Browsing by Author "Mehr, Ali Danandeh"

Filter results by typing the first few letters
Now showing 1 - 2 of 2
  • Thumbnail Image
    Item
    On the calibration of multigene genetic programming to simulate low flows in the Moselle River
    (Uludağ Üniversitesi, 2016-11-27) Mehr, Ali Danandeh; Demirel, Mehmet C.
    The aim of this paper is to calibrate a data-driven model to simulate Moselle River flows and compare the performance with three different hydrologic models from a previous study. For consistency a similar set up and error metric are used to evaluate the model results. Precipitation, potential evapotranspiration and streamflow from previous day have been used as inputs. Based on the calibration and validation results, the proposed multigene genetic programming model is the best performing model among four models. The timing and the magnitude of extreme low flow events could be captured even when we use root mean squared error as the objective function for model calibration. Although the model is developed and calibrated for Moselle River flows, the multigene genetic algorithm offers a great opportunity for hydrologic prediction and forecast problems in the river basins with scarce data issues.
  • Thumbnail Image
    Item
    Streamflow and sediment load prediction using linear genetic programming
    (Uludağ Üniversitesi, 2018-07-17) Mehr, Ali Danandeh; Şorman, Ali Ünal
    Daily flow and suspended sediment discharge are two major hydrological variables that affect rivers’ morphology and ecosystem, particularly during flood events. Artificial neural networks (ANNs) have been successfully used to model and predict these variables in recent studies. However, these are implicit and cannot be simply used in practice. In this paper, linear genetic programming (LGP) approach has been suggested to develop explicit models to predict these variables in two rivers in Iran. The explicit relationships (prediction rules) evolved by LGP take the form of equations or program codes, which can be checked for its physical consistency. The results showed that the LGP outperforms ANNs to get global maximum and minimum discharges providing lowest root mean squared error and higher coefficient of efficiency both for training and validation periods.