Author(s): Biswajit Nayk; Janhabi Meher; Minakshee Mahananda

Linked Author(s):

Keywords: MLR; ANFIS; ANN

Abstract: Outflow from a reservoir is generally various forms of use for different purposes like irrigation power generation, industrial water, and domestic water. Therefore, outflow depends on parameters like inflow water, requirement in various proposes, evaporation discharge during high flood level. In this paper, multiple linear regression (MLR) and adaptive neuro fuzzy inference system (ANFIS) have been developed in prediction of outflow of Hirakud Reservoir, Odisha. Three models were developed by taking input as reservoir inflows and outflows current and past time steps. The inputs are current and previous day inflows and previous day outflows which were selected through autocorrelation of inflows as well as outflows and cross-correlation between inflows and outflows, among which variables with high correlation values (> 0.7) were selected as input. The developed models were trained, tested, and validated using data of 21 years spanning from 1995 to 2015. Performance of MLR models was evaluated using four indicators such as R2, R, RMSE, and MAE.

DOI: https://doi.org/10.1007/978-981-97-6009-1_45

Year: 2022