The Transactions of the Korean Institute of Electrical Engineers C (대한전기학회논문지:전기물성ㆍ응용부문C)

The Korean Institute of Electrical Engineers (대한전기학회)
권호 표지

Optimization of Polynomial Neural Networks: An Evolutionary Approach

다항식 뉴럴 네트워크의 최적화 : 진화론적 방법

  • 김동원 (고려대학교 공과대학 전기공학과) ;
  • 박귀태 (고려대학교 공과대학 전기공학과)
  • Published : 2003.07.01
⟨ Previous Next ⟩

Abstract

Evolutionary design related to the optimal design of Polynomial Neural Networks (PNNs) structure for model identification of complex and nonlinear system is studied in this paper. The PNN structure is consisted of layers and nodes like conventional neural networks but is not fixed and can be changable according to the system environments. three types of polynomials such as linear, quadratic, and modified quadratic is used in each node that is connected with various kinds of multi-variable inputs. Inputs and order of polynomials in each node are very important element for the performance of model. In most cases these factors are decided by the background information and trial and error of designer. For the high reliability and good performance of the PNN, the factors must be decided according to a logical and systematic way. In the paper evolutionary algorithm is applied to choose the optimal input variables and order. Evolutionary (genetic) algorithm is a random search optimization technique. The evolved PNN with optimally chosen input variables and order is not fixed in advance but becomes fully optimized automatically during the identification process. Gas furnace and pH neutralization processes are used in conventional PNN version are modeled. It shows that the designed PNN architecture with evolutionary structure optimization can produce the model with higher accuracy than previous PNN and other works.

Keywords

References

  1. Automatica v.7 System identification-a survey K.J.Astrom;P.Eykhoff https://doi.org/10.1016/0005-1098(71)90059-8
  2. IEEE Trans. Syst., Man, Cybern. v.SMC-1 no.1 Polynomial theory of complex systems A.G.Ivakhnenko https://doi.org/10.1109/TSMC.1971.4308320
  3. Sov. Automat. Contr. v.7 Long-term prediction by GMDH algorithms using the unbiased criterion and the balance-of-variables criterion A.G.Ivakhnenko;N.A.Ivakhnenko
  4. Sov. Automat. Contr. v.8 Long-term prediction by GMDH algorithms using the unbiased criterion and the balance-of-variables criterion, part 2 A.G.Ivakhnenko;N.A.Ivakhnenko
  5. Sov. Automat. Contr. v.11 Principal version of the minimum bias criterion for a model and an investigation of their noise immunity A.G.Ivakhnenko;V.N.Vysotskiy;N.A.Ivakhnenko
  6. Theoretical Systems Ecology: Advances and Case Studies Identification of the mathematical model of a complex system by the self-organization method A.G.Ivakhnenko;G.I.Krotov;N.A.Ivakhnenko;E.Halfon(ed.)
  7. Self-Organizing Methods in Modeling, GMDH Type-Algorithms S.J.Farlow
  8. IEEE Trans. Syst., Man, Cybern. v.28 no.3 Generating Optimal Adaptive Fuzzy-Neural Models of Dynamical Systems with Applications to Control S.Barada;H.Singh https://doi.org/10.1109/5326.704574
  9. Trans. KIEE v.49D no.3 다항식 뉴럴네트워크 구조의 최적 설계에 관한 연구 오성권;김동원;박병준
  10. Master's thesis, Dept. Control Instrum., Wonkwang Univ. 자기구성 다항식 뉴럴네트워크의 진화론적 설계 김동원
  11. Adaptation in Natural and Artificial Systems J.H.Holland
  12. International Journal of System Science v.29 no.3 GA-based reinforecement learning for neural networks C.T.Lin;C.P.Jou;C.J.Lin https://doi.org/10.1080/00207729808929517
  13. IEEE Trans. Fuzzy Syst. v.3 Simulaneous design of membership functions and rule sets for fuzzy controllers using genetic algorithms A.Homaifar;E.McCormick https://doi.org/10.1109/91.388168
  14. Time Series Analysis : Forecasting and Control(2nd ed.) G.E.P.Box;F.M.Jenkins
  15. pH and pION Control in Process and Waste Streams F.G.Shinskey
  16. 전자공학회지 v.22 no.11 유전 알고리즘이론 및 응용 장병탁
  17. Trans. on Control, Automation and Systems Engineering v.3 no.1 Advanced Polynomial Neural Networks Architecture with New Adaptive Nodes Sung-Kwun Oh;Dong-Won Kim;Byoung-Jun Park;Hyung-Soo Hwang
  18. Fuzzy Sets Syst 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
  19. IEEE Trans. Fuzzy Syst. v.1 no.1 A fuzzy-logic-based approach to qualitative modeling M.Sugeno;T.Yasukawa https://doi.org/10.1109/TFUZZ.1993.390281
  20. Fuzzy Sets Syst. v.13 An identification algorithm in fuzzy relational system W.Pedrycz https://doi.org/10.1016/0165-0114(84)90015-0
  21. Fuzzy Sets Syst. v.108 A new artificial neural networks based fuzzy inference system with moving consequents in if-then rules and selected applications J.Leski;E.Czogala https://doi.org/10.1016/S0165-0114(97)00314-X
  22. IEEE Trans. Fuzzy Syst. v.8 no.1 Evolutionary Design of Fuzzy Rule Base for Nonlinear System Modeling and Control S.J.Kang;C.H.Woo;H.S.Hwang;K.B.Woo
  23. Inf. Sci. v.110 A simple identified Sugeno-type fuzzy model via double clustering E.Kim;H.Lee;M.Park;M.Park https://doi.org/10.1016/S0020-0255(97)10083-4
  24. IEEE Trans. Fuzzy Syst. v.3 no.2 A new approach to fuzzy-neural modeling Y.Lin;G.A.Cunningham Ⅲ https://doi.org/10.1109/91.388173
  25. IEEE Trans. Syst. Man Cybern. Part B-Cybern. A Design of EA-based Self-Organizing Polynomial Neural Networks using Evolutionary Algorithm for Nonlinear System Modeling D.W.Kim;G.T.Park
  26. Fuzzy Sets Syst. v.115 Identification of fuzzy systems by means of an auto-tuning algorithm and its application to nonlinear systems S.Oh;W.Pedrycz https://doi.org/10.1016/S0165-0114(98)00174-2