Experimentally validated predictive fractional-order internal model control for time-delayed chemical processes and reactors

Abstract Fractional-order controllers are recognized for their enhanced tuning flexibility, while the Smith predictor (SP) is a widely used approach for compensating time delays in control systems. This study explores a hybrid approach that integrates the advantages of both fractional-order control and SP structure to address challenges in time-delay process control. In this study, a predictive modified fractional-order internal model control (MFOIMC) strategy is proposed, wherein the fractional-order controller C FOC ( s ) is designed based on maximum sensitivity and phase margin specifications to achieve desired robustness and performance. The proposed controller is tested in four benchmark case studies: a proton exchange membrane fuel cell (PEMFC), a bioreactor, a continuous stirred tank reactor (CSTR), and a cryogenic distillation process. Simulation results demonstrate the superior performance of the proposed MFOIMC approach in terms of tracking, disturbance rejection, and robustness, compared to existing methods. The effectiveness is quantitatively validated using error indices and the integral square of control input (ISU). Robust stability under parametric uncertainties is also verified. Practical viability is confirmed through real-time implementation on a two-tank level control setup, showcasing the industrial applicability of the proposed method.