application of soft computing methods in the analysis of Velocity field in dividing channel

The simplest water- diversion method in irrigation systems, agriculture and drainage systems is using intake. Measuring the mean velocity is one of the essential hydraulic parameters in increasing the efficiency of intake. In this study the mean velocity will be predicted for different width ratios of a intake using ANN- MLP neural network model. In order to do that the flow field within a 90- degree intake was first three- dimensionally simulated using ANSYS- CFX. k- ω turbulence model has been used in this simulation. The neural network used in this study includes 3 inputs coordinates (Y*), the ratio of the branch channel to the main channel, and the linear mean velocity (V*line). V*line is the average velocity in the horizontal column of the branch channel which has been measured by numerical model. A comparison between the numerical and experimental results indicate the proper accuracy of the numerical model in predicting the specifications of the flow field within the intake in addition the comparison between the obtained results from ANN model prediction and the experimental results indicates the acceptable degree of accuracy of the artificial neural network model in predicting the flow mean velocity in intake and for different width ratios.

Keywords:

intake , mean velocity , width ratios , ANN model