• Search Research Projects
  • Search Researchers
  • How to Use
  1. Back to previous page

Learning Networks with Structure Easy-to-Use

Research Project

Project/Area Number 13650491
Research Category

Grant-in-Aid for Scientific Research (C)

Allocation TypeSingle-year Grants
Section一般
Research Field Control engineering
Research InstitutionKYUSHU UNIVERSITY

Principal Investigator

FURUZUKI Takayuki  KYUSHU UNIVERSITY, Graduate School of Information Science and Electrical Engineering, Research Associate, 大学院・システム情報科学研究院, 助手 (50294905)

Co-Investigator(Kenkyū-buntansha) MURATA Junichi  KYUSHU UNIVERSITY, Graduate School of Information Science and Electrical Engineering, Associate Professor, 大学院・システム情報科学研究院, 助教授 (60190914)
HIRASAWA Kotaro  Waseda University, Graduate School of Information, Production and Systems, Professor, 大学院・情報生産システム研究科準備室, 教授 (70253474)
Project Period (FY) 2001 – 2002
Project Status Completed (Fiscal Year 2002)
Budget Amount *help
¥2,400,000 (Direct Cost: ¥2,400,000)
Fiscal Year 2002: ¥1,100,000 (Direct Cost: ¥1,100,000)
Fiscal Year 2001: ¥1,300,000 (Direct Cost: ¥1,300,000)
KeywordsNonlinear model / Neural networks / Neuro-fuzzy networks / Learning / Prior knowledge / Nonlinear control / Robust control / Fault diagnosis / 線形構造 / 線形制御技術
Research Abstract

Neural networks have recently attracted much interest in system control community because they learn any nonlinear mapping. However, from a user's point of view, neural networks are not user friendly, That is, they are not easy-to-use ; more specifically they do not have structures favorable to the applications of system control and fault diagnosis. To solve these problems, the following studies have been carried out.
1. A modeling scheme has been developed, which consists of two parts : a macro-net part and kernel-net part. The macro-net part is a user-friendly interface constructed using application specific knowledge and the nature of network structure. The kernel-net part is a flexible multi-input-multi-output (HIMO) nonlinear model such as neural networks and neurofuzzy networks.
2. An optimization scheme has been developed. The scheme consists of two learning loops. It has been studied to increase robustness of the algorithm to local minima and over-fitting, by using such as homotopy and hierarchical techniques.
3. Applications of the proposed modeling scheme to controller design and fault detection of nonlinear systems have been studied. Some new approaches are proposed and confirmed through numerical simulations.

Report

(3 results)
  • 2002 Annual Research Report   Final Research Report Summary
  • 2001 Annual Research Report
  • Research Products

    (43 results)

All Other

All Publications (43 results)

  • [Publications] K.Hirasawa: "A New Control Method of Nonlinear Systems Based on Impulse Responses of Universal Learning Networks"IEEE Trans on Systems, Man and Cybernetics, B. 31. 362-372 (2001)

    • Description
      「研究成果報告書概要(和文)」より
    • Related Report
      2002 Final Research Report Summary
  • [Publications] J.Hu: "A Homotopy Approach to Improving PEM Identification of ARMAX Models"Automatica. 37. 1323-1334 (2001)

    • Description
      「研究成果報告書概要(和文)」より
    • Related Report
      2002 Final Research Report Summary
  • [Publications] H.Chen: "A New Minimax Control Method for Nonlinear Systems Using Universal Learning Networks"電気学会論文誌. 121-C. 1471-1478 (2001)

    • Description
      「研究成果報告書概要(和文)」より
    • Related Report
      2002 Final Research Report Summary
  • [Publications] J.Hu: "A Quasi-ARMAX Approach to the Modeling of Nonlinear Systems"International Journal of Control. 74. 1754-1766 (2001)

    • Description
      「研究成果報告書概要(和文)」より
    • Related Report
      2002 Final Research Report Summary
  • [Publications] K.Hirasawa: "Improvement of Generalization Ability for Identifying Dynamical Systems by Using Universal Learning Networks"Neural Networks. 14. 1389-1404 (2001)

    • Description
      「研究成果報告書概要(和文)」より
    • Related Report
      2002 Final Research Report Summary
  • [Publications] J.Hu: "Overlapped Multi-Neural-Network and Its Training Algorithm"電気学会論文誌. 121-C. 1949-1956 (2001)

    • Description
      「研究成果報告書概要(和文)」より
    • Related Report
      2002 Final Research Report Summary
  • [Publications] S.M.Kyu: "Behavior Learning of Autonomous Robots by Modified Learning"計測自動制御学会論文集. 37. 1162-1168 (2001)

    • Description
      「研究成果報告書概要(和文)」より
    • Related Report
      2002 Final Research Report Summary
  • [Publications] J.Hu: "An Embedded Sigmoidal Neural Network for Modeling of Nonlinear Systems"Proceedings of IEEE International Joint Conference of Neural Networks. 3. 1698-1703 (2001)

