Answers are grounded in this paper and cite their sources.
In this study, energy dissipation of flow over stepped spillways in skimming flow condition was modelled and predicted using multivariate adaptive regression splines (MARS) techniques. To develop MARS model-related data-set published in the literature was applied. To compare the performance of MARS model with other types of soft computing techniques, multilayer perceptron neural network (ANN) was prepared as a common type of soft computing method. To define the most effective parameters on energy dissipation, the Gamma Test and sensitivity analysis of ANN along with the development of MARS model were carried out. Results showed that the drop number and ratio of critical depth to the height of step are the most important parameters in energy dissipation. Results of modelling and predicting energy dissipation using MARS and ANN showed that both models are accurate, whereas MARS model is a bit more accurate compared to ANN.
No takes yet. Share an insight, caveat, or question.
Parsaie et al. (2016) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: