Lattice bum-up calculations in thermal reactors are complicated by the necessity
for use of transport theory to represent fuel rods, control rods, and burnable absorbers,
by many time-dependent variables which must be considered in the analysis, and by
geometric complexity which introduces time-dependent, spatial-spectral variations.
Representation of lattice structure in a core is further complicated by fuel materials and
loading patterns which can be non-symmetric, and by the type of material used as the
moderator.
The incore thermionic reactor system developed under the Advanced Thermionic
Initiative (ATI) is an example of such a reactor. In this design, the fuel is highly enriched
uranium dioxide and the moderating medium is zirconium hydride. The traditional bum-up
and fuel depletion analysis codes have been found to be inadequate for these
calculations, largely for the reasons mentioned above and because the neutron spectra
assumed for the codes such as LEOPARD and ORIGEN do not even closely fit that for
a small, thermal reactor using ZrH as moderator. A more sophisticated codes such as the
transport lattice type code WIMS is suitable for the terrestrial commercial reactors.
However it lacks some materials, such as ZrH, needed in special applications and it is not
capable of performing calculations with highly enriched fuel. Thus a new method which
could accurately calculate the neutron spectrum and the appropriate reaction rates within
the Thermionic Fuel Elements (TFE) is needed to be developed. The method developed
utilizes and interconnects the accuracy of the Monte Carlo Neutron/Photon (MCNP)
method to calculate reaction rates for the important isotopes, and a time dependent
depletion routine to calculate the temporal effects on isotope concentrations within the
TFEs. This required the modification of the MCNP itself to perform the additional task of
accomplishing burn-up calculations. The modified version called, MCNPBURN, evolved
to be a general dual purpose code which can be used for standard calculations as well
as for burn-up. The of burnable absorber Gadolinium which adds complications both in
the physical model and the numerical analysis requires frequent spatial and spectral reevaluations
as a function of burn-up. This difficulty is overcome by the application of
MCNPBURN by assuming that the burnable poison is uniformly mixed in the fuel.
MCNPBURN was benchmarked using a standard Pressurized Water Reactor fuel
element against the LEOPARD and WIMS computer codes. The results from
MCNPBURN show good agreement with LEOPARD and WIMS. The maximum difference
between MCNPBURN and either of the two codes was approximately 9%. The
differences can be accounted for by the appropriate treatment of the accumulated fission
product.
Application of the MCNPBURN for the ATI reactor core, which consists of 165
TFEs and operates at 375 kW of thermal power, showed a system lifetime greater than
the projected lifetime of 7 years at full power. The average efficiency is about 5.86% and
the change in the overall efficiency over the life time is 0.2%. The percentage of fuel
mass burned is estimated to be about 3.6% of the initial mass. Another calculation
includes the influence of burnable poisons mixed in the peak pins to flatten the overall
core radial power distribution was performed. The efficacy of this change is quite
apparent in reducing the power effectively in the peak pins though it may give rise in
power elsewhere in the core. / Graduation date: 1994
| Identifer | oai:union.ndltd.org:ORGSU/oai:ir.library.oregonstate.edu:1957/35833 |
| Date | 30 September 1993 |
| Creators | Abdul-hamid, Shahab A. |
| Contributors | Klein, Andrew C. |
| Source Sets | Oregon State University |
| Language | en_US |
| Detected Language | English |
| Type | Thesis/Dissertation |
Page generated in 0.0047 seconds