SCALING OF SMALL BREAK LOCA IN VVER 1000 SYSTEM · International Conference Nuclear Energy for New Europe 2005 Bled, Slovenia, September 5-8, 2005 039.1 SCALING OF SMALL BREAK LOCA

International ConferenceNuclear Energy for New Europe 2005

Bled, Slovenia, September 5-8, 2005

039.1

SCALING OF SMALL BREAK LOCA IN VVER 1000 SYSTEM

Dino Araneo, Alessandro Del Nevo, Francesco D’Auria, Giorgio M. Galassi DIMNP, University of Pisa,

Via Diotisalvi 2, 56126 Pisa, Italy [email protected], [email protected], [email protected], [email protected]

ABSTRACT

This paper deals with the problem of scaling complex scenario measured in experimental facilities and replicated by the use of Cathare2 code. ‘Kv-scaled’ calculation plays an important role in the frame of the “on-transient qualification” of the nodalization (following the procedure envisaged at Pisa University) and to evaluate the importance of some scale-distortions in boundary and initial conditions, which are foreseeable for the plant in comparison with the transients, performed in the facilities. This is the way in order to demonstrate that the “nominal” calculation can be considered the "best estimate" prediction of the plant performance in the case of the transient considered. These steps are part of the methodology developed with the purpose to set up a tool suitable for application of the BE codes to DBA and BDBA conditions (including AM scenarios) as needed for the licensing process.

The reference data base for the activity is the Small Break Loss Of Coolant Accident (SBLOCA) transient, counterpart of the LOBI facility experiment, that has been performed in the Integral Test Facility (ITF) PSB-VVER, installed at EREC (Electrogorsk, Russia), in the framework of the OECD PSB-VVER Project.

The calculation (PSB and Nuclear Power Plant) has been performed using Cathare2 v1.5b computer code and its accuracy has been demonstrated by the qualitative evaluation reported in the present document.

1 INTRODUCTION

The prediction of plant scenarios during postulated accident transient conditions is the main goal of research in nuclear reactor thermal hydraulics. Consequently large experimental facilities have been built, complex computer codes developed and the latter are undergoing a qualification process [1]. The capability of the experimental results and code calculations to reproduce the reactor situations needs to be proven because distortions are introduced by the adopted scaling laws and there are approximations in the basic models implemented in the codes. In order to quantify the uncertainties related to the code calculation of a NPP transient scenario, a methodology has been developed in several years at the DIMNP, this methodology is called UMAE (Uncertainty Methodology based on Accuracy Extrapolation). The approach has been a reliable starting point for the

039.2

Proceedings of the International Conference “Nuclear Energy for New Europe 2005”

development of larger project. It includes the methodology for code assessment, the nodalization qualification and the uncertainty evaluation by the creation of the database of error (kernel of the Code with the capability of Internal Assessment of Uncertainty procedure).

A simplified flow diagram summarizing this methodology is shown in figure 1 and a comprehensive discussion of the methodology is in reference [2] and [3].

This paper is focused on the dashed (ijk) path. Following the directive of the block “j” of the UMAE methodology, two plant calculations must be carried out: the ‘facility Kv scaled” calculation and the “realistic condition” calculation (outside of the scope of this work). In the former calculation, boundary and initial conditions utilized as input are derived from those of an available experimental facility, following a scaling study. In the latter case the reference NPP nominal conditions are used.

If the similarity between the plant calculation and the experimental data is demonstrated, by the Kv scaled calculation it is possible to conclude that the code and the nodalization are able to predict the relevant thermal-hydraulic phenomena occurring in the reference NPP for the same accident scenario. Acceptability of the Plant Calculation (block k) are achieved following the qualitative judgment of the phenomena and RTA (qualitative accuracy evaluation). The quantitative accuracy evaluation, by the application of the Fast Fourier Transform Based Method (FFTBM), is not strictly required. This method quantifies the level of agreement between the system code prediction and the experimental facility measured data by the calculation of the accuracy [4], [5]. When the demonstration of the similarity is achieved, an Analytical Simulation Model (ASM), i.e. a qualified NPP nodalization, is available and the reference ‘realistic condition’ calculation can be performed. If some differences are identified, the results of the calculation are acceptable if the reasons are understood and predictable.

