NUREG-0825, Safety Evaluation Supporting Util 840709,1231 & 851024 Repts Re Evaluation of Plant for Wind & Tornado Events as Requested in Integrated Plant Safety Assessment Rept, Sections 4.5 & 4.8.Risk from Wind/Tornado Events Assessed
| ML20214F434 | |
| Person / Time | |
|---|---|
| Site: | Yankee Rowe |
| Issue date: | 05/13/1987 |
| From: | Office of Nuclear Reactor Regulation |
| To: | |
| Shared Package | |
| ML20214F429 | List: |
| References | |
| RTR-NUREG-0825, RTR-NUREG-825 NUDOCS 8705260085 | |
| Download: ML20214F434 (11) | |
Text
{{#Wiki_filter:, - . - . . . -_
# 4
\ UNITED STATES NUCLEAR REGULATORY COMMISSION
$ I WASHINGTON, D. C. 20556
' \.....
SAFETY EVALUATION BY OFFICE OF NUCLEAR REACTOR REGULATION RELATING TO NUREG-0825 SECTIONS 4.5 WING AND TORNADO LOADIN'G AND 4.8 TORNADO MISSILES YANKEE NUCLEAR POWER STATION YANKEE ATOMIC ELECTRIC COMPANY DOCKET NO. 50-029
.1. 0 INTRODUCTION i 4
-i Yankee Atomic Electric Company (YAEC), licensee for the Yankee Nuclear Power l Station, has performed a. cost-benefit analysis of plant modifications for I reduction of the risk from wind and tornado events. The-analysis includ;s an assessment of the risk for the existing plant configuration (which includes the dedicated safe shutdown .;ystem), and of the potential risk reduction of various upgrades.
l
2.0 BACKGROUND
1 The Yankee plant was originally designed for straight wind pressures corre:ponding to an 80 mph wind. Durin tornado hazard at Yankee (Topic II-2.A)g the'SEP review of the wind and
, the staff developed a hazard function
. for wind and tornado events, which was transmitted to the licensee in r
Reference 1. The hazaM function included a median (50% confidence' limit) i curve and an upper 95% confidence limit curve. For a given wind speed, there
- can be an order of magnitude difference in the probability of exceeding that
' I speed depending on whether the 50% curve or the 95% curve is consider 9d. The topic evaluation recommended that the tornado with probability of 10~ / year
, , (95% limit) be used. The staff's review under Topics III-2, Wind and Tornado Loadings and III-4.A, Tornado Missiles, transmitted by References 2 and 3 respectively, concluded that many structures and components could not withstand this hazard, which corresponds to a 300 miles per hour (mph) tornado. ssment Report (IPSAR) for Yankee, NUREG-0825 In the4),Integrated (Ref. YAEC proposed PlantthatSafety a 10- Assg/ year tornado-(median) of 110 mph was a . appropriate design basis and proposed to satisfy the following objectives t , ggate adequate tornado protection, rather than to upgrade for the 10-y/ year (1) Maintain integrity of the reactor coolant pressure boundary - (2) Maintain integrity of secondary system pressure boundary as a-heat sink (3) Assure capability for steam generator feedwater and primary system makeup. 8705260005 870513 PDR ADOCK 05000029. p PDR s
+r qmr- m + - - ' y yp m+ w+--r+mv&** , , - - - 9 g -ey y--%w- t+w- ,, yme.
- W-e, en , -w- N --W's
x .- j The staff concluded that the general method was acceptable but that the following. specific recomendations should be used: (1) Determine the capacity of structures, systems and components required for reaching a ot shutdown e ndition at windspeeds corresponding to the 10_g/ year and 10 g/ year upper 95% confidence limits. (2) Detemine the modifications needed to upgrade the plant to protect against both windspeeds. . (3) Estimate the costs of such modifications.
-(4) Perfom a cost-benefit evaluation to decide what modifications to make.
