Sliding mode control with neural network for active magnetic bearing system

Zhi Cao, Jianning Dong, Faisal Wani, Henk Polinder, Pavol Bauer, Fei Peng, Yunkai Huang

Research output: Chapter in Book/Conference proceedings/Edited volumeConference contributionScientificpeer-review

6 Citations (Scopus)
92 Downloads (Pure)

Abstract

A novel controller design procedure is proposed for a 5-degree-of-freedom (DOF) active magnetic bearing (AMB) system, based on sliding mode control (SMC) and neural network (NN). The SMC is used to achieve high robustness and fast response while the NN can compensate unmodeled uncertainty and external disturbance by on-line tuning algorithm. The proposed controller is compared with the well-tuned PID controller by simulations. The simulation results show the superior performance of the proposed controller.
Original languageEnglish
Title of host publicationProceedings
Subtitle of host publicationIECON 2019 - 45th Annual Conference of the IEEE Industrial Electronics Society
Place of PublicationPiscataway, NJ, USA
PublisherIEEE
Pages744-749
Number of pages6
ISBN (Electronic)9781728148786
ISBN (Print)978-1-7281-4878-6
DOIs
Publication statusPublished - 2019
EventIECON 2019: IEEE 45th Annual Conference of the Industrial Electronics Society - Lisbon, Portugal
Duration: 14 Oct 201917 Oct 2019
Conference number: 45th

Publication series

NameIECON Proceedings (Industrial Electronics Conference)
Volume2019-October

Conference

ConferenceIECON 2019
Country/TerritoryPortugal
CityLisbon
Period14/10/1917/10/19

Bibliographical note

Green Open Access added to TU Delft Institutional Repository ‘You share, we take care!’ – Taverne project https://www.openaccess.nl/en/you-share-we-take-care  Otherwise as indicated in the copyright section: the publisher is the copyright holder of this work and the author uses the Dutch legislation to make this work public.

Keywords

  • magnetic bearing
  • neural network
  • sliding mode control

Fingerprint

Dive into the research topics of 'Sliding mode control with neural network for active magnetic bearing system'. Together they form a unique fingerprint.

Cite this