Modeling of non-linear dynamic systems via discrete-time recurrent neural networkks and variational training algorithm

被引：0

作者：

Minchev, SV ^{[1
]}

Venkov, GI ^{[1
]}

机构：

[1] Tech Univ Sofia, Fac Appl Math & Informat, Sofia, Bulgaria

来源：

2004 2ND INTERNATIONAL IEEE CONFERENCE INTELLIGENT SYSTEMS, VOLS 1 AND 2, PROCEEDINGS | 2004年

关键词：

current transformer; learning algorithm; hysteresis; recurrent neural networks; non-linear system;

D O I：

10.1109/IS.2004.1344645

中图分类号：

TP18 [人工智能理论];

学科分类号：

081104 ; 0812 ; 0835 ; 1405 ;

摘要：

This paper proposes a discrete-time recurrent neural network architecture and parameter adaptation algorithm for modeling of non-linear dynamic systems. The learning algorithm is based on variational calculus and operates off-line. A neural network based current transformer non-linear 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.

引用

页码：105 / 108

页数：4

共 14 条

[11]

RUIZ V, 1995, IEEE T NEURAL NETWOR, V6, P657

[12]

Sage A. P., 1971, SYSTEM IDENTIFICATIO, V80

[13] MEMORY NEURON NETWORKS FOR IDENTIFICATION AND CONTROL OF DYNAMICAL-SYSTEMS [J].

SASTRY, PS ;

SANTHARAM, G ;

UNNIKRISHNAN, KP .

IEEE TRANSACTIONS ON NEURAL NETWORKS, 1994, 5 (02) :306-319

[14] IDENTIFICATION AND CONTROL OF A DC MOTOR USING BACK-PROPAGATION NEURAL NETWORKS [J].

WEERASOORIYA, S ;

ELSHARKAWI, MA .

IEEE TRANSACTIONS ON ENERGY CONVERSION, 1991, 6 (04) :663-669

← 1 2 →