Continuous flow modeling of soil and sand water mixtures
CEDA Dredging Days 2015 - Innovative Dredging Solutions for Ports, Rotterdam
Goeree JC, Keetels GH and van Rhee C - Delft University of Technology, the Netherlands
Abstract: In dredging large volumes of sand are transported mostly hydraulically. The sand is suspended in water and forms a mixture. A thorough understanding of the flow behavior of sediment for different volume concentrations is eminent. The volume concentrations of sand in these mixtures can vary from 0% to 50%, i.e. from water to dense suspensions. At this volume concentration mixtures still behave as a liquid. At volume concentrations of solids ranging from 50% to 60% the mixture is densely packed and the sediment behaves more as a solid. This is the case for dense granular flows.
It is a considerable challenge describing the wide range of flow behaviors commonly known in dredging. For instance in a settling mixture under gravity, particles accumulate at the bottom of the domain forming a granular bed. Therefor, a solid/mixture interface needs to be tracked distinguishing the fluid and solid part. Furthermore a remeshing algorithm needs to be used discretization and solving both the solid and liquid part of the mixture. This approach has been followed by van Rhee (2002). However, implementation of this method is cumbersome. In this paper another approach is introduced. A numerical model has been developed in which the sediment/water mixture and dense granular flow behavior is described. This is done using a continuous flow model. The sediment water mixture, at lower volume concentrations, is described with the Navier-Stokes equations. The interactions of the water and sediment are modeled using closure relations. Here the closure relation of Richardson and Zaki is incorporated.
A granular flow model is used modeling the mixture at higher volume concentrations. The continuous flow model describes sediment water mixtures ranging from densely packed sediments to dilute suspensions. The model has been validated with several experiments found from literature. The computational and experimental results agree well.
Key words: Sediment, particle, numerical, granular flow, Bingham, numerical