Evgeni Borisov, Risk Engineering Ltd., MELCOR team ... · CV060 CV050 CV010 18.519 030-040 x09-010...

Evgeni Borisov, Risk Engineering Ltd., MELCOR team, Bulgaria

5th Meeting of European MELCOR User Group (EMUG)KTH - Albabova, Roslagstullsbaken 21, Stockholm, Sweden

May 2-3, 2013

Reliability, Safety and Management

Engineering and Software Development Services

RISK ENGINEERING LTD



� Objectives;

� Variants of the nodalizations considered;

� Conditions and assumptions;

� Cases analyzed, initial and boundary conditions;

� Important parameters for the sensitivity studies;

� Summary results from the sensitivity studies;

� Issues identified during the sensitivity studies;

� Conclusions;

� References.

5th Meeting of European MELCOR User Group (EMUG)May 2-3, 2013 2




Main objectives of the presentation are as follows:o To discuss the results from sensitivity studies which are performed in

order to be analyzed WWER-1000 plant model response during severe accident progression when three different MELCOR 2.1 nodalizations for the core region are used. These three nodalizations are respectively with one, five and thirty hydrodynamic volumes in the reactor core region;

o To discuss the results from sensitivity studies which are performed in order to be analyzed WWER-1000 plant model response during severe accident progression when two different MELCOR 2.1 nodalizations for the steam generators secondary side are used. These two nodalizationsare respectively with one and three hydrodynamic volumes for the steam generators secondary side;

o To represent and discuss various MELCOR 2.1 and/or plant model issues which have been identified during the performance of the sensitivity studies.





The nodalization schemes for the reactor pressure vessel and reactor

internals, the primary loops and the steam generators secondary side are

shown on Figure 1, Figure 2 and Figure 3 respectively.





5th Meeting of European MELCOR User Group (EMUG)May 2-3, 2013

010-020010-020010-020

CV013

CV035

CV046

CV032

CV037

CV034

CV055

CV050

CV027

CV054CV014

CV026

CV025

CV033

CV024

CV053CV043CV023

CV056

CV044

CV022CV012

CV047

CV042

CV016

CV057

CV045

CV017

CV052

CV036

CV015

CVX00

CVX09(CV110*)

CV

040

X=1,2,3,4

035-060034-060

033-060032-060

031-060 027-060

057-050

x09-010 (110-010*)

29,440

035-040

CVX09(CV110*)

16.899

18.519

060-070

070-050

18.419

040-060

24.930

020-040

22.928

020

-06

0

CV020

CV

040

CV070

CV050CV091

CV

060

CV

010

22.928047-060

CV

032

033-034 031-032034-035 032-033

035-040

23.728

CV

031

CV

033

091-092

057-060

047-050

CV

034

CV

035

CV092

020-031020-040

26.753

020-032020-033

020

-06

0

020-034020-035

050-091

24.647

040-050

040-060 040-050

070-092

037-060

CV

010

CV

060

CV070

031-050032-050

020-012

033-050034-050

16.899

035-050

050-x00

020-022

017-050

24.930

x09-010 (110-010*)

CVX00

29,440

CVX09(CV110*)

16.899

26.953 26.953

020-032

030-060

18.519

020-040

020-030

060-070

070-050

18.419

060-070

020-042

027-050

24.930

040-050

020

-06

0

040-060

29,440

030-050

CV030

CV020

CV

040

020-052

CV070

CV050CV

060

CV

010

18.519

030-040

050-x00

x09-010 (110-010*)

22.928

CV020

017-060

037-050

050-x00 CVX00

o Figure 1: Reactor pressure vessel and reactor internals nodalizations (for all the

cases 5 radial rings in the COR package are modeled)

