Application of artificial neural network and Soil and Water Assessment Tools in evaluating power generation of small hydropower stations

Abstract

The objective of this study is to estimate the river’s capacity power generation and identify proper sites to install small hydropower stations The innovation in this study is estimating the river flow rate using different versions of the Soil and Water Assessment Tools (SWAT) model. Evaluating the flow rivers using different models helps to increase the accuracy of estimating river discharge and it can be effective in accurately identifying the river’s hydroelectric potential. In this investigation, an Artificial Neural Network (ANN) alongside the SWAT model has been used to increases the model’s ability to estimate streams flow. Also, an optimized algorithm called Improved the Pathfinder Optimizer (IPFO) have been utilized to reduce error during the learning process. The results of this study showed that among the three proposed models, SWAT-ANN-IPFO has the best estimate for river flow. Then, using the SWAT-ANN-IPFO model and Geographic Information system (GIS) software assessed the power potential sites along the rivers. In this study, the selection of proper sites for the installation of small hydropower stations based on physiographic and hydrological characteristics has been evaluated. Two criteria, slope and discharge for this purpose have been investigated. In this study, in total, 2,031 suitable points identified for installing hydroelectric power stations. The class 4 waterways are the most prone to installing hydroelectric power plants, and the identified sites are almost 846 because it has more discharge volume and more head differences than other order streams.