spacer
spacer Main Page
spacer
spacer Call For Papers
spacer
spacer Location
spacer
spacer Chair-Committee
spacer
spacer Deadlines
spacer
spacer Paper Format
spacer
spacer Fees
spacer
spacer SUBMIT A PAPER
spacer
spacer SUBMIT A SPECIAL SESSION
spacer
spacer SEND THE FINAL VERSION
spacer
spacer Conference Program
spacer
spacer Presentation Information
spacer
spacer Call for Collaborators
spacer
spacer Relevant WSEAS Conferences
spacer
spacer REVIEWERS
spacer
spacer CONTACT US
Past Conferences Reports
Find here full report from previous events


Impressions from previous conferences ...
Read your feedback...


History of the WSEAS conferences ...
List of previous WSEAS Conferences...


Urgent News ...
Learn the recent news of the WSEAS ...

 



 

spacer


Identification of Non-Linear Systems Using Neural Networks, with Application at Fuzzy Systems

Professor  Constantin Volosencu
Department of Automatics and Applied Informatics
Faculty of Automatics and Computers
“Politehnica” University of Timisoara
Bd. V. Parvan nr. 2
Timisoara, 300223
Romania
E-mail: constantin.volosencu@aut.upt.ro, http://www.aut.upt.ro/~cvolos/
 


Abstract: The paper present a short review of the ways of using artificial neural networks with continuous values for non-linear identification, with application at the fuzzy systems. Feedforward neural networks with continuous values may be seen as general approximation functions. Using adequate training methods and according Kolmogorov’s theorem, we may approximate any kind of multivariable non-linear functions. Fuzzy systems, developed based on different membership functions, inference methods, rule bases and defuzzification methods are non-linear systems. The paper presents some study cases of using feedforward neural networks with hidden layer and neurons with continuous values to approximate fuzzy systems with two input and one output variables. Some quality criteria for the training of the neural networks are introduced. Based on usage of different training methods, training parameter of neural networks a comparison is presented, emphasising the quality of training.
 

Brief Biography of the Speaker: Prof. Constantin Volosencu graduated in 1981 the Faculty of Electrotechnics, “Traian Vuia” Polytechnic Institute of Timisoara, Romania, as an engineer in automatics and computers and he is doctor in control systems at “Politehnica” University of Timisoara. In present he is professor at “Politehnica” University of Timisoara, Faculty of Automatics and Computers, Department of Automatics and Applied Informatics. His research interest is in linear control systems, fuzzy control, neural networks, control of electrical drives, modelling, simulation, identification and sensor networks. He is author of 9 books and more then 100 scientific papers, published at international conferences and journals. He was manager of over 30 national an international research projects.
Constantin Volosencu worked from 1981 to 1990 at “Electrotimis” Enterprise Timisoara, in the field of the control systems for industrial machines, where he developed control equipments for a large scale of machineries, which are the objects of 27 patents.

Copyright © www.wseas.org                        Designed by WSEAS