Skip to main navigation Skip to search Skip to main content

Magnetically suspend flywheel rotor speed nonlinear control

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

Abstract

In this paper, a nonlinear speed control scheme which combines nonlinear optimal controller, nonlinear disturbance observer and adaptive kalman filter is proposed for Magnetically Suspend Flywheel (MSFW). Firstly, in order to improve the dynamic and steady-stage performances of motor driving system, the state feedback exact linearization model of BUCK converter is derived, and a nonlinear optimal controller is proposed. Considering that the runout of the suspended rotor of MSFW would induce periodic disturbance to the system, a nonlinear disturbance observer is designed to suppress the sinusoidally and synchronously disturbance. While the general disturbance that exist in the system such as circuit noise can be rejected by the improved Kalman filter which adds a fading factor to prediction covariance matrix. Finally, the validity of the proposed method has been proved by simulation and experiment.

Original languageEnglish
Title of host publication2014 IEEE International Conference on Robotics and Biomimetics, IEEE ROBIO 2014
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages2255-2261
Number of pages7
ISBN (Electronic)9781479973965
DOIs
StatePublished - 20 Apr 2014
Event2014 IEEE International Conference on Robotics and Biomimetics, IEEE ROBIO 2014 - Bali, Indonesia
Duration: 5 Dec 201410 Dec 2014

Publication series

Name2014 IEEE International Conference on Robotics and Biomimetics, IEEE ROBIO 2014

Conference

Conference2014 IEEE International Conference on Robotics and Biomimetics, IEEE ROBIO 2014
Country/TerritoryIndonesia
CityBali
Period5/12/1410/12/14

Keywords

  • converters
  • flywheel
  • optimal control
  • speed control
  • vibration

Fingerprint

Dive into the research topics of 'Magnetically suspend flywheel rotor speed nonlinear control'. Together they form a unique fingerprint.

Cite this