Advertisement

Hierarchical Competitive Net Architecture

  • Theresa W. Long
  • Emil L. Hanzevack
Part of the Advances in Industrial Control book series (AIC)

Abstract

Development of hypersonic aircraft requires a high degree of system integration. Design tools are needed to provide rapid, accurate calculations of complex fluid flow patterns. This project demonstrates that neural networks can be successfully applied to calculation of fluid flow distribution and heat transfer in a six leg heat exchanger panel, typical of the type envisioned for use in hypersonic aircraft.

We used training data generated from fluid flow and heat transfer equations in explicit form to train a neural net to solve the associated inverse dynamics problem. We developed an improved competitive net architecture. Finally, we successfully implemented a hierarchical neural network scheme to link multiple heat exchanger panels together. This method is direct, fast, and accurate.

Keywords

Hide Layer Mass Flow Rate Heat Load Heat Transfer Equation Hypersonic Vehicle 
These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

References

  1. [1].
    Jacobs, R.A., M.I. Jordan, S.J. Nowlan, & G.E. Hinton, “Adaptive Mixtures of Local Experts”, Neural Computation, 3, 79–87, 1991.CrossRefGoogle Scholar
  2. [2].
    Lange, T.E., “Dynamically-Adaptive Winner-Take-All Networks”, J. Moody, S. Hanson, & R. Lippmann, Neural Information Svstems,4, pp. 341–348, San Mateo, CA, Morgan Kaufmann, 1992.Google Scholar
  3. [3].
    Jacobs, R.A., M.I. Jordan, “Learning Piecewise Control Strategies In a Modular Connectionist Architecture”, in preparation, 1992.Google Scholar
  4. [4].
    Pao, Y.H., “Adaptive Pattern Recognition and Neural Networks”, Addison Wesley, pp.57–82, 1989.zbMATHGoogle Scholar
  5. [5].
    Bridle, J., “Probabilistic Interpretation of Feedforward Classification Network Outputs with Relationships to Statistical Pattern Recognition”, F. Fogelman-Soulie & J. Herault (eds.), Neuro-computing: Algorithms Architectures, and Applications, New York, Springer-Verlag, 1989.Google Scholar
  6. [6].
    Jordan, M.I., R.A. Jacobs, “Hierarchies of Adaptive Experts,”, J. Moody, S. Hanson, & R. Lippmann, Neural Information Svstems,4, pp.985–992, San Mateo, CA, Morgan Kaufmann, 1992.Google Scholar
  7. [7].
    White, D.A., A. Bowers, K. Iliff, G. Noffz, M. Gonda, and J. Menousek, “Flight, Propulsion, and Thermal Control of Advanced Aircraft and Hypersonic Vehicles,” D.A. White and D.A. Sofge, (eds.), Handbook of Intelligent Control, New York, Van Nostrand Reinhold, 1992.Google Scholar

Copyright information

© Springer-Verlag London Limited 1995

Authors and Affiliations

  • Theresa W. Long
    • 1
  • Emil L. Hanzevack
    • 2
  1. 1.NeuroDyne Inc.WilliamsburgUSA
  2. 2.University of South CarolinaColumbiaUSA

Personalised recommendations