An Evaluation of Methods for the Selection of Inputs for an Artificial Neural Network Based River Model

  • G. J. Bowden
  • G. C. Dandy
  • H. R. Maier

Abstract

Artificial Neural Network (ANN) models are highly flexible function approximators, which have shown their utility in a broad range of ecological modelling applications. The rapid emergence of ANN applications in the field of ecological modelling can be attributed to their advantages over standard statistical approaches. Such flexibility provides a powerful tool for forecasting and prediction, however, the large number of parameters that must be selected only serves to complicate the design process. In most practical circumstances, the design of an ANN is heavily based on heuristic trial-and-error processes with only broad rules of thumb to guide along the way.

Keywords

Phosphorus Phytoplankton Turbidity Anabaena 

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

References

  1. ASCE Task Committee on Application of Artificial Neural Networks in Hydrology (2000) Artificial neural networks in hydrology. IIGoogle Scholar
  2. BioComp Systems, I. (1998) NeuroGenetic Optimizer (NGO). Redmond, WA. Hydrologic applications. Journal of Hydrologic Engineering, ASCE, 5(2): 124–137.Google Scholar
  3. Bowden GJ, Maier HR, Dandy GC(2000) Optimal division of data for neural network models in water resources applications. submitted to Water Resources Research.Google Scholar
  4. Cai S, Toral H, Qiu J, Archer JS (1994) Neural network based objective flow regime identification in air-water two phase flow. The Canadian Journal of Chemical Engineering, 72: 440–445.CrossRefGoogle Scholar
  5. Chakraborty K, Mehrotra K, Mohan CK, Ranka S (1992) Forecasting the behavior of multivariate time series using neural networks. Neural Networks, 5: 961–970.CrossRefGoogle Scholar
  6. Chon TS, Park YS, Moon KH, Cha EY (1996) Patternizing communities by using an artificial neural network. Ecological Modelling, 90: 69–78.CrossRefGoogle Scholar
  7. Cybenko G (1989) Approximation by superpositions of a sigmoidal function. Mathematics of Control Signals and Systems, 2: 203–314.CrossRefGoogle Scholar
  8. Dandy GC, Simpson AR, Murphy LJ (1996) An improved genetic algorithm for pipe network optimisation. Water Resources Research, 32(2): 449–458.CrossRefGoogle Scholar
  9. Downing K (1998) Using evolutionary computation techniques in environmental modelling. Environmental Modelling & Software, 13(5–6): 519–528.CrossRefGoogle Scholar
  10. Fernando DAK, Jayawardena AW (1998) Runoff forecasting using RBF networks with OLS algorithm. Journal of Hydrologic Engineering, 3(3): 203–209.CrossRefGoogle Scholar
  11. Foody GM (1999) Applications of the self-organising feature map neural network in community data analysis. Ecological Modelling, 120: 97–107.CrossRefGoogle Scholar
  12. Goldberg DE (1989) Genetic Algorithms in Search, Optimization and Machine Learning, Addison-Wesley, Reading, MA, 412 pp.Google Scholar
  13. Howard LM, D’Angelo DJ (1995) GA-P: a genetic algorithm and genetic programming hybrid. IEEE Expert, 10(3): 11–15.CrossRefGoogle Scholar
  14. Islam S, Kothari R (2000) Artificial neural networks in remote sensing of hydrologic processes. Journal of Hydrologic Engineering, 5(2): 138–144.CrossRefGoogle Scholar
  15. Jolliffe IT (1986) Principal Component Analysis, Springer-Verlag New York Inc., New York, 271 pp.CrossRefGoogle Scholar
  16. Kaski S, Kangas J, Kohonen T (1998) Bibliography of Self-Organizing Map (SOM) Papers: 1981–1997. Neural Computing Surveys, 1: 102–350.Google Scholar
  17. Kohonen T (1982) Self-organized formation of topologically correct feature maps. Biological Cybernetics, 43: 59–69.CrossRefGoogle Scholar
  18. Kohonen T (1990) The Self-Organizing Map. Proc. IEEE, 78(9): 1464–1480.CrossRefGoogle Scholar
  19. Lachtermacher G, Fuller JD (1994) Backpropagation in hydrological time series forecasting. In K. W. Hipel, A. I. McLeod, U. S. Panu, and V. P. Singh (Eds), Stochastic and Statistical Methods in Hydrology and Environmental Engineering, Kluwer Academic Publishers, Dordrecht, pp. 229–242.CrossRefGoogle Scholar
  20. Maier HR, Dandy GC (1997) Determining inputs for neural network models of multivariate time series. Microcomputers in Civil Engineering, 12(5): 353–368.CrossRefGoogle Scholar
  21. Maier HR, Dandy GC (2000a) Application of Artificial Neural Networks to Forecasting of Surface Water Quality Variables: Issues Applications and Challenges. In R. S. Govindaraju and A. R. Rao (Eds), Artificial Neural Networks in Hydrology, Kluwer Academic Publishers pp. 348.Google Scholar
  22. Maier HR, Dandy GC (2000b) Neural networks for the prediction and forecasting of water resources variables: a review of modelling issues and applications. Environmental Modelling and Software, 15: 101–124.CrossRefGoogle Scholar
  23. Maier HR, Dandy GC, Burch MD(1998) Use of artificial neural networks for modelling cyanobacteria Anabaena spp. in the River Murray, South Australia. Ecological Modelling, 105: 257–272.CrossRefGoogle Scholar
  24. Maier HR, Sayed T, Lence BJ(2000) Forecasting cyanobacterium Anabaena spp. using Bspline neurofuzzy models, 2nd International Conference on Applications of Machine Learning to Ecological Modelling, Adelaide, Australia.Google Scholar
  25. Masters T (1993) Practical Neural Network Recipes in C++, Academic Press, San Diego, 493 pp.Google Scholar
  26. Masters T (1995) Neural, Novel and Hybrid Algorithms for Time Series Prediction, John Wiley and Sons, New York, 514 pp.Google Scholar
  27. NeuralWare (1998) Neural Computing: A Technology Handbook for NeuralWorks Professional II/PLUS and NeuralWorks Explorer, Aspen Technology Inc., USA, 324 pp.Google Scholar
  28. Simpson AR, Dandy GC, Murphy LJ (1994) Genetic algorithms compared to other techniques for pipeline optimisation. Journal of Water Resources Planning and Management, ASCE, 120(4): 423–443.Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2003

Authors and Affiliations

  • G. J. Bowden
  • G. C. Dandy
  • H. R. Maier

There are no affiliations available

Personalised recommendations