Modeling the distribution of bubbles in liquid gas floating bed system by computational fluid dynamics (CFD) method.

Dissertation for M.Sc."

(Chemical Engineering - Thermodynamics and Kinetics)

Abstract:

Key words: computational fluid dynamics, bubbling, continuity and failure, floating bed reactors, multiphase systems, vertical flows Faoghi, PBM, DQMOM,

Recently, liquid gas floating bed reactors have received attention due to their wide application in chemical, petrochemical and environmental processes. On the other hand, with the progress made in computational methods and computer science, computational fluid dynamics (CFD) has become a new method with great potential to understand the effect of fluid dynamics in the performance of chemical reactors. . In this type of reactors, the role of bubbles in multi-phase processes caused by the breaking and collision of bubbles justifies the importance of this study.

Among the different numerical models for two-phase flow, the two-fluid model that tracks two phases using two sets of transition equations is one of the most suitable models. In the present study, the ability of the two-phase model and closed terms to simulate vertical and horizontal bubbly flow conditions has been investigated. It is also adopted from an equilibrium model to track the collision mechanisms between bubbles/bubbles and bubbles/vortices.

Also, in this study, the performance of the empirical relation of drag coefficient presented by Simont et al. (2007) is investigated. This model is compared with the distributed fluid particle model presented by Ishii-Zuber, which has been widely used in commercial software packages. Also, three variables including gas void fraction, average bubble diameter, and gas velocity were selected to evaluate the model with the experimental data of Hibiki et al. (2001). Previously, the class method presented by the model (MUSIG) is considered as one of the best methods to check the population balance equation (PBM) by discretizing the range of continuous sizes into sequences of discrete size classes. Compared to the MUSIG model, the main advantage of the (DQMOM) method is that the number of moments to be solved is generally small. In this thesis, the DQMOM model is developed and implemented in ANSYS FLUENT software to match the joining and breaking of gas bubbles. In the end, the simulation results have been confirmed with MTLOOP and TOPFLOW experimental data. Chapter 1: Generalities 1-1) Introduction Although in the early days of science development, mathematicians were looking for relationships that govern the performance of existing systems instead of predictions, but today, with the progress made, compared to experimental scientists They are ahead of the curve. Although experimental scientists are familiar with the mathematical solution of phenomena, they face many problems for their experiments. The most important problem related to fluid dynamics has not been solved mathematically for a long time, and those that have been solved are also facing the problem of taking a lot of time to perform mathematical operations. With the development of computers day by day this problem becomes easier and easier. And now, the most complicated of these problems, which are the important issues of heat transfer and fluids, can be solved by computer. Today, the science of computational fluid dynamics has become a powerful tool for researchers and engineers to analyze the behavior of fluid flow and heat transfer in systems with complex geometry and complex governing equations. The complexity of the equations governing the problem, the mutual influence of different physical phenomena, the transience of most engineering problems, the high costs of laboratory equipment and the limitation of using measuring devices in many scientific problems are among the reasons that limit the use of analytical and laboratory methods compared to numerical methods. Although modeling the reactor, determining its hydrodynamic parameters seems essential. The hydrodynamics of these reactors is strongly affected by their performance scale. (Due to the wide applications of these reactors in the industry, many efforts have been made to provide a reliable method for scaling up.) In the past, researchers used laboratory experiments to obtain the hydrodynamics of these reactors.) in the past, researchers used to conduct laboratory experiments to obtain the hydrodynamics of these reactors. The results of these experiments were not necessarily accurate in large scales and therefore could not be used as scaling laws. For example, the effects of the walls of a small reactor on the movement, formation and breaking of bubbles are known. It is also clear that this effect is different in larger reactors. However, the extent and manner of these differences are not known, and therefore the best way to achieve the hydrodynamics of large diameters is to conduct experiments in reactors with the same diameter, which of course is very expensive. With the help of CFD [1], the reactor can be simulated in real size and according to the results, the configuration and suitable conditions of the reactor can be predicted.

1-2) Analysis of fluid behavior

These efforts led to the emergence of fluid mechanics[9]. In other words, fluid mechanics is the basis of the results and findings of the study, which were obtained in a laboratory manner and as a result of extensive trials and repetitions.

Using the results obtained from various experiments and the extensive use of differential equations and mathematical relations, the governing equations of applied and modern theory were obtained. Many scientists collected and generalized fluid mechanics equations. So, in general, there are two methods for analyzing the behavior of fluids:

Laboratory and experimental method

Theoretical method (use of governing equations)

As mentioned, theoretical methods are obtained from laboratory and real studies of scientific phenomena. Using mathematical methods, it is possible to solve theoretical equations. Analytical solutions of mathematical equations are very accurate solutions provided that the equation in question can be solved by analytical method according to the geometry of the problem. The conditions of the problem such as two-dimensional or three-dimensional geometry, boundary conditions, two-phase problem, the size of the geometric dimensions of the problem and so on. It caused the use of numerical methods to solve the equations.

Numerical methods in the form of finite elements solve the problem with appropriate approximation and give us ideal and acceptable answers.

Abstraction :

chemical, petroleum and biological progressions have been focused. Todays, with developments in accounting ways and computer, CFD have been changed. In to a new way with large number of potential thunder standing fluid dynamic affection in application of chemical reactors. Bubbles in multiplay phase progression has an important role in these reactors. Division of bubbles indifferent size because of their breaking, explains these readings. conditions. An additional population balance model has been adopted to trace the interaction mechanisms between bubbles/bubbles and bubbles/eddies.

      Due to extensive expectations of simulating practical industrial scenarios under highly concentrated gas bubbly flows, plentiful researches have been carried out to explore neighboring influence on drag reduction in a highly packaged swarm of bubbles.

     In this study, the performance of an experimental drag coefficient correlation proposed by Simonnet et al. by considering representation such neighboring effects in terms of local void fraction has been assessed. This model has been compared with a densely distributed fluid particles drag model proposed by the Ishii-Zuber model which is used in commercial software boxes.