- 183 views
- 237 downloads
Simulation and performance characterization of air dense medium fluidized bed for coal beneficiation
-
- Author / Creator
- Azimi, Ebrahim
-
The dry coal beneficiation method, Air Dense Medium Fluidized Bed (ADMFB) system, can offer an efficient solution for removal of ash forming minerals from run of mine (ROM) coal to improve coal quality and alleviate application issues and footprint. The investigation has been performed in several steps; ROM beneficiation studies, optimization of key operating parameters to reach optimum beneficiation levels, integrating coal beneficiation and drying, specifying effect of beneficiation on product quality, and finally, simulating particle segregation in ADMFB. For beneficiation studies, batch and continuous ADMFB apparatus were used to investigate segregation pattern of low and high ash ROM particles once added to a bed of fluidized Geldart type B particles. Design of experiment methods were used to study the effect of main operating parameters (superficial air velocity, separation time and bed height) and their mutual interactions on the performance of batch or continuous ADMFBs. Product (clean coal) ash content, combustible material recovery and system separation efficiencies were considered as process evaluation criteria and desired levels of them were considered for process optimization. Based on the developed mathematical models, several significant mutual interactions were revealed for any of the evaluation responses, sometimes effective than the direct effect of main parameters. Considerably better separation results were obtained for high ash coal (more than 62% ash rejection) than low ash feed (at most 27% ash rejection). Beneficiation performance of ADMFB showed improvement once coarsest particles (5.6-13.2 mm) were fed to the bed instead of finer size fractions, regardless of feed ash content. Application of finer sand particles as fluidization media reduced number of bubbles in bed and its effective density, resulting further promotion in separation quality. Continuous beneficiation experiments on 2.8-5.6 mm high ash feed revealed that, almost the same level of separation (or even better) was achievable in continuous mode as the batch bed. Optimization of mathematical model for minimum clean coal ash content suggested superficial air velocity, separation time and bed height of 19.5 cm/s, 76 cm (full bed length) and 15 cm, respectively for the 2.8-5.6 mm coal particles and the range of operating parameters used in experiments. Repeating experiments showed that, it was possible to produce a clean coal with ash content of 10% from a feed (5.6-13.2 mm) of 29.1% ash and ash rejection, combustible material recovery and system separation efficiency of 65.4, 89.11 and 67.42%, respectively. Staged coal separation and drying experiments presented promising results for combining two processes since acceptable particle separation could be reached in a short time interval. Moisture removal of 33.8 to 52.5% was obtained for 7.5 min fluidized bed (U=18 cm/s) coal drying. Wide range of coal characterization techniques such as ultimate analysis, ICP-MS, Hg analysis, TGA, XRD, XRF and ash fusion temperature were applied. Characterization results indicated that due to beneficiation by ADMFB, HHV and reactivity (burn out rate) of clean coal products had increased (significantly for low ash coal) regardless of feed ash content. On the other hand, the amount of most hazardous elements and mercury content of clean coal products showed different (sometimes severe) levels of reduction. It was concluded that, Na, Fe and Ca are associated with coal phase. Pyritic type S content was not abundant in either of coal samples. Diagnostic experiments and available models predicted an increase in slagging propensity and decrease in molten slags viscosities. CFD simulation of particles fluidization and segregation were investigated considering Euler-Euler approach and using commercial fluid dynamic software. Several stages considered before preparing final three phase model. After conducting several 2D simulations for grid sensitivity, drag function and solid-solid restitution coefficients studies, prediction of a 3D model was compared with the results of reference experiment. The predictability of 3D (3phase) model was found to be 89.33% which was 29.1% better than its equivalent 2D simulation model.
-
- Graduation date
- Fall 2014
-
- Type of Item
- Thesis
-
- Degree
- Doctor of Philosophy
-
- License
- This thesis is made available by the University of Alberta Libraries with permission of the copyright owner solely for non-commercial purposes. This thesis, or any portion thereof, may not otherwise be copied or reproduced without the written consent of the copyright owner, except to the extent permitted by Canadian copyright law.