1. The discrete element method (DEM) is a versatile numerical method for modelling granular dynamics, but is computationally intensive.
2. Data-driven model order reduction can be used to accelerate DEM simulations by predicting new system states from detailed simulations in advance.
3. This paper explores the feasibility of this approach and tests it on two different systems, with promising results.
The article provides an overview of data-driven model order reduction as a way to accelerate DEM simulations of granular media. The authors present their own implementation of this technique and test it on two different systems, with promising results. The article is well written and provides a clear explanation of the concept and its potential applications.
However, there are some points that could be improved upon in terms of trustworthiness and reliability. Firstly, the article does not provide any evidence for the claims made about the accuracy and speed-up achieved by using this technique; instead, it simply states that the results were “promising” without providing any further details or analysis. Secondly, while the authors do mention possible risks associated with using this technique (e.g., monetary cost and energy consumption), they do not explore these risks in detail or provide any recommendations for mitigating them. Finally, while the authors do discuss potential applications for this technique (e.g., real-time simulators), they do not explore other potential uses or consider how this technique might be applied in other contexts or industries.
In conclusion, while the article provides an interesting overview of data-driven model order reduction as a way to accelerate DEM simulations of granular media, it could benefit from more detailed evidence for its claims and a more comprehensive exploration of potential risks and applications associated with this technique.