Collaborative optimization of the wet screening process: A CFD-DEM based modeling approach with bi-objective optimization

Abstract

Wet Vibrating Screening Process (WVSP) is an intricate physical process with interactions between screen surface, fluid and particles. In this paper, the linear WVSP was modeled with the CFD-DEM and optimized based on virtual experiments. Firstly, a numerical model was built to explore the impact of water on the particle behavior. Secondly, virtual experiments were conducted to establish the response surface for the screen efficiency and the screen surface load. Finally, the NSGA-II (Non-dominated Sorting Genetic Algorithm-II) was employed to collaboratively optimize the screening parameters for both screen efficiency and screen surface load. The modeling results indicate that the water stream is not conducive to the screen efficiency and the optimization results show that with the screen efficiency equal to 93.7%, 96.2% and 98.5%, the optimizable range of the screen surface load is 61.0%, 48.9% and 40.1% respectively. While the numerical model provides a revealing insight into the WVSP, the collaborative optimization significantly enhances screening performance and extends the screen's service life.