Redesign of an Intelligent Adaptive Control Scheme for the Cement Milling Process

The re-engineering of the process control systems within the legacy cement plants in the countries in transition is a very important problem because of the existing necessity to ensure compliance with the increasingly demanding cement quality standards, to reduce the energy and the operating costs. The paper presents a redesigned method for intelligent adaptive control of a cement milling circuit. Estimates of the one-step-ahead errors in control signals are calculated through a developed fuzzy neural network-based 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 results from simulations conducted with an experimentally verified model of a cement milling circuit show the fast convergence ability of the proposed control scheme and its good performance on reducing mapping errors, leading to an improvement of the transient response of the closed-loop system. The proposed approach allows handling of mismatches, uncertainties and parameter changes in the plant model.