5





o Figure 2: Primary loops nodalization (4 separate loops are modeled)

x05-x06

x00-x04

25.275

CV

x01

CV

x05

32,17

x24-x25

x06-x07

x26-x27

x07-x08

x23-x24

22,18

CVx23

x20-x21CV

x02

CV

x03

x21-x22

x27-x28

109-110*

26.125

23.475

x04-x05

CVx25

CV

x08

CVx27

CVx00

24.000

30,128

29.140

x28-x05

20,215 20,215

CV

x06

32,17

29.140

CV

x04

CVx09

30,128

27,93827,938

24.325

x25-x05

x01-x02

CVx28

CVx20

x02-x23

CVx26

CVx07

23.900

x08-x09

x02-x26

CVx21 CVx22

CVx24

25.700

20,640

x02-x03

CV110

x22-x05x02-x20

21,065

23.900x09-010 (110-010*)

25.700050-x00

28,0395

28.7565

29.5545

CV050

CV010

CV458CV458

110-458 109-458

X=1,2,3,4

6





o Figure 3: Steam generators secondary side nodalizations

CVX28

CVX24

CVX27

CVX01CVX22

CVX02

CVX25

FLX03

CVX20

FLX05

FLX04

CVX23

X=5,6,7,8

CVX21

CVX26

Riser

Steam Dome

Downcomer

CVX00

CVx12

x00-

x10

CVx10x0

0-x1

0

X=5,6,7,8

x10-x12CVx10 CVx12

x02-x06

29.155

30.380

31.930

35.275

34.695

35.275

30.380

27.930

34.695

31.930

27.930

7




For the performance of the sensitivity studies, the following conditions and assumptions take place:

o Subversion 5026 of MELCOR 2.1 has been used for the calculations;

o For the flow paths which connect the reactor lower plenum to the reactor core, flow paths inside the core, flow path to the reactor core bypass and the artificial flow paths between the core and the core shroud, the flow blockage option has been activated (FL_BLK). The default value of the minimum porosity which is used in calculation of the flow resistance is chosen. The exception of that assumption is a case with double ended break of a main circulation pipeline (case 3B from Table 2) for which this coefficient is set to 0.5 in order to avoid failure of MELCOR 2.1;

o No safety systems are available with the exception of the case with reactor pressure vessel bottom break (as initiating event);

o HS degasing option is activated (HS_DG) – for the steel structures inside the vessel;

o No operator actions are considered.





For the purposes of sensitivity studies, the following cases have been selected: o Total loss of power supply to the unit – For WWER-1000 (V320) this transient

is characterized by high primary pressure and relatively slow primary water inventory decrease. Input decks with 1, 5 and 30 volumes in the reactor core are used. One and three volumes for the steam generators secondary side are separately used;

o Double ended break of a main circulation pipeline with DN 2x850 mm (cold leg break) – this case is characterized by the most dynamic change in the primary side parameters and very fast degradation of the core in case of unavailability of safety systems. Input decks with 1, 5 and 30 volumes in the reactor core are used. One volume for the steam generators secondary side is used because the secondary side influence on the severe accident progression for that case is not expected to be significant;

o Reactor pressure vessel partial break with DN 1130 mm (cross section area around 1.0 m2) with active and passive safety systems availability – this case is analyzed in order to assess the core blockage behavior for 1, 5 and 30 volumes in the reactor core region with safety systems availability. One volume for the steam generators secondary side is used, because the secondary side influence on the severe accident progression for that case is not expected to be significant;





o Middle LOCA of a cold circulation pipeline (close to the reactor vessel) with

DN60 mm – this case is analyzed in order to assess the reactor installation

response in case of severe accident progression as a result of middle LOCA.

Input decks with 1, 5 and 30 volumes in the reactor core are used. One and

three volumes for the steam generators secondary side are separately used;

o Primary to secondary LOCA with DN 43 mm (PRISE 43 mm) – this case is

specifically analyzed in order to be assessed the secondary side influence on

the primary parameters in severe accident conditions.

Summary information for the cases which have been selected for the sensitivity

studies is presented in Table 1.






