Modeling of a magnetorheological damper by recursive lazy learning Articles
Overview
published in
publication date
- April 2011
start page
- 479
end page
- 485
issue
- 3
volume
- 46
Digital Object Identifier (DOI)
International Standard Serial Number (ISSN)
- 0020-7462
Electronic International Standard Serial Number (EISSN)
- 1878-5638
abstract
- Nowadays dampers based on magnetorheological (MR) fluids are receiving significant attention specially for control of structural vibration and automotive suspensions systems. In most cases, it is necessary to develop an appropriate control strategy which is practically implementable when a suitable model for MR dampers is available. It is not a trivial task to model the dynamic of MR dampers because of their inherent non-linear and hysteretic dynamics. In this paper, a recursive lazy learning method based on neural networks is considered to model the MR damper behavior. The proposed method is validated by comparison with experimental obtained responses. Results show the estimated model correlates very well with the data obtained experimentally. The method proposed learns quickly that it is only necessarily a learning cycle, it can learn on-line and it is easy to select the network structure and calculate the model parameters.
Classification
keywords
- magnetorheological damper; recursive lazy learning; neural network; modeling