In the design of equipment for the chemical industry, mathematic models are used for scaleup from laboratory scale to full-size systems. The models consist of mass, momentum, energy balances and closure models that describe the phenomena inside the investigated region. The phenomena that these closures describe are related to include flow conditions, interfacial areas, or thermodynamic properties. Experimental data is needed for the development and validation of these models. The first part of the thesis compares various analysis techniques to investigate fluid flow. The focus is to understand the benefits and drawbacks of each technique, how the raw data should be converted so that a comparison between experimental work modeling is possible. Experimental work consisted of single-phase and multiphase experiments. Single-phase experiments were made to validate the flow fields and mixing times in the systems. Multiphase experiments were needed to obtain information about the size distributions of the bubbles or particles, determine the parameters of mass transfer between the phases and the drag forces between the phases. The experimental work was conducted in laboratory-scale stirred vessels (ranging from 14 to 200 liters) and in a pipe flow. Various experimental techniques were used to obtain essential information about the phenomena occurring in the investigated systems. The experimental results showed that the inhomogeneities in fluid velocities and the interfacial areas between the phases inside the systems can be significant. In traditional scaleup, these inhomogeneities cannot be taken into account, and the models depend on the geometry and operating conditions of the system. This can easily lead to inefficient systems. Computational Fluid Dynamics (CFD) can provide information on local conditions in the equipment. However, we are still far away from the possibility of modeling all the phenomena in industrial-scale vessels, due to the computational capacity required. A possible trade-off between traditional modeling and direct numerical simulation is to develop closure models for CFD that are independent of the systems’ geometry and operating conditions. In the modeling section of this thesis, the focus was to compare the closure models that were implemented for CFD and validate them against experimental data. The investigation was focused on size distribution models of the dispersed phase, mass transfer models, drag models between the phases, viscosity models of the continuous phase and turbulence models. The results show that the closure models that have been developed for CFD still have a dependency on the system that they were developed for and validated in. Therefore, if these models are used, it must be ensured that the designed system is similar to the system where the model was developed.
|Translated title of the contribution||CFD mallien ja simulointien kokeellinen validointi kemiantekniikan sovelluksissa|
|Publication status||Published - 2012|
|MoE publication type||G5 Doctoral dissertation (article)|
- experimental validation
- fluid dynamics
- single phase