Empirical model of predicting longitudinal velocity in sharp bend

Natural rivers are rarely straight and usually tend to take a winding course, called as meandering. Due to mutual interaction between flow and bed in river meanders, understanding the flow characteristics is an interesting issue in river engineering field. The main problem with describing the characteristics of flows in river bends is the existence of strong secondary flows and its impact on the distribution of longitudinal and transverse velocity. The issues of flow in river bends has been investigated by several researchers, and a variety of different equations for identifying the parameters of flow have been proposed. However, the proposed equations do not provide a good solution; this is especially the case in sharp bends in which strong secondary flows influence velocity patterns. In this research, firstly, taking the properties of longitudinal velocities in sharp bends into consideration, a new model for computing the velocity was proposed which includes two power and sinous sections. Then, the new model is compared with some longitudinal velocity models on bend. The results achieved from the study on 1930 bend showed that the new model not only has the properties of a longitudinal velocity in sharp bends but also enjoys less error than the other models. Also, comparing the new model with 3 series of data from 3 bends with different curvature radius to width, showed that the new model is able to have good predictions in different bends.