Estimation of the longitudinal dispersion coefficient via a fusion of optimized models

Abstract Determination of the longitudinal dispersion coefficient (LDC) is fundamental to development strategies for environmental management river systems. This paper presents an integrated model estimation by a fusion optimized intelligent models (optimized neural network (ONN), fuzzy inference system (OFIS), and support vector regression (OSVR)) via committee machine (CM), with optimization done Bat-inspired algorithm (BA). The eliminates associated loss accuracy models, which direct consequence improper adjustment parameters (weights biases in network, membership's functions system, user-defined regression). Data gathered from literature are employed validate proposed model. A comparison between machine, based on statistical parameters, shows that can attain high accuracy. Sensitivity analysis (SA) contribution each ranks ascending order as regression, significantly correlated prediction.