Numerical Modelling of Multiphase Flow With Application to Industrial Processes

Multiphase flows are witnessed often in nature and the industry. Simulating the behaviour of multiphase flows is of importance to scientists and engineers for better prediction of phenomena and design of products. This thesis aims to develop a multiphase flow framework which can be applied to industrial applications such as placement of concrete in construction and proppant transport in oil and gas. Techniques available in literature to model multiphase flows are systematically introduced and each of their merits and demerits are analysed. Their suitability for different applications and scenarios are established. The challenges surrounding the placement of fresh concrete in formwork is investigated. Construction defects, the physics behind these defects and existing tests used to monitor fresh concrete quality are evaluated. Methods used to simulate fresh concrete flow as an alternative to experiments are critically analysed. The potential benefits of using numerical modelling and the shortcoming of the existing approaches are established. It is found that the homogeneous Bingham model is currently the most widely used technique to model fresh concrete flow. Determining the Bingham parameters for a given concrete mix remains a challenge and a novel method to obtain values for them is demonstrated in this work. The Bingham model is also applied to a full-scale tremie concrete placement procedure in a pile. Knowledge on the flow pattern followed by concrete being placed using a tremie is extracted. This is used to answer questions which the industry currently demands. The need for a more sophisticated model is emphasised in order to obtain an even greater understanding of fresh concrete flow behaviour. A CFD-DEM framework in which the multiphase nature of concrete is captured is developed. To validate this framework a new benchmark test is proposed in conjunction with the fluidised bed experiment. A comparative study of the drag models used in CFD-DEM approaches is performed to systematically assess each of their performances. CFD-DEM modelling is then applied to model fresh concrete flow and its potential to model defect causing phenomena is demonstrated. A model to capture more complex behaviours of concrete such as thixotropy is introduced and demonstrated.

Experimental Insights into Concrete Flow-Regimes Subject to Shear-Induced Particle Migration (SIPM) during Pumping

In this paper, the authors have focused on shear-induced particle migration (SIPM), its effect on concrete flow patterns, and lubricating layer formation during pumping. For this purpose, various volume-fractions ϕ of aggregates were selected. The particle migration was analyzed by applying two methods: sampling hardened concrete exposed to pumping and performing X-ray microcomputed tomography (μCT) and image analysis to determine the thickness of the lubricating layer due to SIPM. The results indicate that the first approach is unsuitable due to the nearly equal molecular density of particles and matrix. The second approach indicated that the actual thickness of the lubricating layer depends on the discharge rate as well as on ϕ and viscosity of concrete bulk; hence, it cannot be defined as a constant parameter for all concrete mixtures. Additionally, the concrete pipe-flow pattern, i.e., plug versus shear flow, was captured and studied while considering pumping pressure and discharge rate. It was concluded that particle migration is essential in the cases of both flowable and very flowable concretes with a high volume-fraction of solids. The changes in rheological properties caused by SIPM are severe enough to influence the definition of the flow pattern as plug or shear and the discharge rate of pumped concrete as well.

Numerical simulation of fresh concrete flow: insight and challenges

Recent developments in concrete technology are advancing into a scientific-based approach, where both experimental and numerical simulations are utilised to achieve an optimum mix design and an effective placement into formwork at the jobsite. Since the load carrying capacity and service life of concrete structures is fully dependent on the success of the placement process, researchers all over the world have started to work on casting prediction tools using different numerical software. However, a lot of work is still to be done in order to properly model the large-scale flow processes. This is because fresh concrete is a very complex material and its simulations involve complex material models and extensive computations. An exact material model of fresh concrete does not exist, and the researchers use diverse approximations to depict concrete flow. In this paper, we identify the main challenges for modelling fresh concrete and review the existing simulation methods. The advantages, disadvantages and application fields are discussed, including future perspectives for having numerical tools for practical use.

Numerical Modelling of Self-Compacting Concrete Flow

With the appearance of Self-Compacting Concrete (SCC) that streams uninhibitedly, under the sole impact of gravity, the desire for issue free and unsurprising castings even in complex cases, spurged the recreation of solid stream as a way to demonstrate and anticipate solid functionality. To accomplish total and dependable structure loading up with smooth surfaces of the solid, the fortified formwork geometry must be perfect with the rheology of the new SCC. Anticipating stream conduct in the formwork and connecting the required rheological parameters to stream tests performed on the site will guarantee an improvement of the throwing procedure. In this theory, numerical reproduction of solid stream is explored, utilizing both discrete just as constant approaches. The discrete molecule model here fills in as a way to mimic subtleties and marvels concerning totals demonstrated as individual items. The here gave cases are reenacted round particles. Be that as it may, it is conceivable to utilize nonspherical particles too. Total surface harshness, size and viewpoint proportion might be modeles by molecule erosion, size and bunching a few circles into framing the ideal molecule shape. The consistent methodology has been utilized to mimic huge volumes of cement. The solid is displayed as a homogeneous material, specific impacts of totals, for example, blocking or isolation are not represented. Great correspondence was accomplished with a Bingham material model used to reenact solid research center tests (for example droop stream, L-box) and structure filling. Stream of cement in an especially clogged segment of a twofold tee chunk just as two lifts of a multi-layered full scale divider throwing were reenactedsucessfully. A huge scale quantitative investigation is performed rather easily with the constant methodology. Littler scale subtleties and marvels are better caught subjectively with the discrete molecule approach. As PC speed and limit always develops, recreation detail and test volume will be permitted to increment. A future converging of the homogeneous liquid model with the molecule way to deal with structure particles in the liquid will highlight the progression of concrete as the physical suspension that it speaks to. One single ellipsoidal molecule falling in a Newtonian liquid was considered as an initial step.