Neural Computing and Applications

, Volume 29, Issue 10, pp 721–731 | Cite as

The isolation layered optimization algorithm of MIMO polygonal fuzzy neural network

Original Article
  • 37 Downloads

Abstract

The single-input single-output or multi-input single-output polygonal fuzzy neural network can accomplish some information disposing based on a finite number of points of polygonal fuzzy number. Although it does not depend on a precise mathematical model, it involves logical reasoning, numerical calculation and nonlinear functional approximation. The multi-input multi-output (MIMO) polygonal fuzzy neural network model is proposed for the first time in this article. The two different algorithms are designed in the input layer and hidden layer, respectively, and some parameters of the connection weights in the isolation layered manner can be optimized. Particularly, the neurons in the hidden layer are optimized one by one. Results showed that the isolation layered optimization algorithm of MIMO polygonal fuzzy neural network could improve the computational efficiency and convergent rate.

Keywords

Polygonal fuzzy number Polygonal fuzzy neural network Isolation layered optimization algorithm Generalized inverse minimum norm Least square method 

Notes

Acknowledgment

This work has been supported by National Natural Science Foundation China (Grant No. 61374009).

References

  1. 1.
    Aliev RA, Guirimov BG, Fazlollahi B, Alive RR (2009) Evolutionary algorithm-based learning of fuzzy neural networks. Part 2: recurrent fuzzy neural networks. Fuzzy Sets Syst 160(17):2553–2566. doi: 10.1016/j.fss.2008.12.018 MathSciNetCrossRefMATHGoogle Scholar
  2. 2.
    Anitha J, Jude HD (2013) An efficient Kohonen-fuzzy neural network based abnormal retinal image classification system. Neural Netw World 23(2):149–167. doi: 10.14311/NNW.2013.23.011 CrossRefGoogle Scholar
  3. 3.
    Chung FL, Duan JC (2000) On multistage fuzzy neural network modeling. IEEE Trans Fuzzy Syst 8(2):125–142. doi: 10.1109/91.842148 CrossRefGoogle Scholar
  4. 4.
    Er MJ, Wu S, Gao Y (2001) A fast approach for automatic generation of fuzzy rules by generalized dynamic fuzzy neural networks. IEEE Trans Fuzzy Syst 9(4):578–594. doi: 10.1109/91.940970 CrossRefGoogle Scholar
  5. 5.
    Chen TP, Chen H (1993) Approximation of continuous functionals by neural networks with applications to dynamic systems. IEEE Trans Neural Netw 6(4):910–918. doi: 10.1109/72.286886 CrossRefGoogle Scholar
  6. 6.
    Feuring T, Lippe WM (1999) The fuzzy neural network approximation lemma. Fuzzy Sets Syst 102(2):227–237. doi: 10.1016/S0165-0114(97)00125-5 MathSciNetCrossRefMATHGoogle Scholar
  7. 7.
    Liu PY (2001) Universal approximation of continuous analyses fuzzy valued functions by multi-layer regular fuzzy neural networks. Fuzzy Sets Syst 119(3):313–320. doi: 10.1016/S0165-0114(99)00132-3 CrossRefMATHGoogle Scholar
  8. 8.
    Rumelhart DE, Hinton CE, Wiliams RJ (1986) Learning representations by back-propagating errors. Nature 323(9):533–536. doi: 10.1038/323533a0 CrossRefMATHGoogle Scholar
  9. 9.
    Tian CJ, Wei G (2001) A new layer-wise linearized algorithm for feedforward neural network. Acta Electron Sin 29(11):1495–1498Google Scholar
  10. 10.
    Xie H, Cheng HZ, Niu DX, Zhang GL (2005) A fast learning algorithm for feedfoward neural network based on the layer-by-layer and neuron-by-neuron optimizing procedure. Acta Electron Sin 33(1):111–114Google Scholar
  11. 11.
    Tao YQ, Cui DW (2011) Load identification of algorithm based on dynamic fuzzy granular neural network. Control Decis 26(4):519–523+529MathSciNetGoogle Scholar
  12. 12.
    Wu JL, Zhang BH, Wang K (2012) Application of adaboost-based BP neural network for short-term wind speed forecast. Power Styst Technol 36(9):221–225 Google Scholar
  13. 13.
    Zhou Y, Wang S, Gao CS, Sun WC (2010) An improved PSO-based fuzzy neural network and its application in short-term weather forecast. Comput Appl Softw 27(5):234–237Google Scholar
  14. 14.
    Liu PY (2002) A new fuzzy neural network and its approximation capability. Sci China Ser E 32(1):76–86Google Scholar
  15. 15.
    He CM, Ye YP (2011) Evolution computation based learning algorithms of polygonal fuzzy neural networks. Int J Intell Syst 262(4):340–352. doi: 10.1002/int.20469 CrossRefMATHGoogle Scholar
  16. 16.
    Wang GJ, Li XP (2011) Universal approximation of polygonal fuzzy neural networks in sense of \(K\)-integral norms. Sci China Inf Sci 54(11):2307–2323. doi: 10.1007/s11432-011-4364-y MathSciNetCrossRefMATHGoogle Scholar
  17. 17.
    Wang GJ, Li XP (2014) Construction of the polygonal fuzzy neural network and its approximation based on K-integral norm. Neural Netw World 24(4):357–376. doi: 10.14311/NNW.2014.24.021 CrossRefGoogle Scholar
  18. 18.
    He Y, Wang GJ (2012) Conjugate gradient algorithm of the polygonal fuzzy neural networks. Acta Electron Sin 40(10):2079–2084. doi: 10.3969/j.issn.0372-2112.2012.10.029 Google Scholar
  19. 19.
    Sui XL, Wang GJ (2012) Influence of perturbations of training pattern pairs on stability of polygonal fuzzy neural network. Pattern Recognit Artif Intell 26(6):928–936Google Scholar
  20. 20.
    Yang YQ, Wang GJ, Yang Y (2014) Parameter optimization of polygonal fuzzy neural networks based on GA-BP hybrid algorithm. Int J Mach Learn Cybernet 5(5):815–822. doi: 10.1007/s13042-013-0224-y CrossRefGoogle Scholar
  21. 21.
    Wang GJ, He Y, Li XP (2015) Optimization algorithms for MISO polygonal fuzzy neural networks. Sci China Inf Sci 45(5):650–667. doi: 10.1360/N112013-00214 Google Scholar

Copyright information

© The Natural Computing Applications Forum 2016

Authors and Affiliations

  1. 1.School of Mathematics ScienceTianjin Normal UniversityTianjinChina
  2. 2.School of MathematicsTonghua Normal UniversityTonghuaChina

Personalised recommendations