Appraisal of Runoff Through BPNN, RNN, and RBFN in Tentulikhunti Watershed: A Case Study

摘要

Three unlike neural network models, (i) radial basis fewer network (RBFN) model, (ii) recurrent neural network (RNN), and (iii) backpropagation neural network (BPNN) model, are employed to guesstimate runoff at Tentulikhunti watershed, Odisha, India. Scenarios with minimum temperature, maximum temperature, and precipitation are considered for experiencing the impact on runoff. In Tentulikhunti watershed, RNN executes preeminent by means of architecture 4-3-1 succeeding tangential sigmoid transfer function. Equally, RBFN and BPNN perform in parallel with small deviation of prediction for predicting runoff.