At BIOMATH, where bioprocesses are studied, CFD (Computational Fluid Dynamics) modelling is used for system analysis and design optimization. Reactor design is one of the most overlooked issues in bioprocess engineering, in contrast to chemical engineering applications. The fluid matrix in bioprocesses is very complex (activated sludge, digester sludge, fermentation broth) which makes modelling a challenge, specifically with respect to rheology and turbulence modelling. The latter is currently a stumbling block in CFD simulations of biosystems and has not received ample attention. More precisely, the use of LES (Large-Eddy Simulations), a technique that is becoming feasible due to increased computing power, will allow substantial progress in the analysis of biosystems and the proper choice of simpler turbulence models to reduce computational efforts. Another challenge is the coupling with kinetics especially due to the fact that processes have very different time constants, turning the system into a stiff numerical system to solve. In this project, the focus will be on the anaerobic digestion process.
At LCT the focus is on development of novel three-dimensional (3D) chemical reactor geometries by accurately modelling turbulence and accounting for the full complexity of the occurring chemical reactions. New manufacturing techniques such as three-dimensional (3D) printing allow the creation of 3D geometries that were previously impossible to make and that could enhance radial mixing, resulting in higher selectivities and reduced fouling. An additional challenge is implementing detailed kinetic models in CFD, because they typically contain thousands of reactions involving hundreds of intermediates, and hence pose a strong challenge to numerically solve these large systems of differential equations within a reasonable amount of time. Therefore in most cases simple 1D reactor models are still applied although they present several important shortcomings and make it impossible to account for complex 3D turbulent flow patterns.
At FloHeaCom, numerical simulations are at the core of the research as well. In the Transport Technology unit, most of the research is focused on developing numerical tools to aid the R&D in internal combustion engines. Because of pollutant emission limits and CO2 targets, engines are developing quickly but are also becoming increasingly complex. For example, a state-of-the-art gasoline engine uses turbocharging, direct in-cylinder injection with multiple injections per cycle, stratified combustion, while employing variable valve timing, exhaust gas recirculation, … Thus, numerical tools have become indispensable to either replace or support experimental efforts. In the present proposal, special attention will be given to “dual fuel” combustion modes. Dual fuel engines are essentially diesel engines running on (mainly) natural gas: the natural gas is mixed with the combustion air and ignited by a small quantity of injected diesel (pilot injection). Dual fuel is currently seen as an economic way to meet upcoming emission standards. However, numerical tools are lacking due to the modelling complexity: pilot injection of a liquid fuel, followed by atomization, evaporation, mixing and auto-ignition, which causes the gas in the immediate vicinity of the pilot to auto-ignite, which is followed by flame propagation through the bulk of the gas. There is thus a combination of different phases (liquid/gas), different fuels, so different chemical kinetics, burning in mixed combustion modes (non-premixed/premixed), in a turbulent flow field, in a varying geometry (moving piston and valves).
In the Combustion, Fire and Fire Safety unit, CFD does not only serve the purpose of improving combustion devices and fire safe designs, it is also an essential tool in the further development of understanding fundamental combustion and fire related phenomena. The use of CFD has nowadays become an integral part of fire protection design of real life applications. The wide range of applicability of CFD models in areas such as e.g. fire modelling, design of smoke control systems, performance based design or structural fire engineering has made them a useful tool and their use by research scientists and fire safety engineers has substantially increased over the last decades. CFD models can be used, not only to evaluate the effectiveness of current or future fire protection systems, but also to answer “what if” questions and be used towards a cost-effective design without compromising fire safety in buildings. In the present proposal, the focus is on fundamental research in the field of spray combustion, as well as fire dynamics with variable ventilation conditions and the possible interaction with water sprays. This will serve technological developments on the longer term. In the context of fire forecasting, CFD is too slow. Faster, and consequently less accurate, techniques must be applied. This inaccuracy can be remedied by providing real-time multi-sensor data. The latter is within the field of IDLab’s multi-sensor fire detection research, that will further examine the potential to forecast a fire in terms of size and geometry by combining real-time analyzed volume sensor data, such as video, with fast numerical simulations. The main goal is to use the measured data from the fire, e.g., obtained by video fire analysis, in order to replace or correct model predictions. The information delivered by this real-time predictive tool will help fire fighters and decision makers consider the necessary actions to take.