Adaptive PID Control Strategy with Online Model Parameter Identification for Dynamic Systems: Design and Simulation

Proportional-Integral-Derivative (PID) control is widely employed in automation. However, dynamic systems often face changing conditions and uncertainties, necessitating adaptive control strategies. This paper addresses this need by introducing an innovative adaptive PID control strategy capable of real-time parameter adjustments to maintain system stability and performance amidst evolving dynamics. Our approach combines Recursive Least Squares with a Forgetting Factor (RLS-FF) for online recursive system identification and Internal Model Control (IMC) to determine PID controller gains. RLS-FF ensures accurate and up-to-date system parameter estimates, while IMC provides precise setpoint tracking and disturbance rejection. Key findings demonstrate that the adaptive PID controller, implemented as a Self-Tuning Regulator (STR), outperforms conventional PID and IMC-based controllers in various aspects. It exhibits superior setpoint tracking, reduced overshoot, faster settling times, and enhanced disturbance rejection. These improvements are empirically validated through benchmark testing in a refrigeration cycle. In conclusion, this paper presents a novel adaptive PID control strategy integrating online model parameter identification and IMC. The results reveal substantial performance enhancements over traditional control methods, positioning it as a promising solution for dynamic systems.