ML20056D947
| ML20056D947 | |
| Person / Time | |
|---|---|
| Site: | Davis Besse  |
| Issue date: | 07/29/1993 |
| From: | Leisure M CENTERIOR ENERGY |
| To: | Gillespie F NRC - REGULATORY REVIEW GROUP |
| References | |
| FRN-58FR33285 2163, NUDOCS 9308190048 | |
| Download: ML20056D947 (4) | |
Text
- _
.-w 4g9-249-2302 NUCLEAR LICEFSDG 775 P01 JUL 29 '93 33:18
.s.
7
==rm s
Louis F. Sforz 300 Madison Avenue Vce President - Nucesor Toledo, DH 43652-0001 i
Davis 4 esse (419)245 2300 l
Docket Number 50-346 Post.it' brand fax transmittat memo 7671
- of pas.*
- 4 License Number NPF-3
Fuot ce russw '"'h unwu
- ' rJ R4.
$blID > FZ)f.3od Serial Number 2163 oog* PN3 enon.'(JI *dkME \\ *7tG July 29, 1993
[io 504 Igg 7
'""(116 2 M Z 30 L Mr. Frank P. Gillespie, Director PMAS - Mail Stop 12-G18 i
United States Nuclear Regulatory Commission j
Vashington, D.C.
20555 j
Subject:
Comments on Regulatory Review Group Report
Dear Mr. Gillespie:
i i.
' The Toledo Edison Company, operator of the Davis-Besse Nuclear Power I
Station, has reviewed the Regulatory Reviev Group Report which vas placed in the NRC Public Document Room on May 28, 1993, for a 30-day comment period (later extended to a 60-day comment period ending July l
29, 1993 by 58 FR 33285). Toledo Edison strongly supports the NRC's and industry's efforts to identify and eliminate regulatory requirements and commitments that are economically burdensome yet provide little or no safety value. Toledo Edison concurs that the elimination of burdensome regulatory requirements could indirectly benefit safety, in that the freed up resources that vould result may be redirected to more safety-significant vork.
As you are avare, the resources which Toledo Edison, or any utility, can make available to initiate burden reduction requests are limited.
Toledo Edison recognizes that the NRC's resources are also limited.
As-our resources permit, Toledo Edison vill continue to initiate or participate in industry. activities designed to reduce regulatory burden, provided the potential benefits outveigh the cost, and provided l
that there is a reasonable opportunity for NRC acceptance. The Reviev Group Report provides an excellent summary of potential opportunities l
in this area.
Specific comments on the report are included in the attachment to this letter.
9308190048 930729 T@[
goa nooCx 0300034s PDR
.i ocews common.es j
cg-. ogmoi %
1
/JM'Y
g 419-249-2302 NJCLEAR LICENSIN3 775 P03 JUL 29 '93 13:19 Dock t Nu ber 50-346 Lic:ns] Nu:ber NPT-3 serial Number 2163 Attachment 3
Tage 1 Comments on the Regulatory Reviev Group Risk Technology Application, Volume 4 Use the term "Probabilistic Safety Assessment" (PSA) vice the term Probabilistic Risk Assessment throughout the report. While PSA techniques and applications are still re3atively new, consistent terminology should be used within the industry and regulators. NUMARC, EPRI and utilities are using the term PSA; use of terminology with the vord safety instead of risk may allow for a vider acceptance of these techniques.
In response to the first bullet on page 4-2 that discusses reliance on quantitative results from multiple plant-specific PSAs and the tse of generic failure data, is one particular generie data base being proposed? Most PSA analysts have their own generic data base comprised of the generic sources available to them.
The second bullet on page 4-2 discusses the reliance on single-plant-specific PRA quantitative results in selected areas. Examples are provided for this type of application. An important example that is not included in this listing and should be are justifications for continued plant operation (JCOs).
An important point to note regarding the human interactions write-up in the second paragraph on page 4-7 is the methodology developed by EPRI that utilizes simulator data for human interaction rates.
Plant-specific operator training simulator exercises are observed and the data from these is used to calculate plant-specific human interaction rates. This methodology is described in EPRI NP-6937, Volumes 1, 2 & 3, Operator Reliability Experiments Using Power Plant Simulators.
The last paragraph on page 4-7 and continued on page 4-8 discusses that the use of PSA may be more appropriately applied to the potential for severe core damage or system availability than to public risk. This is certainly the case for most of the utilities who performed the minimum requirements for the IPE because, only a Level 1 PSA along with a containment analysis was performed i.e., a level 3 PSA vas not performed.
Section 4.2.1 (page 4-12) and section 4.2.2 (page 4-13) discuss the elements of a PRA. The paragraphs that discuss initiating events do not mention steam generator tube ruptures or internal floods as initiating events. Both of these initiating events are included in current PSAs.
The Initiating Event Analysis portion of Section 4.2.2 (page 4-13) notes that Boolean models depicting various systems and components contributing to the initiating event are generally not developed. This is not necessarily alvays the case.
For some BOP initiating events, like losses of specific /both trains of service vater, component cooling vater and makeup, Boolean models vere developed and used as the basis for the initiating event frequency.
m
n.
d39-2d*-rTO - te.m LEM t.!CEt45 3 4G L775 Pod JuL 2993 13:2e Dockot-Numb:r 50-346 J*-
Liecnso Nurber NPT-3 Serial Number 2163 Attachment t
Page 2' Table 4.2-1-is confusing. It seems to contain a lot of extra information if the only purpose is.to identify some of the plant In some systems that are modeled in the PSA and those that are not.
cases, the systems that are identified in the table as not explicitly modeled or evaluated are not consistent with several PSAs. For example, most PSAs do explicitly model component cooling vater, normal service water and required ventilation systems.
i
'Page 4-30 discusses hov often the PRA needs to be updated for specific PSA applications. While it is certainly reasonable-to update the PSA-following an outage or in response to a specific design: change, it is i
unreasonable.to make it 'real-time driven'. This is especially true of-the plant specific data analysis. Each application should consider the j
status of hov up-to-date the PSA is; in the majority of cases, a
'real-time driven' PSA is not necessary.
-t i
on the top of page 47, an equation:is provided that calculates the -
+
maximum pre-determined A0T axtension for any single SSC. An example using real component data vould be helpful; it is not obvious that the denominator vould end up as a positive number.
l Differentiation needs to be clarified with respect to PSA application criteria for one-time changes or exemptions and those that.will be i
implemented permanently.
[
)
As discussed in several sections of this report, the PSA application i
criteria (sections 4.4, 4.5 and 4.6) vill need to be applied in several pilot studies before it is implemented. Pilot' studies are necessary to a
evaluate the existing review criteria and further redefine.the guidelines as appropriate. Purthermore, similar to the efforts r
involved in maintaining a 'living' PSA, this Risk Application j
4 Technology vill need to be updated to take into account nov applications and criteria along.vith new state-of-the art PSA techniques and methodology.
y
'I
..i
~
I i
^!
l i
=.
- ~
J:% '
~
sn Lice 451N3 775 PC2 31 29 *93 13,gg g
I
. ~ '
pocket Number 50-346 License Number NPF-3 6
Serial Number Page 2 If you have any questions regarding this matter, please contact Mr. Dale R. Uuokko, Manager - Regulatory Affairs (Acting), at (419) 249-2366.
Sincere 1y, duc.c MKL/dle Attachment cc J. B. Hopkins, NRC Senior Project Manager i
J. B. Martin, Regional Administrator, NRC Region III S. Stasek, DB-1 NRC Senior Resident Inspector Utility Radiological Safety Board e
t l
l l
1
)
1 I
m.
m