Wavelet-ANN and wavelet-ANFIS based runoff modeling
Loading...
Date
2018-01
Authors
Journal Title
Journal ISSN
Volume Title
Publisher
G.B. Pant University of Agriculture and Technology, Pantnagar - 263145 (Uttarakhand)
Abstract
Land and water are two vital natural resources for sustaining life on earth. These resources are getting degraded by an alarming rate, due to over exploitation and unscientific management of the resources owing to increasing population and industrialization. For the scientific and proper management of these resources, it is needed to having a tool for precise prediction of future events. The present study is an attempt to develop a mathematical model which will be efficient in data correctness and prediction of future events. Present study has been carried out to simulate, forecast and compare runoff yield from a watershed using a hybrid artificial neural network modeling technique combined with wavelet and fuzzy logic analysis. In the present study the Nagwan and Vamsadhara watershed of the south-east monsoon were selected. The data of the entire monsoon period, starting from June 1st to September 30th of each year were used for the model development and verification of developed models. One computer progamme in C language was developed for resolving the observed daily data of runoff. For modeling of Artificial neural network and Adaptive neuro fuzzy inference system, a software NEUROSOLUTION was used. The model performance was checked through seven selected performance evaluation criteria, viz. Mean square error, Normalize mean square error, Correlation coefficient, Percentage error, Nash-Sutcliffe efficiency, Percent bias and RMSE-observations Standard deviation ratio. The results of the study reveal that, resolution of all input data makes the model more acceptable at the same time one should avoid increasing complexity of model which may adversely affect the performance of the model.
Description
Keywords
null