The learning process of a multilayered feedforward neural network involves extracting a desired function from the training data presented through an appropriate training algorithm. To achieve the desired function, the generation of good training data is necessary. A closed-loop methodology for neural network training for control of drives with nonlinearities is presented. Problems associated with the more common open-loop training scheme, and how these are addressed by the proposed closed-loop method, are discussed. An inverse nonlinear control using a neural network (INC/NN), a control strategy which incorporates the neural network for control of nonlinear systems, is described and used to demonstrate the effectiveness of the closed-loop training scheme. Simulation studies and experimental results are presented to verify the improvement achieved by the closed-loop training methodology.<>
Description
A methodology for neural network training for control of drives with nonlinearities | IEEE Journals & Magazine | IEEE Xplore
%0 Journal Article
%1 222646
%A Low, T.-S.
%A Lee, T.-H.
%A Lim, H.-K.
%D 1993
%J IEEE Transactions on Industrial Electronics
%K control neuralnetwork nonlinearity todo:read
%N 2
%P 243-249
%R 10.1109/41.222646
%T A methodology for neural network training for control of drives with nonlinearities
%U https://ieeexplore.ieee.org/document/222646
%V 40
%X The learning process of a multilayered feedforward neural network involves extracting a desired function from the training data presented through an appropriate training algorithm. To achieve the desired function, the generation of good training data is necessary. A closed-loop methodology for neural network training for control of drives with nonlinearities is presented. Problems associated with the more common open-loop training scheme, and how these are addressed by the proposed closed-loop method, are discussed. An inverse nonlinear control using a neural network (INC/NN), a control strategy which incorporates the neural network for control of nonlinear systems, is described and used to demonstrate the effectiveness of the closed-loop training scheme. Simulation studies and experimental results are presented to verify the improvement achieved by the closed-loop training methodology.<>
@article{222646,
abstract = {The learning process of a multilayered feedforward neural network involves extracting a desired function from the training data presented through an appropriate training algorithm. To achieve the desired function, the generation of good training data is necessary. A closed-loop methodology for neural network training for control of drives with nonlinearities is presented. Problems associated with the more common open-loop training scheme, and how these are addressed by the proposed closed-loop method, are discussed. An inverse nonlinear control using a neural network (INC/NN), a control strategy which incorporates the neural network for control of nonlinear systems, is described and used to demonstrate the effectiveness of the closed-loop training scheme. Simulation studies and experimental results are presented to verify the improvement achieved by the closed-loop training methodology.<>},
added-at = {2021-05-05T17:32:23.000+0200},
author = {Low, T.-S. and Lee, T.-H. and Lim, H.-K.},
biburl = {https://www.bibsonomy.org/bibtex/24e55e98ab99121518e1c17c63a6cf63c/annakrause},
description = {A methodology for neural network training for control of drives with nonlinearities | IEEE Journals & Magazine | IEEE Xplore},
doi = {10.1109/41.222646},
interhash = {fbc53fc3d234a78812c2f28ea79ac560},
intrahash = {4e55e98ab99121518e1c17c63a6cf63c},
issn = {1557-9948},
journal = {IEEE Transactions on Industrial Electronics},
keywords = {control neuralnetwork nonlinearity todo:read},
month = {April},
number = 2,
pages = {243-249},
timestamp = {2021-05-05T17:32:23.000+0200},
title = {A methodology for neural network training for control of drives with nonlinearities},
url = {https://ieeexplore.ieee.org/document/222646},
volume = 40,
year = 1993
}