Current Research Activities

Discrete Particle Model (DPM)

The novel and innovative character of XDEM is to  consider a packed or moving bed to be composed of a finite number of discrete particles which may have different properties, shapes or sizes. Motion of particles is described by the traditional Discrete Element Method (DEM). Additionally, each particle undergoes a sequence of thermodynamic processes that are described by a set of one-dimensional and transient conservation equations for mass, species momentum and energy by the Discrete Particle method (DPM) with sufficient accuracy. Heat, mass and momentum transfer couples the flow within the void spaces to the particles. Thus, the sum of all particle processes represents the overall packed bed process such as heat-up, drying, pyrolysis and gasification.  This approach is applicable to a large range of different particle processes due to its generic concept and predicts results on a particle level with a large degree of details. The DPM module already contains relevant and validated kinetic data that allows predicting both temperature distribution and chemical reactions e.g. emissions for an individual particle shown in the following fig. 1 as a temperature distribution:

Image temperature profile
Figure 1: Temperature distribution in a particle dependent on time and space

Fluidised Bed

Image Fluidised bed
Figure 2: Fluidised bed

The objectives of the current proposal are to investigate numerically into the phenomena of drying of sewage sludge and biomass on a vibrating grate and subsequent gasification in a fluidised bed. Contrary to entirely based continuous phase approaches, the current methodology relies on the Extended Discrete Element Method (XDEM) that resolves the particles as a discrete phase coupled via heat, mass and momentum transfer to the gas phase, and thus, represents an advanced and innovative simulation technology. Consequently the state of each particle during thermal conversion in conjunction with the surrounding gaseous phase is predicted, that furnishes detailed results for both the conversion history of individual particles and the gas phase. The proposed approach will be validated with measurements from the Valortech project to a large extent and if necessary with additional data from literature. Hence, predictions obtained by numerical simulation complement excellently experiments and contribute to a deeper understanding of the underlying physics of the processes involved. These findings feed back into the Valortech project, for which drying and gasification processes form the core part to valorise a fuel derived from sewage sludge and biomass. Analysis of predictions of both the drying vibrating grate and the fluidised bed lead to an improved design and operating conditions, and thus, strongly support the Valortech project.

Partners: M. Demouling, SoilConcept
Contact: MSc M. Mohseni (mohammad.mohseni@uni.lu)

Thermal Conversion of a Moving Bed on a Forward Acting Grate

Image bed_temperature_100s
Figure 3: Particle temperatures on a forward acting grate

The objective of this research is to investigate into heat-up of a moving or fixed bed of solid e.g. fuel particles. The dominant mechanisms of heat transfer of a packed bed include conduction between particles in contact, radiative heat transfer between neighbouring particles and convective heat transfer to a fluid in the void space. Thus, heat transfer in reacting fixed and moving beds influences widely progress of reactions and completeness of conversion e.g. burn-out. These parameters including motion on a forward acting grate are subject to investigations by the Discrete Particle Method (DPM).
The neighbours of each particle change versus time, and thus, determine the time-varying heat transfer from neighbouring particles. Heat transfer due to conduction is evaluated by a contact surface and the particles conductivity whereas radiative heat transfer is described by the dense body approximation. Convective heat transfer between the particles and the fluid flow is described by well derived empirical correlations for heat transfer coefficients.

Partners: Dr. A. Dziugys, Lithuanian Energy Institute
Contact: Prof. B. Peters (bernhard.peters@uni.lu)

Fluiddynamic Instability

Image injection
Figure 4: Injection into quiescent air

In order to investigate the stability of liquid sheets both a linear and non-linear stability analysis will be performed. Break-up o liquid jets is important for injector nozzles of a Gasoline Direct Ignition (GDI) engines. The various parameters affecting the breakup of the liquid sheet in to droplets will be identified and compared with experimental results and theoretical results. A Computational Fluid Dynamics (CFD) analysis of the liquid sheet breakup will also performed using the OpenFOAM code based on a VOF-LES method. The results of the stability analysis and experimental data will be compared with the predictions obtained by a CFD LES model. Thus, the study aims at the improvement of the design of the fuel injection nozzles used in GDI engines.

