Neurofuzzy-Based GMDH-PSO to Predict Maximum Scour Depth at Equilibrium at Culvert Outlets

摘要: AbstractIn this study, the neurofuzzy-based group method of data handling (NF-GMDH) as an adaptive learning network was utilized to predict maximum scour depth at equilibrium downstream culvert outlet structures. The NF-GMDH developed using particle swarm optimization (PSO). Effective variables on included those sediment size outlets, geometry and flow characteristics upstream culvert. Training testing performances NF-GMDH-PSO were carried out nondimensional sets that collected from literature. results model compared with gene-expression programming (GEP) traditional equations. produced a lower error prediction than obtained other models. Also, most effective parameter wa...