Adaptive neuro-fuzzy inference technique for estimation of light penetration in reservoirs

摘要: An adaptive neuro-fuzzy inference technique has been adopted to estimate light levels in a reservoir. The data were collected randomly from Doganci Dam Reservoir over number of years. input set is matrix with vectors time, depth, sampling location, and incident solar radiation. output vector representing measured at various depths. Randomization logarithmic transformations have applied as preprocessing. One-half the utilized for training; testing validation steps one-fourth each. system (ANFIS) built prediction model penetration. Very high correlation values between predictions real on measurements relatively low root mean square error obtained training, test, sets. Elimination overtraining problem was ensured by satisfying close all