On Monday 25th June 2001 (MNM Week 26) we reported the Annual Seminar and Summer Management Meeting of the EuroFlam consortium at Cardiff - Wales, on Thursday 21st and Friday 22nd June. Sixteen Graduates/Visiting Investigators who presented their completed reports or progress reports attended the meeting.

Last week (MNM Week 28) we commenced the publication of abstracts of the presentations made. This week, we are publishing the second series of abstracts.

Assessment of the energy content of French/UK household waste
by Adeline Bisson - Cardiff

Abstract

The increasing amount of waste generated in modern societies is of major concern to environmental policies. Consequently, we have to change the way waste are managed in our countries.

In 1998, in France and in the UK, about 28 million tonnes of household waste were produced. Respectively 82% and 48% of the waste stream was landfilled in the UK and in France, and, in both countries recycling and composting schemes only concern about 10% of the waste stream.

However a waste classification project lead in the Rhonddha Cynon Taff County Borough Council (South Wales) has shown that recycling components represent 43% of the waste stream and about 73% of the waste stream could be recovered by recycling and composting schemes. Thus the aim of the European directives for waste management is to reduce significantly the part of the waste stream landfilled.

A key step towards more sustainable development is recognising waste as a potential resource. Energy could be recovered in incinerator plant as the calorific value of the waste stream is about 15 MJ/kg (mostly due to plastic components). Recycling and composting options also contribute to save energy and raw materials.

Moreover the waste biodegradation process that occurs in landfill site produces gas (methane and carbon dioxide) and leachate (with heavy metals) that could be potentially harmful to the environment and to the population.

All this goes to show that they are real possibilities to manage waste in a more sustainable way that would ensure a better quality of life for the future generations.

Fuel Oil Characterization by Fourier-transform Infrared Analysis and a Multivariate Statistical Regression
by Maria Lourdes Ruiz Villar - ENEL

Abstract

Fuel oil is a mixture of various organic compounds and its reactivity depends on its average composition.

The products solid generated during the combustion of the fuel oil depends on its composition. To know in advance, and in an immediately way, this composition will help us to optimise the combustion process.

The scope of this work is to find a model that analyses the fuel oil composition in a quantitative and very quick way in order to reduce the fly ash produced on its combustion.

A quick determination of the fuel-oil properties used in the power station represents a very important goal because this knowledge can:

• help in the choice of a fuel
• help in optimising combustion parameters
• Reduce the emission of solid particles from the power station roofs.

In order to apply the regression technique and evaluate a relationship between the IR , absorption and the oil composition at least 40 spectra of analysed oil are necessary.

15 oils were analysed by a modern FTIR; and about 25 spectra of oil analysed in the past were reconstructed using the digitiser UN-SCAN-IT.

Multivariate calibration means determining how to use many measured variables simultaneously for quantifying some target variable.

The X-variables will be spectroscopic measurements, and the target variable will be analyte concentration.

It is necessary to find a calibration by correlation technique which is able to relate the composition of fuel oil with the FTIR patterns.

Spectral data consisting of hundreds and even thousands of absorbency values per spectrum can now be routinely collected in matter of seconds. These data can be converted into valuable information using multivariate calibration techniques such as partial least squares (PLS) or principal component regression (PCR).

The technique that will be used in order to find a relationship between oil composition and FTIR spectra is PLS (Partial Least Squares) after the selection of the most significant absorption lines by using a Genetic Algorithm (GA).

A data set made of 40 FTIR spectra and 40 fuel oil compositional string was obtained.

A genetic algorithm will be applied to the reconstructed spectra in order to select the most important absorption wave lengths.

PLS will be applied finally to find a relationship between the oil composition and its FTIR spectrum.

Developing Use of IFRF Mathematical Modelling Facility
by Dennis Brauckmann - IFRF

Abstract

The cement manufacturing industry is one major consumer of heat energy delivered by combustion. The specific energy consumption is high indeed since the process of clinker formation is highly endothermic. Although the specific energy consumption for the production of clinker has been decreased during the last decades, it is still important to reduce the energy consumption even further and simultaneously to reduce the pollutants emissions to achieve a cleaner and more efficient process.

To examine the influences of the burner design, operation and fuel type on thermal efficiency and NOx formation, the CEMFLAME project was founded. In this studies, flames from the CEMFLAME project have been modelled using the commercial FLUENT package for fluid flow calculations. The grid was constructed with GAMBIT containing around 35000 cells.

As a main conclusion from the discussion of the MVB-Coal flame it can be pointed out that the general flame shape and most of the furnace exit parameters have been predicted with a reasonable accuracy. It has been observed, that a correct fuel characterisation is very important for a good quality prediction of the flame. The release rate of volatiles has been proven to be an important factor in computing the stand off distance, heat release profile and furnace exit parameters. The volatile matter combustion in the model seems to be too fast in comparison with the measurements. This caused a lack of oxygen in some flame regions. Even though the ignition point is predicted very well, the over-predicting peak of the temperature causes several problems.

Generally the correct description of the reaction of the volatiles and also of the char are of enormous importance for the description of the whole flame.