No description available.

Pulverized coal injection (PCI) technology is widely practised in blast furnace ironmaking due to economic, operational and environmental benefits. High burnout of pulverized coal in the tuyere and raceway is required for high PCI rate operation. A comprehensive review reveals that although there have been a variety of PCI models, there is still an evident need for a more realistic model for PCI operation in blast furnace. Aiming to build a comprehensive PCI model of a full-scale blast furnace, this thesis presents a series of three-dimensional mathematical models, in terms of model development, validation and application, in a sequence from a pilot-scale to a full-scale, from a simple to complicated geometry, from a coal only system to a coupled coal/coke system. Firstly a three-dimensional model of pulverized coal combustion is developed and applied to a pilot-scale PCI test rig. This model is validated against the measurements from two pilot-scale test rigs in terms of gas species composition and coal burnout. The gas-solid flow and coal combustion are simulated and analysed. The results indicate that the model is able to describe the evolutions of coal particles and provide detailed gas species distributions. It is also sensitive to various parameters and hence robust in examining various blast furnace operations. This model is then extended to examine the combustion of coal blends. The coal blend model is also validated against the experimental results for a range of coal blends conditions. The overall performance of a coal blend and the individual behaviours of its component coals are analysed. More importantly, the synergistic effect of coal blending on overall burnout is examined and the underlying mechanisms are explored. It is indicated that such synergistic effect can be optimized by adjusting the blending fraction, so as to compensate for the decreased burnout under high coal rate operation. The model provides an effective tool for the optimum design of coal blends. As a scale-up phase, the coal combustion model is applied to the blowpipe-tuyereraceway region of a full-scale blast furnace, where the raceway is simplified as a tube with a slight expansion. The in-furnace phenomena are simulated and analysed, focusing on the main coal plume. The effect of cooling gas conditions on combustion behaviours is investigated. Among the three types of cooling gas (methane, air, and oxygen), oxygen gives the highest coal burnout. Finally, a three-dimensional integrated mathematical model of pulverized coaVcoke combustion is developed. The model is applied to the blowpipe-tuyere-raceway-coke bed region of a full-scale blast furnace, which features a complicated raceway geometry and coke bed properties. The model is validated against the measurements in terms of coal burnout from a test rig and gas composition from a blast furnace, respectively. The model gives a comprehensive full-scale picture of the flow and thermo-chemical characteristics of PCI process. The typical operational parameters are then examined in terms of coal burnout and gas composition. It is indicated that the final burnout along the tuyere axis is insensitive to some operational parameters. The average burnout over the raceway surface can better represent the amount of unburnt coal particles entering the surrounding coke bed and it is also found to be more sensitive to the changes of most parameters. In addition, the underlying mechanisms of coal combustion are obtained. The coal burnout strongly depends on both oxygen availability and residence time. The existence of recirculation region gives a more realistic coal particle residence time and burnout. Compared with the fore-mentioned two models, this model is considered as a more comprehensive model of PCI operation for understanding the infurnace behaviours and provides more reliable information for the design of operational parameters.

Coal continues to be burned by direct combustion in packed or moving bed in small size domestic furnaces, medium size industrial furnaces, as well as small power stations. Recent stringent restrictions on exhaust emissions call for a better understanding of the process of combustion of coal in beds. The present study is a prelude to developing methods of analysis to obtain this improved understanding. A one-dimensional steady-state computational model for combustion of a bed of solid fuel particles with a counterflowing oxidant gas has been developed. Air, with or without preheating, is supplied at the bottom of the bed. Spherical solid fuel particles (composed of carbon and ash) are supplied at the top of the bed. Upon sufficient heating in their downward descent, the carbon in particles reacts with oxygen of the flowing gas. The governing equations of conservation of mass, energy, and species are integrated numerically to obtain the solid supply rate whose carbon content can be completely consumed by a given gas supply rate. The distributions of solid and gas temperatures, of concentrations of various gas species, of carbon content in solid, and of velocity and density of gas mixture are also calculated along the bed length. The dependence of these distributions on the solid and gas supply rates, the air supply temperature, the size of solid fuel particle, and the initial carbon content in solid is also investigated. The calculated distributions are compared with the available measurements from literature to find reasonable agreement. More gas supply is needed for complete combustion at higher solid supply rate. At a given gas supply rate, more solid fuel particles can be consumed at higher gas supply temperature, for larger particle size, and for lower initial carbon content in solid. The temperature of the bed becomes higher for higher solid supply rate, higher gas supply temperature, larger solid particle diameter, or lower initial carbon content in solid. These reasonable results lead one to encourage extension of the model presented here to more complex problems involving combustion of coals in beds including the effects of drying and pyrolysis. / Graduation date: 1994

