Vol. 6, Issue 6 (2018)
Evapotranspiration modeling using different Heuristic neural network Approaches
Author(s): Sirisha Adamala and Monisha Perli
Abstract: To schedule irrigation properly, a grower must know the environmental demand for surface water. For the grower, this surface water loss occurs primarily through evapotranspiration (ETo), which is simply the amount of water returned to the atmosphere through evaporation (moisture loss from the soil, standing water, etc.) and transpiration (biological use and release of water by vegetation). In this study, the potentiality of different ANN models: multi-layer perceptrons (MLPs), radial basis neural networks (RBNNs) and generalized regression neural networks (GRNNs) were tested for different climatic locations in India. The performance indices used for comparing the above models include root mean squared error (RMSE), coefficient of determination (R2), mean absolute error (MAE), and the ratio of average output to the average target ETo values (R). The results revealed that though all models performed well in estimating or modelling ETo, the performance of GRNN models was superior with respect to low RMSE and MAE errors and high R2 values as compared to MLP and RBNN models.
Pages: 314-318 | 657 Views 102 Downloads
download (5831KB)
How to cite this article:
Sirisha Adamala, Monisha Perli. Evapotranspiration modeling using different Heuristic neural network Approaches. Int J Chem Stud 2018;6(6):314-318.
