Closed-Form Solution for Flow Field in Curved Channels in Comparison with Experimental and Numerical Analyses and Artificial Neural Network

In this study the velocity field in a 90° bend channel was investigated by using artificial intelligence, analytical, experimental, and numerical methods on the basis of an extensive review of literature. First, a neural network approach was proposed for predicting the velocity components. Then a closed-form solution for a flow field of a curved channel was imparted. After that ANSYS-CFX software was used for 3D simulation of the flow field in the considered bend in 3 phases (air+water+sediment). Finally, the results of all applied methods were compared with each other. Results showed that in general, all used models are in good agreement with experimental results and have some advantages and disadvantages. But in most cases, Artificial Neural Network (ANN) model and numerical method had better performance than analytical solution.