A wavelet-linear genetic programming model for sodium (Na+) concentration forecasting in rivers

Created by W.Langdon from gp-bibliography.bib Revision:1.3872

@Article{Ravansalar:2016:JH,
  author =       "Masoud Ravansalar and Taher Rajaee and 
                 Mohammad Zounemat-Kermani",
  title =        "A wavelet-linear genetic programming model for sodium
                 (Na+) concentration forecasting in rivers",
  journal =      "Journal of Hydrology",
  volume =       "537",
  pages =        "398--407",
  year =         "2016",
  ISSN =         "0022-1694",
  DOI =          "doi:10.1016/j.jhydrol.2016.03.062",
  URL =          "http://www.sciencedirect.com/science/article/pii/S0022169416301767",
  abstract =     "Summary The prediction of water quality parameters in
                 water resources such as rivers is of importance issue
                 that needs to be considered in better management of
                 irrigation systems and water supplies. In this respect,
                 this study proposes a new hybrid wavelet-linear genetic
                 programming (WLGP) model for prediction of monthly
                 sodium (Na+) concentration. The 23-year monthly data
                 used in this study, were measured from the Asi River at
                 the Demirkoepru gauging station located in Antakya,
                 Turkey. At first, the measured discharge (Q) and Na+
                 datasets are initially decomposed into several
                 sub-series using discrete wavelet transform (DWT).
                 Then, these new sub-series are imposed to the ad hoc
                 linear genetic programming (LGP) model as input
                 patterns to predict monthly Na+ one month ahead. The
                 results of the new proposed WLGP model are compared
                 with LGP, WANN and ANN models. Comparison of the models
                 represents the superiority of the WLGP model over the
                 LGP, WANN and ANN models such that the Nash-Sutcliffe
                 efficiencies (NSE) for WLGP, WANN, LGP and ANN models
                 were 0.984, 0.904, 0.484 and 0.351, respectively. The
                 achieved results even points to the superiority of the
                 single LGP model than the ANN model. Continuously, the
                 capability of the proposed WLGP model in terms of
                 prediction of the Na+ peak values is also presented in
                 this study.",
  keywords =     "genetic algorithms, genetic programming, Water
                 quality, Na+ concentration, Data driven model",
}

Genetic Programming entries for Masoud Ravansalar Taher Rajaee Mohammad Zounemat-Kermani

Citations