tailieunhanh - Suspended sediment concentration distribution in an alluvial channel flow
In this paper a numerical solution for the problem on suspended sediment concentration distribution in an alluvial channel flow has been computed on the base of PROFILE model proposed by L. 9· Van Rijn for the case of uniform flow. | I M ! ~ T~p H9c JournM of Mechanics, NCNST of Vietnam T. XVI, 1994, No 1 (15 - 19) SUSPENDED SEDIMENT CONCENTRATION DISTRIBUTION IN AN ALLUVIAL CHANNEL FLOW DANG HUU CHUNG Institute of Mechanics, NCNST of Vietnam SU1v1MARY. In this paper a ~umerical solution for the problem on suspended sediment concentration dist!'ibution in an alluvial channel flow haS been computed on the base of PROFILE model proposed by L. 9· Van Rijn for the case of uniform flow. Although, in the present case, mathematical model is quite simple, but it is exact enough to apply to somme problems in practice. The aiJ? of the paper is that the author would like to use the finite differenc;;e method for the same problem. The input data was used from experiment flume. The result showed that the concentration distribution fast decreased along the channel and concentration gradients became very small at the sections situated far enough from the upstream. Besides, a computing programme with the ability of graphic expression was established. §1. INTRODUCTION The computation of vertical distribution of suspended sediment in a flow is one of the problems relating ·to sedimentation and morphological processes. It really plays important role in many problems in practice, specially, in dealing with the deposit at estuaries as well as the management of the quality water. Nowadays, it is interesting many scientists in the world. The mathematical model consists of hydrodynamics equations for velocity field and the sediment diffusion-convection equation. The overview of this we can find in many papers (such as \3]). In this paper, however, only a special case of steady u'fiiform flow was studied. In point of View of economics it is fully reasonable to study such ·a problem. In fact, the errors in input data measurement are too big, according to L. C. Van Rijn it's about 25% \4]. Therefore in some cases we can consider the problem with a simpler model. In the following the implicit finite .
đang nạp các trang xem trước