Modeling of nonlinear dynamic systems via discrete-time recurrent neural networks and variational training algorithm

Author

Minchev, Stefan V. ; Venkov, Gancho I.

Author_Institution

Fac. of Appl. Math. & Informatics, Tech. Univ. of Sofia, Bulgaria

Volume

1

fYear

2004

Firstpage

105

Abstract

This paper proposes a discrete-time recurrent neural network architecture and parameter adaptation algorithm for modeling of nonlinear dynamic systems. The learning algorithm is based on variational calculus and operates off-line. A neural network based current transformer nonlinear model is presented as a demonstration of the proposed architecture and learning algorithm. It is designed for power engineering needs in power systems and is suited for real-time applications in digital relay protections.

Keywords

current transformers; learning (artificial intelligence); nonlinear dynamical systems; power engineering computing; power systems; recurrent neural nets; relay protection; transformer protection; digital relay protections; discrete-time recurrent neural networks; hysteresis; learning algorithm; neural network based current transformer nonlinear model; nonlinear systems; parameter adaptation algorithm; power engineering; variational calculus; variational training algorithm; Adaptation model; Calculus; Current transformers; Neural networks; Power engineering; Power system dynamics; Power system modeling; Power system protection; Power system relaying; Recurrent neural networks;

fLanguage

English

Publisher

ieee

Conference_Titel

Intelligent Systems, 2004. Proceedings. 2004 2nd International IEEE Conference

Print_ISBN

0-7803-8278-1

Type

conf

DOI

10.1109/IS.2004.1344645

Filename

1344645