Magnetic bearing synthesis for rotating machinery

Design of magnetic bearing suspensions for rotating machinery is investigated with special attention to the issues of modeling, actuator capacity requirements, and feedback controller design. A consistent approach to modeling the various components of a magnetic suspension system: rotor, amplifiers, actuators, sensors, and controller is developed based on the state space two-block form. This permits easy and automatic assembly of the various systems needed for this analysis. Actuator capacity requirements are determined with reference to anticipated loads, specific rotor performance requirements, and rotor dynamic character. The method developed is based on optimal open loop control strategies and generates a lower bound on the capacity which would ultimately be demanded by a feedback system. Controller design grows from the solution of the linear quadratic Gaussian regulator problem with extensions to provide load accommodation and to improve robustness of the closed loop systems. The resulting controller is an output feedback multivariable device, extracting its input from physically realizable position sensors and driving the rotor through physically realizable amplifiers and magnetic actuators. The controller is also a function of the shaft speed, reflecting the strong effect shaft speed has on the rotor model as well as the dominant forcing mechanisms.

Note: Abstract extracted from PDF file via OCR.