039.3


Figure 1: Flow diagram of UMAE

The paper is focused on the demonstration of the ASM availability by the analysis of SBLOCA conducted by EREC in the PSB-VVER ITF. This test, commissioned in the framework of the OECD PSB-VVER Project [6], is the counterpart test of a LOBI facility experiment. The design of the experiment is the result of collaboration between Pisa University and EREC, following the methodology proposed by Pisa University [7].

The analysis has been accomplished using suitable and qualified input deck for Cathare2 v1.5b system code of the VVER1000 NPP [8] [9].

2 PSB-VVER INPUT DECK DESCRIPTION

The PSB-VVER facility nodalization scheme for Cathare2 code is shown in Figure 2, and information about the code resources is given in Table 1. The correspondence between the zones of the facility and the nodes of the code model are exposed in reference [10].

The vessel model is composed of 19 hydraulic components that are connected through 23 junctions. The lower part of the down-comer is duplicated to simulate as well as possible the inner and the outer part of the element. The heat structures utilized in the RPV scheme are

Code assessment

039.4


represented by 30 wall type elements (excluding core heat structures) to simulate the heat release from structures during the transient; heat losses are foreseen. To take into account the heat losses of the facility a Heat Transfer Coefficient (HTC) is based on EREC data [11]. A wall structure is provided to generate the electric thermal power to simulate power generation along the FRS (Fuel Road Simulator) with uniform power distribution. The heated zone of the core bypass is simulated by an equivalent heat flux imposed to the passive structure.

Four loops represent with geometrical fidelity the real hydraulic configuration of the experimental facility. All four loops are modelled separately: each loop includes a hot leg, a SG, a pump, a loop seal and a cold leg. The heat structures in the primary side piping are represented by 8 wall type elements (excluding horizontal tubes connected heat structures) for each loop to simulate the heat release from structures during the transient. The pumps are schematized with ‘pumpchar’ components and they work with the detailed data provided by EREC [12].

The SG horizontal tubes are schematized by a 6 axial element representing all the 34 tubes of the PSB facility SG. The height and the length of the elements are the same of the tubes in the average position. The only exception is the first and the last tube in order to reproduce the same heated length in the secondary side of the SG. The pressurizer is connected to the hot leg loop 4 via the surge-line. The heat structures in the pressurizer and surge line are represented by 3 wall type elements to simulate the heat release from structures during the transient and the heaters foreseen in PRZ.

The Emergency Core Cooling Systems (ECCS) are simulated by a source component connected to hot legs and cold legs. It must been stressed that in this analysis only 3 LPIS are assumed available and they are connected to the loops 1, 3 and 4. The four SIT are simulated by four accumulator components (only two available in this experiment) and they are connected in pairs to the down-comer and to the upper plenum. The line connecting the SIT and the DC or UP is implicit in the definition of the accumulator component.

The secondary side is very simple and straightforward. Each SG model is composed of 3 components which are connected through 2 junctions between them and with other two with the FW (boundary condition) and with the “small” SL. Two “bcondit” components represent the boundary conditions: one upstream the SG simulates the FW component, the second, downstream, is a boundary condition for the pressure simulating the SL discharge tank. The heat structures in the secondary side are represented by 6 wall type elements for each SG to simulate the heat release from structures during the transient. An “exchanger” operator defines the heat structures that regulate heat exchange between primary and secondary circuits. The geometrical features of the SL have been simulated in detail. Four axial components represent the small steam lines connected to a volume and another axial (the main steam line). Each BRU-a valve is linked to correspondent small steam line using the ‘teebranch’ component, located in different position like in the facility. An ‘axial’ component is foreseen between the boundary condition and the ‘teebranch’ component. Moreover the special component used for the secondary side cool-down system is also represented in the nodalization. It must been stressed that it has been insulated using “fictitious” isolation valves in this calculation because it was not part of the facility during the experiment. Finally all valves (i.e. BRU-a, steam line discharge valve, MSIV) are schematized as link in the command block of the input deck. A deep investigation on the secondary side heat losses has been also addressed [13].

