ELECTRiC POWERar~r i IRESEARCH INSTITUTE

New Plant Seismic IssuesResolution Program - S2.1Incoherence Project - SASSI-CLASSI Comparisons

May 31, 2007

Bob KennedyJim JohnsonSteve ShortFarhang OstadanDan GhiocelGreg Hardy

Topics for Discussion

" Background

* SASSI-CLASSI Comparisons for Incoherent SeismicResponse

* Effect of Rock Coherency Function

" NRC Comments

02006EI.000P~. R06f~fl 00101. 61 01I~2111 16601 a ~II

1

Background

" Past NRC Interaction

" Representative NPP model for SSI analyses

" CLASSI-SASSI Methodologies

fl..,O i.na. i~ou ,~ ,.,~o 3~rl iI

Past NRC Interaction on Seismic Wave Incoherence

" EPRI Report 1013504

" Dec. 20-21, 2006 NRC Meeting @ EPRI

- CLASSlinco was validated for use in analysis of incoherentground motion by comparison to analytical solutions frompublished literature" Luco & Mita, 1987; Veletsos & Prasad, 1989; Luco & Wong, 1986

" March 1, 2007 NRC Meeting

- SASSI results for rigid massless foundations agreed closely withCLASSlinco and analytical solutions from published literature(Luco & Mita, 1987)

- Preliminary SASSI and CLASSI comparisons presented forrepresentative NPP stick model with SSI

0-O-e EP..1 -....o .. - -...

2

! Conclusions from March 2007 NRC Meeting

* Response spectra and transfer functions forCLASSlinco, SASSI-SRSS, & SASSI-LC were presented

• Very close agreement for coherent ground motion aswould be expected.

* All methods agreed closely for incoherent responsespectra & transfer function amplitudes at the foundation

* There was generally good agreement at structurelocations, but there are some exceptions requiring moreinvestigation

* Overall, good agreement between CLASSlinco andeither SASSI method was demonstrated

* Recommended action - Examine outrigger response notin agreement

I 417

ASB

Representative NPPStructure StickModel with Outriggersand Offset MassCenters

406

scv

401

L.

* ISO

ha -

I~t~2II -Q~

Rock Site ProfileShear Wave Velocities vs. Depth

i

a r-f aiCI;CbJ rAr-T

a I- E... - - ..O~OO. ~kt PAAMO A...wch A~flA a. b.o. AM abA ,nm'~. 7

4

CLASSI-SASSI Methodologies

" CLASSlinco- Deterministic phasing

" CLASSlinco-SRSS- Structure response to each foundation input motion combined by

SRSS" SASSI-Simulation

- Spatial modes assigned random phasing- Mean of structural response to spatial modes computed

* SASSI-SRSS- Structural responses to each spatial mode are combined by

SRSS* SASSI-LC

- Linear combination (algebraic sum) of spatial modes used tocompute structural response

5

CLASSI-SASSI Agreement Not Adequate for CISOutrigger (Node 229) Z response due to Z input(March meeting)

Node 229-CIS x response due to z input

{V -CLASShnc

SASSI-SRSS '

Fr~queonoy (8w)

700.- 00 I 05,.<50<,50 M,-, ,,5ASIf,

CLASSI-SASSIValidation for Incoherent

Ground Motion

12 1-- d 1 0*0 ,,* ,, ....

6

Examination of the Differences

" Characteristics of foundation/structure/solution techniquesthat lead to differences have been identified

" Good agreement between CLASSI and SASSI forfoundation transfer function amplitude and responsespectra for all six DOF

- The relative phasing of foundation response componentsis different between CLASSI and SASSI solutions suchthat the structural response is different

* Reevaluated the treatment of phasing in both CLASSI andSASSI to capture the random incoherence nature

020S ft.U..,, hh. ~ A ~gn ... r.~.13 ~W I

7

CLASSI-SASSI Methodologies

" CLASSlinco- Deterministic phasing

" CLASSlinco-SRSS- Structure response to each foundation input motion combined by

SRSS" SASSI-Simulation

- Spatial modes assigned random phasing- Mean of structural response to spatial modes computed

* SASSI-SRSS- Structural responses to each spatial mode are combined by

SRSS* SASSI-LC

- Linear combination (algebraic sum) of spatial modes used tocompute structural response

15 a- r-, ra 1, ,14 CýX. T 4!