For tornado missiles, the staff position in the IPSAR was that the missiles described in the staff's August 31, 1982 safety evaluation (Ref. 3), a steel rod and a utility pole, should be considered in the licensee's analysis of the effects of winds and tornadoes. By letter dated July 9,1984 (Ref. 5), the Tornado Cost-Benefit Analysis was i submitted. In response. to staff questions, Revision 1 was submitted on December 31, 1984 (Ref. 6). Sample structural calculations, including the fire water tank, block walls and structural steel, were submitted on February 13, 1985(Ref.7). A more detailed report addressing the risk analysis methodology was subsequently provided on October 24, 1985 (Ref. 8). Responses to staff requests for clarification were provided in References 9 and 10 l Three meetings with the licensee were also conducted for discussion of questions and for staff audit of calculations. In the sections below, the licensee's analyses and conclusions are evaluated by the staff. 3 3.0 DISCUSSION 3.1 TIATANTURCTION The wind and tornado hazard probabilities used in the licensee's analysis are the following: Tornado .. Annual 50% Upper 95% Straight Wind Probability Value Value Upper 95% Value 10-4 40 85 110 10-5 110 165 --- i These wind and tornado values are slightly different from those presented in Ref. I because an adjustment was made for the local region area used in the hazard analysis. At the 10-4 level, both the straight wind and tornado (differential pressure) effects were separately considered to obtain the limiting case. The NRC staff posit on is that wi dspeeds corresponding to 1 upper 95% confidence limits for 10-{/ year and 10-g/ year annual probability--be ; used for evaluations. l
- 3-3.2 DEDICATED SAFE SHUTDOWN SYSTEM As part of the resolution of SEP Topic III-6 (Seismic Design Considerat' -) and of Appendix R to 10 CFR~Part 50 (Fire Protection), the licensee has implemented a dedicated Safe Shutdown System (SSS), which is designed to enable the plant'to reach a safe shutdown condition in the event that the systems normally used for shutdown are not operational. Included within this scope is the reactor coolant pressure boundary, main steam and main feedwater piping, a _means of primary system makeup, a means of decay heat removal via the steam generators, and nonitoring instrumentation. Some existing and some newly installed plant structures, piping and equipment are include. This system could also be used in the event a wind or tornado disabled other plant systems and equipment. Use of the SSS, including the instrumentation for monitoring plant .
conditions (reactor power, steam generator level, pressure etc.), is credited in the analysis at windspeeds below its capacity.. 3.3 PROBABILISTIC ASSESSMENT A plant-specific,' internal events probabilistic' safety study (PSS) was' performed by the licensee and submitted in 1983 for the-Yankee plant. For the cost-benefit evaluation of tornado modifications, this study was adapted. The highlights of the licensee's study include the following:
- 1) Both hazard-induced failures and non-hazard-induced random failures were modeled, in order that a realistic assessment of;the impact of plant modifications could be made.
d
- 2) Ultimate structural capacities of critical components and structures t
were determined and depending on the assumed wind speed, a failure probability of zero or unity was assigned to the operability of I equipnent judged to be adversely impacted by the particular j , structure's failure. i
- 3) A list of initiating events developed for the Yankee Probabilistic Safety Study (PSS) was analyzed for their applicability to this wind j hazard study. Those initiating events with low probabilities that
- could not be caused by the wind speeds of interest (e.g. reactor coolant pipe break inside the containment) were not considered further. Further reduction of this list due to similar plant responses resulted in two main categories of initiating events, namely, loss of offsite power and loss of main system coolant inventory (e.g. through stuck-open relief valve (s)). .*
- 4) The critical safety functions, and systems which provide those functions, needed to mitigate the effects of the two initiating events were identified and event trees were developed.
4 4 m _ __ . . ~ -. .nm........, --. . . - w - ,ww ,_r.-y .~ - . - . _ _ _ _ _ , _ , , _ , , . _ . . , , ,, _ _ . , _ ., ..y,..,,_ . , - , _ ,, ,,_.__.-._,_,,_m _ _ _ . _ , , , , . . , . . , _ _ _ , , . _ , , , , , , , , ,
- p. d.
- 4-
- 5) The quantification of theLlogic expressions resulting from the event
' trees was done in stages to account for . location failures at various
~
wind speeds. Specifically. the hazard wind / tornado speeds were I divided into several ranges with appropriate locations failed in - i each range. Once the conditional probability of core me)t given : a.specified wind' speed -range is detennined, the core melt frequency- l can be calculated by multiplying the. conditional core melt frequency l by the wind hazard frequency. Summing these products yields the l total annual frequency of core melt for the event in question. ,
- 6) YAEC selected the Cable Spreading Room and the Safe ~ Shutdown System (SSS) building as the most'likely candidates for. cost-effective improvements to the plant's . safety. The same . logic expressions used to calculate the " base case" core melt frequency were used-to calculate the core melt frequency for the potential plant improvements as well. This was achieved by changing the wind failure ,
limits for those terms in the event tree. logic expressions that ,
, represent equipment adversely impacted by the failure of the cable .
spreading room or the SSS building or both. I j 7) The PSS release category information was used to evaluate.the consequences of the core melt accidents referenced above. The release category distributions used-for this study were based on the types of core melt scenarios that are expected for the initiating events mentioned above.