    • Description
      「研究成果報告書概要(和文)」より
    • Related Report
      2002 Final Research Report Summary
  • [Publications] J.Hu: "A Hierarchical Method for Training Embedded Sigmoidal Neural Networks"Lecture Notes on Computer Science 2130. 937-942 (2001)

    • Description
      「研究成果報告書概要(和文)」より
    • Related Report
      2002 Final Research Report Summary
  • [Publications] 平澤 宏太郎: "一般化学習ネットワークのインパルス応答に基づく非線形制御方式"電気学会論文誌. 122-C. 105-115 (2002)

    • Description
      「研究成果報告書概要(和文)」より
    • Related Report
      2002 Final Research Report Summary
  • [Publications] W.Wan: "Enhancing the Generalization Ability of Backpropagation Algorithm through Controlling the Outputs of Hidden Layers"計測自動制御学会論文集. 38. 411-419 (2002)

    • Description
      「研究成果報告書概要(和文)」より
    • Related Report
      2002 Final Research Report Summary
  • [Publications] Q.Xiong: "Universal Learning Networks with Branch Control"電気学会論文誌. 122-C. 1812-1820 (2002)

    • Description
      「研究成果報告書概要(和文)」より
    • Related Report
      2002 Final Research Report Summary
  • [Publications] J.Hu: "A Quasi-ARX Model Incorporating Neural Networks for Control of Nonlinear Systems"Proceedings of the 16^<th> World Congress. (CDROM). (2002)

    • Description
      「研究成果報告書概要(和文)」より
    • Related Report
      2002 Final Research Report Summary
  • [Publications] J.Hu: "A Method for Applying Multilayer Perceptrons to Control Nonlinear Systems"Proceeding of 9^<th> International Conference on Neural Information Processing. 3. 1267-1271 (2002)

    • Description
      「研究成果報告書概要(和文)」より
    • Related Report
      2002 Final Research Report Summary
  • [Publications] 古月 敬之: "非線形システムの制御のためのニューラルネットワークモデル"計測自動制御学会論文集. 39(印刷中). (2003)

    • Description
      「研究成果報告書概要(和文)」より
    • Related Report
      2002 Final Research Report Summary
  • [Publications] D.Li: "Hybrid Universal Learning Networks"電気学会論文誌. 123-C(印刷中). (2003)

    • Description
      「研究成果報告書概要(和文)」より
    • Related Report
      2002 Final Research Report Summary
  • [Publications] K.Hirasawa: "A New Control Method of Nonlinear Systems Based on Impulse Responses of Universal Learning Networks"IEEE Trans on Systems, Man and Cybernetics, Part B. Vol.31, No.3. 362-372 (2001)

    • Description
      「研究成果報告書概要(欧文)」より
    • Related Report
      2002 Final Research Report Summary
  • [Publications] J.Hu: "A Homotopy Approach to Improving PEM Identification of ARMAX Models"Automatica. Vol.37, No.9. 1323-1334 (2001)

    • Description
      「研究成果報告書概要(欧文)」より
    • Related Report
      2002 Final Research Report Summary
  • [Publications] H.Chen: "A New Minimax Control Method for Nonlinear Systems Using Universal Learning Networks"Trans of IEE Japan. Vol.121-C, No.9. 1471-1478 (2001)

    • Description
      「研究成果報告書概要(欧文)」より
    • Related Report
      2002 Final Research Report Summary
  • [Publications] J.Hu: "A Quasi-ARMAX Approach to the Modeling of Nonlinear Systems"International Journal of Control. Vol.74, No.18. 1754-1766 (2001)

    • Description
      「研究成果報告書概要(欧文)」より
    • Related Report
      2002 Final Research Report Summary
  • [Publications] K.Hirasawa: "Improvement of Generalization Ability for Identifying Dynamical Systems by Using Universal Learning Networks"Neural Networks. Vol.14, No.10. 1389-1404 (2001)

    • Description
      「研究成果報告書概要(欧文)」より
    • Related Report
      2002 Final Research Report Summary
  • [Publications] J.Hu: "Overlapped Multi-Neural-Network and Its Training Algorithm"Trans of IEE Japan. Vol.121-C, No.12. 1949-1956 (2001)

    • Description
      「研究成果報告書概要(欧文)」より
    • Related Report
      2002 Final Research Report Summary
  • [Publications] S.M.Kyu: "Behavior Learning of Autonomous Robots by Modified Learning"Trans of the Society of Instrument and Control Engineering. Vol.37, No.12. 1162-1168 (2001)

    • Description
      「研究成果報告書概要(欧文)」より
    • Related Report
      2002 Final Research Report Summary
  • [Publications] J.Hu: "An Embedded Sigmoidal Neural Network for Modeling of Nonlinear Systems"Proceedings of IEEE International Joint Conference of Neural Networks. Vol.3. 1698-1703 (2001)

    • Description
      「研究成果報告書概要(欧文)」より
    • Related Report
      2002 Final Research Report Summary
  • [Publications] J.Hu: "A Hierarchical Method for Training Embedded Sigmoidal Neural Networks"Lecture Notes on Computer Science 2130 (Proc.of ICANN'01). 937-942 (2001)

