest that autoencoders can be used to encode the structures. This is particularly useful for generating new structures with modified yield stress. We are interested in creating an understanding of how these neural networks reflect the underlying physics with minimal direct guidance.
Speaker
Fahimeh Najafi is a PhD candidate at the Condensed Matter Physics Group and NJORD. She studies the frictional properties of material coupled with machine learning methods.
Program
11:30 – Doors open and lunch is served
12:00 – "Modeling mechanical properties of material using neural networks" by Fahimeh Najafi (PhD Candidate, NJORD)
This event is open for all students, PhD candidates, postdocs, and everyone else who is interested in the topic. No registration needed.
About the seminar series
Once a month, dScience will invite you to join us for lunch and professional talks at the Science Library. In addition to these, we will serve lunch in our lounge in Kristine Bonnevies house every Thursday. Due to limited space (40 people), this will be first come, first served. See how to find us here.
Our lounge can also be booked by PhDs and Postdocs on a regular basis, whether it is for a meeting or just to hang out – we have fresh coffee all day long!