Skip to main content

Login for students

Login for employees

Publication detail

Piecewise Linear Neural Models for Process Control
Authors: Doležel Petr | Taufer Ivan | Mareš Jan
Year: 2011
Type of publication: ostatní - článek ve sborníku
Name of source: Proceedings XI Mizhnarodnaia nauko-tehknichna koferencia aspirantov i studentov Avtomatizacia technologicheskich obektoi ta procesiv. Pošuk molodikh
Publisher name: Donetsk national technical university
Place: Donetsk
Page from-to: 158-160
Titles:
Language Name Abstract Keywords
eng Piecewise Linear Neural Models for Process Control The aim of the contribution is to explain how to use neural network with piecewise-linear activation functions in hidden layer in process control. To be more specific, there is described technique of controlled plant linearization using neural nonlinear model. Obtained linearized model is in a shape of linear difference equation and can be used for any controller tuning. Nonlinear process control; Neural networks