Decoding powder flowability: Machine learning pioneers the analysis of particle-size distribution effects

Abstract

Powders play a crucial role in a wide array of industrial applications such as powder coating, food, and pharmaceutics. The flowability of powder, both in static and dynamic conditions, is paramount for its practical utility. Among various factors influencing powder flowability, the particle-size distribution (PSD) stands out as a significant feature. Nevertheless, the impact of PSD on powder flowability remains empirical and not fully characterized. This study undertakes an in-depth investigation into the intricate relationship between PSD and powder flowability, utilizing advanced machine learning models. The angle of repose (AOR) and outflow mass rate (OMR) were employed to measure the static and dynamic flowability, respectively. Artificial neural network (ANN) and decision-tree models identify the impactful particle-size bins with positive and negative effects on static and dynamic flowability. These established models unveil the specific influence regions of PSD on powder flowability and hold potential for broader applications across various industries.