Evaluation of selected formulas and neural network model for predicting the Longitudinal Dispersion Coefficient in river

Pollution transmission mechanism in rivers is more complex. Longitudinal dispersion coefficient (LDC)is one of the most important parameter in the field of studding on computer modeling of river water quality that leads to improve managing the river water quality for human healthily problems. Several ways as empirical formulas and artificial intelligent techniques have been proposed for predicting the LDC by researcher. This is necessary to evaluate the performance of these equations and models for LDC prediction. In this study, 12 of empirical formulas were collected and have been evaluated by data set. The best accuracy is related to the Tavakollizadeh and Kashefipour formula( ). And to assess the performance of these equation in a case study problem calculating the LDC for Severn River was considered. For Severn River the Tavakollizadeh and Kashefipour equation has best accuracy ( ). To reach more accuracy in calculating the LDC; the MLP model has been developed by data set and it used again to predict the LDC for Severn River. The result shows that the Multilayer Perceptron (MLP) neural network model has acceptable accuracy ( ) in using to case study problem.

Keywords:

Longitudinal dispersion coefficient , empirical formula , Multilayer Perceptron (MLP) neural network , dispersion routing method