Neuro modelling and vibration control of flexible rectangular plate structure
The demand for soft computing techniques in the modeling and control of dynamic system has increased in recent years especially for flexible structures. Flexible plate structures are extensively used in many space applications, however this type of structure leads to high vibration problems. The aim...
Saved in:
| Main Author: | |
|---|---|
| Format: | Thesis |
| Language: | English |
| Published: |
2006
|
| Subjects: | |
| Online Access: | http://eprints.utm.my/id/eprint/35156/1/RainahBintiIsmailMFKM2006.pdf |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Summary: | The demand for soft computing techniques in the modeling and control of dynamic system has increased in recent years especially for flexible structures. Flexible plate structures are extensively used in many space applications, however this type of structure leads to high vibration problems. The aim of this investigation is to modelling and control of two dimensional flexible plate structures. This will involve an identification system including least squares, recursive least squares, and neural networks within an active vibration control framework. A thin rectangular plates with all edges clamped is considered. A simulation algorithm characterising the dynamic behaviour of the plate is developed through a discretisation of the governing partial differential equation formulation of the plate dynamics using finite difference methods. The simulation algorithm thus developed and validated forms a suitable test and verification platform in subsequent investigations for development of vibration control techniques for flexible plate structures. The design and analysis of an active vibration control (AVC) system utilizing conventional and soft computing methods with single-input single-output AVC structure is presented to suppressing the vibration of the flexible plate structures. Finally a comparative performance of the algorithm in implementing AVC system using recursive least square (RLS), Multilayer perceptron neural networks (MLP-NN) and Elman Neural networks (ENN) is presented and discussed. |
|---|