Abstract
The continuous stirred tank reactor (CSTR) tends to reveal unstable severe nonlinear behavior when the operating level changes. As investigated in an earlier work, the hard nonlinearity nature of the CSTR originates from multiple probable sets of states for the same reaction in the same CSTR under identical ongoing inlet conditions. The previous work concluded, and this paper will discuss, that nonlinear processes with dynamic trajectories will involve a degrading control performance whenever the measured process variable evolves away from the designed level of the desired output trajectory. In this work, this problem will be examined for temperature tracking control on a CSTR with an adaptive fuzzy gain-scheduling proportional-integral-derivative (AFGS-PID) controller scheduled for a dynamic output trajectory. It will be proven that the AFGS-PID has better tracking performance than the fuzzy gain scheduling (FGS-PID) and conventional PID. Although all three controllers demonstrate the same level of control efforts, AFGS-PID has the smallest IAE and ISE. The lowest settling time being exhibited by AFGS-PID also proves that the intended regulator can rapidly track the desired coolant jacket temperature. A Lyapunov analysis is presented to prove the stability of the closed-loop and support the simulated result in the comparative study.