Modified Smith predictor-based fuzzy self-tuning FOPID controller for mean arterial pressure regulation

Abstract

Abstract The simultaneous regulation of hemodynamic variables, particularly mean arterial pressure (MAP) and cardiac output (CO), is challenging due to time-varying delays, cross-coupling effects, and parameter uncertainty inherent in drug–response dynamics. A robust control framework is developed that integrates a fractional-order PID (FOPID) controller with a fuzzy logic self-tuning mechanism (FLST) and a modified Smith predictor (MSP) to enable adaptive gain scheduling and explicit compensation of bounded delays. The patient is modeled as a two-by-two first-order time-delay MIMO system actuated by sodium nitroprusside (SNP) and dopamine (DPM). Besides, transport and recirculation delays are assumed variable and conservatively bound at 60 s. While the inclusion of fractional orders provides additional degrees of freedom, resulting in improved transient performance and enhanced steady-state accuracy, the MSP maintains stability and tracking performance in the presence of time-varying delays. Extensive simulation studies under nominal and uncertain conditions demonstrate faster rise and settling times, reduced overshoot, and lower integral error indices than those of conventional PID and fractional-order PID controllers. Robustness to inter-patient variability in the presence of disturbances and noise measurement, and clinically acceptable performance, indicating feasibility for real-time closed-loop drug infusion. Overall, the proposed FLST–FOPID–MSP architecture offers a promising and safety-oriented solution for automated MAP and CO regulation and motivates future comparisons with predictive and -based control strategies.