A dissertation submitted to the Faculty of Engineering, University of the Witwatersrand, Johannesburg, in fulfilment of the requirements for the degree of Master in Science in Engineering, Johannesburg September 1998 / Combustion modelling of utility furnace chambers provides a cost efficient means to extrapolate the combustion behaviour of pulverised fuel (pf) as determined from drop tube furnace (DTF) experiments to full scale plant by making use of computational fluid dynamics (CFD). The combustion model will be used to assimilate essential information for the evaluation and prediction of the effect of • changing coal feedstocks • proposed operational changes • boiler modifications. TRI comrnlssloned a DTF in 1989 which has to date been primarily used for the comparative characterisation of coals in terms of combustion behaviour. An analysis of the DTF results allows the determination of certain combustion parameters used to define a mathematical model describing the rate at which the combustion reaction takes place. This model has been incorporated into a reactor model which can simulate the processes occurring in the furnace region of a boiler, thereby allowing the extrapolation of the DTF determined combustion assessment to the full scale. This provides information about combustion conditions in the boiler which in turn are used in the evaluation of the furnace performance. Extensive furnace testwork of one of Eskom's wall fired plant (Hendrina Unit 9) during 1996, intended to validate the model for the ar plications outlined above, included the measurement {If : • gas temperatures • O2, C02, CO, NOx and S02 concentrations • residence time distributions • combustible matter in combustion residues extracted from the furnace • furnace heat fluxes. The coal used during the tests was sampled and subjected to a series of chemical and other lab-scale analyses to determine the following: • physical properties • composition • devolatilisation properties " combustion properties The same furnace was modelled using the University of Stuttgart's AIOLOS combustion code, the results of Which are compared with the measured data. A DTF derived combustion assessment of a coal sampled from the same site but from a different part of the beneficiation plant, which was found to burn differently, was subsequently used in a further simulation to assess the sensitivity of the model to char combustion rate data. The results of these predictions are compared to the predictions of the validation simulation. It was found that the model produces results that compare well with the measured data. Furthermore. the model was found to be sufficiently sensitive to reactivity parameters of the coal. The model has thereby demonstrated that it can be used in the envisaged application of extrapolating DTF reactivity assessments to full scale plant. In using the model, it has become apparent that the evaluations of furnace modifications and assessments of boiler operation lie well within the capabilities of the model. / MT2017

We investigate several aspects of the numerical solution of the radiative transfer equation in the context of coal combustion: the parallel efficiency of two commonly used opacity models, the sensitivity of turbulent radiation interaction (TRI) effects to the presence of coal particulate, and an improvement of the order of temporal convergence using the coarse mesh finite difference (CMFD) method. There are four opacity models commonly employed to evaluate the radiative transfer equation in combustion applications; line-by-line (LBL), multigroup, band, and global. Most of these models have been rigorously evaluated for serial computations of a spectrum of problem types [1]. Studies of these models for parallel computations [2] are limited. We assessed the performance of the Spectral-Line- Based weighted sum of gray gasses (SLW) model, a global method related to K-distribution methods [1], and the LBL model. The LBL model directly interpolates opacity information from large data tables. The LBL model outperforms the SLW model in almost all cases, as suggested by Wang et al. [3]. The SLW model, however, shows superior parallel scaling performance and a decreased sensitivity to load imbalancing, suggesting that for some problems, global methods such as the SLW model, could outperform the LBL model. Turbulent radiation interaction (TRI) effects are associated with the differences in the time scales of the fluid dynamic equations and the radiative transfer equations. Solving on the fluid dynamic time step size produces large changes in the radiation field over the time step. We have modifed the statistically homogeneous, non-premixed flame problem of Deshmukh et al. [4] to include coal-type particulate. The addition of low mass loadings of particulate minimally impacts the TRI effects. Observed differences in the TRI effects from variations in the packing fractions and Stokes numbers are difficult to analyze because of the significant effect of variations in problem initialization. The TRI effects are very sensitive to the initialization of the turbulence in the system. The TRI parameters are somewhat sensitive to the treatment of particulate temperature and the particulate optical thickness, and this effect are amplified by increased particulate loading. Monte Carlo radiative heat transfer simulations of time-dependent combustion processes generally involve an explicit evaluation of emission source because of the expense of the transport solver. Recently, Park et al. [5] have applied quasidiffusion with Monte Carlo in high energy density radiative transfer applications. We employ a Crank-Nicholson temporal integration scheme in conjunction with the coarse mesh finite difference (CMFD) method, in an effort to improve the temporal accuracy of the Monte Carlo solver. Our results show that this CMFD-CN method is an improvement over Monte Carlo with CMFD time-differenced via Backward Euler, and Implicit Monte Carlo [6] (IMC). The increase in accuracy involves very little increase in computational cost, and the figure of merit for the CMFD-CN scheme is greater than IMC. / Graduation date: 2012