Intelligent control for operation of iron ore magnetic separating process

In the high intensity separating process of hematite ore, the concentrate grade and tailing grade cannot be measured continuously. Itpsilas difficult to use any accurate mathematical models to describe the relationship between these two indices and other technique indices, i.e. the exciting current, the rinsing water flow, the feed density for its strong nonlinearity and uncertainty. Therefore, an intelligent control method for the process operation has been presented which comprises a loop control layer and an optimization setting layer for the rinsing water flow, the exciting current and the feed density. The optimization setting layer consists of a setting model based on CBR and a feedback compensator using Rule-Based Reasoning. In the setting model, the method of mechanism modelling and CBR are combined to compute the set point of exciting current. The proposed control method has been successfully applied to the high intensity magnetic separating process in the China. It can control the actual concentration grade and tailing grade within their targeted ranges, when the ore separability are fluctuating. The method greatly increases concentration grade, and successfully reduces tailing grade.