A Parametric Model of the LARCODEMS Heavy Media Separator by Means of Multivariate Adaptive Regression Splines

• Main Authors: Mario Menéndez Álvarez, Héctor Muñiz Sierra, Fernando Sánchez Lasheras, Francisco Javier de Cos Juez
• Format: Article
• Language: English
• Published: MDPI AG 2017-06-01
• Series: Materials
• Subjects: heavy media separation; density separations; multivariate adaptive regression splines (MARS); LARCODEMS
• Online Access: http://www.mdpi.com/1996-1944/10/7/729

Modeling of a cylindrical heavy media separator has been conducted in order to predict its optimum operating parameters. As far as it is known by the authors, this is the first application in the literature. The aim of the present research is to predict the separation efficiency based on the adjustment of the device’s dimensions and media flow rates. A variety of heavy media separators exist that are extensively used to separate particles by density. There is a growing importance in their application in the recycling sector. The cylindrical variety is reported to be the most suited for processing a large range of particle sizes, but optimizing its operating parameters remains to be documented. The multivariate adaptive regression splines methodology has been applied in order to predict the separation efficiencies using, as inputs, the device dimension and media flow rate variables. The results obtained show that it is possible to predict the device separation efficiency according to laboratory experiments performed and, therefore, forecast results obtainable with different operating conditions.