Caple AUF900 Manual de usuario descargar pdf (Pagina 201)

199

ENSI Erfahrungs- und Forschungsbericht 2011

a very satisfactory performance with nodal RMS

approaching the 4% level (below which further

enhancements are not considered as easily achiev-

able when taking into account measurement bias-

es and uncertainties). The performance is however

not significantly better than with the predecessor

C4E except when implementing the new option

(vers2) to model more precisely the water crosses

in SVEA fuel (although this enhancement remains

moderate). For the PWR KKB1 and KKB2 plants, a

similar performance was achieved in terms of 3-D

power distributions. However, for the predicted

critical boron concentration, the vali-dation results

confirm the bias already observed for KKG. This is

illustrated on the lower part of Fig. 5 where the

C5M results are compared to C4E solutions based

on various library combina-tions. On the basis of

these results and taking note that C5M is a fairly

direct descendant of C4E, the increased bias seen

with C5M is believed to be mainly attributed to the

new E7 library although further investigations are

required to confirm this.

Also related to lattice physics, STARS is within the

EU 7

Framework NURISP project participating to

the development and verification of computational

schemes for the CEA APOLLO-2 solver foreseen

to be one of the main 2-D assembly depletion

code within the NURESIM platform. During 2011,

a computational scheme for pin-cell calculations

was finalised, integrating the new APOLLO-2.8

code along with the 281-group library based on

JEFF-3.1 and updating consequently the scheme

with regards to the resonance self-shielding mod-

els, decay chains as well as the output processing

capabilities. For the assessment, a series of bench-

mark models including PWRs and BWRs as well as

UOX and MOX fuel, were set-up and analysed. To

assess the APOLLO-2 solutions for these models,

the C5M code was used as reference solution and

an overview of the obtained agreement in terms

of the lattice multiplication factor k-inf is shown in

Fig. 6. A few trends are noted among which the

most noticeable is the tendency for APOLLO-2 to

produce increasingly lower reactivities as function

of burnup for the UOX cases. However, although

not shown here, this was found to be related to

C5M predicting lower reactivity losses along deple-

tion, something indicative of a generally harder

spectrum compared to APOLLO-2. But on average,

the agreement is found to be around 500 pcm

which is well within the expected variation range

when considering for instance that only nuclear

cross-section uncertainties could contribute to

such differences if not higher. Therefore, this level

of agreement is considered as sufficiently satisfac-

tory, noting that additional analyses not shown

here have indicated that the employed neutron

group structure might in fact contribute more to

the differences than the computational schemes

or even the neutron data libraries.

Figure 6: Differences

between APOLLO-2

(AP2) and CASMO-5M

(C5M) for different

LWR pin-cell models.

-900

-800

-700

-600

-500

-400

-300

-200

-100

100

200

0 20 40 60 80 100

relative burnup [%]

k-inf difference AP2-C5 [pcm]

PWR-UOX PWR-MOX PWR-3cycles

BWR-VOI40 BWR-VOI80

1 2 ... 196 197 198 199 200 201 202 203 204 205 206 ... 303 304

Caple AUF900 Manual de usuario Pagina 201

Comentarios a estos manuales

Caple AUF900 Manual de usuario Pagina 201

Comentarios a estos manuales

Relacionado con productos y manuales para No Caple AUF900