Abstract
Prediction of boundary shear force distributions in open channel flow is crucial in many critical engineering problems such as channel design, calculation of losses and sedimentation. During floods, part of the discharge of a river is carried by the simple main channel and the rest is carried by the floodplains. For such compound channels, the flow structure becomes complicated due to the transfer of momentum between the deep main channel and the adjoining floodplains. The complexity further increases when dealing with a compound channel with non-prismatic floodplains. Knowledge of momentum transfer at the different interfaces originating from the junction between the main channel and floodplain can be acquired from the distribution of boundary shear in the subsections. The calculation of boundary shear and depth average velocity in non-prismatic compound channel flow is more complex and simple conventional approaches cannot predict the boundary shear and depth average velocity with sufficient accuracy. Hence, in this area, an easily implementable technique, the Artificial Neural Network can be used for predicting the boundary shear and depth average velocity at different sections of a converging compound channel for different geometry and flow conditions. The model’s performance has lead satisfactory results. Statistical error analysis is also carried out to know the degree of accuracy of the model.
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have