Neural network based adaptive internal model control for nonlinear plants

• Autorzy: Ma Z. , Bajić V. B. , Ho D. W. C.
• Języki publikacji: EN

Abstrakty

EN

A novel non-parametric adaptive control method for nonlinear plants is proposed. It combines neural network (NN) based identification and internal model control (IMC) strategy. The NN is used to determine on-line an approximation of the unknown nonlinear process model. The NN parameters are updated according to the error between the plant output and the NN output. The NN can track the system output very well, so that an adaptive IMC can be implemented successfully. The design does not require computation of the inverse of the internal model of the process. Instead, it uses only system input-output data and NN output. The effectiveness of the proposed method is illustrated by a simulation experiment.

Słowa kluczowe

EN

neural networks; nonlinear plants; internal model control; adaptive control

PL

sieci neuronowe; układy nieliniowe; sterowanie wewnętrzne; sterowanie adaptacyjne

• Wydawca: Wydawnictwo Politechniki Łódzkiej
• Czasopismo: Journal of Applied Computer Science
• Rocznik: 1999
• Tom: Vol. 7, nr 1
• Strony: 63--78
• Opis fizyczny: Bibiogr. 26 poz.

Twórcy

autor

Ma Z.

autor

Bajić V. B.

autor

Ho D. W. C.

• Centre for Engineering Research, Technikon Natal, Durban 4000, P.O. Box 953, South Africa