Defense by Abdelkader Izerroukyene (Mechanical Department)

In the current environmental context, marked by climate change and the energy crisis, biomass is attracting growing interest due to its carbon-neutral potential. However, it is crucial to assess its impact on human health and air quality.

In the course of this thesis, experimental measurements and CFD numerical modeling were carried out in the chimney of a domestic biomass boiler in order to analyze pollutant gas and particle emissions during wood pellet combustion. A boiler with a rated output of 10 kW and a feed burner was used. Various types of sensors (thermocouples, flow meters, relative humidity sensors, anemometer) were installed in the boiler.

Particulate emissions were measured using the EEPS (Engine Exhaust Particle Sizer) technique. In addition, gaseous emissions (CO2, CO, total hydrocarbons HCT, O2) were determined by infrared spectroscopy (IR) and flame ionization detector (FID). Emissions were recorded during the stationary phase of the boiler, and repeatability was ensured over several tests. Four measurement points in the stack were studied to assess the evolution of particle size and their potential deposition on the walls along the stack.

The experimental results were compared with a CFD numerical model built in OpenFOAM incorporating particle transport. The modeling of turbulent flow in the stack uses a 1st-order RANS (Reynolds-Averaged Navier-Stokes) approach or based on the Boussinesq hypothesis. To take account of the anisotropy of the flow, the turbulence model was selected for this study. The effect of turbulent fluctuations on the discrete phase is integrated via the DRW (Discrete Random Walk) turbulent dispersion model. The results obtained enable the topology of the carrier phase flow to be visualized, as well as the complete distribution of the particle field in the stack. Advanced pollutant emission measurements and particle transport modelling have been developed for the first time in a domestic biomass boiler operating under real conditions.

The experimental results reveal several relevant insights. CO and HCT emissions show a similar evolution over time. The particle size distribution measured along the stack highlights the phenomenon of particle agglomeration. In addition, the CFD model and experimental results provide concordant results in terms of flow characteristics and particle size distribution.

Reporters

• Valérie TSCHAMBER, Professeur des Universités, Université Haute-Alsace
• Xesús NOGUEIRA, Professeur des Universités, Université de la Corogne

Examiners

• Denis PETITPREZ, Professeur des Universités, Université de Lille
• Mathieu SPECKLIN, Senior Lecturer, Conservatoire National des Arts et Métiers - Paris

Thesis director

• Céline MORIN, University Professor, INSA Hauts-de-France

Co-director

• Sofiane KHELLADI, University Professor, ENSAM Paris

Co-supervisors

• François BEAUBERT, Senior Lecturer, INSA Hauts-de-France
• David UYSTEPRUYST, Senior Lecturer, INSA Hauts-de-France
• Damien MÉRESSE, Maître de Conférences, INSA Hauts-de-France