Partners: Prof. U. Müller, University of Karlsruhe
Contact: Prof. B. Peters (bernhard.peters@uni.lu)

Atmospheric Dispersion of Pollutants and Air Quality

Image NOx Luxembourg
Figure 5: Distribution of NOx over the region of Luxembourg

Air quality and energy are two important topics, both socio-economically and politically, however, energy and the environment are often 'competitive' in nature. The evolution of energy networks and the implementation of emission standards are simultaneous global concerns. This work explores the possibilities of air quality improvement in Luxembourg using a coupled energy - air quality model. It aims at analyzing concentration levels of the most problematic pollutants in Luxembourg and surroundings, including SO, NO, VOC, CO, CO, O, PM and PM. The work consists of developing the model AYLTP (AsYmptotic Level Transport Pollution Model), using a core calculator as a basis (AUSTAL2000). The AYLTP model is designed to explore urban ozone levels using a fast ozone calculator and pollutant transport model, appropriate for the optimization framework for which it is designed.

Partners: Ass. Prof. D. Zachary, Centre de Ressources des Technologies pour l'Environnement (CRTE)
Contact: Prof. B. Peters (bernhard.peters@uni.lu)

Enhanced Design for High Performance Parallel Execution

Image ghost cells
Figure 6: Plimpton scheme for ghost cell communication

This research proposes an extension of the previous work on Domain Decomposition which showed the benefit of parallel execution for the Dynamics module. The new design will allow parallel and distributed execution of coupled simulation modules of DPM. The targeted execution platforms are clusters of multiprocessor computers.
The primary goal of this enhanced design is to provide:
- Parallel execution of coupled modules: the new design supports internal DPM modules (like Dynamics and Conversion simulations) and external modules as well (like OpenFoam and Diffpack).
- Flexibility and Maintainability: Current simulation modules can be selected at runtime and the integration of new modules in the framework is made easy. A common codepath is used for parallel and sequential execution: the sequential execution is just seen as a special case of the parallel execution.
- High Performance distributed execution: Code and data structures are re-organized in order to provide the best performance. The MPI library offers efficient communications, portability and the ability to use high speed networks like InfiniBand.
In a second step, further improvements will be investigated. We will consider, among others, offloading part of the simulation onto GPUs, and using new dynamics load-balancing algorithms.

Partners: Prof. P. Bouvry, Department of Computer Science, University of Luxembourg
Contact: Dr. X. Besseron (xavier.besseron@uni.lu)

Digital Twin of a Blast Furnace

Image Blast Furnace
Figure 7: Processes in a blast furnace

The visionary objective is to generate an integral multi-physics model for the entire blast furnace that integrates ultra-high fidelity predictions into a digital twin of the blast furnace. A requirement for a Digital Twin results from numerous engineering challenges that necessitate a shift from current empirical-based practice to an advanced multi-physics simulation technology including multi-physical models on different length-scales to mirror accurately the state of a blast furnace. This motivates an encompassing approach to describe various processes such as the tripping zone, raceway or cohesive zone with best available physical models. An analysis of predicted results will unveil underlying physics and thus, impacts design and operation of a blast furnace. The Digital Twin can also predict responses of the blast furnace to safety critical events and uncover previously unknown issues before they become critical. Taking a generic approach for the modelling framework allows applying the simulation platform also to similar processes such as COREX or FINEX.

Partners: Dr. J.-P. Simoes, Mr. L. Hausemer, Mr. K. Mutschler, Paul Würth, Luxembourg
Contact: Prof. B. Peters (bernhard.peters@uni.lu)

Incineration of Biomass in Furnaces

Image packed bed
Figure 8: Gas and particle temperatures in a packed bed

The objective of this research is develop a software tool that predicts thermal conversion of solid fuels such as biomass in a variety of engineering devices e.g. in a packed and moving bed on a forward or backward acting grate. Thermal conversion consists essentially of a reaction process of solid fuel particles coupled to the surrounding gas phase by mass and heat transfer and their motion within the reaction e.g. combustion chamber. As a novelty, the proposed approach considers a packed bed to be composed of a finite number of discrete particles which may have different properties, shapes or sizes. Each of these particles undergoes a sequence of processes that are described by a set of one-dimensional and transient conservation equations for mass, species momentum and energy within the Discrete Particle method (DPM) with sufficient accuracy. Thus, the sum of all particle processes represents conversion of a packed bed including processes such as heat-up, drying, pyrolysis and gasification. Motion of fuel particles is described by the Discrete Element Method (DEM), while the state of the gaseous phase is described by differential conservation equations for mass, momentum and energy of a compressible reacting flow by well established CFD methods. Heat and mass transfer couples the flow within the void spaces to the particles e.g. packed bed. The predictions of the numerical model are compared to both experiments with single particles and measurements of packed and moving beds. Due to the nature of the process of a packed bed of fuel particles, a large number of parameters of the packed bed such as particle temperatures or local composition are inaccessible during an experiment. Therefore, this effort will complement experimental investigations and provide a deeper insight into the process.

