Multi-objective optimization design of spiral demister with punched holes by combining response surface method and genetic algorithm

Abstract

Spiral separator with punched holes was numerically analyzed and the correlation between structural variables and performance index was obtained based on the response surface method. Pitch, diameter, pipe length, and inlet velocity were selected as investigated variables, and the separation efficiency and pressure drop were chosen as objective functions. The results demonstrated that increasing the pitch or reducing the diameter results in greater separation efficiency. Pressure drop factor rises with the increase of vane diameter and the decrease of the pitch. Pitch, pipe length, and inlet velocity all have positive correlations with the separation efficiency. The optimization results were obtained by the Non-dominated Sorting Genetic Algorithm-II. These optimal points achieve greater separation efficiency and less pressure drop. Compared with original design, the separation efficiency of optimal point is enhanced by 1.92%–6.46%, while EU is reduced by 41.83%–58.18%. The maximal separation efficiency of optimal points achieves 0.99 with EU of 4.