Abstract
Particle-based simulations constitute a powerful alternative to continuum methods like FEM to investigate the mechanics of complex materials. The key to these methods is the choice of what the 'particle' is, and what are the physical laws that govern the interaction among particles. From single hydrogen atoms to grains of sand or larger, the extreme flexibility in the nature of the particle allows this class of models to investigate almost ten decades of spatiotemporal scales. From all-atom molecular dynamics simulations to macroscale discrete elements methods, particle-based algorithms excel at accurately investigating phenomena at a scale of 2-3 orders of magnitude larger than the size of their discrete constituents, easily incorporating structural heterogeneity.
The main drawbacks of particle-based methods are usually the high computational cost and the limitations in size and timescale that can be simulated, which are limited by the fastesttimescales of the system. The incorporation of elements of very differentsizesisthen not efficient, noris the investigation of multiscale phenomena like fracture, where emergent phenomena are of critical importance and the macroscale mechanics behavioris not easily predicted by the nano- or micro-scale particle interactions.
To overcome these limitations one can (i) develop multiscale simulation frameworks where particles of different resolutions are seamlessly coupled together, efficiently simulating multiple scales, and (ii) establish general and accurate coarse-graining techniquesto extract and move information across models of increasing size, with the goal of capturing the key microscopic properties that are predictive of the macroscale material mechanics.
In this talk, we will discuss the current trends and challenges in particle-based mechanics simulations across the scales, with a focus on emerging strategiesto tackle multiscale problems and ongoing efforts by the research community in the Netherlands and abroad.
The main drawbacks of particle-based methods are usually the high computational cost and the limitations in size and timescale that can be simulated, which are limited by the fastesttimescales of the system. The incorporation of elements of very differentsizesisthen not efficient, noris the investigation of multiscale phenomena like fracture, where emergent phenomena are of critical importance and the macroscale mechanics behavioris not easily predicted by the nano- or micro-scale particle interactions.
To overcome these limitations one can (i) develop multiscale simulation frameworks where particles of different resolutions are seamlessly coupled together, efficiently simulating multiple scales, and (ii) establish general and accurate coarse-graining techniquesto extract and move information across models of increasing size, with the goal of capturing the key microscopic properties that are predictive of the macroscale material mechanics.
In this talk, we will discuss the current trends and challenges in particle-based mechanics simulations across the scales, with a focus on emerging strategiesto tackle multiscale problems and ongoing efforts by the research community in the Netherlands and abroad.
| Original language | English |
|---|---|
| Title of host publication | Twenty-fifth Engineering Mechanics Symposium, October 25-October 26, 2022. Hotel Papendal, Arnhem |
| Editors | R.A.M.F. van Outvorst, A.J.J.T. van Litsenburg |
| Publisher | Eindhoven University of Technology |
| Pages | 15-15 |
| Number of pages | 1 |
| Publication status | Published - Oct 2022 |
| Event | 25th Engineering Mechanics Symposium, EM 2022 - Hotel Papendal, Arnhem, Netherlands Duration: 25 Oct 2022 → 26 Oct 2022 Conference number: 25 https://engineeringmechanics.nl/symposium/ |
Conference
| Conference | 25th Engineering Mechanics Symposium, EM 2022 |
|---|---|
| Abbreviated title | EM 2022 |
| Country/Territory | Netherlands |
| City | Arnhem |
| Period | 25/10/22 → 26/10/22 |
| Internet address |