AUTOTUNING OF GAIN-SCHEDULED PH CONTROL - AN EXPERIMENTAL-STUDY

pH control is known to be a difficult control problem. In this work, the gain-scheduled proportional-integral (PI) controller is used to overcome process nonlinearity. Using the gain-scheduled PI controller the nonlinear control problem becomes the identification of the titration curve and the design of the PI controller. With necessary instrumentation and appropriate model parameterization, the titration curve can be found in a robust manner. Autotune identification is employed to estimate the titration curve first and goes on to find the nominal tuning constants. This is the gain-scheduled autotune identification (g-ATV) procedure. The g-ATV procedure is tested in a laboratory scale pH neutralization system. The results show that robust estimation of the titration curve can be achieved from continuous cycling. Furthermore, reliable and effective nonlinear pH control can be obtained in a realistic operating condition as shown in this experimental study.