ML20059D971
| ML20059D971 | |
| Person / Time | |
|---|---|
| Issue date: | 12/16/1993 |
| From: | Wilkins J Advisory Committee on Reactor Safeguards |
| To: | Taylor J NRC OFFICE OF THE EXECUTIVE DIRECTOR FOR OPERATIONS (EDO) |
| References | |
| ACRS-R-1549, NUDOCS 9401100168 | |
| Download: ML20059D971 (4) | |
Text
O OREQu
/
'o UNITED STATES l'
NUCLEAR REGULATORY COMMISSION ACRSR-1549 7
n h
ADVISORY COMMITTEE ON REACTOR SAFEGUARDS PDR o
WASHINGTON, D. C. 20555
%...* /
December 16, 1993 Mr. James M.
Taylor Executive Director for Operations U.S.
Nuclear Regulatory Commission Washington, D.C.
20555
Dear Mr. Taylor:
SUBJECT:
INDIVIDUAL PLANT EXAMINATICs PROGRAM During the 404th meeting of the Advisory Committee on Reactor Safeguards, December 9-11, 1993, we discussed the status of the Individual Plant Examination (IPE) program and some aspects of the resolution of Unresolved Safety Issues (USIs) and Generic Safety Issues (GSIs) by the IPE and the Individual Plant Examination of External Events (IPEEE) programs.
These matters were also discussed during a meeting of our Subcommittee on Individual Plant Examinations on November 18, 1993.
During this review, we had the benefit of discussions with representatives of the NRC staff.
We also had the benefit of the documents referenced.
We share the conclusion expressed by the staff that the IPE program appears to have exceeded expectations.
We are particularly encouraged that (1) licensees are actually using their IPE results to effect safety improvements at their plants and (2) most, if not all, licensees plan to maintain their IPEs as an ongoing current assessment of plant risks (although not required), and to use the resultc as an important adjunct in making decisions with potential safety implications.
We also note that the definition of the database structure to be used in the collection and correlation of IPE/IPEEE program results has received careful attention early in the program.
Both the format for entering results into the database and flexible retrieval capabilities will be provided for users.
A users handbook is scheduled to be available in mid-1994.
We do have two concerns about the IPE process that we will provide later in this letter.
With respect to the IPEEE process, we are in general agreement with the staff approach.
We would recommend, however, that the staff consider the possibility of developing some method for converting the qualitative approaches for evaluating external events (such as the Fire Induced Vulnerability Evaluation and the Seismic Margin Assessment methodologies) into quantitative equivalents.
This would facilitate determinations of relative significance, provide a more definitive framework for decisionmaking, and aid in an 9401100168 931210
?
\\N
a
. Mr. James M. Taylor 2
December 16, 1993 overall assessment of the status of the population of plants with respect to the safety goals.
i Regarding the Accident Management program, we expect to comment when the program is more nearly complete.
We are favorably impressed, however, with the extensive and constructive interaction between the staff and industry during the development of accident I
management strategies.
We encourage a continuation of this interaction.
Returning to the IPE process, we have two concerns, as foll'ows:
1.
In our limited review of several IPE results, we were per-plexed by the wide variation in reported values for condition-al containment failure probability (CCFP).
The values ranged i
from a few percent to 80 percent, and this was based on only a few IPE samples.
Since in many cases neither public risks nor large release probabilities are provided (nor were they required) in the IPEs, it is difficult to decide how to determine the existence of a containment performance vulnera-bility.
Part of the large CCFP range appears to be related to a variation in the definition of " containment failure."
For example, it appears that different views have been taken on whether basemat meltthrough, deliberate venting, or interfac-ing system LOCA events constitute containment failure.
Furthermore, the mode and timing of containment failure, as well as the precursor events postulated to occur during a core melt sequence, can cause vast variations in the estimated atmospheric source term (not computed in many of the IPEs).
For example, an early overpressure failure in a PWR is obviously of much greater concern in terms of adverse public impact than a late containment basemat meltthrough. Yet, both failure modes may show up in the IPEs as equivalent contribu-tors to CCFP.
As a result of the inconsistent definition of containment failure and the potential for an exceedingly wide variation in the source term resulting from different containment failures, it is not clear how the staff can draw any meaningful conclu-sions regarding containment vulnerabilities for an individual plant. We recommend that the staff give this matter addition-al consideration and try to establish a framework for evaluat-ing containment performance for severe accidents from the IPE information.
Part of this might include the formulation of an impact index to describe the relative risk significance of various modes of containment failure described in the IPEs based on equivalent containment failure parameters from the NUREG-1150 results as well as the results from other Level III PRAs.
4
,l Mr. James M. Taylor 3
December 16, 1993 i
2.
We are concerned that the resolution of safety issues (USIs/
i GSIs) by the IPE process is not being tracked and evaluated by the staff.
We agree, as we have stated in the past, that the
~
c IPE/IPEEE process is an appropriate mechanism for the resolu-tion of those USIs/GSIs that appear to be highly plant-specific.
However, we were informed that the IPE reviews are not to be focused on the treatment and corresponding results for the USIs/GSIs which are expected to be included in the IPEs.
(This position is consistent with the information contained in your letter to the ACRS Chairman dated September 22, 1993.)
We are concerned that inadequate or incorrect models and assumptions might be used in the treatment of the USIs/GSIs. This can obscure the significance of the USIs/GSIs for a particular plant and would not be discovered because of the incomplete review process.
We note in this regard that the latest IPE Review Guidance Document does not specify review requirements for USIs/GSIs.
The staff previously determined that the USIs/GSIs had potential safety signifi-cance (or they would not be safety issues), based on the results of staff research and assessment.
It therefore seems i
important that the IPEs be reviewed specifically to ensure that (a) the USIs/GSIs have been considered by the IPE assessments, and (b) the models and assumptions associated with the treatment of the USIs/GSIs in the IPE be consistent with the staff's understanding of these issues based on the evaluations performed to establish the significance of the USIs/GSIs.
It seems to us that this review could be facili-tated by incorporating specific USIs/GSIs into the database being created by Brookhaven National Laboratory from the IPE results.
We look forward to continued interaction with the staff on these important topics.
Sincerely, Mb.
y J.
Ernest Wilki o,
Jr.
Chairman
References:
1.
U.S. Nuclear Regulatory Commission, Generic Letter No. 88-20, November 23, 1988, " Individual Plant Examination for Severe Accident Vulnerabilities - 10 CFR 550.54 (f)"
2.
U.S. Nuclear Regulatory Commission,-Generic Letter No. 88-20, Supplement 4, June 28, 1991, " Individual Plant Examination of External Events (IPEEE) for Severe Accident Vulnerabilities -
i
f Mr. James M. Taylor 4
December 16, 1993 3.
Letter dated September 22, 1993, from Mr. James M.
- Taylor, Executive Director for Operations, for Dr. J.
Ernest Wilkins, Jr.,
ACRS Chairman,
Subject:
ACRS Letter Dated August 11, 1993: Proposed Resolution of Generic Issue 143, " Availability of Chilled Water System and Room Cooling" 4.
Memorandum dated November 18,
- 1993, from Warren Minners, Office of Nuclear Regulatory Research, for John Larkins, ACRS,
Subject:
IPE Review Guidance Document (Draft Predecisional) 4 T
c
--