Application of NBP algorithm in system identification

MaRDI QIDQ2735397zbMATH OpenFDO

Authors Yingmin Liu, Cangpu Wu

Publication date 4 November 2002

Published in Journal of Beijing Institute of Technology. Chinese Edition (Search for Journal in Brave)

zbMATH Keywords

feedforward neural networksnonlinear BP algorithmsystem identification

Mathematics Subject Classification ID

Neural networks for/in biological studies, artificial life and related topics (92B20) System identification (93B30) Nonlinear systems in control theory (93C10)

Review text

The authors study the problem for identifying nonlinear dynamic processes using multiplayer feedforward Neural Networks (NN) via Nonlinear Back Propagation (NBP) training algorithms. They present an algorithm for identifying Single-Input Single-Output (SISO) serial-parallel nonlinear systems based on NBP networks that avoids computation of the derivatives of neural activations and give a numerical simulation example to illustrate the effectiveness of their algorithm, which is better than the normal BP algorithm.

Recommendations

• On ``An improved approach for nonlinear system identification using neural networks.
• Non-linear system identification using neural networks
• Neural networks for nonlinear dynamic system modelling and identification
• An improved approach for nonlinear system identification using neural networks
• Recursive identification using feedforward neural networks

Cited in

(2)

• An improved approach for nonlinear system identification using neural networks
• Identification of restoring forces in non-linear vibration systems using fuzzy adaptive neural networks

This page was built for publication: Application of NBP algorithm in system identification

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q2735397)