Kalman filter based MPC with dead band for a ball mill

Abstract

This paper addresses the issues of regulating the fineness and mill load of ball mill by using Kalman filter and linear Model Predictive Control. The MPC is designed by various simulation results with the help of MPC toolbox. The main objective of this paper is to reduce the energy consumption during production. This can be achieved by reducing the energy used or by increasing the product units for different fineness. In this paper, two MPCs are designed, i.e., MPC with dead band type I and type II to reduce the energy consumption in different manner. Usually, normal MPC is used to take control action to force the output to reach a set point. But in this paper, the designed MPC is used in a different manner to maintain the output within a range of set point and Kalman Filter is used to estimate sensor output state as none of the sensor gives accurate measurement. Furthermore, the performance of developed methodology has been compared with the normal MPC by simulation.