| |
Search Options | |||
| Index | Site Map | FAQ | Facility Info | Reading Rm | New | Help | Glossary | Contact Us | |
|||
May 10, 2002 The Honorable Richard A. Meserve SUBJECT: CORE POWER UPRATE FOR THE BRUNSWICK STEAM ELECTRIC PLANT, UNITS 1 AND 2 Dear Chairman Meserve: During the 492nd meeting of the Advisory Committee on Reactor Safeguards, May 2-3, 2002, we completed our review of the Carolina Power and Light Company (CP&L) license amendment request for an increase in core thermal power for the Brunswick Steam Electric Plant (BSEP), Units 1 and 2. Our Subcommittee on Thermal-Hydraulic Phenomena also reviewed this matter during a meeting held on April 23, 2002. During our review, we had discussions with representatives of the applicant and the NRC staff. We also had the benefit of the documents referenced. Recommendation The CP&L application for an increase in core thermal power from 2558 MWt to 2923 MWt (14.3%) for the Brunswick Steam Electric Plant, Units 1 and 2, should be approved. Discussion The BSEP reactors are BWR/4 Mark 1 units, originally licensed to a power level of 2436 MWt. A 5% power uprate to 2558 MWt was approved by the NRC in 1996. The requested power uprate is similar to those already approved for the Duane Arnold Energy Center, Dresden Nuclear Power Station, and Quad Cities Nuclear Power Station. The CP&L application follows the General Electric (GE) Nuclear Energy and NRC-approved ELTR1 and ELTR2 extended power uprate (EPU) licensing topical report framework, with a few exceptions that are consistent with those previously granted to other applicants and described in GE topical report NEDC-33004P, "Constant Pressure Power Uprate" (CPPU). In our letter of April 17, 2002, we recommended approval of the CPPU topical report for application to BWR power increases of up to 20% of the original licensed thermal power. CP&L has committed to modify the standby liquid control system (SLCS) in which the boron solution is sufficiently enriched with Boron-10. This modification will allow the shutdown capability to be met in the event of an anticipated transient without scram (ATWS) with the use of only one of the two available SLCS pumps. The licensee calculates that this modification will reduce the plant's internal events core damage frequency (CDF) and large early release frequency (LERF) by 9% and 28%, respectively. Without the use of enriched boron, the ATWS risk increases slightly due to shortened times for operator decisions. Because of the significant safety benefit that is obtained by offsetting the most significant risk increase associated with EPU, we agree that this modification to the SLCS should be implemented. The staff has determined that the application meets all of the requirements of the regulations, uses approved codes, and follows the required procedures. As in the case of previous staff evaluations of EPUs, these determinations could have benefitted by including the results of independent computations and detailed checks of calculations to support the staff's review and audits of the procedures and conclusions described by the applicant. We encourage the staff to continue to pay close attention to the details of core reload analyses at Brunswick and other BWR EPU plants. This is particularly important with regard to the ways that core thermal success criteria will continue to be met as more sophisticated fuel design and reload management techniques are implemented. The staff should assess the need for more detailed thermal-hydraulic models of the core, replacing the current "averaging" approaches, to complement present neutronic analyses that model the wide variations in fuel composition and power level throughout the core. This review demonstrates an inherent problem in the "two-tier"
regulatory system. The application for the EPU was not risk-informed,
yet a PRA was submitted. This creates a situation in which the PRA is
not seriously reviewed, although it is part of the record. Also, the uncertainties
in human reliability analysis are significant, but there is no mention
of them. The applicant used human reliability models that have not been
reviewed by the staff. The staff acknowledges that large uncertainties
are present and that the models have not been reviewed. However, the staff
concludes that insights regarding the relative importance of operator
actions can be gained. In addition, the potential increases in the change
in core damage frequency ( One can claim that the actual value of PRA quality is essential for risk-informing the regulations. Improvements in PRA quality, such as inclusion of the effects of margin reductions on risk and improving human reliability models, may be discouraged as long as important decisions such as granting power uprates are made by "accepting" PRAs without criticism because the application is not risk-informed. Drs. F. Peter Ford and Victor H. Ransom did not participate in the Committee's
deliberations regarding this matter. Additional comments by ACRS Member Thomas S. Kress and ACRS Member George E. Apostolakis are presented below.
Additional Comments by ACRS Member Thomas S. Kress I agree with the Committee's position on the way the PRA results are used in evaluating non-risk-informed submittals for changes to the licensing basis. I have an additional related concern that the concepts in Regulatory Guide 1.174 are not being properly implemented in the guidance on how to view these submittals in a risk-informed manner. For example, the Brunswick PRA submittal reports a LERF value of 4.27
x 10-6/yr and a There are a number of things wrong with this view of Regulatory Guide 1.174.
If these missing conditions were included, the most appropriate site LERF at Brunswick should be about: (2)(4.27 x 10-6) (3) (1.15) = 3.0 x 10-5/yr The assessed These two values (LERF = 3.0 x 10-5/yr and It can be claimed that the modification to SLCS in combination with the power uprate actually results in a decrease in risk and therefore ought to be automatically acceptable. We must be careful with this approach to dealing with aggregate changes. Clearly, the modifications to SLCS that result in a risk decrease are acceptable. However, this change at Brunswick does not by itself decrease the LERF enough to take Brunswick out of the Region I. Therefore, even with this modification, the plant is still in the Region for which increase in risk (due to the power uprates) should not be allowed without additional justification. While I am convinced that a proper Level 3 risk analysis for the Brunswick site would justify approving the power uprate request, I am disturbed by the staff's cavalier use of the risk-informed decisionmaking process. Additional Comments by ACRS Member George E. Apostolakis Dr. Kress' comments are an excellent example of what happens when the staff does not subject the submitted risk information to serious review because the application is not risk-informed. They also demonstrate how conclusions can be affected when the PRA is incomplete and/or poorly done. References:
|
CDF),
that could arise if the PRA were capable of modeling the effect of margin
reductions on risk, are not included.