Partners: Prof. T. Nussbaumer, Hochschule Luzern
Contact: MSc A. Mahmoudi (amir.mahmoudi@uni.lu)

Snow-Tire Interaction

Image tire
Figure 9: Tire-ground interaction

The objective of this research is to develop a simulation approach that describes the interaction of a tire surface on a snow-covered road. Contrary to the continuum mechanics approach snow is considered to exist of discrete grains, that interact wit each other. Therefore, the Discrete Particle Method (DPM) representing a discrete approach based on the Discrete Element Method is applied to the snow. This tool supports the development of a micro-mechanical model that describes the behaviour of snow such as motion of snow grains and deformation. The predicted results obtained by the simulation tool are compared to experimental data for validation. The forces acting between individual snow grains and the tire e.g. tread will determine the strength of traction. It depends to a large extent on the surface structure of the tire such as shape, orientation and geometry of the tread blocks. In order to predict the viscoelastic deformation of the tire when it comes into contact with snow a Finite Element analysis is applied. The coupling between the discrete approach to characterize snow and the FEM approach for the tire represents an accurate model to assess the traction. Similar to the development of the discrete model describing snow, these predictions will be compared to experimental data.

Partners: Prof. Nicot, IRSTEA, France
Contact: Prof. B. Peters (bernhard.peters@uni.lu)

Community Identification

Image Community
Figure 10: Identification of temperature clusters on a forward acting grate

The objective of this research is to develop and apply community identification algorithms to predictied results of the Discrete Particle Method (DPM). The latter provides a detailed information about the processes in granular matter, however,in order to gain a broader understanding and insight, additional postprocessing of the simulation and, possibly, experimental data is necessary. Since granular matter consists of a large number of interacting particles, one approach is to represent the particle interactions by a graph where the particles are represented as the graph vertices and the relevant interactions between the particles are represented as the graph edges. The algorithms known from graph theory can be thus applied to analysis of particle sets. A problem of identifying clusters of similar particles, e.g., groups of particles having similar values of temperature e.g. hot spots and cold spots, or groups of particles of similar size e.g. size segregation, can be made equivalent to a known problem of community detection in graphs. A number of community detection algorithms has been proposed, and this is currently an active area of research. Application of community detection algorithms for analysis of granular matter has its own specifics, therefore, the standard techniques have to be specifically adjusted, and new algorithms and approaches taylored for the particular tasks of structure identification in granular matter are the subject of the current research.

Partners: Dr. Robertas Navakas, Lithuanian Energy Institue
Contact: Prof. B. Peters (bernhard.peters@uni.lu)

Hardmetal Production

Image Hardmetal
Figure 11: Morphological changes associated with a hydrogen reduction of tungsten oxides path under industrial conditions. Src: Wolfram Bergbau and Huttenges.m.b.H.

Hardmetal production isinvestigated for conversion of pulverised tungsten oxide to metallic tungsten in a packed bed. Reduction of tungsten oxide to tungsten in a hydrogen atmosphere is described by a reaction scheme, for which temperature and reaction progress is described by the Discrete Particle Method (DPM). Hydrogen as a gaseous phase is introduced as a reducing agent that streams over a packed bed of tungsten oxide particles in a push-type furnace. The flow over and penetration of hydrogen into the bed of tungsten particles is represented by advanced two-phase CFD-tools for a porous media. The study will provide a deeper insight into the process, because particle temperatures and interaction of particles with the fluid are inaccessible in a packed bed during experiments.

Partners: Dr. Ralph Useldinger and Dr. Frankie Hippe, CERATIZIT-Luxembourg
Contact: M.-Ing. A. Estupinan Donoso (alvaro.estupinan@uni.lu)

Analysis of the Impact of ROS in Networks Describing Neurodegenerative Diseases

Neurodegenerative diseases such as Parkinson's or Alzheimer are highly complex in nature and involve various physiological processes. There is clinical evidence that generation of reactive oxygen species (ROS) in dopaminergic neurons induces the apoptosis of mitochondria. Therefore, a kinetic model that describes apoptotic and non- apoptotic states of mitochondria and the regulation of the transition between those two states in neurodegenerative diseases is developed and its dynamic behaviour, in particular the role of reactive oxygen species, is investigated in detail within the proposed project as sketched in fig. 9. The network developed is validated by experimental data so that it reflects accurately the behaviour of neurodegenerative diseases.