- 8) The reduction in risk from each potential modification is combined
' with the cost of such a modification to arrive at the cost / benefit value of each upgrade. i 4.0 EVALUATION l 4.1 STRUCTURAL CAPACITIES l The licensee's analysis assumed failure of structures and components when their ultimate capacity is reached. - The staff and a consultant, Franklin Research Center (FRC), reviewed the licensee's structural analyses to confirm that the assumed capacities were reasonable. The FRC technical evaluation report (TER) discussing the results of this review is attached. j 4.1.1 VENT STACK The plant primary vent stack is not required to function following a high wind / tornado event. However, its failure might endanger. safety-related equipment. Therefore, an analysis was performed of the capacity of the. vent t 4
. - . - - - - . . - - . . . . _ . . ~ . . - . _ , . . ... -
, v, ,.v.,-3, . -
,y , ,,- - . , , , , , , ,4 _4 _ . . , _ . . . - . . . .._.._.,m . . , - . . _4- . ,,_.w.. ,,- , ,. .-,, , - . ., - .
i s-stack under wind / tornado loading. The results of the staff consultant's review _ are providedLin the attached.TER. This analysis demonstrates that the stack 165 mph-(10~g/ year).is well:with n allowable stress -limits under a static torna Dynamic effects, such as vortex; shedding werelalso considered.
- acceptable. . Based on this review, the staff finds the stack performance j 4.1.2 FIRE WATER TANK The licensee's' analysis of the. capacity of the fire water tank was also reviewed as discussed in the TER. Based.on this review, the staff concludes that the windspeed capacities reported by the licensee (161 mph tornado) are valid and based on acceptable criteria.
4.1.3 BLOCK WALLS Many of the exterior walls of the plant buildings are. masonry block.- The
^
ultimate capacity of the walls under tornado loading wasl calculated using by a_ factorcriteria acceptable of 1,67.except that the " design" code allowables were increased ~ criteria. The licensee This factor is higher than that normally allowed by NRC provided results of _ tests performed on existing block and mortar which show a higher actual strength than the minimum specified by the ACI code._ With this value, the ultimate capacities determined .in the - i analysis are consistent with the NRC criteria.- The staff consultant also reviewed the licensee's analysis results and modification designs, as discussed' in the attached TER. Based on this review, the staff concludes that the windspeed capacities claimed by the licensee for both the existing walls and those with design modifications are acceptable.
, 4.1.4 OTHER STRUCTURAL ELEMENTS
~
Other elenents such as structural steel, roof decking, connections and siding were also evaluated. The licensee has committed to evaluate and upgrade if needed the' primary auxiliary building (PAB) roof connections between column lines 6 and 8 to ensure that their capacity is equal to that of the. roof deck. The staff concludes accpetable. that the licensee's resolutions for these issues are 4.2 PROBABILISTIC ASSESSMENT The staff's review of the core melt frequency calculations included examining the initiating events selected and their frequencies; the event trees developed for the initiating events; the hazard-related terns and non-hazard related (i.e., random failures) terms of the logic expressions derived from the event trees; the vital equipment. relied upon in the study and the effect of failed
- walls or structures on this vital equipment; the failure data used for the logic expression basic events; and the structures selected by YAEC as likely candidates withstand high forwinds effective or tornados. cost-beneficial upgrading of the plant's ability to f
4 l
.-...-e - - .A..~-,-.%~ ~+ . ~--~m, . . -- - -~. ~ - - ~ - -
- - . - . , , . - _ . , , . - - , , , - . < . , . ,,,. , , _ _ _ . . . . - ..m __ . . _--m_., _ _ , _ . , _ . _ . . . . .,. . .~ , . , . ,--.-- -
The staff relied'mainly on the CRAC-2 results from NUREG/CR-2723 (Ref.11) as the review basis for the licensee's offsite dose calculation results. The cost estimates provided by the licensee were for construction and design ! costs only, and did not include the costs associated with the preparation-and defense of the cost / benefit study or shutdown costs. The staff finds the costs used in the YAEC cost / benefit study to be conservative and acceptable, j lhe staff also finds the core melt frequency calculations and risk estimates provided by the licensee to be acceptable with the provision that the commitments made by the licensee in the September 5,1986 and December 17, 1986 letter to the NRC are completed. These comitments were made in part to respond to specific staff concerns regarding the cost / benefit analysis , and are detailed in the CONCLUSION section below. 4.3 RESULTS The results of the YAEC study,- for the 95% confidence level wind hazard probabilities, are sumarized below: Core-Melt Reduction in I Frequency Person-Rem / ! (peryear) Plant Life Upgrade Cost $/ Person-Rem l l Base Case 1.07 x 10-5 , , , Cable Spreading , Room Upgrade 2.77 x 10-6 52.9 $108,000 $2,060 , l, SSS Upgrade 9.22 x 10-6 9.9 296,000 30,000
