Some aspects of neural network state variable estimator improvement in induction motor drive

Abstract

Some aspects of state variable estimator improvement is proposed in the paper. The estimator approximates stator current components in the rotor flux reference frame with the help of neural networks. Some modification of the training procedure is considered that leads to the estimator accuracy improvement. Provided tests confirmed this feature but further steps are necessary to increase state variables estimation in the low supplying frequency range. © 2013 Springer-Verlag.