Fuzzy tuned model based control for level and temperature processes

At present, model based controllers are being extensively used in process industries due to their simple tuning strategy. Internal model control (IMC) is one of the widely accepted model based controller design methodologies in close-loop control applications with considerable process lag. But, similar to the other model based controller design techniques, availability of an appropriate linear model of the concerned process is essential for IMC design. However, in reality, most of the industrial processes are nonlinear in nature. Hence, designing of an IMC controller for such cases is truly a difficult task especially for ensuring satisfactory performance during transient as well as steady state operating conditions simultaneously. In this study, to achieve an overall acceptable process response, we propose an auto-tuning scheme for the conventional IMC-PID controller by varying its sole tuning parameter depending on the latest process operating conditions. Superiority of the proposed auto-tuned IMC-PID controller is observed for real-time level and temperature processes where the close-loop time constant (λ) is varied with the help of twenty-five fuzzy rules defined on the values of process error (e) and change of error (\(\Delta e\)).