Modeling and simulation of fluidized bed gasifier of biomass
The ever increasing energy demand and the polluting nature of existing fossil fuel energy sources demonstrate the need for non-polluting and renewable sources of energy. Decentralized energy production from renewable sources of energy can be a feasible long term solution for this problem. The object...
Saved in:
| Main Author: | |
|---|---|
| Format: | Thesis |
| Language: | English English |
| Published: |
2010
|
| Subjects: | |
| Online Access: | http://psasir.upm.edu.my/id/eprint/26672/1/FK%202010%2090R.pdf |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Summary: | The ever increasing energy demand and the polluting nature of existing fossil fuel energy sources demonstrate the need for non-polluting and renewable sources of energy. Decentralized energy production from renewable sources of energy can be a feasible long term solution for this problem. The objectives of this research were (i) to estimate kinetics parameters of gasification process (including pyrolysis and combustion processes) using the ermogravimetric analysis (TGA) (ii) to develop a simulation block of a fluidized bed gasifier using Aspen Plus software (iii) and to compare the simulation results with the experimental results. Three biomass samples were selected, namely palm kernel shell, coconut shell and bagasse. The results of non-isothermal thermogravimetric analysis of biomass samples were analyzed. The samples were heated at three different heating rates, namely 10°C/min, 20°C/min and 50°C/min using a step wise temperature program initialized at 30°C and ended at 1000°C. The TGA studies were carried out in three different atmospheres: nitrogen rich atmosphere for pyrolysis, atmosphere containing air for combustion and carbon dioxide rich atmosphere for gasification. The values for all kinetic parameters in Arrhenius equation were estimated by three different models namely Kissinger model, Least Square Estimation (LSE) for first-order reaction models and distributed activation energy model (DAEM). The estimated values obtained from each model were compared. The results showed that LSE for first-order reaction model agree well with the experimental results, indicating that lignocellulosics components in the mixture behave in the same way as they do separately. The estimated activation energy for pyrolysis of hemicellulose and cellulose contents in coconut shell and palm kernel shell were close. The values for both were 118kJ/mol and 157kJ/mol, respectively. The activation energy estimated for cellulose and hemicelluloses contents found in bagasse were lower. This indicates that shell part of biomass have similar thermal conversion behaviors. Also it is seen that the decomposition process shifts to higher temperatures at higher heating rates as a result of the competing effects of heat and mass transfer to the material. The estimated kinetics data were then used to simulate the operation of a fluidized bed gasifier using Aspen Plus software. The fluidized bed gasifier was divided into a number of blocks. The main block was developed in ASPEN CUSTOM MODELER. The simulation results showed that approximately 15-35% of hydrogen can be produced when the operating temperature is set between 750-1000°C. It is also found that higher operation temperature increases the amount of hydrogen produced. The same trend were observed in experimental results conducted by AlipourMoghadam(2010). For equivalence ratio (ER) values ranging between 0.23-0.27, it is observed that hydrogen production rates reduced as the ER value increased. The amount of methane produced in the lab was higher than the simulation values.Simulation block used to emulate the actual pyrolysis process can be the contributing factor for this discrepancy. To conclude, the kinetic data were obtained from TGA experiments for Malaysia-based agricultural wastes. Estimated data via LSE for first order reaction models fit well with the TGA data. Simulation and experimental results for hydrogen production were comparatively similar, yet, some discrepancies for the production rates of other volatile organic compound (VOC)were observed. |
|---|