Neural-network-based identification and control of a thin plate using piezoelectric actuators and sensors (Q4828709)
From MaRDI portal
!
This is the item page for this Wikibase entity, intended for internal use and editing purposes. Please use the normal view instead:
scientific article; zbMATH DE number 2119309
| Language | Label | Description | Also known as |
|---|---|---|---|
| default for all languages | No label defined |
||
| English | Neural-network-based identification and control of a thin plate using piezoelectric actuators and sensors |
scientific article; zbMATH DE number 2119309 |
Statements
Neural-network-based identification and control of a thin plate using piezoelectric actuators and sensors (English)
0 references
26 November 2004
0 references
piezoelectric identifications
0 references
control plate
0 references
thin plate
0 references
response
0 references
stability
0 references
Riccati matrix inequalities
0 references
neural network
0 references
0 references
0.7927504777908325
0 references
0.7359689474105835
0 references
0.7359687089920044
0 references
0.7269752621650696
0 references
0.7259871959686279
0 references
The authors consider a vibrating rectangular thin plate obeying the classical model of thin plate theory, but with inclusion of the piezo-electric surface patches: \(\rho h\partial^2/ \partial t^2(w(R,t))+DAw (R,t)=\mu \sum_{i=1}^d V_{ai} (t) \nabla^2(R,R_{ai})\). Here \(\rho\) is density of the plate, \(h\) its thickness, \(R=\{x,y\}\) denotes the position on the surface of the plate, \(R_{ai}\) the position of the \(i\)-th actuator, \(D\) is the plate stiffness, \(A\) is the self-adjoint operator: \(A=\partial^4/ \partial x^4+2\gamma^2 \partial^4/\partial x^2\partial y^2+ \gamma^4 \partial^4/ \partial y^4\), and \(\gamma\) is the length to width ratio of the plate. The damping ratio for this system was assumed to be \(0.5\%\). The authors investigate the response of these piezoelectric patches attached to the surface of the plate. They construct a Lyapunov matrix checking the stability of a system of Riccati matrix inequalities. A neural network is suggested for the purpose of system identification. Extensive simulations were conducted, producing about 10 pages of small size, densely printed responses of the system.
0 references