Modeling runoff using feed forward-back propagation and layer recurrent neural networks

Abstract

This work describes the function of (i) feed forward-back propagation network (FFBPN) model and (ii) layer recurrent network (LRN), to predict runoff as a function of rainfall, temperature, and evapotranspiration loss. For model architecture, the criteria for evaluation are mean square error training, testing, root mean square error training, testing, and coefficient of determination. Overall results found that LRN performs best as compared to FFBPN for predicting runoff in the watershed. This result will help for planning, design, and management of hydraulic structures in the vicinity of the watershed.