Identification and modeling for non-linear dynamic system using neural networks type MLP

Hernán González Acuña, Max Suell Dutra, Omar Lengerke

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

Abstract

In control systems, the model dynamics of linear systems is the principal and most important phase of a project, but when working with dynamic of non-linear systems obtain the model becomes a very complex task can be used techniques of system identification. This article show the use of one technique for identification and modeling of dynamic linear systems and non-linear systems using dynamics neural networks type multilayer perceptron, obtaining approximate results in the identification of non-linear system.

Original languageEnglish
Title of host publicationProceedings of the 2009 Euro American Conference on Telematics and Information Systems
Subtitle of host publicationNew Opportunities to Increase Digital Citizenship, EATIS '09
DOIs
StatePublished - 2009
Externally publishedYes
Event2009 Euro American Conference on Telematics and Information Systems: New Opportunities to Increase Digital Citizenship, EATIS '09 - Prague, Czech Republic
Duration: 3 Jun 20095 Jun 2009

Publication series

NameProceedings of the 2009 Euro American Conference on Telematics and Information Systems: New Opportunities to Increase Digital Citizenship, EATIS '09

Conference

Conference2009 Euro American Conference on Telematics and Information Systems: New Opportunities to Increase Digital Citizenship, EATIS '09
Country/TerritoryCzech Republic
CityPrague
Period3/06/095/06/09

Keywords

  • Algorithms
  • Dynamic backprogation
  • LP
  • Modeling
  • Multilayer perceptrons
  • Neural networks dynamics
  • Non-linear dynamics
  • Training

Fingerprint

Dive into the research topics of 'Identification and modeling for non-linear dynamic system using neural networks type MLP'. Together they form a unique fingerprint.

Cite this