Improvement of an Industrial Cement Precalciner Simulation Application of
the Zone Method
Hugh Jones
Supervisors: Prof K.D. King, Dr D. Desai
Background and Significance
May (1999) developed a process simulation of an existing cement precalciner. The
simulation calculated a number of operating parameters including the extent of
calcination and gas and solids temperatures for a range of initial conditions.
The model was developed for both plug flow and back mixed flow conditions. The
radiative heat transfer was calculated using a correlation developed by Elsner
et al.(1988).
Aims and Objectives
The major aim of the project was to improve the accuracy and applicability
of the simulation developed by May (1999). This was done by targeting the following
two areas:
- Extend the flow pattern to include multiple CSTRs in series
- Introduce the Zone Method to accurately calculate the radiative heat transfer.
Method
Multiple CSTRs
The first improvement was to introduce the multiple CSTRs in
series flow pattern to the simulation. The energy and conversion balances were
solved assuming
- Each CSTR section was fully mixed
- Inlet conditions for each section were set by the outlet conditions from the
previous section
- The effect of the number of CSTRS on calcination was investigated for a number
of initial conditions. The results were compared to those calculated using the
back mixed and plug flow models. The Figure opposite shows the extent of calcination
with respect to the number of CSTRs in series for the default conditions of the
simulation.
Zone Method
The Zone Method was then used to calculate the radiative heat transfer within
the calciner. This involved splitting the calciner into 10 surface and 10 gas
zones, which were assumed to be isothermal. First the direct interchange factors
and the total interchange areas were calculated between each zone. A temperature
profile along the calciner was assumed and then temperature dependent interchange
areas were calculated. Energy balances were written over each zone and solved
for gas temperature and calcination extent. If the calculated temperature profile
was sufficiently different from the assumed values, the last two steps were iterated
until the assumed and calculated profiles converged.
The gas properties were approximated by a three grey gas model using the method
described by Jenkins and Moles (1981). The overview of the calculation method
used is shown.
Direct Interchange Factors
The direct interchange factors, zizj, are the fraction of total energy leaving
zone i being absorbed in zone j. They were calculated using the Monte Carlo method,
where random numbers were used to determine emission angles and beam length. Depend
on the gas absorption coefficient and the geometry of the precalciner.
Total Interchange Areas
The total interchange areas were calculated from direct interchange areas as per
Guruz and Bac (1981). They give the total energy interchange between each zone
within the precalciner. They depend on the wall emissivities and the areas and
volumes of each zone.
Temperature Dependent Areas
Since the radiative interchange changes with temperature, the
interchange areas must be adjusted for the temperature. These temperature dependent
interchange areas were calculated using the total interchange areas and a temperature
dependent weighting factor. The figure opposite shows an example of the surface
to gas temperature dependent interchange areas
Energy Balances
The energy balances were written over each zone using the temperature dependent
interchange factors to calculate the radiative heat transfer. The 20 simultaneous
non linear equations were attempted to be solved using Newton's Method.
Conclusions
- The extent of calcination increased with the number of CSTRs modelled. The
calcination approached the value calculated using the PFR model but did not reach
it.
- Radiative heat transfer from each gas zone was primarily intrazone, with some
transfer to gas and surface zones up to 2 away.
- Radiative heat transfer from surface zones was primarily to adjacent gas zones
with some transfer to gas zones up to 3 away.
- The energy balances could not be solved using the code written. This may be
due to the difficulty of solving for both calcination and temperature, being three
orders of magnitude different, or due to errors in the code.
- The primary aim of further work on this area would be to achieve convergence
of the energy balances. Increasing the number of zones and the complexity of the
fluid flow model within the zones would also add to the accuracy of the simulation.
|
