Design and statistical robustness analysis of FOPID, IOPID and SIMC PID controllers for the control of an input-output linearized plant model

This paper presents the design and robustness analysis of various controllers for the control of a non-linear industrial process with parametric uncertainties and the presence of external perturbations, which is linearized using an input-output linearization technique. The comparison includes integer order PID (proportional-integral-derivative), fractional PID and SIMC (simple internal model control) PID controllers. The robustness analysis uses a 23 factorial experimental design. The experiment has as its input factors the uncertainty in gains of the plant, the presence of random noise in the feedback loop and the existence of external perturbations. The outputs of the experiment measure the performance of each control system through the temporal input step response of the system and measure the control action using the mean value and the standard deviation of each signal. The results obtained show that the fractional order PID controller exhibits better performance in the presence of the analysed experimental factors, reflected in the control action signal, which indicates greater robustness and lower energy consumption.