(184b) Modified IMC-PID Controller for Stable and Time Delayed Process

Internal Model Control (IMC) is a commonly used technique that provides a transparent mode for the design and tuning of control law for various type of processes. The parameters of a physical system vary with operating conditions, hence it is essential to design a control system that yields robust performance and better set-point tracking. PID controllers are widely used in process industries as they provide satisfactory results for set-point tracking as well as for disturbance rejection, however their responses are generally oscillatory in the face of time delay and/or non-minimum phase characteristics in general, which in turn results in mechanical fatigue. In this study, we propose a modification in IMC structure whereby the controller incorporates a traditional PID controller in its loop to gain crucial advantage in its performance. The proposed method is termed as modified IMC-PID (or simply IMC-PID) controller and it incorporates the effect of model uncertainties and disturbances entering into the process too. A PID controller is used as a slave to a specially tuned IMC as the master. The control loop is structured accordingly. The simulation results on a stable second order overdamped process with time delay show the efficacy of the IMC-PID controller over the other available control laws such as PID controller by Shamsuzzoha and Lee (2008), IMC controller by Rivera et al. (1986), and traditional PID controller tuning through Ziegler-Nichol’s technique. Our method yields better set-point tracking as well as more efficient disturbance rejection. In a servo problem the modified IMC-PID technique yields an ISE value of 0.09 as opposed to 0.11, 3.91 and 2.89 respectively with IMC by Rivera et. al. (1986), PID by Shamsuzzoha and Lee (2008) and traditional PID controller tuning through Ziegler-Nichol’s technique. Modified IMC-PID yields 2.58% overshoot as opposed to 3.65% and 13.07% of IMC by Rivera et. al. (1986) and traditional PID controller tuning through Ziegler-Nichol’s technique respectively, whereas it yields a rise time of 2.87 sec as opposed to 4.82 sec and 4.13 sec by IMC by Rivera et. al. (1986) and PID by Shamsuzzoha and Lee (2008) respectively. We can further interpret from the process response that the proposed IMC-PID method behaves more like PID controller from response point of view, whereas it behaves more like an IMC from robustness point of view. Since it contains the best behaviours of both the controllers, it would be the perfect method to control any process. There are however a couple of limitations in the proposed technique such as the technique is applicable only to an open-loop unstable process and it should be individually controllable by both IMC and PID controller.

References

[1] Shamsuzzoha M. and Monnyong Lee (2008). “Analytical design of enhanced PID filter controller for integrating and first order unstable processes with time delay”, Chemical Engineering Science, 63(10), pp. 2717-2731

[2] Rivera D E., M. Morari, S. Skogestad (1986). “Internal Model Control: PID controller design”, Industrial & Engineering Chemistry Process Design and Development, 25(1), pp. 252-265