Neural networks based active vibration control of flexible linkage mechanisms

Citation
Ym. Song et al., Neural networks based active vibration control of flexible linkage mechanisms, J MEC DESIG, 123(2), 2001, pp. 266-271
Citations number
18
Language
INGLESE
art.tipo
Article
Categorie Soggetti
Mechanical Engineering
Journal title
JOURNAL OF MECHANICAL DESIGN
ISSN journal
1050-0472 → ACNP
Volume
123
Issue
2
Year of publication
2001
Pages
266 - 271
Database
ISI
SICI code
1050-0472(200106)123:2<266:NNBAVC>2.0.ZU;2-W
Abstract
An investigation is presented into the neural networks based active vibrati on control of flexible linkage mechanisms. A smart mechanism featuring piez oceramic actuators and strain gauge sensors is designed. A nonlinear adapti ve control strategy named Neural Networks based Direct Self-Tuning Control (NNBDSC) is employed to suppress the elastodynamic responses of the smart m echanism. To improve the initial robustness of the NNBDSC, the Dynamic Recu rrent Neural Network (DRNN) controllers are designed off-line to approximat e the inverse dynamics of the smart mechanism. Through on-line control, the strain crest of the flexible link is reduced 60 percent or so and the dyna mic performance of the smart mechanism is improved significantly.