AN INTEGRATED APPROACH TO LIVING LEVEL 2 PSA R. Himanen and H. Sjövall Teollisuuden Voima Oy, FIN-27160 Olkiluoto, Finland Presented at: INTERNATIONAL.

AN INTEGRATED APPROACH TO AN INTEGRATED APPROACH TO LIVING LEVEL 2 PSALIVING LEVEL 2 PSA

R. Himanen and H. SjövallTeollisuuden Voima Oy, FIN-27160 Olkiluoto, Finland

Presented at:

INTERNATIONAL WORKSHOPONLEVEL 2 PSA AND SEVERE ACCIDENT MANAGEMENT

COLOGNE, GERMANY

29TH TO THE 31ST OF MARCH 2004

Severe Accident Severe Accident Management in Olkiluoto Management in Olkiluoto

1 and 2 NPP1 and 2 NPP

• Asea-Atom BWR• Reactor thermal power 2500 MW • Net electric power 840 MW• Reactor pressure 7 MPa • Safety systems 4x50 %• Automatic liquid boron system for

ATWS and ATWC• Pressure suppression containment• Containment inerted with nitrogen

during normal operation• Drywell gas volume 4300 m3 • Wetwell gas volume 3000 m3 • Condensation pool volume

2700 m3

Severe Accident Severe Accident Management in Management in

Olkiluoto 1 and 2 NPPOlkiluoto 1 and 2 NPP• Containment design pressure

0.47 MPa • Containment ultimate capacity

1.01 MPa at 100 oC (95 % non-exceedance probability)

• Primary containment surrounded by reactor building acting as secondary containment

•Severe accidents were not included in the original design.•The provisions for severe accident management were installed in Olkiluoto 1 and 2 BWRs during the SAM project, which was finished in 1989.•The SAM approach is hardware oriented.•Plant modifications in order to prevent/withstand severe accident loads and minimize environmental consequences.

• Emergency Operating Procedure for severe accidents

The Emergency Operating Procedure for severe accidents contains instructions for severe accident management and covers all phases of severe accident including a full core melt:

- Primary system depressurisation- Flooding of lower drywell- Containment water filling- Procedures for filtered containment venting- Instructions to recover active core and containment cooling systems


Olkiluoto 1 and 2 NPPOlkiluoto 1 and 2 NPP

PDS 10-6/ry Description

CBP 0.41 Containment by-pass (refuelling only)

RCO 1.3 Reactivity control lost.

ROP 0.13 Very early reactor overpressurization

COP 0.0072 Very early containment overpressurization

HPL 0.045 LOCA initiated core melt begins early at high pressure

HPT 3.6 Transient initiated core melt at high pressure

LPL 0.61 LOCA initiated core melt at low pressure

LPT 8.5 Transient initiated core melt at low pressure

RHL 0.22 LOCA initiated late core melt due to loss of RHR

RHT 2.2 Transient initiated late core melt due to loss of RHR

VLL 0.00005 Unsuccessful RHR using containment venting

VEN (51.) Successful RHR using containment venting (no CD)

FCF (11.) Fuel cladding failure (no CD)

CM 17. Total core damage frequency

Plant damage states and their frequencesPlant damage states and their frequences (Jan 2004)(Jan 2004)

Severe accident phenomena Severe accident phenomena studied in level 2 PSAstudied in level 2 PSA

In-vessel issues: Steam explosion and other in-vessel fuel-coolant interactionsRecriticalityHydrogen generationModes of vessel failure

Ex-vessel issues:

Direct containment heatingSteam explosion and other ex-vessel fuel-coolant interactionsGeneration of noncondensible gasesDebris coolability in the lower drywellCore-concrete interaction

Containment issues:

Non-inert containment during start-upDirect containment bypassContainment venting, leakage and failureBasemat penetration

Integrated simulation of physical and Integrated simulation of physical and probabilistic modelsprobabilistic models

• Simple graphical presentation of CET– ”if–then–else” –statements inside the branching points

• Physical parameters transferred and modified in accident sequences

• Simulation of the phenomenon at branching point– as a function of the input parameter set – production of the output parameter set for next b.p.