039.5


Table 1 Adopted code resources for Cathare2v1.5b PSB nodalization

PARAMETER VALUE NUMBER HYDRAULIC MODULES

Primary Side 65 (1679) Secondary side 30 (434) Total 95 (2113)

NUMBER OF JUNCTIONS Primary Side 93 Secondary side 30 Total 123

NUMBER T.H. STRUCTURES Primary Side 118 Secondary side 38 Total 156 N. CORE ACTIVE STRUCT. 35

Figure 2. Catahre2 nodalization of PSB-VVER1000 facility: general scheme

039.6


Hot collector region

Downcomer region

Cold collector region

Tubes region

Figure 3: PSB SS SG description and SG #1 PS and SS nodalization scheme

3 VVER1000 CATHARE2 NODALIZATION

Table 2: Main design parameters of VVER 1000 and PSB facility Name VVER -1000 PSB-VVER Scale Factor Scale 1 1:300 -

Number of loops 4 4 - Heat losses, % 0.063 1.8 -

Heat power, МW 3000 10 1:300 Primary circuit volume, m3 370 1.23 1:300

Primary circuit pressure, МPа 15.7 15.7 1:1 Secondary circuit pressure, МPа 6.3 6.3 1:1

Coolant temperature, K 563/593 563/593 1:1 Core length, m 3.53 3.53 1:1

Number of fuel/elctrically heated rods 50856 169 1:300 Core volume, m3 14.8 4.9*10-2 1:302

Upper plenum volume, m3 61.2 20.0*10-2 1:306 Down-comer volume, m3 34.0 11.0*10-2 1:309

Hot legs volume, m3 22.8 8.0*10-2 1:285 Cold legs volume, m3 60.0 24.0*10-2 1:250

Number of steam generators 4 4 - Heat exchanging surface, m2 6115 18.2 1:336

Water volume in SG for primary circuit, m3

21.0 6.8*10-2 1:309

PRZ volume, m3 79 26.3*10-2 1:300 Number of hydro accumulators 4 4 -

Number of pumps 4 4 - Volume of hydro accumulators, m3 240 80*10-2 1:300

Water volume in ACCU, m3 200 66.6*10-2 1:300 The VVER 1000 NPP nodalization has been derived adopting the data and the specific

supporting documents, see reference [14], [15], [16], [17] which contain a detailed description of the Relap5 code nodalization available at the DIMNP. A summary of the main design parameters of VVER 1000 are reported in table 2 where they are compared with the same

039.7


parameters of the PSB facility. The sketch showing the overall view of the nodalization, suitable for identification of nodes is given in figure 4.

Table 3 Adopted code resources for Cathare2v1.5b PSB nodalization

PARAMETER VALUE NUMBER HYDRAULIC MODULES

Primary Side 48 (1636) Secondary side 21 (452) Total 69 (2088)

NUMBER OF JUNCTIONS Primary Side 68 Secondary side 25 Total 93

NUMBER T.H. STRUCTURES Primary Side 96 Secondary side 29 Total 125 N. CORE ACTIVE STRUCT. 35

Figure 4: VVER1000 NPP nodalization by Cathare2 code

Four loops are modeled separately, each loop including hot leg, steam generator, main coolant pump and cold leg. The pressurizer is connected with loop 4 via the surge-line. The relief valve is modeled on the top of the pressurizer.

Two parallel down-comer stacks (DCA, DCB) are used to simulate the down-comer region that extends till to the bottom of the RPV to avoid the presence of nodes with stagnant

039.8


fluid. Down-comer stacks are connected in the upper part with two volumes (DOWNA, DOWNB) for the connection of the four cold legs.

The core model consists of 50856 fuel rods modelled with a WALL component connected to an AXIAL module. The core bypass region is simulated. For the active core region a neutron kinetics model of Cathare is used. For the decay power, the same 11 groups energy and the same decay constants for the fissions products from the ANS79-3 implemented in Relap5, have been used.

