Successful industrial application of advanced control theory to a chemical process

This paper reports the experiences and results of a project aimed at designing an automatic control scheme for titanium dioxide rotary kilns. The process was studied by way of computer simulations, both steady-state and dynamic, from which a low order model was derived by matching the input-output frequency responses. Use of
LQG
theory then led to the conclusion that the kiln can be considered as a single-input, single-output process. Plant trials and simulation studies finally led to the adoption of a control scheme incorporating a self-tuning regulator in a feedback loop around a kiln controlled by a discrete regulator designed on minimum-variance principles. This scheme has been in use for three years and resulted in great improvement in control performance. Long term industrial results are presented. Practical considerations concerning implementation and acceptance by plant personnel are given.