Detailed modelling of particle reactivity in turbulent combustion

Simulations help developers of industrial combustion and gasification applications to bring down costs, ensure safe operation and decrease ecological impact. An Integrated Gasification Combined Cycle power plan combines the economic advantages of coal fuels with the ecological viability of a gas turbine. CFD is used to make such a plant reality, but current CFD software still lacks a model to accurately describe the combustion or gasification behaviour of solid particles which is integral to the process. The goal of this work is to implement such a model into widely used CFD applications. The end product is a reaction model that can predict gas phase reactions, turbulence, solid phase reactions and the interactions between the three. To achieve this goal, the current model has to be verified, made more robust and fast, its performance and precision analyzed in a DNS code and implemented into RANS or LES code.

This work is part of the Cenbio project of NTNU and SINTEF.

Participants:

Jonas Krüger (NTNU)

Terese Løvås (NTNU)

Nils Erland L. Haugen (SINTEF Energy Research)

 

 

 

 

 

 

Low-NOx combustion technology (Numerical part)

 

Combustion of hydrogen-rich fuels, such as syngas, have received increasingly attention in the context of climate change and the urgent need for alternative fuels.  Syngas with high hydrogen levels and maximal relative concentrations of the carbon species as CO₂ is particularly interesting for carbon capture and storage (CCS), which eliminates or at least minimizes CO₂ emissions. One technology that can provide such gas and remove CO₂ fromthefuel mixture is pre-combustion CCS. On the other hand, the high reactivity and combustion temperature of the remaining hydrogen-rich fuel generates harmful and regulated mono-nitrogen oxides (NOx) during combustion. Reduction NOx emissions is therefore a major topic in combustion research. Low NOx burners designed for natural gas can, however, not be operated safely with hydrogen-rich fuels, due to the significand different transport properties of hydrogen, which have a strong impact on the likelihood of flashbacks. 

These challenges led to the development of a patented low NOx burner at the department of Energy and Process Engineering at NTNU and SINTEF. The burner features a conical frustum shaped bluff body generating a flame stabilizing recirculation zone.  Fuel is injected into an accelerated crossflow to restrict the potential for flashback. The fuel and air is partially premixed before reaching the combustion zone. The burner has proven good emission performance at laboratory scale. However, the scalability of the system to prototype/industrial size is not well understood yet. A numerical model of the burner is therefore currently been build up making a detailed analysis of the burner flow characteristics accessible. Different turbulence and combustion models as well as other sub-models have been employed and validated against experimental data aiming for a numerical model that is capable to describe the burner characteristics sufficiently. Based on the insight in the burner dynamics gained from the numerical simulations a scaling attempt can be made.

 

Participants:

Christoph Meraner (NTNU)

Terese Løvås (NTNU)

 

 

Numerical simulations in Biofuel and engine applications

The development and application of numerical tools for quantitatively assessing biofuel use in internal combustion engines and facilitating the identification of optimum operating parameters and improved emission strategy is a great challenge. To minimize the time and cost of better understanding the combustion processes in modern state of the art engines fueled with alternative fuels, the availability of reliable and realistic chemical models is the way forward. Yet such models are scarce especially in the case of FT biofuels, and those available are not of practical use in terms of computational time since the fuel is complex and likewise the chemical model. It is an ongoing work to develop and validate the kinetic models suitable for use in engine applications.

The numerical tools are provided by a collaborating partner LOGE.

 

Participants:

David Emberson (NTNU)

Terese Løvås (NTNU)

 

 

 

 

 

CFD−DPM/DEM Simulation of Multiphase Reacting Flow

 

Computational fluid dynamic (CFD) models are powerful predictive tools in multiphase reacting flow research. Generally, all the CFD models developed can be broadly categorized into Eulerian-Eulerian and Eulerian-Lagrangian approaches. For Eulerian-Eulerian approach, both particle and fluid phases are treated as interpenetrating continua. It can predict the macroscopic characteristics of a system with relatively low computational cost. However, in addition to the difficulty of providing closure models for interaction terms between phases within its continuum framework, Eulerian-Eulerian approach does not recognize the discrete character of the particle phase and thus has trouble in modeling flows with a distribution of particle types and sizes. These difficulties can be naturally overcome by Eulerian-Lagrangian approach in which the gas is treated as continuous and particle as discrete phase. When the particle phase is solved by discrete parcel method (DPM) which is applicable to dilute flows, the Eulerian-Lagrangian approach is called CFD-DPM model. When the particle phase is solved by discrete element method (DEM) which is suitable for dense flows, the Eulerian-Lagrangian approach is called CFD-DEM model.

 

This work was carried out within the Cenbio and GasBio project, funded by Research Council of Norway and industry partners.

Participants:

Tian Li (NTNU)

Terese Løvås (NTNU)