Modeling and multi-objective Pareto optimization of new cyclone separators using CFD, ANNs and NSGA II algorithm

Abstract In this article, Multi-Objective Optimization (MOO) of new cyclone separators namely Karagoz cyclones is performed using Computational Fluid Dynamics (CFD), Artificial Neural Networks (ANN) and Non-dominated Sorting Genetic Algorithms (NSGA II). The design of this cyclone is based on the idea of improving cyclone performance by increasing the vortex length. This cyclone differs from a conventional cyclone with the separation space. Instead of conical part, the separation space of this cyclone consists of an outer cylinder and a vortex limiter. For multi-objective optimization process at first, the flow field is solved numerically in various Karagoz cyclones using CFD techniques and collection efficiency ( η ) and pressure drop ( Δ P ) in cyclones are calculated. In this step the Reynolds averaged Navier–Stokes equations with Reynolds stress turbulence model (RSM) are solved. The Eulerian-Lagrangian computational procedure is used to predict particles tracking in the cyclones and the velocity fluctuations are simulated using the Discrete Random Walk (DRW). In the next step, numerical data of the previous step will be applied for modeling η and Δ P using Grouped Method of Data Handling (GMDH) type ANNs. Finally, the modeling achieved by GMDH will be used for Pareto based multi-objective optimization of geometrical parameters in new cyclones using NSGA II algorithm. It is shown that the achieved Pareto solution includes important design information on new cyclones.