CaseCaseCaseCaseInitial Initial Initial Initial

ConditionConditionConditionConditionSafety systems Safety systems Safety systems Safety systems

availabilityavailabilityavailabilityavailabilityVariantVariantVariantVariant

Number of Number of Number of Number of Core/SG Core/SG Core/SG Core/SG volumesvolumesvolumesvolumes

1. Station blackout Nominal Power No

1A 1V/1V

2A 5V/1V

3A 30V/1V

4A 30V/3V

2. LB LOCA with DN2x850 mm

Nominal Power No

1B 1V/1V

2B 5V/1V

3B 30V/1V

3. RPV bottom partial failure with DN 1130

mmNominal Power

Yes (one train from each safety

system)

1C 1V/1V

2C 5V/1V

3C 30V/1V

4. MB LOCA with DN60 mm

Nominal Power No

1D 1V/1V

2D 5V/1V

3D 30V/1V

4D 30V/3V

5. PRISE with DN43 mmNominal Power No

1E 30V/1V

2E 30V/3V

Table 1: Summary information for the cases analyzed

11




For the sensitivity studies performed, the following parameters during the severe accident progression are analyzed: o CPU time consumption (tCPU) – up to the moment of vessel breach;

o Total amount of hydrogen generated in the reactor core (COR-DMH2-TOT);

o Onset of gap release (tGR);

o Time when the maximal fuel cladding temperature reaches 1200oC (tCD);

o Onset of reactor core degradation (tDEG);

o Onset of UO2 relocation into the reactor lower plenum (tMUO2);

o Total mass of UO2 relocated into the lower plenum before the vessel breach (MUO2);

o Onset of the reactor pressure vessel breach (tVBR);

o Total amount of molten material ejected to the reactor cavity (COR-MEJEC-TOT);

o Maximal temperature of the hot legs (Thl_max) – important for station blackout scenarios due to the creep rupture phenomenon (it is assessed 30 min before the vessel breach because HPME/DCH is concurrent to the creep rupture phenomenon);






CaseCaseCaseCase VariantVariantVariantVariant

Number Number Number Number of of of of

Core/SG Core/SG Core/SG Core/SG volumesvolumesvolumesvolumes

CORCORCORCOR----DMH2DMH2DMH2DMH2----

TOTTOTTOTTOT

[kg][kg][kg][kg]

tCPUtCPUtCPUtCPU

[h][h][h][h]

tStZrtStZrtStZrtStZr

[s][s][s][s]

tGRtGRtGRtGR

[s][s][s][s]

tCDtCDtCDtCD

[s][s][s][s]

tDEGtDEGtDEGtDEG

[s][s][s][s]

1. Station blackout

1A 1V/1V 582 10.8 10500 10946 12500 154302A 5V/1V 912912912912 11.3 11500 11691 15000 244352443524435244353A 30V/1V 760 16.9 11000 11347 13000 158444A 30V/3V 668 17.5 10500 10727 12500 14997

2. LB LOCA with DN2x850

mm

1B 1V/1V 114 10.1 5 40 160 8002B 5V/1V 149 10.0 5 45 160 800

3B 30V/1V 131 19.8 5 47 240 720


mm

1C 1V/1V 95 108 6 48 175 797

2C 5V/1V 208 157 6 48 175 807

3C 30V/1V 2.3 331* 120 278 NONONONO NONONONO

4. MB LOCA with DN60

mm

1D 1V/1V 273 11.2 2000 1800 2500 28262D 5V/1V 263 9.8 2000 2000 2500 33413D 30V/1V 295 16.5 2000 2077 2500 31134D 30V/3V 246 17.4 2000 2038 2500 3067

5. PRISE with DN43 mm

1E 30V/1V 454 22.3 8000 8417 9500 105762E 30V/3V 401 22.9 8000 8242 9500 10271

Table 2: Summary results from the sensitivity studies performed – part 1

*Calculated to the 24-th h (accident progression time) after the accident initiation

13





Table 2: Summary results from the sensitivity studies performed – part 2

CaseCaseCaseCase VariantVariantVariantVariantNumber of Number of Number of Number of Core/SG Core/SG Core/SG Core/SG volumesvolumesvolumesvolumes

tMUO2tMUO2tMUO2tMUO2

[s][s][s][s]

MUO2MUO2MUO2MUO2

[tons][tons][tons][tons]

Thl_maxThl_maxThl_maxThl_max********

[K][K][K][K]

tVBRtVBRtVBRtVBR

[s][s][s][s]

