Multi-FNN Identification Based on HCM Clustering and Evolutionary Fuzzy Granulation

  • Park, Ho-Sung (Department of Electrical Electronic & Information Engineering, Wonkwang University) ;
  • Oh, Sung-Kwun (Department of Electrical Electronic & Information Engineering, Wonkang University)
  • Published : 2003.06.01

Abstract

In this paper, we introduce a category of Multi-FNN (Fuzzy-Neural Networks) models, analyze the underlying architectures and propose a comprehensive identification framework. The proposed Multi-FNNs dwell on a concept of fuzzy rule-based FNNs based on HCM clustering and evolutionary fuzzy granulation, and exploit linear inference being treated as a generic inference mechanism. By this nature, this FNN model is geared toward capturing relationships between information granules known as fuzzy sets. The form of the information granules themselves (in particular their distribution and a type of membership function) becomes an important design feature of the FNN model contributing to its structural as well as parametric optimization. The identification environment uses clustering techniques (Hard C - Means, HCM) and exploits genetic optimization as a vehicle of global optimization. The global optimization is augmented by more refined gradient-based learning mechanisms such as standard back-propagation. The HCM algorithm, whose role is to carry out preprocessing of the process data for system modeling, is utilized to determine the structure of Multi-FNNs. The detailed parameters of the Multi-FNN (such as apexes of membership functions, learning rates and momentum coefficients) are adjusted using genetic algorithms. An aggregate performance index with a weighting factor is proposed in order to achieve a sound balance between approximation and generalization (predictive) abilities of the model. To evaluate the performance of the proposed model, two numeric data sets are experimented with. One is the numerical data coming from a description of a certain nonlinear function and the other is NOx emission process data from a gas turbine power plant.

Keywords

References

  1. Inf. Control v.8 Fuzzy sets L. A. Zadeh https://doi.org/10.1016/S0019-9958(65)90241-X
  2. Int. J. of Approximate Reasoning v.5 no.3 NN-driven fuzzy reasoning H. Takagi;I. Hayashi https://doi.org/10.1016/0888-613X(91)90008-A
  3. IEEE trans. Neural Networks v.3 no.5 On fuzzy modeling using fuzzy neural networks with the back propagation algorithm S. Horikawa;T. Furuhashi;Y. Uchigawa https://doi.org/10.1109/72.159069
  4. Journal of Japan Society for Fuzzy Theory and Systems v.4 no.5 A fuzzy rule structured neural networks N. Imasaki;J. Kiji;T. Endo
  5. 4th IFSA'91 A self-tuning method of fuzzy control by descent methods H. Nomura;Wakami
  6. Optimization & Machine Learning Genetic Algorithm in search D. E. Golderg
  7. Genetic Algorithms + Data Structures = Evolution Programs Z. Michalewicz
  8. Parallel Problem Solving from Nature 2 Are Genetic Algorithms Function Optimizers? D. Jong, K. A.;Manner, R.(ed.)Manderick, B.(ed.)
  9. Fuzzy Sets and Syst. v.115 no.2 Identification of fuzzy systems by means of an auto-tuning algorithm and its application to nonlinear systems S. K. Oh;W. Pedrycz https://doi.org/10.1016/S0165-0114(98)00174-2
  10. Trans. SICE v.23 no.6 Fuzzy modeling G. Kang;M. Sugeno
  11. Fuzzy Control M. Sugeno
  12. Tran. of the Society of Instrument and Control Engineers v.22 no.9 Revised GMDH algorithm estimating degree of the complete polynomial T. Kondo https://doi.org/10.9746/sicetr1965.22.928
  13. Technical Report, Georgia Institute of Technology Prediction of Gas Turbine NOx Emissions using Polynomial Neural Network G. Vachtsevanos;V. Ramani;T. W. Hwang
  14. The Institute of Control, Automation and Systems Engineers(ICASE) v.6 no.3 The design of optimal fuzzy-neural networks structure by means of ga and an aggregate weighted performance index S. K. Oh;K. C. Yoon;H. K. Kim
  15. IEE Proc.-Control Theory and Applications v.148 no.Issue 05 Identification of fuzzy models with the aid of evolutionary data granulation B. J. Park;W. Pedrycz;S. K. Oh
  16. Fuzzy Model & Control System by CProgramming S. K. Oh
  17. Neural Networks, and Genetic Algorithms Computational Intelligence by Programming focused on Fuzzy S. K. Oh