Control rods (actually control rods ‘equivalents’) guide tubes (CRGT) are modelled into the UP. This characterizes a ‘tube–internal’ flow path separated from the main flow–path into the UP. Full mixing is assumed for the main flow path in the UP and the four hot legs are connected to a unique node.

Two parallel stacks of nodes have also been distinguished in the UH to separate the flows in the open region and inside the CRGT. This nodalization structure prevents dead zone and allows fluid from the top of the core to flow into the UH.

The various bends (or ‘loop seals’) in the vertical plane present in the cold legs have been considered. The primary side of steam generators has been modelled by two vertical stacks of nodes simulating the hot and the cold header and by three groups of horizontal tubes.

The secondary side is simulated simply by three hydraulic components: a VOLUME for the lower part in order to connect the feed water and auxiliary feed water, an AXIAL for the of the thermal exchange region, and another VOLUME to simulate the upper part of the SG. A pipe simulating the steam line is connected to the upper part of the SG. A valve (BRUA) is connected by means of a pipe to the steam line. This valve (one for each steam line) controls the pressure in the SG, discharging steam in the environment when the opening set point is reached. All the four steam lines are connected to a collector . The turbine is not simulated.

The three LPIS (TQ12), the three HPIS (TQ13) and the HHPIS (TQ14) are simulated by an external source foreseen in Cathare. The two lines of the LPIS inject directly into the UP and the third one into the upper part of the Downcomer region. The HPIS and HHIS are connected downstream the main coolant pump in the loop 1, 3 and 4. The accumulators and the accumulators lines are schematized by means of suitable 0-dimensional object foreseen in Cathare2 code .

4 SBLOCA COUNTERPART TEST TRANSIENT SCENARIO

The experiment is the CT of the experiment carried out in LOBI, SPES, BETHSY and LSTF facilities. Starting from these past experiments the SBLOCA test has been designed in collaboration with EREC, following as well as possible the scaling law for deriving boundary and initial conditions in order to carry out a correct counterpart test [18].

The relative scaling factors are based on the following characteristics of the facilities [19]:

• nominal core power; • the facility's volumetric (power) scaling factor; • primary mass inventory (total primary mass and primary mass without pressurizer

and surge line mass); • primary system volume (total primary volume and primary volume without

pressurizer and surge line); • break location. Due to the availability of detailed LOBI experimental data and configuration, EREC

chose these conditions as guidelines. Adopting the factor defined as the ratio between the primary side volume of the facilities (without the pressurizer), all the conditions were derived:

039.9


pressurizer level, accumulators level, LPIS flow rate and so on. The break is “up-ward oriented” in cold leg of loop #4 and the size has been modified (equivalent to 4.1% of the area of the main pipe in VVER1000) compared with the LOBI test because the different diameter of the cold leg in PWR and VVER1000 prototype reactors. It should be also noted that owing to limitations in the maximum power available in the PSB facility, the initial conditions have been established at a core power around 10% of the reference reactor nominal power (increase of the heat losses estimated in steady state condition). Moreover the choice was to preserve the initial value of the fluid temperature upstream and downstream the core (mass flow-rate reduced of the same amount as power). In order to achieve this, it was necessary to modify the initial temperature and pressure of the SG with respect to the nominal conditions.

The experiment can be subdivided into four main periods from the phenomenological point of view:

1) sub-cooled blow-down and first core dryout – rewet (0 – 113 s); 2) saturated blow-down and primary to secondary side pressure decoupling (113 s –

accumulators emptying); 3) mass depletion in primary loop (accumulators emptying – final dryout); 4) intervention of LPIS that quench the core. Relevant thermal-hydraulic phenomena during the experiment are: mass distribution in

the primary system, heat transfer with secondary side with degraded primary conditions including reversal heat flux, loop seal behavior, core heat up and rewet, accumulator performance and stratification in horizontal pipes.

