Performance Evaluation of Artificial Neural Networks for Predicting Rivers Water Quality Indices (BOD and DO) in Hamadan Morad Beik River

One of the important factor for development in each region is the availability of appropriate water resources. In addition to water quantity quality is also of great importance. The aim of this study is to medel the qualitative indices (BOD, DO) of river water using multi-layer perceptron neural network. In this paper, the information and data from Morad Beik river of hamadan including 10 monthly parameters of water quality in a one-year period and at six stations were used to predict biological exygen demand (BOD) and dissolved oxygen (DO), as indices affecting water quality. Efficiency of the neural network model was evaluated by some statistical criteria including correlation coefficient (R), root mean square error (RMSE) and mean absolute error (MAE). In the optimum structure of neural network the correlations coefficient for BOD and DO were 0.986 and 0.969, and root mean square errors were 8.42 and 0.84 respectively. The results indicated the ability of multi-layers perceptron neural network as a suitable technique for simulating changes in BOD and DO indices.