Advanced model-based control for continuous process industries

Advanced control is an economical method to maximize return on existing capital investment in plants while minimizing the production. The challenge is to also minimize the cost of the material and human resources needed to implement advanced control on real world applications. This paper describes an innovative adaptive process controller based on an unstructured process modeling technique called dynamic modeling technology (DMT). This modeling method reduces the effort required to implement advanced control strategies. The controller is able to automatically determine the structure of the process model as well as adapt the model parameters as operating conditions change. The problems associated with existing classical model based methods, such as unmodeled dynamics, long setup time, changing dynamics and dead time, and the need for detailed process knowledge are greatly reduced. The advantages of an adaptive controller (AC) based on DMT include the ability to control processes with long and changing time delays and the ease of incorporating adaptive feedforward compensation into a control strategy. The limitations of adaptive control strategies that exist today will be reviewed and the results of applications in the glass manufacturing and the oil and gas industries will be presented