Estimation of Critical Velocity for Slurry Transport through Pipeline Using Adaptive Neuro-Fuzzy Interference System and Gene-Expression Programming

Abstract

One of the important trends of development of hydromechanization in hydraulic engineering is the transport of solids in the form of slurries. Slurry is a thick suspension of solids in a liquid. Clogging of the pipeline carrying slurry will not occur if the velocity of the slurry is more than some critical value. Critical flow velocity, which is the minimum velocity to maintain all solid particles in a suspension condition, is the important design parameter in slurry transport through pipelines. Gene-expression programming (GEP) and adaptive neuro-fuzzy inference system (ANFIS) models are developed in this study for the estimation of critical velocity. The estimated critical velocity by GEP and ANFIS models are compared with existing empirical equations and it is found that the ANFIS model produces better results compared with GEP and other existing equations.