A simplified nonlinear modeling for a two-stage grinding process circuit

The grinding is a typical complex industrial process, which can be influenced by the properties of the materials, operating conditions, equipment size and many other stochastic factors. In this paper, a simplified nonlinear model for a two-stage grinding process of a plant in China is constructed based on the particle volume balance principle. The grinding process of the mill is described by the power consumption and the rheological properties of the pulp in the mill. And, the state equations of the grinding circuit are constructed through the connections of the pulp flow between each process. Given that the operating parameters cannot be measured, the initial value and the parameters of state equations are estimated by a least squares method based on the collection data of the grinding process in the steady state. The overflow concentration, the particle size and the flow rate are used to verify the validity of the model, and the experimental results indicate that, the estimated values reflect the actual trends of the process data, and the proposed model helps to monitor the status of the grinding process.