• Simulation of the probability at branching point– conditional probability of the branch – as a function of the result of the simulation of the physical

model

Integration of accident progression and Integration of accident progression and nuclide transportation models (1)nuclide transportation models (1)

• The analysis of source term and transportation of radio nuclides integrated into the simulation of each accident sequence

• No need for binning the CET sequences for this analysis

Integration of accident progression and Integration of accident progression and nuclide transportation models (2)nuclide transportation models (2)

•Time dependent transportation model

•Four dynamically sized control volumes– LDW, UDW, WW gas volume, and reactor building

•Time dependent gas flow between volumes – input parameters from MAAP

•Decontamination factors with uncertainty distributions

– pools – filter– containment spray– deposition on surfaces

The strength of containment weak The strength of containment weak pointspoints

Figure 8.5.4.1-1b: Containment weak points' break pressure as a function of temperature (median and confidence limits)

0,00

0,50

1,00

1,50

2,00

2,50

0 50 100 150 200 250 300 350 400

Temperature oC

Pre

ssu

re [

MP

a]

5% Door

Door

95% Door

5%Eq

Equipm

95%Eq

5%Ho

Hoop

95%Ho

5%DoBY

DomeBoltY

95%DoBY

5%DoSI

DomeSeInst

95%DoSI

5%DoS70

DomeSe70h

95%DoS70

5% 361

361

95% 361

5% 362

362

95% 362

Level 2 PSA showed that the containment may break due to sum pressure of steam and noncondensible gas

Modification in procedures:- Venting line isolation valve to be

left open after initiating event.- Possibility to fast automatic

venting through the rupture disk line


Olkiluoto 1 and 2 NPPOlkiluoto 1 and 2 NPP

No modifications

Air lockstrengthened

Flooding trainedcurrent status

Inert start-up

Early - aggressivephenomena

Early - no flooding

Not inert - Early

Not Inert - Very early

Refueling

Early Venting

Figure 5: Impact of modifications, summaryFigure 5: Impact of modifications, summary

Figure 1: Venting line to be left open after Figure 1: Venting line to be left open after IE(1997). IE(1997).

Total LERF 7.9E‑6/ry, unfiltered 7.0E‑6/ry (89%)Total LERF 7.9E‑6/ry, unfiltered 7.0E‑6/ry (89%)

E_NFL 55%

L_CF 0%

REFUEL 5%

L_VENT_U 0%

E_VENT_U 11%

E_CF 8%

VE_VB_O2 2%

E__VB_O2 18%

VE_UD_ 0%VE_CBP 0%E_CF 8%E_NFL 55%E__VB_O2 18%VE_VB_O2 2%L_CF 0%REFUEL 5%L_VENT_U 0%E_VENT_U 11%L_VENT_W 0%

Severe Accident ManagementSevere Accident Management in Olkiluoto 1 and 2in Olkiluoto 1 and 2 MODE MODE

PROJECTPROJECT

• Energetic ex-vessel fuel coolant interactions

The range of the dynamic loading of steam explosions is estimated to be 10 to 30 kPas.

Regarding steam explosion loads the concrete structures are relatively stiff, particularly during the short period when the pressure waves are reflected.

Severe Accident ManagementSevere Accident Management in Olkiluoto 1 and 2in Olkiluoto 1 and 2 MODE MODE

PROJECTPROJECT

The median ultimate load impulse for the containment concrete structures, i.e. for the liner in the lowermost drywell wall sections corresponds to a rigid wall impulse of 54 kPas. The median ultimate load impulse for the personnel access lock was 6.3 kPas.

The lower drywell access lock of Olkiluoto 1 was modified in 2001 and Olkiluoto 2 in 2002 so that it will sustain a steam explosion of 54 kPas. The personnel lock tube is fixed to the concrete wall so that the connection can resist a steam explosion.

0.00

E+00

1.00

E-06

2.00

E-06

3.00

E-06

4.00

E-06

5.00

E-06

6.00

E-06

7.00

E-06

8.00

E-06

No modifications



Inert start-up


Early - no flooding

Not inert - Early


Refueling

Early Venting


Figure 2: Figure 2: Lower containment air lock strenghtened (2001). Lower containment air lock strenghtened (2001).

