Indian Journal of Science and Technology
Year: 2016, Volume: 9, Issue: 44, Pages: 1-7
N. Vetrivel* and K. Elangovan
*Author for correspondence
Department of Civil Engineering, PSG College of Technology, Coimbatore - 641004, Tamil Nadu, India; [email protected]
Objective: To evaluate the ability of prediction performance of ANFIS (Adaptive Neuro Fuzzy Inference System) model on groundwater level fluctuation in Lower Bhavani River Basin (LBRB). Method: To improve the accuracy of prediction of ANFIS model, WT (Wavelet Transform) and CWTFT (Continuous Wavelet Fast Fourier Transform) are performed as preprocessing techniques. Model development is carried out through three different input parameters (duration, groundwater recharge and groundwater discharge) and one output parameter (groundwater fluctuation) for the period of 2009-2015 on monthly stress basis. Findings: Based on the comparative prediction by the models CWTFT-ANFIS and WTANFIS over conventional ANFIS model through calibration and validation process on different reaches of the study area, optimum model is identified. Statistical indices are performed to measure the best fitting and trend identification for the each prediction.
Keywords: ANFIS, CWTFT, Fluctuation, Groundwater, Lower Bhavani River Basin, Wavelet
Subscribe now for latest articles and news.