Neuro-adaptive modeling and control of a cement mill using a sliding mode learning mechanism

A novel neural network adaptive control scheme for cement milling circuits is presented. Estimates of the one-step-ahead errors in control signals are calculated through a neural predictive model of the plant and used for controller tuning. A robust on-line learning algorithm, based on the direct use of sliding mode control (SMC) theory is applied to both: to the controller and to the model as well. The proposed approach allows handling of mismatches, uncertainties and parameter changes in the plant model. The results from simulations show that both the neural model and the controller inherit some of the advantages of SMC, such as high speed of learning and robustness. Fast convergence ability and good performance on reducing mapping error are observed, leading to an improvement of the transient response of the closed-loop system.