Hysteresis Modeling and Compensation of Fast Steering Mirrors with Hysteresis Operator Based Back Propagation Neural...

Area of Science:

• Optics and Photonics
• Control Systems Engineering
• Materials Science

Background:

• Fast steering mirrors (FSMs) are critical for high-precision beam control.
• Piezoelectric ceramics drive FSMs, but exhibit complex hysteresis nonlinearity.
• Existing models struggle to capture the intricate hysteresis behavior of FSMs.

Purpose of the Study:

• To propose a systematic method for modeling the hysteresis nonlinearity of FSMs.
• To develop an inverse hysteresis model to mitigate FSM nonlinearity.
• To validate the proposed modeling approach with experimental data.

Main Methods:

• A Madelung's rules based symmetric hysteresis operator was employed.
• A cascaded neural network was utilized to modify the basic hysteresis motion.
• Wiping-out and congruency properties of the model were analyzed.

Main Results:

• The proposed model accurately describes the complex hysteresis characteristics of FSMs.
• The inverse hysteresis model effectively reduces the observed nonlinearity.
• Experimental validation confirmed the model's effectiveness.

Conclusions:

• The developed modeling approach provides a robust solution for FSM hysteresis.
• The inverse model offers a practical method for improving beam control precision.
• This work advances the understanding and control of piezoelectric actuators in optical systems.