A tuned approach of the predictive–adaptive GPC controller applied to a fed-batch bioreactor using complete factorial design

Abstract

This work presents a novel tuned approach of the Generalized Predictive Control controller in both adaptive and nonadaptive configurations applied to a fed-batch penicillin process using the complete factorial design method. The controller stabilizes the dissolved oxygen concentration through agitation manipulation. The operating process variables were calculated by a mathematical model solved numerically. This new approach employed complete factorial design methodology to provide the optimal set of parameters, estimating the design parameters influence on the integral of the absolute error between the controlled variable and the set point (IAE). The controller parameters analyzed were prediction and control horizons, suppression factor, reference trajectory and integral factor. Moreover, the robustness performance was evaluated using the white noise presence in the measured variable, different polynomial orders of the internal model, several delay time and sampling periods of the controlled variable, and the presence of the internal polynomial. This controller performed better than PID and DMC controllers