Effects of Width Ratios and Deviation Angles on the Mean Velocity in Inlet Channels Using Numerical Modeling and Artificial Neural Network Modeling

Abstract

Dividing open channels are varied types of open channel structures used to provide water for irrigation channels, agriculture and wastewater networks. In the present study, artificial neural network (ANN) and computational fluid dynamic models are used to calculate the mean velocity in different dividing angles within branch channels. First, the ANSYS-CFX model is used to simulate the flow pattern within the branch at a 90° angle. Results of the CFX model correspond fairly well to the results of the experimental model with a mean absolute percentage error (MAPE) of 5 %. After verification, two CFX models are generated in 30° and 60° angles in different width ratios of 0.6, 0.8, 1, 1.2, and 1.4, and the mean velocities are obtained by a flowmeter. The ANN model is then trained and tested by a set of experimental and CFX data. The ANN model presented an acceptable level of accuracy in predicting the dividing open channel mean flow velocity with a mean value R 2 of 0.93. A comparison of the results indicated that the ANN model with 1.8 % MAPE performs better under 0.8 m width ratio. The MAPE within the 0.8 width is equivalent to the ratios 1.58, 1.87, and 2.04 % in 30°, 60°, and 90° deviation angles, respectively, and therefore the model performs better at the 30° angle.