Application of Soft Computing Techniques for Particle Froude Number Estimation in Sewer Pipes

Abstract

Sedimentation in sewer networks is a major problem in urban hydrology. In comparison to the well-known classic sediment transport models, this study investigates the capabilities of soft computing methods, including multigene genetic programming (MGGP), gene expression programming, and multilayer perceptron to derive accurate sewer design models. A wide range of experimental data sets comprising fluid, flow, sediment, and pipe features was used to develop new models under the nondeposition with a deposited bed self-cleansing condition. The results showed better performances of the new models compared to the conventional ones in terms of statistical performance indices. The proposed MGGP model was found superior to its counterparts. It is an explicit model motivated to be used for self-cleansing sewer pipes design in practice.