A method merging the features of variable structure control and neural network design is presented for speed control of a permanent magnet synchronous motor. The proposed control approach is based on a discrete-time variable structure control and a robust digital differentiator for speed estimation. Radial basis function neural networks are used to learn about uncertainties affecting the system. A stability analysis is provided and the ultimate boundedness of the speed tracking error is proved. Control performance has been evaluated by simulations using the model of a commercial permanent magnet synchronous motor drive.
A discrete-time VS controller based on RBF neural networks for PMSM drives / Ciabattoni, Lucio; Corradini, Maria Letizia; Grisostomi, Massimo; Ippoliti, Gianluca; Longhi, Sauro; Orlando, Giuseppe. - In: ASIAN JOURNAL OF CONTROL. - ISSN 1561-8625. - 16:2(2014), pp. 396-408. [10.1002/asjc.715]
A discrete-time VS controller based on RBF neural networks for PMSM drives
CIABATTONI, LUCIO;GRISOSTOMI, Massimo;IPPOLITI, Gianluca;LONGHI, SAURO;ORLANDO, Giuseppe
2014-01-01
Abstract
A method merging the features of variable structure control and neural network design is presented for speed control of a permanent magnet synchronous motor. The proposed control approach is based on a discrete-time variable structure control and a robust digital differentiator for speed estimation. Radial basis function neural networks are used to learn about uncertainties affecting the system. A stability analysis is provided and the ultimate boundedness of the speed tracking error is proved. Control performance has been evaluated by simulations using the model of a commercial permanent magnet synchronous motor drive.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