Total LERF 7.4E‑6/ry, unfiltered 5.8E‑6/ry (79%)Total LERF 7.4E‑6/ry, unfiltered 5.8E‑6/ry (79%)VE_UD_ 0%

L_CF 0%

REFUEL 6%

L_VENT_U 0%

E_VENT_U 21%

E__VB_O2 19%

VE_VB_O2 2%

E_NFL 47%

E_CF 4%

L_VENT_W 0% VE_CBP 0%


Severe Accident ManagementSevere Accident Management in Olkiluoto 1 in Olkiluoto 1 and 2and 2 SIMULATOR TRAINING SIMULATOR TRAINING

•Failure to flood the LDW in time has almost 50% contribution to the LERF.

•Full scope simulator on site•All shifts were trained on the simulator once, and the

flooding seems to succeed in time (2001)•Flooding of LDW trained also to the emergency

organization in full scope emergency exercise (2002)

0.00

E+00

1.00

E-06

2.00

E-06

3.00

E-06

4.00

E-06

5.00

E-06

6.00

E-06

7.00

E-06

8.00

E-06

No modifications



Inert start-up


Early - no flooding

Not inert - Early


Refueling

Early Venting


Figure 3: LDW flooding – operators trained Figure 3: LDW flooding – operators trained (2001). (2001).

Total LERF 6.6E‑6/ry, unfiltered 3.6E‑6/ry (54%)Total LERF 6.6E‑6/ry, unfiltered 3.6E‑6/ry (54%)VE_UD_ 0%

L_VENT_U 0%

E_CF 9%VE_CBP 0%L_VENT_W 0%

L_CF 0%VE_VB_O2 3%

REFUEL 6%

E_VENT_U 46% E__VB_O2 21%

E_NFL 14%


What if?What if?

• Inert start-up from refueling• Several negative effects, like more difficult

leakage check at start-up• Benefit rather small

0.00

E+00

1.00

E-06

2.00

E-06

3.00

E-06

4.00

E-06

5.00

E-06

6.00

E-06

7.00

E-06

8.00

E-06

No modifications



Inert start-up


Early - no flooding

Not inert - Early


Refueling

Early Venting


Figure 4: Inert cmnt when start-up (option). Figure 4: Inert cmnt when start-up (option). Total LERF 6.4E‑6/ry, unfiltered 2.9E‑6/ry Total LERF 6.4E‑6/ry, unfiltered 2.9E‑6/ry

(46%)(46%)

E_NFL 26%

E_VENT_U 54%

E_CF 11%

REFUEL 6%

VE_VB_O2 0%

E__VB_O2 0%


Summary of parts of level 2 PSASummary of parts of level 2 PSA

Structural – analysis of the strength of the containment

– details, strength against static and dynamic loads

– uncertainties before cut off

Physics– thermal hydraulics, phenomena, loads

– sequence specific source terms

– use of several codes, comparison of results

– not to be limited in ”representative” or ”worst” cases

– uncertainties before cut off

Probabilistic– accident sequences

– treatment of uncertainties (not cut off)

– importance ranking

SummarySummary

Structural • Omission of detailed and realistic analyses with

uncertainties may lead to biased risk profile

Physical• Omission of detailed plant and accident

sequence specific analyses with sensitivity studies may lead to misunderstanding of uncertainty and biased risk profile

Probabilistic• Next page

SummarySummary

Probabilistic

• Level 2 PSA in SAM is like map and compass in orienteering

• Without them one can– loose his way in the forest of structures

or – go deep to the endless morass of

physical phenomena

AN INTEGRATED APPROACH TO LIVING LEVEL 2 PSA R. Himanen and H. Sjövall Teollisuuden Voima Oy, FIN-27160 Olkiluoto, Finland Presented at: INTERNATIONAL.

Documents

severe accident phenomena

phases of severe accident

m3 severe accident management

accident sequences simulation

m3 wetwell gas volume

ww gas volume

nuclide transportation

m3 condensation pool