This week, in this Volume 4, we are publishing the last series of abstracts of the presentations made at the Seminar.

Study And Mathematical Model Of Ultra Low-Nox Gas Burner
by Albena GUEORGUIEVA - ENEL

This report has the aim to examine of Ultra-Low NOx gas burner.

The main objective of the study, is the prediction and reduction of NOx emissions under levels recommended from European standards for gas combustion processes.

The work includes: study the effectiveness of low-NOx concepts and understand the impact of combustion and swirl air distribution, flue gas recirculation on peak flame temperatures, flame structure and fuel/air mixing.

The experimental part of the project consists of participation at combustion tests on experimental facility located in Livorno - ULNG 5 MWt burner on D-shape boiler. The results of the experiments are used, to obtain better vision for combustion processes on small-scaled design and to collect the necessary input data for Fluent simulations.

A mathematical model of burner and combustion chamber is developed based on interacting fluido-dynamics processes: turbulent flow, gas phase chemical reactions, heat and radiation transfer. The NOx prediction model for prompt and thermal NOx is developed The validation of CFD (Computer fluido-dynamics) simulations corresponds to 5 MWt burner type - TEA, installed on CASPER boiler. This burner is three-stream air distribution burner with swirl effect, designed by ENEL to meet future NOx emission standards.

For performing computer modelling of the combustion process, FLUENT CFD code has been used, because of its capabilities to provide accurately description of large number of rapid interacting processes: turbulent flow, phase chemical reactions, heat and radiation transfer and for its possibilities to present wide range of calculation and graphical output reporting data.

CONCLUSIONS

• Pluses: Good Approximation of experimental and simulation data.
• Minuses: Bi-Dimensional axisymmtric swirl space simulate the real 3-D furnace space with some inaccuracy.

FUTURE GOALS, PLANS

• 2-nd Stage Combustion Tests.
• 3-D Simulation with different Models and Parameterisations.

Potential for Integration of CFD with Detailed Combustion Chemistry
by Sinead Nolan - Cardiff

Summary

Background Information

A rapid compression machine is an instrument to simulate a single cycle in an internal combustion engine. The RCM was built in the 60's by the shell Oil Company at their Thornton Research facility in Chester. The motivation was to investigate the chemical processes that precede ignition and result in an induction time call ignition delay time.

• RCM’s allow the study of spontaneous ignition at high pressures very close to the true working environment of internal combustion engines without the complications that one encounters with real working engines such as oil droplets, abraded metal, particles, constantly changing condition etc.
• The twin opposed pistons are a unique feature of the RCM. In a single piston RCM the speed of the pistons are limited as aerodynamic heating occurs at high piston speeds. High piston speeds are necessary to obtain short compression times. The twin opposed piston design extends this limit by halving the compression time.

Aims of Time Spent in Cardiff

The aims of my time spent in Cardiff are to redesign the piston heads. Vortex formation is a direct result of vigorous motion generated by the piston along the wall which promotes the mixing of cooler gases from the wall with the adiabatic core. The net effect is a non-uniform temperature distribution in the reaction thus making interpretations and modelling of results very problematic.

• Redesigning of the pistons essentially involves the machining of a crevice into the side of the piston head to accommodate the boundary layer during compression and suppress the formation of the corner vortex.
• Redesigning of the reaction chamber is to install a rapid sampling system; this will be based on a fast piezoelectric valve so that a sample of the post compression gases can be taken at predefined intervals during the experiments. To render the compression processes as ideal as possible the ‘dead volume’ must be reduced to a minimum.
• The CFD code that will be employed is Vectis, which was created by Ricardo. Initial training such as grid generation will take place at Ricardo in Shoreham-on-Sea.
• The culmination of the project will be to implement a CFD model coupled with a suitable kinetic model to conditions experienced in the RCM.

Combustion Calculations of the Primary Stage of a 2-stage
by Klaus Hoerzer - Cardiff

Summary

Due to environmental problems caused by the combustion of fossil fuels there is a need to use renewable fuels just like wood or wood dust. One possibility is to combust wood dust in a two-stage multi – inlet combustion chamber. It was designed by the Institute of Thermal Turbomachines and Power Plants at the Vienna University of Technology.

