The 7th National Conference on Computational Mechanics of Granular Materials in conjunction with The 4th International Symposium on Computational Particle Technology (CMGM-2024 & CPT-2024)

Recent studies have proposed several novel inlet types for hydrocyclones, which have improved certain performance objectives compared to traditional inlets when their geometric parameters are fixed. However, the extent of overall improvement in key performance objectives relative to optimal traditional inlet geometric variables remains unknown, impacting the practical application of these new inlet types. Therefore, this study optimizes the inlet parameters, specifically the aspect ratio and height, of two novel inlet types, Tangential Circular Inlet (TCI) and Tapered Spiral Inlet (TSI), as well as the traditional inlet (TI). The aim is to enhance overall performance objectives including pressure drop (∆P), separation sharpness (E_p), cut size (d₅₀), and water split (R_f). Initially, NSGA-II is employed to generate a Pareto optimal set for the four objectives, followed by TOPSIS to determine the best overall performance for each inlet type. To expedite the optimization iteration process, a response surface methodology based on experimentally validated computational fluid dynamics (CFD) simulations is utilized to predict all considered performance objectives. The results show that different inlet types for hydrocyclones exhibit unique optimization potentials. When E_p is prioritized, TI demonstrates the best overall performance. TCI achieves globally superior d₅₀ performance, albeit at the expense of reducing R_f and E_p compared to TI and TCI. TSI exhibits strong overall separation performance, albeit with a significantly higher ∆P compared to the other two inlet types. Additionally, the TI demonstrates advantages in terms of ∆P. These findings provide novel insights into the overall performance optimization of hydrocyclones.