Self-Tuning Neuro-PID Controller for Piezoelectric Actuator

The purpose of this study was to design a tracking controller for micro-piezoelectric motion platform applications. The hysteresis effect is an inherent element of piezoelectric actuated platforms often leading to nonlinearity in the behavior of the system. We constructed a Prandtl-Ishlinskii model to describe the hysteresis behavior of piezoelectric actuators, and derived the weights of the model using the LMS (Least-Mean-Square) algorithm. Based on the Prandtl-Ishlinskii model, we developed a feed-forward controller to compensate for hysteresis nonlinearity, and implemented a self-tuning neuro-PID controller to suppress tracking error due to modeling inaccuracy. The efficacy of this approach was numerically and experimentally verified, demonstrating the performance and applicability of the proposed designs under a variety of operating conditions.