Estimation of maksimum velocity of flow around a horizontal circular cylinder with artificial neural network

28058

Two-dimensional submerged circular cylinders are widely used structural elements in engineering practices, and therefore the studies on the interaction of such bluff bodies with the fluid flow are important for design considerations. Velocity distribution around a cylindrical body is an important parameter in determining the boundary layer development, separation point and boundary shear stress. In this study, the velocity field of flow around a horizontal circular cylinder is investigated for the three different Reynolds number (ReD= 840, 4150 and 9500) and five different gap ratio (G/D= 0.2, 0.3, 0.6, 1.0 and 2.0). PIV (Particle Image Velocimetry) technique is used to measure the flow velocities. The Multiple-Linear Regression (MLR) and Artifical Neural Network (ANN) model were developed to determine the relationship among the gap ratio (G/D), Reynolds number, (ReD), and α for estimating the dimensionless maximum value of tangential velocity of flow around an isolated horizontal circular smooth cylinder in an open channel. The comparisons of results show that the ANN model gives more accurate results than the MLR model.

63

Velocity Distribution , Boundary Layer , MLR , ANN

Erciyes University Journal Of The Institute Of Science an‎d Technology

70

Erciyes University Journal Of The Institute Of Science an‎d Technology