Mathematical Expression of Energy Dissipation of Flow over Stepped Spillways Using Genetic Programing

Stepped spillways are affective structures for energy dissipation. In this study, to mathematical expression of effect of involved parameters on energy dissipation, a genetic programing (GP) technique was used. The GP categorized in smart function fitting methods that automatically define the most effective parameters and assigned them more weight in final model derived. Involved parameters on energy dissipation of flow are drop number, number of steps, ration of critical depth to height of steps, longitudinal slop of weir, and Froude number of flow at toe of spillways. To compare the performance of GP with other type of soft computing techniques, the multilayer perceptron neural network was developed as well. Results indicated that derived model from GP model has suitable to performance to mathematical expression of involved parameters on energy dissipation. Reviewing the structure of derived model from GP indicated that drop number and ratio of critical depth to the height of steps are the most effective parameters on energy dissipation. Comparison the performance of MLP model with GP showed that the accuracy of MLP is a bit more than the GP. Sensitivity analysis of MLP upheld the results of GP in term of effective parameters.