~
l Combined Upgrade 1.22 x 10-6 62.7 404,000 6,440 l The only upgrade that is near the $1000 per person-rem guideline generally used by the NRC is the cable spreading room upgrade according to the analysis perfomed by YAEC. YAEC proposed to upgrade the cable spreading room to resolve this SEP issue. The cable spreading room had a fairly low capacity for winds, about 70 mph. Since failure of this structure was assumed to lead to loss of all vital instrumentation in that room, and since a core melt was assumed when all monitoring instrumentation was lost, these sequence were dominant contributors to the tornado / wind risk and, therefore, upgrading of this room represented a l substantial risk reduction. ! 4
)
It should be noted that the above cost-benefit ratios are based on ten years of remaining-plant life. However, as the licensee has upgraded the cable spreading room and as the cost-benefit ratio for the SSS upgrade is substantially higher, the overall conclusions of the study would not be altered 4 if a longer plant lifetime was assumed. 1 _,w.. . ~.%.-,m -.=sL..,-+=w+-.a m-e+ .
-m-e-.n y,-- -
w-.-r-- - a.-- w,
,-- - - .,,, y,. v., ., q , , yiy rwy- wy-+ g.,-
e . .. t The analysis was performed for two cases; namely, for the tornado corresponding to the 50% confidence limit and for the tornado corresponding to the 95% confidence limit. At the.same windspeed, for these two cases, there is an
- order of magnitude , - at difference in-the probability the 50% confidence limit, a 110ofmph exceeding wind is e that 10- speeg/ year Thus, for event, not examp1g/
a 10 year wind event. The staff has used the 95% limit curve for evaluating the licensee's proposal to account for uncertainties. Based on their analysis, the licensee concluded that modifications to the Safe Shutdown System design to provide capability beyond its existing strength are not justified, but that the cable spreading room should be upgraded. In
. particular, addition of structural steel bents to provide lateral supports for-the' masonry. walls would . increase the ultimate capacity of the structural to 186
- mph (tornado). Modifications to certain block walls in the PAB,~ upper pipe
- chase and tuttine building were also identified by the licensee's analysis.
b These modifications have already been implemented.
- 4.4_ TORNADO MISSILES 4
The licensee has addressed the issue of tornado missiles 1n two ways. First, 4 the licensee notes that at tornado windspeeds in the 10-5/ year range, there is insufficient energy to pick up the steel rod and utility pole missiles and to infifct damage. Secondly, the licensee perfonned some scaling studies from other tornado missile probabilistic analyses t demonstrate that the likelihood of tornado , missile damage is very small (10-6/ year) . Based on the review in the attached TER, the staff concludes that the risk from tornado missile is sufficiently low that missiles need not be part-of the plant design basis. : l 5.0
SUMMARY
AND CONCLUSIONS Based on th position in for the 10~g/ year and 10 g/ year tornadoes.he Based IPSAR, the licensee on a cost-benefit evaluated structura study, , plant modifications to improve plant capacity were identified. .. For tornado windspeeds higher than 165 mph, substantial plant damage _would - i most likely occur. The containment itself can withstand higher windspeeds but systems for decay heat removal would be affected. Thus, as shown by the licensee's calculation, the residual risk from tornad events even after the- ' upgrading proposed by this analysis is 'about a 3 x 10-6/ year core melt frequency (upper 95% confidence limit). The' locations that would-be affected at such windspeeds are not localized; thus, numerous upgrades would be needed to reduce the risk even further. Above 250 mph, integrity of the containment would be affected. Thus, for such an event, core melt with containment failure and the resulting release was i consider d9 . Because of the low probability of a tornado greater than 250 mph (4 x 10 / year for the upper 95% confidence limit), the risk is small. t j e,- -
,...%-. - :---..-,m- .
