A novel methodology for data analysis of dynamic angle of repose tests and powder flow classification

We present a new post-processing methodology to analyse powder flow image data gathered via dynamic angle of repose tests. We aim to expand the flow descriptors, allowing a more detailed and nuanced measurement of flow rheology. This makes the data extraction reliable even if the free surface profile is not clearly identifiable. After defining 30 flow descriptors to be measured from powder flow snapshots, we use Principal Component Analysis to understand their relations with the physics of the system. The set of descriptors is optimised, and the most significant ones are identified, allowing the physics to be captured with fewer essential parameters. We demonstrate that a comprehensive picture of powder flow is achievable simply with the centre of mass and the flow merging point. Our research demonstrates that traditional data extraction methodologies are insufficient to fully describe the flow, making our framework ideal for enhancing the understanding of flow properties.