CORCORCORCOR----MEJECMEJECMEJECMEJEC----

TOTTOTTOTTOT

[tons][tons][tons][tons]

1. Station blackout

1A 1V/1V 15684 69 1260126012601260 25236 1952A 5V/1V 24490244902449024490 52 1428142814281428 25615 2183A 30V/1V 16389 63 1015 22351 2224A 30V/3V 15946 53 994 20525 194

2. LB LOCA with DN2x850 mm

1B 1V/1V 1940 77 Not needed 8166 154

2B 5V/1V 2470 74 Not needed 9024 178

3B 30V/1V 2480 78 Not needed 7983 157


mm

1C 1V/1V 797 83 Not needed 22350223502235022350 144

2C 5V/1V 9208 76 Not needed 40500405004050040500 114

3C 30V/1V NO NO Not needed NONONONO NONONONO

4. MB LOCA with DN60 mm

1D 1V/1V 5704 70 Not needed 12845 186

2D 5V/1V 4290 77 Not needed 14936 173

3D 30V/1V 5923 72 Not needed 12057 156

4D 30V/3V 5506 76 Not needed 13258 162

5. PRISE with DN43 mm

1E 30V/1V 10698 66 Not needed 17972 194

2E 30V/3V 10422 70 Not needed 18235 180

14





o Figure 4: Total amount of the hydrogen generated for the in-vessel phase of severe accident (cases A, B,)

0

100

200

300

400

500

600

700

800

900

1000

0 10000 20000 30000 40000 50000 60000

Mass

[kg

]

Time [s]

COR-DMH2-TOT_1ACOR-DMH2-TOT_2ACOR-DMH2-TOT_3ACOR-DMH2-TOT_4A

0

20

40

60

80

100

120

140

160

0 10000 20000 30000 40000 50000

Ma

ss

[k

g]

Time [s]

COR-DMH2-TOT_1B

COR-DMH2-TOT_2B

COR-DMH2-TOT_3B

15





o Figure 5: Total amount of the hydrogen generated for the in-vessel phase of severe accident (cases D, E)

0

50

100

150

200

250

300

0 20000 40000 60000 80000 100000

Ma

ss

[k

g]

Time [s]

COR-DMH2-TOT_1DCOR-DMH2-TOT_2DCOR-DMH2-TOT_3DCOR-DMH2-TOT_4D

0

50

100

150

200

250

300

350

400

450

500

0 20000 40000 60000 80000 100000

Ma

ss

[k

g]

Time [s]

COR-DMH2-TOT_3E

COR-DMH2-TOT_4E

16





o Figure 6: Total amount of the hydrogen generated –RELAP/SCDAP – Mass (tons) vs. Time (s) [2]

17





o Figure 7: Maximal gas temperature in the core – station blackout cases

18

500

1000

1500

2000

2500

3000

10000 12000 14000 16000 18000 20000 22000 24000 26000 28000 30000

Tem

per

atu

re[K

]

Time [s]

CFVALU_2357_1V1VCFVALU_2357_5V1VCFVALU_2357_30V1VCFVALU_2357_30V3V

0

500

1000

1500

2000

2500

3000

0 10000 20000 30000 40000 50000 60000

Tem

per

atu

re[K

]

Time [s]

CFVALU_2357_1V1VCFVALU_2357_5V1VCFVALU_2357_30V1VCFVALU_2357_30V3V





o Figure 8: Hot legs gas temperatures (left) and hot legs metal temperature (right)

19

400

600

800

1000

1200

1400

1600

1800

0 10000 20000 30000 40000 50000 60000

Tem

per

atu

re[K

]

Time [s]

CVH-TVAP_400_1V1VCVH-TVAP_400_5V1VCVH-TVAP_400_30V1VCVH-TVAP_400_30V3V 400

600

800

1000

1200

1400

1600

0 10000 20000 30000 40000 50000 60000

Tem

per

atu

re[K

]

Time [s]

