Application of Neural-Network-Based Vibration Control to Single-Degree-of-Freedom System Structure with Dynamic Vibration Absorber

• HIRATSUKA Kiyotoshi

  Shingu Laboratory, Nihon University

• SHINGU Kiyoshi

  College of Science and Technology, Nihon University

抄録

Using the neural-network-based vibration control suggested by the authors, the control results differ for each learning rate that is to be considered in this paper. A single-degree-of-freedom (SDOF) system structure with a dynamic vibration absorber (DVA) with its damping ratio controlled using neural network algorithm. In actual situation, it is supposed that the neural network algorithm is operated on real time. In the simulation, the control is carried out at the same sampling time of the seismic waves. An optimum-learning rate of the neural network is estimated comparing to the relation between the maximum absolute value of the relative displacements and learning rates. Ten kinds of seismic waves are used as excitation in the simulations.

収録刊行物

• Theoretical and Applied Mechanics Japan

  Theoretical and Applied Mechanics Japan 51 (0), 133-140, 2002

  日本学術会議 「機械工学委員会・土木工学・建築学委員会合同IUTAM分科会」

詳細情報

• CRID: 1390001205209595904
• NII論文ID: 130004463473
• DOI: 10.11345/nctam.51.133
• ISSN: 13494244
• ISSN: 13480693
• 本文言語コード: en
• データソース種別:
  • JaLC
  • CiNii Articles
• 抄録ライセンスフラグ: 使用不可