A Theory for Multiresolution Signal Decomposition: The Wavelet Representation
IEEE Transactions on Pattern Analysis and Machine Intelligence
Hi-index | 0.00 |
In this investigation, a vector controlled induction motor drive is simulated and the feedback signals of this vector controlled drive are estimated using neural networks. The neural networks receive the machine terminal signals as inputs and estimate the rotor flux and unit vectors cos*** e and sin*** e as outputs. These outputs are used in the vector controlled drive system. The calculated feedback signals by the neural networks are not sensitive to the motor parameter variations. In this paper, three types of neural networks (i.e. multilayer perceptron (MLP), radial basis function (RBF) and wavenet) are used and the obtained results are compared. Finally, on the basis of the advantages of wavenets, the results prove the accuracy and effectiveness of the wavenet based estimator.