Filtern
Dokumenttyp
- Dissertation (2) (entfernen)
Sprache
- Englisch (2) (entfernen)
Referierte Publikation
- nein (2)
Schlagworte
- CFD (2) (entfernen)
This thesis addresses numerical simulations of self-compacting concrete (SCC) castings and suggests a novel modelling approach that treats reinforcement zones in a formwork as porous media.
As a relatively new field in concrete technology, numerical simulations of fresh concrete flow can be a promising aid to optimise casting processes and to avoid on-site casting incidents by predicting the flow behaviour of concrete during the casting process. The simulations of fresh concrete flow generally involve complex mathematical modelling and time-consuming computations. In case of a casting prediction, the simulation time is additionally significantly increased because each reinforcement bar occurring in succession has to be considered one by one. This is particularly problematic when simulating SCC casting, since this type of concrete is typically used for heavily reinforced structural members. However, the wide use of numerical tools for casting prediction in practice is possible only if the tools are user-friendly and simulations are time-saving.
In order to shorten simulation time and to come closer to a practical tool for casting prediction, instead to model steel bars one by one, this thesis suggests to model zones with arrays of steel bars as porous media. Consequently, one models the flow of SCC through a reinforcement zone as a free-surface flow of a non-Newtonian fluid, propagating through the medium. By defining characteristic parameters of the porous medium, the influence on the flow and the changed (apparent) behaviour of concrete in the porous matrix can be predicted. This enables modelling of any reinforcement network as a porous zone and thus significantly simplifies and fastens simulations of reinforced components’ castings.
Within the thesis, a computational model for SCC flow through reinforced sections was developed. This model couples a fluid dynamics model for fresh concrete and the macroscopic approach for the influence of the porous medium (formed by the rebars) on the flow. The model is implemented into a Computational Fluid Dynamics software and validated on numerical and experimental studies, among which is a large-scale laboratory casting of a highly reinforced beam. The apparent rheology of concrete within the arrays of steel bars is studied and a methodology to determine unknown input parameters for the porous medium is suggested. Normative tables defining characteristic porous medium parameters as a function of the topology of the rebar zone for different reinforcement cases are generated. Finally, the major contribution of this work is the resulting numerical package, consisting of the numerical solver and the parameter library. The thesis concludes on the ability of the porous medium analogy technique to reliably predict the concrete casting behaviour, while being significantly easier to use and far less time consuming than existing tools.
A new safety characteristic named “dustiness” according to the German guideline VDI 2263 – part 9 is investigated. Dustiness describes the tendency of a powder to form airborne dust by a prescribed mechanical stimulus. Dusts often behave differently in a dust/air mixture or in the case of a dust explosion, even if they have comparable physical properties such as particle size and density. In order to look into the effects of dustiness on dust cloud Formation and explosion properties experiments and simulations in a 75 L vertical dust Dispersion glass tube apparatus were carried out. In a second step industrial-scale experiments were carried out in a 50 m³ silo.
Experiments showed that particle size and density are not the only factors which influence dustiness, since the chosen dusts with comparable densities and particle size distributions showed very different behavior in the flow. Other dust properties such as particle shape, specific surface area, humidity and agglomeration processes have an influence which can outweigh size and density. Preliminary explosion experiments showed that dustiness has an influence on the reduced explosion pressure and pressure rise in a vented 75 L test apparatus. In order to verify the results for applications in the process industries further tests with different settings were carried out in industrial-scale experiments. First dust concentration
measurements were done in order to evaluate the reproducibility of filling processes.
Experiments showed that single tests differed by 30 % and more from the average depending on dust sample and filling method. First explosion experiments with a worst-case
scenario in terms of high turbulence and homogenous dust distribution showed that the maximum reduced explosion pressures were well below the calculated values. Reduced
explosion pressures and rates of pressure rise of the hree tested dust were as their Explosion characteristics pmax and KSt let suggest.
The Euler/Lagrange and the Euler/Euler approaches were compared simulating dust/air mixtures. Especially sedimentation and the ability of the approaches to simulate the tendency of dust to stay airborne were investigated. The Euler/Lagrange approach is better
suited for simulating local dust concentrations, particle size distributions and particle forces.
With the Euler/Euler method it is possible to achieve fast solutions for one specified diameter.
The computational fluid dynamics code ANSYS CFX R14 was used for all simulations.