Image Neuro-degenerative diseases landscape
Figure 12: A complex three-dimensional landscape depicting stable (A), metastable (B) and unstable (C) state (Huang et al. 2009)

Partners: Prof. Balling and Dr. Kolodkin, Luxembourg Centre of Systems Biomedicine
Contact: MSc A. Ignaenko (andrew.ignatenko@uni.lu)

Dripping Zone of Blast Furnace

Dripping Zone of Blast Furnace
Figure 13: Dripping zone of a blast furnace

Blast Furnace iron making is one of the largest reactor in the world in which several phases of gas and liquids coexist simultaneously. The severe operating conditions of blast furnace made the engineers to use computational tools to discover the exact behavior of fluids inside the furnace. The lower part of blast furnace where coke particles are piled up while liquid iron and slag move downward and blast gas flows upward, is called dripping zone. This part of furnace could be classified as a countercurrent multiphase packed bed reactor. The main object of this project is to extend an up-to-date approach of Extended Discrete Particle Method (XDEM) by multi-phase flow capabilities for reactors specially trickle bed reactors. Rather than expanding the discrete element method by CFD approaches, a coupling with well-established CFD approaches i.e. OpenFoam is preferred due to both time and cost reductions. Another purpose of this project is to consider Interaction and counter-current contact between gas and two liquid phases, slag and iron, which have great effect on heat transfer process, chemical reactions and damage of tuyeres at the lower part of the furnace as well as on the smooth operation of the furnace as a whole.

Partners: Paul Würth, Luxembourg
Contact: M.-Ing. M. Baniasadi (maryam.baniasadi@uni.lu)

Cohesive Zone of Blast Furnace

Cohesive Zone of Blast Furnace
Figure 14: Cohesive zone of a blast furnace

The objective of this research project is to predict the evolution of a cohesive zone in a blast furnace as a transition from layered arrangements of burden of ore and coke material to the start of the dripping zone. The newly gained knowledge supports understanding the internal physics of the cohesive zone and its impact on redistribution of hot gas flow and operating conditions. For this purpose the integrating concept of the Extended Discrete Element Method (XDEM) is employed that deals with advanced multi-physics simulation technology by linking the Discrete Element Method (DEM) and Computational Fluid Dynamics (CFD). This approach allows representing the cohesive zone composed of coke and not yet molten ore particles for which the interstitial space between the particles is occupied by a multi-phase flow of gas, liquid iron and slag. Softening and stickiness of particles is taken into account by an increased overlap and cohesive forces between particles, respectively. The multi-phase flow character due to gas, molten iron and slag is described by the simulation framework of OpenFoam. The results of numerical works will be compared to experimental data.

Partners: Paul Würth, Luxembourg; CRM Group, Belgium; Tata Steel, Nederland
Contact: M.-Ing. M. Baniasadi (mehdi.baniasadi@uni.lu)

Highly Turbulent three phase flows in industrial geometries

Three phase Flow in a nozzle
Figure 15: Gas, Liquid, and particles mixture flow at very High Reynolds Number inside an industrial nozzle

Many engineering applications are characterized by the presence of multiple phases interacting with each other. The simultaneous presence of a fluid, a gas and a solid phase is the standard for most industrial processes. Despite of the great advances in the Computational Fluid Dynamics field, the simulation of multiphase flows still remains a difficult task. In particular when the Reynolds numbers become very high capturing the interfaces dynamics inside the flow become a particularly challenging task. Nevertheless many industrial applications receive enormous benefits from an accurate description of this kind of flow, especially when the density and the physical properties of the various components strongly differ. The objective of this research is to obtain an accurate description of the dynamic behaviour of particulate matter inside a multiphase flow. This task represents the first stage in order to approach the problem of the interaction between three phase flows and the containment structures and then optimize the flow parameters for specific applications.

Partners: Dr. Ralph Useldinger, Dr. Marc Elsen, Dr. Zhongming Xia, CERATIZIT-Luxembourg
Contact: MSc. G. Pozzetti (gabriele.pozzetti@uni.lu)