DSP-based direct neural control for thrust active magnetic bearing learned through adaptive differential evolution

Syuan-Yi Chen, Min Han Song

Research output: Chapter in Book/Report/Conference proceedingConference contribution

2 Citations (Scopus)

Abstract

A digital signal processor (DSP)-based direct recurrent wavelet neural network (RWNN) controller is proposed to control the rotor position of a thrust active magnetic bearing (TAMB) system learned through adaptive differential evolution (ADE). First, the dynamic analysis of the TAMB with differential driving mode (DDM) is derived. Subsequently, due to the exact dynamic model of TAMB system is absent; a RWNN is adopted to deal with the highly nonlinear TAMB system for the tracking of reference trajectory. Moreover, Due to the gradient descent method is used in back propagation (BP) to derive the on-line learning algorithm for the RWNN; it may reach the local optimal solution due to the inappropriate initial values. Therefore, an ADE algorithm is adopted to optimize the initial network parameters including connective weights, translations and dilations for the RWNN controller. Finally, a DSP with PowerPC 440 processor and real time VxWorks OS is used for implementing the RWNN-ADE controller for TAMB system. Experimental results show the high-accuracy control performance of the proposed RWNN-ADE controlled TAMB system.

Original languageEnglish
Title of host publicationCACS 2015 - 2015 CACS International Automatic Control Conference
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages96-101
Number of pages6
ISBN (Electronic)9781467365734
DOIs
Publication statusPublished - 2016 Jan 11
Event9th International Automatic Control Conference, CACS 2015 - Yilan, Taiwan
Duration: 2015 Nov 182015 Nov 20

Publication series

NameCACS 2015 - 2015 CACS International Automatic Control Conference

Other

Other9th International Automatic Control Conference, CACS 2015
CountryTaiwan
CityYilan
Period15/11/1815/11/20

Fingerprint

Magnetic bearings
Digital signal processors
Neural networks
Controllers
Backpropagation
Dynamic analysis
Learning algorithms
Dynamic models
Rotors
Trajectories

ASJC Scopus subject areas

  • Control and Systems Engineering
  • Artificial Intelligence

Cite this

Chen, S-Y., & Song, M. H. (2016). DSP-based direct neural control for thrust active magnetic bearing learned through adaptive differential evolution. In CACS 2015 - 2015 CACS International Automatic Control Conference (pp. 96-101). [7378372] (CACS 2015 - 2015 CACS International Automatic Control Conference). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/CACS.2015.7378372

DSP-based direct neural control for thrust active magnetic bearing learned through adaptive differential evolution. / Chen, Syuan-Yi; Song, Min Han.

CACS 2015 - 2015 CACS International Automatic Control Conference. Institute of Electrical and Electronics Engineers Inc., 2016. p. 96-101 7378372 (CACS 2015 - 2015 CACS International Automatic Control Conference).

