2026/08/06 by Stefan R. Tölle, Lorenz Dörschel, Stefan Klinken-Uth +4
Engineering · Environmental Science · #Agrégation #Coagulation and Flocculation Studies #Exploit #Granular flow and fluidized beds #Granulation #Identification (biology) #Mineral Processing and Grinding #Population #Process (computing) #Robustness (evolution) #Scaling
paper · pdf · doi:10.1016/j.jprocont.2026.103804
published in Journal of Process Control 166, 103804 (Elsevier BV)
openalex publication_date 2026/08/06 · crossref created 2026/08/06 · openalex created_date 2026/08/07 · crossref deposited 2026/08/07 · crossref indexed 2026/08/07 · openalex updated_date 2026/08/09 · crossref issued 2026/10/01 · crossref published 2026/10/01 · crossref published-print 2026/10/01
Population balance models provide a mathematical framework for describing the dynamics of particulate systems. For aggregation processes, such as twin-screw wet granulation, population balance models critically depend on accurate aggregation kernels, yet first-principles derivation is often impractical, and data-driven methods can lack physical interpretability. This work presents a sparse identification framework for learning physically plausible aggregation kernels directly from data. The approach enforces physical constraints, exploits structural properties, applies a systematic scaling strategy, and extends naturally to actuated systems. Validation on experimental data from a continuous twin-screw wet granulation process confirms the method’s robustness and its ability to recover physically meaningful aggregation kernels from real-world measurement data. The identified population balance model was able to predict the median particle size with a mean absolute error of 134.8 µm (10 %).