Prediction of Monthly Mean Flow to Shahid Abaspour dam Reservoir Using Time Series Models (Box- Jenkins) and Artificial Neural Network

The optimal exploitation of water from a reservoir requires a comprehensive knowledge of future availability of water resources. In this case the amount of water that will be available in the future is important. Also, we need to examine the flows at the dam from a short-term perspective. The main aim of the shortterm estimation of the runoff is to evaluate the risk of floods. This is necessary to avoid overflowing and to minimize damage. In order to facilitate forecasting of the water resources, many different techniques have been developed through the years.
In this paper, using monthly data (since 1957 to 2003) obtained from Hydrometric station at Shaloo bridge (upstream of Shahid Abaspour reservoir), the Auto Regressive Integrated Moving Average (ARIMA) model and artificial neural networks (ANNs), prediction of monthly mean flow to shahid Abaspour reservoir was accomplished. On the basis of comparison of the results of the models with measured data and computation of standard deviation, variation coefficient, sum of square error, root mean square error and correlation coefficient of the results, the performance of ARIMA(0,1,1)´(1,1,1) model shows an improvement in comparison with the ANNs model.