Reasons for the use of a two – stage combustion chamber are first of all, that solid fuels are combusted. The produced gas is directly used in a gas turbine. Hence, there is no need for storage facilities for a low heating value gas, which would be produced in gasification. At last, a two – staged combustion (in the primary stage the equivalence ratio is <1, in the secondary stage the equivalence ratio is >1) delivers appropriate magnitudes of combustion temperatures to accomplish low values of nitrogen oxides.

The aim of the investigation was to calculate the gasification/combustion of the primary stage by application of the following steps. First of all, the calculation grid was created. In the second step, the isothermal flow distribution of the primary stage was calculated. Afterwards, the calculation of the particle movement in the isothermal flow took place. In the last step, the complete gasification/combustion was calculated.

The 3D – grid was unstructured, this means that it consisted of tetrahedral elements. The number of nodes was approximately 58,000, the number of the elements about 200,000. The turbulence model was the Reynolds – Stress – Model (RSM) which is supposed to be used in combustion calculations, especially at higher swirl numbers. The species were modelled by means of the mixture fraction/pdf – approach. The radiation was modelled with the so – called P1 – model, which is based on the expansion of the radiation intensity I into an orthogonal series of spherical harmonics. Wood was considered as the material to be combusted.

The primary chamber has six tangential air inlets to provide high swirl flow and hence high residence times of the combusting particles. The wood dust enters axially the combustion chamber through a centre inlet in the bottom and hits a centre diffuser which diverts the particles from axial to radial direction.

The temperature distribution of the primary chamber showed conformity with the distribution of the tangential velocity (which is high near the wall) insofar, that in areas of high tangential velocity appear high static temperatures as well. Higher temperatures of about 1200 K appear near the centre diffuser as well.

The species distribution of solid carbon shows low or even zero values of it close to the wall and in areas of high tangential velocities, which indicates, that the solid carbon is mostly combusted. The solid carbon is not completely combusted in the chamber, some particles can escape through the outflow.

The distribution of carbon monoxide shows high values in some areas near the wall. The content of carbon monoxide decreases in the direction of the outflow. For carbon dioxide there are high values near the bottom. In areas of the high content of carbon monoxide the part of carbon dioxide is low, which indicates, that in these areas the Boudouard – reaction takes place. The Boudouard – reaction describes, that one mole of solid carbon reacts with one mole of carbon dioxide to two moles of carbon monoxide. Considering the fact, that the Boudouard – reaction is an endothermic reaction, the vicinity of the hot wall seems to provide enough energy to make this reaction happen.

Considering the particles for a small particle (0.005 mm), a medium particle (0.15 mm) and a large particle (0.3 mm) the post-processing showed that the small particles do not even hit the centre diffuser, they just slip around and reach very fast the area of the axial outflow. Hence, some particles are able to leave the combustion chamber. Medium particles hit the centre diffuser and are diverted from axial to radial direction, where they are caught by the tangential flow. The large particles hit the centre diffuser as well, and they are caught in the tangential flow near the bottom of the combustion chamber.

Future work will deal with the modelling of the complete combustion chamber, and work will be done in the manner described above.

 

PowerFlam Research Consortium – Substitute Fuels for Power Generation – Opportunity to Participate: Contributed by Lois Brett AFRC Spring Meeting 2002: Contributed by Eddy Chui/Lois Brett TOTeM 17 – The Use of Oxygen for Industrial Combustion – Meeting Conclusions: Contributed by Jacques Dugué, Philipp Staab and Lois Brett Annual EuroFlam Seminar 2001, Abstracts - Vol 4: Contributed by Aristide Mbiock Environmental News: Contributed by Aristide Mbiock

 

This periodical forms part of the group of publications owned by the IFRF
The IFRF Monday Night Mail is published by:
IFRF NET, P O Box 10,000, The Netherlands
Edited: Peter Roberts
ISSN 1562-4781