    • Description
      「研究成果報告書概要(欧文)」より
    • Related Report
      2002 Final Research Report Summary
  • [Publications] W.Wan: "Enhancing the Generalization Ability of Backpropagation Algorithm through Controlling the Outputs of Hidden Layers"Trans of the Society of Instrument and Control Engineering. Vol.38, No.4. 411-419 (2002)

    • Description
      「研究成果報告書概要(欧文)」より
    • Related Report
      2002 Final Research Report Summary
  • [Publications] Q.Xiong: "Universal Learning Networks with Branch Control"Trans of IEE Japan. Vol.122-C, No.10. 1812-1820 (2001)

    • Description
      「研究成果報告書概要(欧文)」より
    • Related Report
      2002 Final Research Report Summary
  • [Publications] J.Hu: "A Quasi-ARX Model Incorporating Neural Networks for Control of Nonlinear Systems"Proceedings of the 16^<th> World Congress. (2002)

    • Description
      「研究成果報告書概要(欧文)」より
    • Related Report
      2002 Final Research Report Summary
  • [Publications] J.Hu: "A Method for Applying Multilayer Perceptrons to Control of Nonlinear Systems"Proceedings of 9^<th> International Conference on Neural Information Processing. Vol.3. 1267-1271 (2002)

    • Description
      「研究成果報告書概要(欧文)」より
    • Related Report
      2002 Final Research Report Summary
  • [Publications] J.Hu: "Neural Network Based Prediction Model for Control of Nonlinear Systems"Trans of the Society of Instrument and Control Engineering. Vol.39, No.2. (2003)

    • Description
      「研究成果報告書概要(欧文)」より
    • Related Report
      2002 Final Research Report Summary
  • [Publications] D.Li: "Hybrid Universal Learning Networks"Trans of IEE Japan. Vol.123-C, No.3. (2003)

    • Description
      「研究成果報告書概要(欧文)」より
    • Related Report
      2002 Final Research Report Summary
  • [Publications] 平澤 宏太郎: "一般化学習ネットワークのインパルス応答に基づく非線形制御方式"電気学会論文誌. 122・C・1. 105-115 (2002)

    • Related Report
      2002 Annual Research Report
  • [Publications] Weishui Wan: "Enhancing the Generalization Ability of Backpropagation Algorithm through Controlling the Outputs of the Hidden Layers"計測自動制御学会論文集. 38・4. 411-419 (2002)

    • Related Report
      2002 Annual Research Report
  • [Publications] Jinglu Hu: "A Quasi-ARX Model Incorporating Neural Networks for Control of Nonlinear Systems"Proceedings of the 16^<th> World Congress. (CDROM). (2002)

    • Related Report
      2002 Annual Research Report
  • [Publications] Qingyu Xiong: "Universal Learning Networks with Branch Control"電気学会論文誌. 122-C・10. 1812-1820 (2002)

    • Related Report
      2002 Annual Research Report
  • [Publications] Jinglu Hu: "A Method for Applying Multilayer Perceptrons to Control of Nonlinear Systems"Proceedings of 9^<th> International Conference on Neural Information Processing. 3. 1267-1271 (2002)

    • Related Report
      2002 Annual Research Report
  • [Publications] 古月 敬之: "非線形システムの制御のためのニューラルネットワーク予測モデル"計測自動制御学会論文集. 39・2(未定). (2003)

    • Related Report
      2002 Annual Research Report
  • [Publications] Kotaro Hirasawa: "A New Control Method of Nonlinear Systems Based on Impulse Responses of Universal Learning Networks"IEEE Trans on Systems, Man and Cybernetics, Part B. 31・3. 362-372 (2001)

    • Related Report
      2001 Annual Research Report
  • [Publications] Jinglu Hu: "A Homotopy Approach to Improving PEM Identification of ARMAX Model"Automatica. 37・9. 1323-1334 (2001)

    • Related Report
      2001 Annual Research Report
  • [Publications] Hongping Chen: "A New Minimax Control Method for Nonlinear Systems Using Universal Learning Networks"電子学会論文誌. 121-C・9. 1471-1478 (2001)

    • Related Report
      2001 Annual Research Report
  • [Publications] Jinglu Hu: "Overlapped Multi-Neural-Network and Its Training Algorithm"電気学会論文誌. 121-C・12. 1949-1956 (2001)

    • Related Report
      2001 Annual Research Report
  • [Publications] Jinglu Hu: "A Quasi-ARMAX Approach to the Modeling of Nonlinear Systems"International Journal of Control. 74・18. 1754-1766 (2001)

    • Related Report
      2001 Annual Research Report
  • [Publications] Shon Min Kyu: "Behavior Learning of Autonomous Robots by Modified Learning Vector Quantization"計測自動制御学会論文集. 37・12. 1162-1168 (2001)

    • Related Report
      2001 Annual Research Report

URL: 

Published: 2001-04-01   Modified: 2016-04-21  

Information User Guide FAQ News Terms of Use Attribution of KAKENHI

Powered by NII kakenhi