A Model Predictive Control for Coal Beneficiation Dense Medium Cyclones

The purpose of the controller presented is to maintain the proportion of carbon content in the products of the coal washing dense medium cyclone (DMC) at a certain level by optimally manipulating the density of the heavy medium. In practice, the DMC processes are mainly controlled by empirical methods which are process specific and cannot satisfy the industrial requirements today. As a result, industries are looking at introducing sensors to monitor the coal quality from the DMC process. Application of such sensors requires the development of new controllers that can handle large delays and sampled measurement. Therefore, a model based closed-loop control approach is presented to improve the separation efficiency of DMC plants. The feed coal quality is taken as the feed forward information and the sampled and delayed measurement from the DMC output is utilized as feedback for the controller to improve the quality of coal product and ensure the robustness of the controller. In addition, a state updating scheme is presented for the controller because the time delay of the measured feedback can be one hour or longer. The effectiveness of the designed controller is demonstrated by simulations.