Characteristic model based all-coefficient adaptive control of an AMB suspended energy storage flywheel test rig
Feedback control of active magnetic bearing (AMB) suspended energy storage flywheel
systems is critical in the operation of the systems and has been well studied. Both the
classical proportional-integral-derivative (PID) control design method and modern control
theory, such as H∞ control and μ-synthesis, have been explored. PID control is easy to
implement but is not effective in handling complex rotordynamics. Modern control design
methods usually require a plant model and an accurate characterization of the uncertainties …
systems is critical in the operation of the systems and has been well studied. Both the
classical proportional-integral-derivative (PID) control design method and modern control
theory, such as H∞ control and μ-synthesis, have been explored. PID control is easy to
implement but is not effective in handling complex rotordynamics. Modern control design
methods usually require a plant model and an accurate characterization of the uncertainties …
Abstract
Feedback control of active magnetic bearing (AMB) suspended energy storage flywheel systems is critical in the operation of the systems and has been well studied. Both the classical proportional-integral-derivative (PID) control design method and modern control theory, such as H∞ control and μ-synthesis, have been explored. PID control is easy to implement but is not effective in handling complex rotordynamics. Modern control design methods usually require a plant model and an accurate characterization of the uncertainties. In each case, few experimental validation results on the closed-loop performance are available because of the costs and the technical challenges associated with the construction of experimental test rigs. In this paper, we apply the characteristic model based all-coefficient adaptive control (ACAC) design method for the stabilization of an AMB suspended flywheel test rig we recently constructed. Both simulation and experimental results demonstrate strong closed-loop performance in spite of the simplicity of the control design and implementation.
Springer