Control of pure time delay processes

Abstract The general problem of controlling a physical process with predominant time delay is analysed. An effective method of controlling such a process is proposed and successfully applied. The control scheme consists of a two-loop configuration: a fast auxiliary loop consisting of a proportional controller and a simple lag element, and a main (process) loop provided with another proportional controller operating on the process. While the auxiliary loop output acts as common feedback to both the controllers, the process output adjusts the set point of the auxiliary controller. Disturbances entering the process, for which compensation is desired, are transmitted to the auxiliary loop where a corrective action is generated even before an error appears in the controlled variable. Digital simulation study reveals that, by suitable choice of the compensator parameters, this control arrangement results in remarkably superior performance compared to other known methods of dead time control. The technique is applied to a real process having a dead time of 2·7 min to demonstrate its easy implementability and to confirm its superiority over existing control schemes.