International Journal of Control, Automation, and Systems

Institute of Control, Robotics and Systems (제어로봇시스템학회)
권호 표지

Self-Organizing Polynomial Neural Networks Based on Genetically Optimized Multi-Layer Perceptron Architecture

  • Park, Ho-Sung (Dept. of Electrical Electronic & Information Engineering, Wonkwang University) ;
  • Park, Byoung-Jun (Dept. of Electrical Electronic & Information Engineering, Wonkwang University) ;
  • Kim, Hyun-Ki (Dept. of Electrical Engineering, Suwon University) ;
  • Oh, Sung-Kwun (Dept. of Electrical Electronic & Information Engineering, Wonkwang University)
  • Published : 2004.12.01
Download PDF
⟨ Previous Next ⟩

Abstract

In this paper, we introduce a new topology of Self-Organizing Polynomial Neural Networks (SOPNN) based on genetically optimized Multi-Layer Perceptron (MLP) and discuss its comprehensive design methodology involving mechanisms of genetic optimization. Let us recall that the design of the 'conventional' SOPNN uses the extended Group Method of Data Handling (GMDH) technique to exploit polynomials as well as to consider a fixed number of input nodes at polynomial neurons (or nodes) located in each layer. However, this design process does not guarantee that the conventional SOPNN generated through learning results in optimal network architecture. The design procedure applied in the construction of each layer of the SOPNN deals with its structural optimization involving the selection of preferred nodes (or PNs) with specific local characteristics (such as the number of input variables, the order of the polynomials, and input variables) and addresses specific aspects of parametric optimization. An aggregate performance index with a weighting factor is proposed in order to achieve a sound balance between the approximation and generalization (predictive) abilities of the model. To evaluate the performance of the GA-based SOPNN, the model is experimented using pH neutralization process data as well as sewage treatment process data. A comparative analysis indicates that the proposed SOPNN is the model having higher accuracy as well as more superb predictive capability than other intelligent models presented previously.reviously.

Keywords

References

  1. IEEE Trans. on Neural Networks v.7 Comparison of adaptive methods for function estimation from samples V. Cherkassky;D. Gehring;F. Mulier https://doi.org/10.1109/72.508939
  2. IEEE Trans. on Systems, Man and Cybernetics v.26 Fuzzy function approximation with ellipsoidal rules J. A. Dicherson;B. Kosko https://doi.org/10.1109/3477.517030
  3. IEEE Trans. on Systems, Man and Cybernetics v.SMC-1 Polynomial theory of complex systems A. G. Ivakhnenko
  4. Inductive Learning Algorithms for Complex Systems Modeling A. G. Ivakhnenko;H. R. Madala
  5. Pattern Recognition and Image Analysis v.5 no.4 The review of problems solvable by algorithms of the group method of data handling (GMDH) A. G. Ivakhnenko;G. A. Ivakhnenko
  6. Pattern Recognition and Image Analysis v.4 no.2 Self-organization of neural networks with active neurons A. G. Ivakhnenko;G. A. Ivakhnenko;J.A. Muller
  7. Information Science v.141 The design of selforganizing polynomial neural networks S.-K. Oh;W. Pedrycz https://doi.org/10.1016/S0020-0255(02)00175-5
  8. Computers and Electrical Engineering v.29 no.6 Polynomial neural networks architecture: analysis and design S.-K. Oh;W. Pedrycz;B.-J. Park https://doi.org/10.1016/S0045-7906(02)00045-9
  9. Journal of the Research Institute of Engineering Technology Development (in Korean) v.22 Optimal design of self-organizing polynomial neural networks by means of genetic algorithms H.-S. Park;B.-J. Park;S.-K. Oh
  10. Studies in Fuzziness and Soft Computing v.8 Evolutionary optimization of fuzzy models in fuzzy logic: A framework for the new millennium W. Pedrycz;M. Reformat;V. Dimitrov(ed.);V. Korotkich(ed.)
  11. Adaptation in Natural and Artificial Systems J. H. Holland
  12. Genetic Algorithm in Search, Optimization & Machine Learning D. E. Goldberg
  13. Parallel Problem Solving from Nature 2 Are genetic algorithms function optimizers? K. A. De Jong;Manner R.(ed.);Manderick, B.(ed.)
  14. Genetic Algorithms + Data Structures = Evolution Programs Z. Michalewicz
  15. Fuzzy Sets and Systems v.115 no.2 Identification of fuzzy systems by means of an auto-tuning algorithm and its application to nonlinear system S.-K. Oh;W. Pedrycz https://doi.org/10.1016/S0165-0114(98)00174-2
  16. Int. J. of Intelligent Systems v.17 no.1 Hybrid identification of fuzzy rule-based models S.-K. Oh;W. Pedrycz;B.-J. Park https://doi.org/10.1002/int.1004
  17. Fuzzy Sets and Systems v.138 Hybrid identification in fuzzy-neural networks S.-K. Oh;W. Pedrycz;H.-S. Park https://doi.org/10.1016/S0165-0114(02)00441-4
  18. IEEE Trans. on Instrumentation and Measurement Fuzzy relation-based neural-networks and their hybrid identification S.-K. Oh;W. Pedrycz;H.-S. Park
  19. pH and pION Control in Proc. and Waste Streams F. G. Shinskey
  20. Proc. ACC Modeling and self-tuning control of a multivariable pH neutralization process R. C. Hall;D. E. Seberg
  21. Ind. Eng. Chem. Process Des. Develop v.11 no.1 Time optimal and Ziegler-Nichols control T. J. McAvoy https://doi.org/10.1021/i260041a014
  22. IEEE Control Systems Magazine v.7 no.1 Comparison of linear and nonlinear adaptive control of a pH-process G. A. Pajunen https://doi.org/10.1109/MCS.1987.1105238
  23. IEEE Trans. on Fuzzy Systems v.1 Fuzzy control of pH using genetic algorithms C. L. Karr;E. J. Gentry https://doi.org/10.1109/TFUZZ.1993.390283
  24. Ind. Engrg. Chem. Process Des. Develop v.11 no.1 Dynamics of pH in controlled stirred tank reactor T. J. McAvoy;E. Hsu;S. Lowenthal https://doi.org/10.1021/i260041a013
  25. Chem. Engrg. Sci. v.38 Dynamic modeling and reaction invariant control of pH T. K. Gustafsson;K. V. Waller https://doi.org/10.1016/0009-2509(83)80157-2
  26. Fuzzy Sets and Systems v.78 Modeling pH neutralization processes using fuzzy-neural approaches J. Nie;A. P. Loh;C. C. Hang https://doi.org/10.1016/0165-0114(95)00118-2
  27. IEEE Trans. on Fuzzy Systems v.10 no.5 Fuzzy polynomial neural networks: hybrid architectures of fuzzy modeling B.-J. Park;W. Pedrycz;S.-K. Oh https://doi.org/10.1109/TFUZZ.2002.803495
  28. Lecture Notes in Artificial Intelligence v.2639 Genetically optimized rule-based fuzzy polynomial neural networks: synthesis of computational intelligence technologies S.-K. Oh;J. F. Peters;W. Pedrycz;T.-C. Ahn
  29. IEEE Trans. on SMC-A v.33 no.2 Selforganizing neurofuzzy networks based on evolutionary fuzzy granulation S.-K. Oh;W. Pedrycz;B.-J. Park
  30. Fuzzy Model & Control System by CProgramming S.-K. Oh
  31. Computational Intelligence by Programming focused on Fuzzy, Neural Networks, and Genetic Algorithms S.-K. Oh