Multi-Objective Optimization Approach for Optimizing the Performance of Double-Stage and Lapple Cyclone Separator

Cyclone separators play a crucial role in various industrial applications, particularly in particle filtration for environmental and health protection. This study employed Response Surface Methodology (RSM), Artificial Neural Network (ANN), Multi Objective Genetic Algorithm (MOGA), and a novel approach called Multi Objective for Ant Lion optimizer (MOALO) to optimize the performance of double-stage and lapple cyclone separators. Through statistical analysis, this study investigates the impacts of cyclone diameter, inlet velocity, and size of the particle on the total efficiency, grade removal efficiency, and pressure drop. The results indicate that the double-stage cyclone separator is preferable to the lapple cyclone separator. The RSM and ANN models exhibit high predictive accuracy, with R2 values indicating strong correlations with actual data. Optimal operating conditions for the double cyclone have been identified with the cyclone diameter 400 mm, inlet velocity 12 m/s, and particle size 1.68 µm, resulting total efficiency of 80.58%, pressure drop of 689.71 Pa, and grade removal efficiency of 61.59%, providing insights for improving cyclone separator performance. This research highlights the effectiveness of RSM, ANN, MOGA and MOALO in optimizing cyclone separators.