~~--+ --.. a.-
.- - m -- .-
, . _ . , .,m. . _ - - . , , -y ~ ....-,,-._v,, ., ..m., - --
...,_.,-m.
i
. .. a t
)
~
As noted above, the licensee has implemented modifications in several areas of 4
.the plant to improve'their resistance to tornadoes, i
Further, the licensee committed to the following evaluation / upgrades during the course of the staff's review: . ,
- 1) Exposed main steam and feedwater piping will be evaluated for a 178 mph tornado. Any upgrades required by this evaluction will be integrated with the seismic upgrades for this piping. ~
(implementation scheduled for 1988 outage) t
- 2) The west wall of the upper level Primary Auxiliary Building between ,
- column lines Fh and E cwill be evaluated for a 134 uph wind /121 mph tornado, and upgraded if reautred. (implementation scheduled for 4
1989)
- 3) The west wall of Diesel Generator Cubicle No.3 will be upgraded for a 134 mph straight wind /121 mph tornado. (implementationscheduled for.1989)
The connections of the PAB upper level roof deck to the supporting 4)- steel between column lims 6 and 8 will be evaluated to ensure that , , their capacity is at least equal to that of the roof deck. (implementation scheduled *or 1989) The upgrades outlined above are designed to remove ambiguities the staff felt were in the YAEC cost / benefit study. The upgrade of the diesel generator-4 west wall, for example, will remove the necessity of relying upon the interior walls between the diesel generators for protection against winds speeds below 134 mph straight wind /121 mph tornado. , The staff concludes that the licensee has adequately addressed the effects of , tornado winds on plant structures. The risk assessment performed by the licensee considered possible initiating events that could be caused by a
; tornado event and which structures and equipment would withstand that event. ; In addition, random failures of other equipment which in conjunction with wind
[ damage could lead to a core melt were also included. Based on the above considerations, the staff concludes that: i 1)- the licensee has appropriately assessed the risk from wind / tornado
, events at Yankee.
- 2) the modifications proposed by the licensee result in an acceptable level of risk from high wind / tornadoes for the Yankee plant.
t t i e, n--.m.s , m m n - m m ew,r awes,e nw e--w mvw~~+~~~ ' ' ' ' * '"
,,---,.-,.--n--, ,--------,-w-- -
- - - r -------,m ,- ,wr--,-- en,
-,,,,v,y - ,- ,, ----,---g-,,ge--- - - , - , - g-wger -, , . - - - -- ,e w e ,,
i 9-
- 3) the licensee has demonstrated that the risk from tornado missiles is sufficiently low that further plant modifications to prevent tornado missile damage are not required.
6.0 PRINCIPAL CONTRIBLITORS: , , P.Y. Chen M. Fields E. McKenna Date: MAY 13 E 4 1 _ . - - _ _ . - _ _ . , . , , . . . . . _ _ . . . ~ _ , _ , w- -- r - g- - w , -- y
7.0 REFERENCES
- 1. Letter from D. M. Crutchfield (NRC) to J. A. Kay (YAEC),
SUBJECT:
SEP Topic -II-2. A, Severe Weather Phenomena dated Decemver 17, 1980.
- 2. Letter from R. Caruso (NRC) to J. Kay (YAEC),
SUBJECT:
-SEP Topic III Wind and Tornado' Loadings dated August 31, 1982.
- 3. Letter from R. Caruso (NRC) to J. Kay (YAEO),
SUBJECT:
SEP Topic III-4.A Tornado Missiles dated August 31, 1982. 1 4. NUREG-0825, Integrated Plant Safety Assessment Report-(IPSAR) for the Yankee Nuclear Power Station, June 1983.
- 5. Letter from J. Kay (YAEC) to D. Crutchfield (NRC),
SUBJECT:
SEP Topics III-2 and III-4.A dated July 9, 1984.
- 6. Letter from J. Kay (YAEC) to J. Zwolinski (NRC), dated December 31, 1984
- 7. Letter from J. A. Kay (YAEC) to J. Zwolinski (NRC),
SUBJECT:
SEP Topics /III-2 and III-4.A dated February 13, 1985.
- 8. Letter from G. Papanic (YAEC) to J. Zwolinksi (NRC), SEP Topics III-2 and III-4.A, Cost-Benefit Evaluation, Additional Information on the
- Tornado dated October 24, 1985.
- 9. Letter from G. Papanic (YAEC) to E. McKenna (NRC), Responses to Requests for Additional Infonnation, SEP Topics III-? and III-4.A. dated September 5, 1986.
l
- 10. Letter from G. Papanic (YAEC) to E. McKenna (NRC), dated December 17, 1986. l 11.' NUREG/CR-2723. Estimates of the Financial Consequences of Nuclear Power Reactor Accidents, November 1982. )
l 1 1 1
---g - - g- , , -.,---
-er -,, ,,v w- - - , -}}