ML20046B463
| ML20046B463 | |
| Person / Time | |
|---|---|
| Site: | Prairie Island  |
| Issue date: | 07/16/1993 |
| From: | Office of Nuclear Reactor Regulation |
| To: | |
| Shared Package | |
| ML20046B461 | List: |
| References | |
| NUDOCS 9308040239 | |
| Download: ML20046B463 (3) | |
Text
,
h, UNITED STATES Q ['*(
(
i !
NUCLEAR REGULATORY COMMISSION g
s,
,/
w Assiscton, o.c. 20sss-oooi
)
j SAFETY EVALUATION BY THE OFFICE OF NUCLEAR REACTOR REGULATION RELATED TO TRANSIENT POWER DISTRIBUTION METHODOLOGY NORTHERN STATES POWER COMPANY PRAIRIE ISLAND NUCLEAR GENERATING PLANT. UNIT NOS. 1 AND 2 DOCKET NOS. 50-282 AND 50-306
1.0 INTRODUCTION
i In a letter dated April 12, 1993 (Ref. 1), Northern States Power Company (NSP) requested U.S. Nuclear Regulatory Comission (NRC) review of the topical report NSPNAD-93003, " Prairie Island Units 1.and 2 Transient Power Distribution Methodology" (Ref. 2). This topical report relates to the NSP method for core power distribution control and presents the methodology used by NSP for determination of V(z) factors. V(z) is the ratio of the transient to equilibrium predicted nuclear hot channel factor (F ", where F " is the maximum local heat flux on the surface of a fuel rod dlv)ided by the average heat flux in the core. This V(z) factor is applied to equilibrium F " values to bound F " values that could be measured at non-equilibrium condit[ons. The V(z) cycle,-dependent values for the Prairie Island Units are incorporated into the plant Core Operating Limits Report (COLR).
A meeting was held on February 9, 1993, between the NRC staff and NSP staff to discuss a change in Prairie Island's V(z) analysis.
Currently, NSP receives this analysis through Westinghouse in accordance with WCAP-8385 (Ref. 3).
It is the intention of NSP to perform its own V(z) analysis based on the information provided in NSPNAD-93003. The main difference between the NSP and Westinghouse analysis is that NSP will be using a three-dimensional calculation versus a one-dimensional /two-dimensional synthesis method performed by Westinghouse. Therefore, the current operational procedures are not affected, only axial offset (AO) limits related to allowed operations i
outside of the normal offset control band. Axial offset is the ratio of the difference in power between the top and bottom halves of the core to the total l
power in the core. The target axial offset (TAO) is a 100% power, unrodde<i, i
equilibrium A0 which is used as a target or reference value for load follow transients.
2.0 EVALUATION The computer methodology used by NSP to analyze the core power distribution is a three-dimensional nodal code called N3P and is based on the EPRI-NODE-P code (Ref. 4). The NRC has approved this methodology for both core design and transient xenon power analysis (Ref. 5). To further demonstrate that N3P modeling is adequate for both core-wide and localized transient xenon I
behavior, N3P predictions were compared to four flux maps taken during a load-9308040239 930716 ADOCK0500g2 PDR
b follow sequence. The comparisons show excellent agreement and confirm the ability of the N3P code to model transient xenon behavior.
In generating a V(z) curve, NSP first established the initial conditions necessary for a transient power distribution analysis. These initial conditions consisted of a set of TA0s, one being a +TA0 which represents an equilibrium core condition in which the power shape is skewed toward the top of the core. The other starting condition is a -TA0 which is an equilibrium core condition where the power shape is skewed toward the bottom of the core.
Although a specified TA0 may be represented by an unlimited number of power j
distributions, each generated in a variety of ways, NSP has found that for a i
given TAO, the resulting V(z) is independent of the method used to generate the reference power distribution. Hence, the V(z) curve is dependent on the TA0, not on the power shape used to generate the TA0.
After establishing the initial conditions, NSP defines the various core operating strategies utilized during load follow maneuvers. From these initial conditions, a 72-hour "3-6-3-12" load follow scenario (used by both Westinghouse and Siemens) using two power ramps, 100-30 and 100-50%, are utilized in four operational modes (Rebound, Float, Plus AO, Minus A0).
In the Rebound mode, the flux difference (41) is maintained as positive as possible (within the target bandwidth) above 90% power, and as negative as possible (within the target bandwidth) at or below 90% power. The Float operating strategy minimizes operator intervention and only requires AI to be maintained within the allowable bandwidth. The Plus A0 strategy maintains al as positive as possible (near top of the target bandwidth) and minimizes control rod insertions at all times but maximizes boration/ dilution system duty. The Minus A0 strategy maintains AI as negative as possible (near bottom of target bandwidth) and maximizes control rod insertions at all times during core operations. NSP has shown that these operating strategies observe the allowable plant technical specification operating regime and induce the most severe xenon oscillations possible through various mechanisms (control rods, boration/ dilution control, or minimal operational intervention). The NRC, therefore, considers these to be the most operationally feasible and bounding conditions under which the most limiting V(z) curve can be constructed.
NSP has analyzed three cycles of data from Prairie Island core designs, P214, P115, and P215. The analysis was performed at middle-of-cycle (MOC) and end-of-cycle (E0C). The resulting V(z) curves display very similar behavior and magnitudes which lead to the generation of bounding (generic) V(z) curves at M0C and E0C. The qualification for use of generic curves requires that the generic case list shown in Table X.A of NSPNAD-93003 be run and confirm that the resultant V(z) curves do not exceed the generic i.urve. Otherwise, a cycle-specific analysis will be required.
3.0 CONCLUSION
Based on the above evaluation, the NRC concludes that the NSP method used to generate a cycle-specific or generic V(z) factor for application to equilibrium F," values to bound F " values that could be measured at non-equilibrium conditions is acceptable with the following requirements.
f
S
+
3
. For cycle-specific V(z) curves:
(1) The cycle specific case list presented in Table IX.A and Figure IX.A must be used.
(2) The cases run must represent allowable Prairie Island Technical Specification operating conditions.
(3) The analyzed TA0 values must bound the measured equilibrium axial offsets over the exposure range of interest.
For generic V(z) curves:
(1) The generic curve must be generated from at least 3 cycles of data with no technical specification changes to power distribution control strategies having occurred since the generation of the generic curve.
(2) The generic case list presented in Table X.A and Figure X.G represents the limiting cases necessary for qualification. of the generic curve.
(3) The analyzed TA0 values must bound the measured equilibrium axial offsets over the exposure range of interest.
This Safety Evaluation (SE), or an approved version of Topical Report NSPNAD-93003 which incorporates this SE, may be refercated in Section 6.7.A.6, " Core Operating Limits Report," of the Prairie Island Technical Specifications as an approved reference for the V(z) curves in the plant COLR.
4.0 R((fRENCES (1) Letter from R. O. Anderson (NSP) to Docurant Control Desk (NRC),
" Submittal of NSPNAD-93003," April 12, 1993.
(2) NSPNAD-93003, " Prairie Island Units 1 and 2 Transient Power Distribution Methodology," Northern States Power Company, April 1993.
(3) WCAP-8385, " Power Distribution Control and Load Following Procedures -
Topical Report," Westinghouse Electric Corporation, September 1974.
(4) Advanced Recycle Nethodology Program (ARMP) System Documentation, EPF.I CCM-3 Research Project 118-1, September 1977.
(5) NSPNAD-8101-A, Rev.1, " Qualification of Reactor Physics Methods for Application to PI Units," Northern States Power Company, December 1982.
Principal Contributor:
L. Kopp Date:
121y16,1!B3