Experimental and modelling study of pulverized biomass combustion
- Autores: María Pilar Remacha Gayán
- Directores de la Tesis: Santiago Jiménez (dir. tes.), Javier Ballester Castañer (dir. tes.)
- Lectura: En la Universidad de Zaragoza ( España ) en 2014
- Idioma: español
- Tribunal Calificador de la Tesis: F. J. Higuera (presid.), César Dopazo García (secret.), Javier Gil (voc.)
- Materias:
- Texto completo no disponible (Saber más ...)
- Resumen
The use of biomass for the generation of electricity has received a growing interest during the last decades due to the increasing concern about the environmental impact of the CO2 emissions from coal-fired boilers. Although the range of alternatives proposed to model the combustion of biomass particles is quite broad, detailed experimental studies are comparatively much scarce and most of them have been performed at conditions that only vaguely resemble those of real combustion systems and do not cover the whole range of applications of practical interest.
The objective of this work is to evaluate different alternatives for the modeling of the combustion of biomass particles under conditions representative of practical systems as well as to validate them with experimental data generated in tests specifically designed. In particular, this work analyzes the relevance of intraparticle temperature gradients in the combustion of biomass particles with sizes up to 15 mm under the high temperatures and heating rates typical of the industrial facilities. Two different models have been implemented to simulate the processes involved in biomass combustion. The essential difference between both theoretical models lies in the consideration (or not) of the intraparticle temperature gradients: in the case of the thermally-thin particle (TNP) model, they are simply neglected and the particle behaves as a whole; on the contrary, the thermally-thick particle (TKP) model accounts for them and each layer of the particle evolves at different rates.
The predictions of these models have been compared with experimental data obtained from different types of tests conducted in an Entrained Flow Reactor (EFR) and in a Flat Flame Burner (FFB) at well controlled combustion conditions (gas temperature, velocity and oxygen concentration). A complete set of devolatilization and oxidation tests has been carried out in the EFR with particles of an energy crop, Cynara Cardunculus (thistle), in the size cut of 300-400 µm. Two different types of particles were used in the FFB experiments: on the one hand, particles of an inert material (alumina) were selected to validate, independently, the heating and drying stages in the combustion of a thermally-thick particle; on the other hand, Buxus sempervirens (boxwood) spherical particles were used in devolatilization tests. The influence of the particle size and the oxygen concentration on the pyrolysis process has been independently analyzed by means of two sets of experiments. In the first case, tests in inert and oxygen-containing atmospheres were conducted for particle sizes ranging from 3 to 15 mm. In the second study, the effect of oxygen concentration was specifically analyzed for particles of 15 mm tested at fixed gas temperature and velocity.
The gas and intraparticle temperatures, as well as the particle size and shape and the volatile flame were monitored in these tests with different instruments; specifically, the temperature at several points inside the particle was determined with fine thermocouples. These data have served as a benchmark for the validation of the TKP model developed and, as part of this process, fuel- and model-dependent devolatilization kinetic parameters have been derived. The experimental data have also been compared with the predictions of the TNP model, which was previously validated with the experiments performed in the EFR with pulverized biomass (thistle) particles.
The results obtained indicate a good agreement between the predictions of the TNP model and the experimental data for particles under ~3 mm, and a progressive deviation for bigger particles, especially regarding the times required to completely release the volatiles. On the contrary, the TKP model reasonably reproduces these devolatilization times, as well as the temperature histories and size evolution of the particles observed for the whole range of test conditions. Neglecting the internal gradients results in a delayed and more 'intense' devolatilization pattern; this, together with the differences in the overall devolatilization times, proves the practical relevance of considering those gradients in facilities where particles above a few millimeters constitute a significant fraction of the fuel particle size distribution.
The measurement of temperatures inside biomass particles demonstrated not to be as straightforward as it was initially expected and required a specific analysis. It was found that the conduction heat transfer along a thermocouple wire may introduce a non negligible error in the temperatures measured inside the particles with `fine¿ thermocouples (50 µm in diameter) which can be up to 300 K for the temperature recorded at the center of a 3 mm biomass particle. A procedure has been developed to estimate the errors and to correct the measured temperatures.
