Representing and Learning Unmodeled Dynamics with Neural Network Memories

  title={Representing and Learning Unmodeled Dynamics with Neural Network Memories},
  author={Tor Arne Johansen and Bjarne A. Foss},
  journal={1992 American Control Conference},
A nonlinear model representation consisting of an interpolation of several local models, which are valid within certain operation regimes, is proposed. Using this representation, first principles models and black-box models like neural networks may be integrated. Only operation regimes of the plant not adequately modeled by first principles are being represented and learned by a neural network memory. The principle is illustrated by simulation examples. 


Publications referenced by this paper.
Showing 1-10 of 21 references

A NARMAX model represntation for adaptiv control based on local models

  • T. A. Jobanen, B. A. Foss
  • Modelig, Identificstiox, and Control,
  • 1992

A combined PID and neural control scheme for nonliniear dynamicalsystems

  • E. Tzirkel-Hancock, F. Fallside
  • 1992

Construetion of composite models from observed data

  • Anders Skeppstedt
  • J . Control
  • 1992

Foes . Nonlinear local model represenation for adaptive systems

  • J. Moody, C. J. Darken.
  • Conf . on Intelligent Control and Instrumentstion
  • 1992

Nonlinear local model represenation for adaptive systems

  • T. A. Johansen, B. A. Foes
  • In Proceeding of the Singapore lat. Conf. on…
  • 1992
1 Excerpt

Intelligent adaptive control of mode-switch proces

  • R. A. Hilkort, J. van Amerongen, P. L8hnberg
  • In Proc. IFAC Internetinal Symposium on…
  • 1991

Neural net based model predictive control

  • L. Ljung A. Skeppstedt
  • Itt . Joumra of Control
  • 1991

Neural network based control of mode-switch prones

  • R. A. Hilhorst, J. van Amerongen, P. L6hnberg
  • In Proc. IFAC Ist. Sgsnpo on Adnneed Contrl of…
  • 1991

Similar Papers

Loading similar papers…