ArticlesNo Access

A REPAIR ALGORITHM FOR RADIAL BASIS FUNCTION NEURAL NETWORK AND ITS APPLICATION TO CHEMICAL OXYGEN DEMAND MODELING

JUN-FEI QIAO

College of Electronic and Control Engineering, Beijing University of Technology, Beijing, 100124, China

Search for more papers by this author

and

HONG-GUI HAN

College of Electronic and Control Engineering, Beijing University of Technology, Beijing, 100124, China

Corresponding author.

Search for more papers by this author

https://doi.org/10.1142/S0129065710002243Cited by:29 (Source: Crossref)

Abstract

This paper presents a repair algorithm for the design of a Radial Basis Function (RBF) neural network. The proposed repair RBF (RRBF) algorithm starts from a single prototype randomly initialized in the feature space. The algorithm has two main phases: an architecture learning phase and a parameter adjustment phase. The architecture learning phase uses a repair strategy based on a sensitivity analysis (SA) of the network's output to judge when and where hidden nodes should be added to the network. New nodes are added to repair the architecture when the prototype does not meet the requirements. The parameter adjustment phase uses an adjustment strategy where the capabilities of the network are improved by modifying all the weights. The algorithm is applied to two application areas: approximating a non-linear function, and modeling the key parameter, chemical oxygen demand (COD) used in the waste water treatment process. The results of simulation show that the algorithm provides an efficient solution to both problems.

Keywords:

References

A. Mathur and J. F. Martin, Lancet 364, 183 (2004), DOI: 10.1016/S0140-6736(04)16632-4. Medline, Web of Science, Google Scholar
S. Ghosh-Dastidar and H. Adeli, International Journal of Neural Systems 19(4), 295 (2009), DOI: 10.1142/S0129065709002002. Link, Web of Science, Google Scholar
A. Karim and H. Adeli, Journal of Transportation Engineering 129(5), 494 (2003), DOI: 10.1061/(ASCE)0733-947X(2003)129:5(494). Web of Science, Google Scholar
S. Ghosh-Dastidar, H. Adeli and N. Dadmehr, IEEE Transactions on Biomedical Engineering 55(2), 512 (2008), DOI: 10.1109/TBME.2007.905490. Medline, Web of Science, Google Scholar
H. Liu, X. Wang and W. Qiang, International Journal of Neural Systems 17(6), 459 (2007), DOI: 10.1142/S0129065707001299. Link, Web of Science, Google Scholar
T. D. Jorgensen, B. P. Haynes and C. C. F. Norlund, International Journal of Neural Systems 18(5), 389 (2008), DOI: 10.1142/S012906570800166X. Link, Web of Science, Google Scholar
D. S. Huang and J. X. Du, IEEE Transaction on Neural Networks 19(12), 2099 (2008), DOI: 10.1109/TNN.2008.2004370. Medline, Web of Science, Google Scholar
N. Jin and D. R. Liu, IEEE Transaction on Neural Networks 19(3), 535 (2008). Medline, Web of Science, Google Scholar
A. Espositoet al., Neural Networks 13(3), 651 (2000), DOI: 10.1016/S0893-6080(00)00035-6. Medline, Web of Science, Google Scholar
X. X. Wang, S. Chen and D. J. Brown, Neurocomputing 62, 441 (2004), DOI: 10.1016/S0925-2312(04)00308-X. Web of Science, Google Scholar
N. Y. Lianget al., IEEE Transaction on Neural Networks 17(6), 1411 (2008), DOI: 10.1109/TNN.2006.880583. Medline, Web of Science, Google Scholar
S. Chen, E. S. Chng and K. Alkadhimi, International Journal of Control 64(5), 829 (1996), DOI: 10.1080/00207179608921659. Web of Science, Google Scholar
M. J. L. Orr, Neural Computation 7(3), 606 (1995), DOI: 10.1162/neco.1995.7.3.606. Web of Science, Google Scholar
K. Li, J. Peng and G. W. Irwin, IEEE Transaction on Automation Control 50(8), 1211 (2005). Web of Science, Google Scholar
A. L. I. Oliveiraet al., International Journal of Neural Systems 6(4), 271 (2006). Google Scholar
J. Gonzalezet al., IEEE Transaction on Neural Networks 14(6), 1478 (2003), DOI: 10.1109/TNN.2003.820657. Medline, Web of Science, Google Scholar
L. Yingwei, N. Sundararajan and P. Saratchandran, IEEE Transaction on Neural Networks 9(2), 308 (1998). Medline, Web of Science, Google Scholar
M. Salmerónet al., Neurocomputing 41, 153 (2001), DOI: 10.1016/S0925-2312(00)00363-5. Web of Science, Google Scholar
I. Rojaset al., Neurocomputing 42, 267 (2002), DOI: 10.1016/S0925-2312(01)00338-1. Web of Science, Google Scholar
P. Hao and J. Chiang, International Journal of Neural Systems 16(4), 283 (2006), DOI: 10.1142/S0129065706000688. Link, Web of Science, Google Scholar
G. B. Huang, P. Saratchandran and N. Sundararajan, IEEE Transaction on Systems, Man, and Cybernetics—Part B: Cybernetics 34(6), 2284 (2004), DOI: 10.1109/TSMCB.2004.834428. Medline, Web of Science, Google Scholar
G. B. Huang, P. Saratchandran and N. Sundararajan, IEEE Transaction on Neural Networks 16(1), 57 (2005). Medline, Web of Science, Google Scholar
Q. Meng and M. Lee, Neurocomputing 71(7–9), 1449 (2008), DOI: 10.1016/j.neucom.2007.05.012. Web of Science, Google Scholar
J. M. Lianet al., IEEE Transaction on Neural Networks 19(3), 460 (2008). Medline, Web of Science, Google Scholar
A. Saltelliet al., Reliability Engineering and System Safety 91(10–11), 1109 (2006), DOI: 10.1016/j.ress.2005.11.014. Web of Science, Google Scholar
J. Cariboniet al., Ecological Modeling 203(1–2), 167 (2007), DOI: 10.1016/j.ecolmodel.2005.10.045. Web of Science, Google Scholar
A. Saltelli, S. Tarantola and K. S. Chan, Technometrics 41(1), 39 (1999), DOI: 10.2307/1270993. Web of Science, Google Scholar
P. Lauret, E. Fock and T. A. Mara, IEEE Transaction on Neural Networks 17(2), 273 (2006), DOI: 10.1109/TNN.2006.871707. Medline, Web of Science, Google Scholar
M. J. Er and S. Wu, Fuzzy Sets and Systems 126(33), 337 (2002), DOI: 10.1016/S0165-0114(01)00034-3. Web of Science, Google Scholar
G. Leng, G. Prasad and T. M. McGinnity, Neural Networks 17(10), 1477 (2004), DOI: 10.1016/j.neunet.2004.07.009. Medline, Web of Science, Google Scholar
M. Graciaet al., Water Research 43(1), 1 (2009). Medline, Web of Science, Google Scholar