Stability analysis for dynamical neural network systems

Author

Lam, S. ; Hung, Y.S.

Author_Institution

Dept. of Electr. & Electron. Eng., Hong Kong Univ., Hong Kong

Volume

2

fYear

1994

fDate

27 Jun-2 Jul 1994

Firstpage

960

Abstract

In this paper, the small gain theorem is used to establish a criterion for the stability of a feedback system containing a feedforward neural network. A method for the determination of the gain of a piecewise-linear feedforward neural network is introduced and applied to the stability analysis for a control system consisting of a LTI SISO system with a dynamic neural network controller

Keywords

closed loop systems; control system analysis; discrete time systems; feedforward neural nets; linear systems; neurocontrollers; stability; state-space methods; SISO system; closed loop system; discrete time system; dynamic neurocontroller; dynamical neural network systems; feedback system; linear time invariant system; piecewise-linear feedforward neural network; stability analysis; state space model; Artificial neural networks; Control systems; Feedforward neural networks; Multi-layer neural network; Neural networks; Neurofeedback; Nonlinear control systems; Piecewise linear techniques; Stability analysis; Stability criteria;

fLanguage

English

Publisher

ieee

Conference_Titel

Neural Networks, 1994. IEEE World Congress on Computational Intelligence., 1994 IEEE International Conference on

Conference_Location

Orlando, FL

Print_ISBN

0-7803-1901-X

Type

conf

DOI

10.1109/ICNN.1994.374311

Filename

374311