Particle Classification Optimization of a Circulating Air Classifier

ABSTRACT

Undoubtedly, wet processing of ores requires huge quantity of water. This provides enough incentive for dry beneficiation of ores which has great promise predominantly from an environmental standpoint and water scarcity in the mining and processing industries. Therefore, the present investigation made an attempt to effectively address the issues related to dry classification of minerals. In this study, a three-factor-three-level Box–Behnken factorial design combined with response surface methodology (RSM) was employed for modeling and optimization of operational parameters of a circulating air classifier. The three main operating parameters studied were air flow rate, feed rate, and guide vane angle. The primary and interaction effects of operating variables were evaluated using RSM while generating the second-order response functions for both the responses, cut size and size selectivity increment. The values of cut size and size selectivity increment obtained using predictive models were in excellent agreement with the observed values. The optimization of these predictive response models resulted in the optimal values of air flow rate, feed rate, and guide vane angle for achieving better classification efficiency. This study establishes that the Box–Behnken factorial design combined with RSM effectively model the performance of circulating air classifier.

KEYWORDS:

• Dry beneficiation
• molybdenum ore
• air classification
• response surface methodology
• modeling
• optimization

Nomenclature

Α	=	Reduced efficiency curve parameter/sharpness of separation
BBD	=	Box-Behnken design
d80	=	80% passing size, μm
d50a	=	Cut size at actual efficiency, μm
d50c	=	Cut size at corrected efficiency, μm
Ea	=	Actual efficiency, %
Ec	=	Corrected efficiency, %
R2	=	Correlation coefficient
RSM	=	Response surface methodology
wc	=	Mass flow rate of coarse fraction, kg/h
wf	=	Mass flow rate of fine fraction, kg/h
Xi	=	Independent variables in actual/uncoded form
xi	=	Independent variables in coded form
Yi	=	Response variables
Δϕc	=	Cumulative size distribution interval for coarse fraction
Δϕf	=	Cumulative size distribution interval for fine fraction
ΔS	=	Size selectivity increment

Acknowledgments

The author wishes to thank the Director of the CSIR-Institute of Minerals and Materials for permitting publication of this work.

Disclosure statement

The authors report no conflicts of interest. The authors alone are responsible for the content and writing of the article.

Additional information

Funding

This work has been financially supported by the European H2020 project “Flexible and Mobile Economic Processing Technologies” (FAME) [Grant No. 641650].