Modeling and Hybrid Pareto Optimization of Cyclone Separators Using Group Method of Data Handling (GMDH) and Particle Swarm Optimization (PSO)

In the present study, a three-step multi-objective optimization algorithm of cyclone separators is utilized for the design objectives. First, the pressure drop (Delta p) and collection efficiency (eta) in a set of cyclone separators are numerically evaluated. Secondly, two meta models based on the evolved Group Method of Data Handling (GMDH) type neural networks are regarded to model the Delta p and eta as the required functions of geometrical characteristics. Finally, a multi-objective (MO) algorithm based on hybrid of Particle Swarm Optimization (PSO), multiple crossover and mutation operator are used for Pareto based optimization of cyclones considering two conflicting objectives Delta p and eta. By comparing the Pareto results of MOPSO with that of multi-objective genetic algorithms (NSGA II) regarding Pareto based multi-objective optimization of the obtained polynomial meta-models, it is shown that there are some interesting and important relationships as useful optimal design principles involved in the performance of cyclone separators.