Modeling and simulation of spiral concentrators

Authors: Sadeghi, Maryam
Advisor: Bazin, ClaudeHodouin, D.
Abstract: Spiral concentrators are gravity separation devices used for the valorization of ores for which the specific gravity of the valuable minerals is significantly different from that of the gangue minerals. This is the case of iron ore processing discussed in this thesis. Spirals are preferred to other mineral separation devices because they are inexpensive, simple of operation, practically maintenance free, low energy demanding and do not require the use of reagents to separate minerals. Despite these appealing qualities and unlike flotation, few studies are dedicated to the analyze the spiral operation. Indeed, it is only recently that basic process analysis has shown that spirals fail to recover coarse dense mineral particles to the heavy product. It has also been recently found that wash water, as the main control variable for the operation of the studied spirals, is partly responsible for the loss of coarse valuable particles. To make up for this loss and to maintain a satisfactory operation of spirals, other control variables such as the concentrate cutter openings and position of the splitter gates at the spiral discharge should be used. The study of alternative control strategies for spirals implies testing these strategies in the industrial environment, which is practically impossible considering that the operation of mineral processing plants is subjected to constraints imposed on the production and on the quality of the plant product. Besides, the impact of disturbances due to ore variations and/or mechanical problems in the plant is unavoidable. Testing new control strategies in pilot plant is possible but it is certainly costly if one can find a pilot plant equipped for such testing. Plant operators are thus seeking for rapid and inexpensive approaches to carry out such investigation to identify strategies that can subsequently be safely tested in a plant. Process simulation offers this possibility, but a literature survey rapidly revealed the absence of simulation tools for spirals and more particularly for gravity separation circuits that use spiral concentrators. The first step in building such circuit simulator is to find or develop mathematical models for the units of the circuit. With the objective of testing control strategies, it is thus important that the model to be developed be able to account for the usual spiral operating variables, namely the cutter opening, the discharge splitter position and the wash water addition. The objective of this research project is to propose such model for spirals. A mathematical model is therefore described to predict the trajectory of particles distributed in mineral/size classes as they flow down a spiral. This is achieved here by introducing the new concept of Particle Displacement Functions (PDF) in spiral modelling to describe the movements of the particles flowing down the spiral. The model is calibrated using data generated from tests conducted on a pilot-scale test rig built around a Mineral Technologies WW6+ spiral, installed in the Mineral Processing Laboratory of Université Laval. The model is further validated with data obtained from a WW6E spiral operated at COREM. The test rig commissioned at Laval University uses a controlled wash water distribution within the spiral turns to overcome the observed problem of unequal wash water distribution when using the spiral built-in wash water distribution system. The test rig is also designed to sample simultaneously the product from each concentrate cutter of the spiral, so that the progress of the heavy mineral concentration down the spiral can be monitored. This information can be used to calibrate the model. The calibrated spiral simulator predicts adequately the spiral response to changes in the states of the operating variables. The model can simulate a cleaner spiral operation although it is calibrated using data collected from a rougher spiral operation. The proposed model is ready to be implemented into a simulator for an iron ore processing plant, although work remains to be done to finalize the calibration methodology to improve the prediction of the mineral size distribution in the spiral middling stream.
Document Type: Thèse de doctorat
Issue Date: 2021
Open Access Date: 14 June 2021
Grantor: Université Laval
Collection:Thèses et mémoires

Files in this item:
Description SizeFormat 
37089.pdf34.75 MBAdobe PDFThumbnail
All documents in CorpusUL are protected by Copyright Act of Canada.