Wyniki wyszukiwania - BazTech

Ograniczanie wyników

Znaleziono wyników: 2

Liczba wyników na stronie

Wyniki wyszukiwania

Wyszukiwano:
w słowach kluczowych: Runge-Kutta neural network

Sortuj według:

Ogranicz wyniki do:

Predictive feedback approach to structural vibration suppresion

Leniowski R., Leniowska L.

Archives of Control Sciences

2006

Vol. 16, no. 1

37-50

The problem of active vibration control of a plate has been vastly researched and described in recent years. Theoretical and experimental results demonstrate the effectiveness of the designed controllers and indicate the potential of control techniques for reducing transient and steady state dynamics in structural acoustic systems. The examples from the computational studies, confirmed that vibration levels could be effectively reduced, however, the implementation procedures are not yet ideal, still exists the gap between experimental and simulations findings. To overcome this problem, autors propose extension for the Fuzzy-PID controller, with an on-line identification technique coupled with a control scheme, for a plate vibrationsupression. It is assumed, that the system to be regulated is unknown, the control schemes presented in this work have the ability to identify and suppress a plate vibrations with only an initial estimate of the system order. A prediction method implemented was designed using a neutral network (NN) identification algorithm, based on the well-known Runge-Kutta methods. This algorithm is similar to described by Wang and Lin [14], but it uses a copmutation structure of Runge-Kutta-3/8. with radial cosine basis neural network.

Application of Prediction Methods for Plate Vibration Suppression

Leniowska L., Leniowski R.

Molecular and Quantum Acoustics

2004

Vol. 25

159-175

In this paper, predictive feedback control is used to suppress circular plate vibrations. It is assumed that the system to be regulated is unknown. The plate is excited by a uniform force over the bottom surface generated by a loudspeaker. The axially-symmetrical vibrations of the plate are measured by the application of the strain sensors located along the plate radius and two centrally placed piezoceramic discs are used to cancel the plate vibrations. The control schemes presented in this work have the ability to predict the error sensor signals, to compute the control effort and to apply it to the actuator within one sampling period. For precise estimation of system behaviour the modified Runge-Kutta-3/8 neural network algorithm has been applied and tested. This control scheme is then illustrated through some numerical examples with simulations modeling the fuzzy predictive P-PI-PD controller and the improvement gained by incorporating a feed-forward path into the controller is demonstrated.