Findings

- Appropriate use of CLASSI

- CLASSlinco for scoping and preliminary evaluation

- CLASSlinco-SRSS for final evaluation

* Appropriate use of SASSI

- SASSI-LC for scoping and preliminary evaluation- SASSI-Simulation or SASSI-SRSS for final evaluation

000 BehIc P0, R-00,0001- . All lighb 2 16 ~I I

8

CLASSI-SASSI Agreement Not Adequate for CISOutrigger (Node 229) Z response due to Z input(March meeting)

Node 229-CIS z response due to z input

"' SASSI-SRSS

7 SASS)I ]

1- 7 .

o 10oo0

Frmquercvy (Hz)

0200 E006 O~n5.66,2 0665. o. 050 ,one.17

CIS Outrigger Response Comparison

Node 229-CIS z response due to x input

10-

10 100

Frequency (Hzo)

16 =-f2I200 - 01, E.- .- 060f 51000.0. . All v--e-002

9

I CIS Mass Center Response Comparison

Node 29-CIS x response due to z input

35

Frequency (H.)

~2ifO~OOC EJ.fl P., ~ b0. n~. a eho d. 19

Conclusions

* CLASSI and SASSI incoherency approaches thatincorporate random phasing are in close agreement for allcases considered to date

* CLASSIinco-SRSS, SASSI-Simulation, and SASSI-SRSSare validated for evaluation of seismic response toincoherent ground motion

- CLASSlinco and SASSI-LC are good approximations

00 r E- tu-u, 1 0•A6;,gfl ... 202

10

Next Steps

* Industry Actions

- Document CLASSI-SASSI comparisons- Develop user guide for running incoherent SSI

analysis

" NRC Action- Reachclosure

- Concurrence that CLASSI and SASSI can be used toevaluate seismic response to incoherent groundmotion

.. ............... i• ..... ~21 I'-I•• 1......

- MEMMOOMMMIL - --, I I -

Effect of Rock Coherency Function

11

!Abrahamson Coherency Functions

* Ground motion coherency functions- Soil - Surface foundations (NAA 2005-2006)

- Soil - Embedded foundations (NAA Dec 2006)

- Rock - Surface and embedded foundations (NAA April 2007)

* Effect of NAA 2007 rock ground motion coherencyfunctions- CLASSlinco seismic analyses performed

- Rock site profile & high frequency input motion

- Three free-field components applied simultaneously

- Comparison of structure response due to coherent motion,incoherent motion (soil), and incoherent motion (rock)

C UU-23 I I ...fe lf.......

12

51 4100U3

210.

SCV

4 17 --ral417out

a .iS

:401 •a...

ASB

75•

120 _~

4Q

Representative NPPStructureStructure StickModel with Outriggersand Offset MassCenters

O0 -gEJ-fkP06-lU I~- hl 0AUg6. 60.-

75'

80

CiS-10' 538 s36g u..40'* 75*

535

I .rz

-ZA-' s -

~l2lt ~25

Comparison of In-Structure Response SpectraHoriz. Y - Top of Auxiliary Building

Node 120mc

5% Damped AP1 000 Top of AB Mass Center - Y Direction

000.4(4

F~q.-.y (Hz)

-Co h.NP112- y .3•nh1-NP 112-y(NAAO 2 CO7) Inco- NP 02-.(. 00 5 20605ý]

0-006 E.- 0 .0.6 3... 4. l, A 3l 6 ll,0,f1- 26 C : I

13

Comparison of In-Structure Response SpectraVert. Z - Outrigger Top of Auxiliary Building

Node 120out

5% Damped AP1000 AB Outrigger- Z Direction

OlIO

01 .0 100 100"

Fr.qu.n.y (H40

I- h • NP212 1ýnh -NP21z(NAA, 2007) -n.- NP21ZzINA, 2005-2006)

02005~~2 a-0P00 r-owo Snot 200f00.2s 7~iI

!Comparison of In-Structure Response SpectraHoriz. X - Outrigger Top of Steel Containment VesselNode 417out

5% Damped APF1OM SCV Outrigger- X Direction

90-

00

70

oo "

0! 10 100 1000

Frequency (0.)

7)h- P45-, flrsh. NP1 40., (SPA/ 2007) lnmoh -74 N I4- (NA).. 2000-2000)J

14