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Chen, S-Y & Song, MH 2016, DSP-based direct neural control for thrust active magnetic bearing learned through adaptive differential evolution. in CACS 2015 - 2015 CACS International Automatic Control Conference., 7378372, CACS 2015 - 2015 CACS International Automatic Control Conference, Institute of Electrical and Electronics Engineers Inc., pp. 96-101, 9th International Automatic Control Conference, CACS 2015, Yilan, Taiwan, 15/11/18. https://doi.org/10.1109/CACS.2015.7378372
Chen S-Y, Song MH. DSP-based direct neural control for thrust active magnetic bearing learned through adaptive differential evolution. In CACS 2015 - 2015 CACS International Automatic Control Conference. Institute of Electrical and Electronics Engineers Inc. 2016. p. 96-101. 7378372. (CACS 2015 - 2015 CACS International Automatic Control Conference). https://doi.org/10.1109/CACS.2015.7378372
Chen, Syuan-Yi ; Song, Min Han. / DSP-based direct neural control for thrust active magnetic bearing learned through adaptive differential evolution. CACS 2015 - 2015 CACS International Automatic Control Conference. Institute of Electrical and Electronics Engineers Inc., 2016. pp. 96-101 (CACS 2015 - 2015 CACS International Automatic Control Conference).
@inproceedings{6401379e64454507baaea19141467cf1,
title = "DSP-based direct neural control for thrust active magnetic bearing learned through adaptive differential evolution",
abstract = "A digital signal processor (DSP)-based direct recurrent wavelet neural network (RWNN) controller is proposed to control the rotor position of a thrust active magnetic bearing (TAMB) system learned through adaptive differential evolution (ADE). First, the dynamic analysis of the TAMB with differential driving mode (DDM) is derived. Subsequently, due to the exact dynamic model of TAMB system is absent; a RWNN is adopted to deal with the highly nonlinear TAMB system for the tracking of reference trajectory. Moreover, Due to the gradient descent method is used in back propagation (BP) to derive the on-line learning algorithm for the RWNN; it may reach the local optimal solution due to the inappropriate initial values. Therefore, an ADE algorithm is adopted to optimize the initial network parameters including connective weights, translations and dilations for the RWNN controller. Finally, a DSP with PowerPC 440 processor and real time VxWorks OS is used for implementing the RWNN-ADE controller for TAMB system. Experimental results show the high-accuracy control performance of the proposed RWNN-ADE controlled TAMB system.",
author = "Syuan-Yi Chen and Song, {Min Han}",
year = "2016",
month = "1",
day = "11",
doi = "10.1109/CACS.2015.7378372",
language = "English",
series = "CACS 2015 - 2015 CACS International Automatic Control Conference",
publisher = "Institute of Electrical and Electronics Engineers Inc.",
pages = "96--101",
booktitle = "CACS 2015 - 2015 CACS International Automatic Control Conference",

}

TY - GEN

T1 - DSP-based direct neural control for thrust active magnetic bearing learned through adaptive differential evolution

AU - Chen, Syuan-Yi

AU - Song, Min Han

PY - 2016/1/11

Y1 - 2016/1/11

N2 - A digital signal processor (DSP)-based direct recurrent wavelet neural network (RWNN) controller is proposed to control the rotor position of a thrust active magnetic bearing (TAMB) system learned through adaptive differential evolution (ADE). First, the dynamic analysis of the TAMB with differential driving mode (DDM) is derived. Subsequently, due to the exact dynamic model of TAMB system is absent; a RWNN is adopted to deal with the highly nonlinear TAMB system for the tracking of reference trajectory. Moreover, Due to the gradient descent method is used in back propagation (BP) to derive the on-line learning algorithm for the RWNN; it may reach the local optimal solution due to the inappropriate initial values. Therefore, an ADE algorithm is adopted to optimize the initial network parameters including connective weights, translations and dilations for the RWNN controller. Finally, a DSP with PowerPC 440 processor and real time VxWorks OS is used for implementing the RWNN-ADE controller for TAMB system. Experimental results show the high-accuracy control performance of the proposed RWNN-ADE controlled TAMB system.

AB - A digital signal processor (DSP)-based direct recurrent wavelet neural network (RWNN) controller is proposed to control the rotor position of a thrust active magnetic bearing (TAMB) system learned through adaptive differential evolution (ADE). First, the dynamic analysis of the TAMB with differential driving mode (DDM) is derived. Subsequently, due to the exact dynamic model of TAMB system is absent; a RWNN is adopted to deal with the highly nonlinear TAMB system for the tracking of reference trajectory. Moreover, Due to the gradient descent method is used in back propagation (BP) to derive the on-line learning algorithm for the RWNN; it may reach the local optimal solution due to the inappropriate initial values. Therefore, an ADE algorithm is adopted to optimize the initial network parameters including connective weights, translations and dilations for the RWNN controller. Finally, a DSP with PowerPC 440 processor and real time VxWorks OS is used for implementing the RWNN-ADE controller for TAMB system. Experimental results show the high-accuracy control performance of the proposed RWNN-ADE controlled TAMB system.

UR - http://www.scopus.com/inward/record.url?scp=84965011034&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84965011034&partnerID=8YFLogxK

U2 - 10.1109/CACS.2015.7378372

DO - 10.1109/CACS.2015.7378372

M3 - Conference contribution

AN - SCOPUS:84965011034

T3 - CACS 2015 - 2015 CACS International Automatic Control Conference

SP - 96

EP - 101

BT - CACS 2015 - 2015 CACS International Automatic Control Conference

PB - Institute of Electrical and Electronics Engineers Inc.

ER -