HS-TEMP_1V1VHS-TEMP_5V1VHS-TEMP_30V1VHS-TEMP_30V3V




Issue 1: Related to station blackout scenario (5 volumes in the core

region):

When nodalization scheme with 5 volumes in the core is used, then the core

degradation starts immediately after the vessel breach or very late after that

moment which is not physically accurate. This leads to initial ejection of a very

small amount of molten material (mainly steel – about 500 kg) and the peak of

the molten corium ejection starts about 4000 s after the vessel breach (see COR-

MEJEC-TOT_2B_1 on Figure 9). The root cause of that issue might be related to

erroneous activation of the lower head mechanical model (by the code or due to

user mistakes). It is not clear so far what the exact reason is.





This issue has been overcome by the following ways:

o The lower head mechanical model is switched off by the sensitivity coefficient

1600 (3) which is set to 1.0E10. When the molten core material in the lower

plenum reaches around 50-70 tons and there is molten material into the lower

plenum region below the barrel perforated bottom (around 2.5 t), then SC

1600 (3) is set back to its default value of 1.0E3. The lower head fails and the

molten corium ejection seems to be quite reasonable (see COR-MEJEC-

TOT_2B_2 on Figure 9). For Variant 2A (from Table 2) this approach has been

applied;

o Nodalization scheme with 30 hydrodynamic volumes in the core – for that

nodalization this issue does not appear at all.





Depending on the time step this issue might cause additional problems

such as:

o Reactor vessel breach with a very small amount of corium ejected – 1.0E-38 kg

followed by a large corium ejection several hundred seconds after the vessel

breach;

o Generation of error massage: Cavity Overfilled (an attempt to be ejected

physically impossible amount of molten material – around 900 tons). This

used to happen for the MELCOR 2.1 subversions before subversion 5026 for

our WWER-1000 plant model specifically.






o Figure 9: Mass of molten corium ejected – left hand side (close plan)

23

0

200

400

600

800

1000

1200

1400

1600

1800

2000

20000 22000 24000 26000 28000 30000

Mas

s [k

g]

Time [s]

COR-MEJEC-TOT_2B_1

COR-MEJEC-TOT_2B_2

0

50000

100000

150000

200000

250000

20000 30000 40000 50000 60000

Mas

s [k

g]

Time [s]

COR-MEJEC-TOT_2B_1

COR-MEJEC-TOT_2B_2




Issue 2: Related to station blackout scenario (5 and 30 volumes in the

core region):

When nodalization scheme with 5 or 30 volumes in the core is used, then, at the

moment of the reactor pressure vessel breach, the primary pressure oscillation

between 14 MPa and 20 MPa occurred (see CVH-P_20_A on Figure 10). It is

followed by a message for lowed head failure due to overpressure. This issue has

been overcome by the following ways:

o 500-1000 s before the vessel breach the sensitivity coefficient 1505 (1) was

set to 1.0 (core blockage porosity coefficient) (see CVH-P_20_B on Figure 10);

o When 3 volumes for the steam generators are used then this issue does not

appear at all.






o Figure 10: Reactor lower head pressure before and shortly after

the vessel breach

13

14

15

16

17

18

19

20

22200 22250 22300 22350 22400

Pre

ssu

re [

MP

a]

Time [s]

CVH-P_20_A

CVH-P_20_B

25




Issue 3: Related to LB LOCA with DN 2x850 mm (30 volumes in the core

region)

o When nodalization scheme with 30 volumes in the core is used then a very

large temperature peak around 10 000 K in a volume from the core region

occurs and therefore MELCOR 2.1 fails (see CFVALU_2357_0.05 on Figure 11).

