Abstract
We investigate the use of reduced-order modelling to run discrete element simulations at higher speeds. Taking a data-driven approach, we run many offline simulations in advance and train a model to predict the velocity field from the mass distribution and system control signals. Rapid model inference of particle velocities replaces the intense process of computing contact forces and velocity updates. In coupled DEM and multibody system simulation, the predictor model can be trained to output the interfacial reaction forces as well. An adaptive model order reduction technique is investigated, decomposing the media in domains of solid, liquid, and gaseous state. The model reduction is applied to solid and liquid domains where the particle motion is strongly correlated with the mean flow, while resolved DEM is used for gaseous domains. Using a ridge regression predictor, the performance is tested on simulations of a pile discharge and bulldozing. The measured accuracy is about 90% and 65%, respectively, and the speed-up range between 10 and 60.
Highlights
Computational modelling of granular dynamics has important applications in both science and engineering, but is challenging due to the complex nature of granular media
By principal component analysis (PCA) of simulation snapshots, sampled in a regular grid covering the mixer interior, models were built for predicting the particle velocity field and blade force as function of the mixer control parameters
The proper orthogonal decomposition (POD) method is a special case of SVD-based model order reduction that is popular in computational mechanics and fluid dynamics
Summary
Computational modelling of granular dynamics has important applications in both science and engineering, but is challenging due to the complex nature of granular media. The computationally expensive process of computing contact forces is substituted by rapid inference of the model to output the velocity field at the particle positions This approach can be expected to perform well in the solid and liquid regime but poorly in the gaseous regime, where individual particle motion is not strongly correlated with the mean flow. The method is tested on two different systems, a pile with controlled feed and gravity-driven discharge flow, and a blade cutting and pushing through a particle bed like a bulldozer blade In both cases, a model is trained to predict the velocity field from the given input signals and the current mass distribution. It is natural to investigate the performance of a plain regression model and building knowledge for employing more advanced machine learning algorithms
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
