Search

This paper focuses on the numerical analysis of the combustion of the liquid fuel Extra Light Fuel Oil (ELFO) in a small combustion device. The calculation was performed by ANSYS CFX computer code for experimental optimal air-fuel ratio (λopt) and the middle power of the oil burner Weishaupt WL5/1-A (37 kW). The goal of the numerical analysis was to establish, whether the experimentally determined λopt ensured the conditions for a complete combustion in each point of the combustion chamber. The comparison of numerical results and measured values confirm the suitability of the chosen numerical model.

An effective air staging strategy is of great importance for achieving low emissions and sensible heat losses through flue gas extraction. In cases where a commercially available system needs to be optimised, often the only viable measure is the modification of process parameters. In this work, the aim is to (1) Discover a combination of the most suitable process parameters based on a multi-criteria decision-making method and (2) To unveil relevant correlations between the two process parameters under study (PA/SA ratio and excess air), emissions and combustion temperatures. A modified commercial small-scale hot water wood pellet boiler was installed into a laboratory heating system. Nine different cases have been addressed within a parametric study, differing in the PA/SA ratio and overall excess air. Emissions, temperatures inside the combustion chamber and the flow rate of air entering the combustion chamber were measured. A low PA/SA ratio of 0.53 (54.7% reduction from factory settings), combined with a low O2 concentration in the flue gases of 5.26% (39.8% reduction from factory settings), and the elimination of infiltration air resulted in a simultaneous reduction of NOx and CO emissions by 14.4% and 93.9% respectively and a flue gas sensible heat loss reduction of 31.6%.

The common combustion disturbances in the fuel bed in a grate-fired boiler pose a great challenge to accurate modelling of the fuel bed conversion. Therefore, it is difficult to describe the lengthwise profiles of the combustibles leaving the fuel bed into the freeboard accurately. This paper investigates how different fuel bed models or grate inlet profiles will affect the Computational Fluid Dynamics simulation of combustion in the freeboard in industrial grate boilers. Two grate inlet conditions, which are very different in the profiles along the grate but yield the same total mass, momentum, species and energy fluxes into the freeboard, are used in the freeboard simulation of a 13 MWth grate boiler, respectively. The key findings from the comparison between the simulation results and measurement data are as follows. Firstly, the fuel bed model or grate inlet condition accounting for the realistic lengthwise biomass conversion pattern can be used reliably for the simulation of a grate boiler and simulation-based boiler optimisation. Secondly, the impacts of different fuel bed models or different profiles of the grate inlet conditions are virtually restricted only to the vicinity of the fuel bed or in the primary combustion chamber. After the secondary combustion chamber in which up-flowing gas is mixed well with the secondary and tertiary air jets, the simulation results are close to each other. For more general and reliable applications, a comprehensive bed model resolving the mixing, reactions, heat and mass transfer in the fuel bed, is needed and under development.

Optimal geometric and process parameters related to air staging, which ultimately lead to lower pollutant formation and a higher thermal efficiency, can be found by utilisation of the CFD simulation method. Simplified and computationally efficient methods for the simulation of complex phenomena which occur in a biomass boiler can, therefore, be of great use, provided they offer reliable and reasonably accurate predictions. A previously developed and validated Computational Fluid Dynamics combustion simulation methodology for large-scale waste to energy grate-fired boilers is modified and implemented for the case of a commercial 32 kW wood pellet hot water boiler. An experimental investigation is performed in order to determine process parameters used in the fuel conversion modelling, case setup and, finally, to enable the validation of the numerical modelling approach. The solid fuel conversion is modelled by a coupled empirical 1D bed model which predicts the composition of the combustible gas mixture released from the fuel bed into the freeboard. Compared to the originally proposed fuel bed conversion model, the modified model includes partial char and volatile oxidation, and relies on gas temperatures measured above the fuel bed. Using the modified fuel bed conversion model, several cases were simulated in order to find the most appropriate parameters of the combined finite rate – eddy dissipation combustion model. Results show that by implementing the modified fuel bed conversion model that relies on measurements, reasonably accurate predictions can be achieved of the gas temperature in the combustion chamber and CO concentration at the flue gas outlet.

Grate-fired boilers are commonly used to burn biomass/wastes for heat and power production. In spite of the recent breakthrough in integration of advanced secondary air systems in grate boilers, grate-firing technology needs to be advanced for higher efficiency and lower emissions. In this paper, innovative staging of combustion air and recycled flue gas in a 13 MW th waste wood-fired grate boiler is comprehensively studied based on a numerical model that has been previously validated. In particular, the effects of the jet momentum, position and orientation of the combustion air and recycled flue gas streams on in-furnace mixing, combustion and pollutant emissions from the boiler are examined. It is found that the optimized air and recycled flue gas jets remarkably enhance mixing and heat transfer, result in a more uniform temperature and velocity distribution, extend the residence time of the combustibles in the hot zone and improve burnout in the boiler. Optimizing the air and recycled flue gas jet configuration can reduce carbon monoxide emission from the boiler by up to 86%, from the current 41.0 ppm to 5.7 ppm. The findings of this study can serve as useful guidelines for novel design and optimization of the combustion air supply and flue gas recycling for grate boilers of this type.

Numerical modelling is widely used in industry for detailed understanding of the combustion process and for appropriate design and optimization of biomass/waste-fired boilers. This paper presents a numerical study of a 13 MWth waste wood-fired grate boiler, based on the coupled in-bed fuel conversion modelling and freeboard combustion modelling methodology. A 1D model is developed for the conversion of the waste wood in the fuel bed on the grate, providing the appropriate grate inlet condition for the 3D simulation of the freeboard region. Since part of the flue gas is recycled into the boiler as an innovative attempt to improve the boiler performance, a refined weighted-sum-of-grey-gases-model of greater accuracy is developed to better address the impacts of the elevated CO2 and H2O vapour concentrations on radiative heat transfer in the boiler. The impacts of full buoyancy on the turbulent flow are also investigated. The temperature profiles at different ports in the furnace are measured to shed some light on the flow and combustion characteristics in the boiler and also to collect some in-flame data for modelling validation. The overall modelling strategy, the new sub-models and the use of recycled flue gas are all of great benefit or reference for modelling and design of grate-fired boilers.

A comparison between two numerical models describing the thermo-chemical conversion process of a solid fuel bed in a grate-fired boiler is presented. Both models consider the incoming biomass as subjected to drying, pyrolysis, gasification and combustion. In the first approach the biomass bed is treated as a 0D system, where the thermo-chemical processes are divided in two successive sections: drying and conversion. Phenomenological laws are written to characterize the syngas release as a function of the main governing parameters. The second model is an empirical 1D approach. Temperature, species concentrations and velocity of the syngas provided by the two models are compared. Sensitivity analyses with respect to the drying agent mass flow rate, the initial moisture content and the composition of the biomass are performed. The relative error between the mean values of the temperature and velocity of the syngas predicted by the two models is equal to about 7%. The application to different types of biomass shows that the difference in the predictions increases as the carbon content grows. The phenomenological model, in fact, generally considers higher conversion rates of this element to volatiles with respect to the analogy model.