This issue is overcome by setting sensitivity coefficient 1505 (1) to 0.5. Values

of that coefficient equal to 0.05 (by default), 0.1 and 0.25 lead to high

temperature peak in the core and code failure.






o Figure 11: Maximal gas temperatures in the core for Issue 3

(right) – blockage porosity 0.05, 0.25 and 0.5

0

1000

2000

3000

4000

5000

6000

7000

8000

9000

10000

11000

600 800 1000 1200 1400

Tem

per

atu

re[K

]

Time [s]

CFVALU_2357_0.05

CFVALU_2357_0.25

CFVALU_2357_0.5

27




Issue 4: Related to RPV bottom partial failure with DN 1130 mm (all

variants);

For that initiating event, quite unexpected results have been obtained. For the

variant with 1 and 5 hydrodynamic volumes in the core, the reactor vessel breach

occurs about 22350 s and 40500 s after the accident initiation respectively (see

Figure 12 (left hand side)). For the case with 30 volumes, core degradation does

not occur at all (to ensure that, the calculation is extended to 24h) (see Figure 12

(right hand side)). At that moment, no reasonable explanation for these results

has been found.






o Figure 12: Mass of molten corium ejected (left) and maximal fuel

cladding temperature (right) for Issue 4

0

20

40

60

80

100

120

140

160

0 50000 100000 150000 200000 250000 300000

Ma

ss [

ton

s]

Time [s]

COR-MEJEC-TOT_1C

COR-MEJEC-TOT_2C

0

500

1000

1500

2000

2500

3000

0 20000 40000 60000 80000 100000

Tem

pe

ratu

re[K

]Time [s]

CFVALU_9017_1CCFVALU_9017_2CCFVALU_9017_3C

29




As a result of the sensitivity studies performed, the following conclusions can be drawn: o The general progression of the severe accident for most of the cases considered is

reasonable. The exceptions are the cases with RPV bottom break (with DN 1130 mm). The obtained results are quite different and it is difficult to draw a clear conclusion about them. This causes great uncertainty in the containment event trees development for these cases;

o Most of the issues which have been identified are related to the nodalization schemes with 5 and 30 hydrodynamic volumes into the core region;

o The variants with 1 hydrodynamic volume in the core have not caused any issues and the code run quite stable. This variant leads to the lowest amount of hydrogen generated for the in-vessel phase of the severe accident in case of station blackout scenario (case 1A). According to the results for the same initiating event which are generated by RELAP/SCDAP, the total amount of hydrogen generated for that phase is around 800 kg (see Figure 6). So the lowest amount of hydrogen for that case may or may not be physically accurate. As it can be seen from Table 2, the COR-DMH2-TOT parameter for cases 2A ÷ 4A is significantly higher than in case 1A which can be explained by the possibility for in-core circulation of water-steam mixture and the different core blockage behavior (more volumes and more flow paths available in the core for 5 and 30 volumes nodalizations). For case 2A, the largest value for the COR-DMH2-TOT parameter comes from the latest core relocation into the lower head and therefore latest vessel breach;





o For LB LOCA with DN 2x850 mm, MB LOCA with DN 60 mm and PRISE 43 mm, the

total amount of hydrogen generated into the core during the in-vessel phase does not

differ significantly for the nodallizations with 1, 5 and 30 volumes in the core. Variant

3B is characterized by physically impossible gas temperature behavior in the core

when the minimal blockage fraction is less than 0.5;

o For the cases 1A and 1B, the maximal hot legs metal temperatures are significantly

higher than in cases 1C and 1D. This result is important when one considers creep

rupture phenomenon which is more likely to occur for the hot legs than for the

pressurizer surge line or SG tube (in the cases where none of the loop seals is

cleared);





1) Actualization of PSA level 2 for KNPP (Kozloduy NPP, Bulgaria) units 5 and

6 (full power and shut-down), Risk Engineering Ltd (project is in

progress);

2) Accident Management Guidelines and Procedures: Improvement of the

Emergency Documentation for Ukrainian NPP, Risk Engineering Ltd.

(project is in progress)






Headquarters:

10, Vihren str.

Sofia 1618

Bulgaria

Tel. + 359 2 8089 703

Fax: +359 2 9507 751

[email protected]

www.riskeng.bg

CONTACT DETAILS




THANK YOU!


Evgeni Borisov, Risk Engineering Ltd., MELCOR team ... · CV060 CV050 CV010 18.519 030-040 x09-010...

Documents

Transcript of Evgeni Borisov, Risk Engineering Ltd., MELCOR team ... · CV060 CV050 CV010 18.519 030-040 x09-010...