A new structure adaptation algorithm for RBF networks and its application (original) (raw)

Abstract

An adaptation algorithm is developed for radial basis function network (RBFN) in this paper. The RBFN is adapted on-line for both model structure and parameters with measurement data. When the RBFN is used to model a non-linear dynamic system, the structure is adapted to model abrupt change of system operating region, while the weights are adapted to model the incipient time varying parameters. Two new algorithms are proposed for adding new centres while the redundant centres are pruned, which is particularly useful for model-based control. The developed algorithm is evaluated by modelling a numerical example and a chemical reactor rig. The performance is compared with a non-adaptive model.

Access this article

Log in via an institution

Subscribe and save

Buy Now

Price excludes VAT (USA)
Tax calculation will be finalised during checkout.

Instant access to the full article PDF.

Similar content being viewed by others

References

  1. Hunt KJ, Sarbaro D, Zbikowsky R, Gawthrop PJ (1992) Neural network for control systems—a survey. Automatica 28(6):1083–1112
    Article MATH MathSciNet Google Scholar
  2. Chen S, Billings SA, Grant PM (1992) Recursive hybrid algorithm for non-linear system identification using radial basis function networks. Int J Control 55:1051–1070
    MATH MathSciNet Google Scholar
  3. Schenker B, Agarwal M (1998) Predictive control of a bench-scale chemical reactor based on neural network models. IEEE Trans Control Syst Technol 6(3):388–400
    Article Google Scholar
  4. Chen S, Cowan CFN, Grant PM (1991) Orthogonal least squares learning algorithm for radial basis function networks. IEEE Trans Neural Netw 2:302–309
    Article Google Scholar
  5. Karayiannis NB, Mi GW (1997) Growing radial basis networks: merging supervised and unsupervised learning with network growth techniques. IEEE Trans Neural Netw 8(6):1492–1506
    Article Google Scholar
  6. Yu DL, Gomm JB, Williams D (1997) A recursive orthogonal least squares algorithm for training RBF networks. Neural Process Lett 5(3):167–176
    Article Google Scholar
  7. Luo W, Karim MN, Morris AJ, Martin EB (1996) Control relevant identification of a pH waste water neutralisation process using adaptive radial basis function networks. Comput Chem Eng 20:1017–1022
    Article Google Scholar
  8. Gomm JB, Yu DL (2000) Selecting radial basis function network centres with recursive orthogonal least squares training. IEEE Trans Neural Netw 11(3):306–314
    Article Google Scholar
  9. Bobrow JE, Murray W (1993) An algorithm for RLS identification of parameters that vary quickly with time. IEEE Trans Autom Control 38(2):35–254
    Article MathSciNet Google Scholar
  10. Narendra KS, Parthasarathy K (1990) Identification and control of dynamical systems using neural networks. IEEE Trans Neural Netw 1(1):4–27
    Article Google Scholar

Download references

Acknowledgments

The project was funded by the U.K. EPSRC with the grant No. GR/N18697.

Author information

Authors and Affiliations

  1. Control Systems Research Group, School of Engineering, Liverpool John Moores University, Byrom Street, Liverpool, L3 3AF, UK
    D. L. Yu
  2. Department of Automation, Northeast University at Qinhuangdao, Qinhuangdao, China
    D. W. Yu

Corresponding author

Correspondence toD. L. Yu.

Appendix

Appendix

See Table 1.

Table 1 Physical parameters of the chemical reactor rig

Full size table

Rights and permissions

About this article

Cite this article

Yu, D.L., Yu, D.W. A new structure adaptation algorithm for RBF networks and its application.Neural Comput & Applic 16, 91–100 (2007). https://doi.org/10.1007/s00521-006-0067-5

Download citation

Keywords