The test starts by opening the break valve, which has an opening time of 0.4 s. With the opening of the break the pressurizer heaters are switched off. When the primary pressure achieves 13 MPa the following actions are imposed:

• closure of the turbine shut valve at the end of the steam header. The steam generators remain aligned to the steam header;

• closure of steam generator feed water; • trip of all MCPs; completely stopped in 4 s; • scram, the core power starts to follow the given power/time curve. When the primary pressure is 4 MPa the accumulators start to inject water in the

downcomer. Accumulators remain connected until the water level is 1.31 m (from the bottom) in order to avoid nitrogen penetration into the primary system. The low pressure ECCS (LPIS) are activated on a heater rod temperature of 500 °C. The water is supplied in cold legs of loops # 1, 3 and 4 with the mass flow rate of 0.248 kg/s in each line. The SG BRU-A valves are opened at a secondary pressure of 7.4 MPa and closed at 7.2 MPa; these set points are derived from the secondary pressure behavior in LOBI facility. During the whole transient, the SG remain aligned to a common steam header therefore the secondary pressures are expected to be equal. To avoid different behavior in the SG (due to a insignificant uncertainty in measuring the pressures), all the BRU-A valves are assumed to be dependent on SG-1 pressure. The experiment is terminated when a steady core cooling conditions after the final core rewet has been achieved.

5 KV-SCALED CALCUALTION AND RESULTS

A requirement of the UMAE methodology, as already mentioned, is constituted by the ‘demonstration of similarity’ block k in figure 1. The available qualified nodalization of the VVER1000 NPP must be capable of simulating the selected transient. This requirement is relevant in order to applied the nodalization to the selected transient in nominal condition and to use the extrapolation of the accuracy (this phase is not the scope of this paper).

A Kv-scaled calculation is carried out with the following steps:

039.10


1. calculation of the Kv-ratio considering the primary system volumes of the concerned facilities or plants;

2. achievement of steady state in the NPP nodalization, ‘scaling’ the relevant quantities on the basis of the related values;

3. performing the Kv-scaled calculation with the scaled boundary conditions, comparison between calculated and reference (experimental in this case) data and conclusion about the acceptability of the results.

The results of step one shows that the generic VVER1000 primary system is 300 times bigger than PSB primary system volume. This value is used for scaling, to preserve the break area and the ratio of core power over primary system volume. Given the consideration above mentioned initial and boundary conditions are strictly derived from the corresponding parameters of the considered test in the PSB facility. It must be stressed that this step has been fulfilled easily because the PSB is a VVER simulator scaled 1 to 300.

The steady state conditions (step two), envisaged by the target values of the ‘Kv-scaled’ calculation, have been reached running a transient calculation for 100 s. During this time the various time trends are stabilized before starting the CL-4.1-03 test in VVER1000. The relevant results regarding initial and boundary conditions of the experiment and of NPP calculation are given in table 4, where the values reported in the fifth column are related to the end of steady state.

The main results of step three are given in table 5 and 6. A systematic procedure consisting in the identification of phenomena and of Relevant Thermal-hydraulic Aspect (RTA) has been applied. In details:

subdivision of the considered transient into “phenomenological windows” (i.e. time spans in which a unique relevant physical process mostly occurs and a limited set of parameters controls the scenario): phenomena consequent to the physical process characterize each phenomenological window;

for each Ph. W. the identification of the RTA (events or phenomena consequent to the physical process and characteristic of each transient) and the selection of the parameters characterizing the RTA have been taken into account;

qualitative analysis of results obtained by evaluating and ranking the comparison between measured and calculated values.

The qualitative analysis, is based on five levels of judgment (E, R, M, U and -) whose meaning is essentially derived from a visual observation of the experimental and predicted trends:

E the code predicts qualitatively and quantitatively the parameter (Excellent – the calculation is within experimental data uncertainty band);

R the code predicts qualitatively, but not quantitatively the parameter (Reasonable – the calculation shows only correct behavior and trends);

M the code does not predict the parameter, but the reason is understood and predictable (Minimal – the calculation lie within experimental data uncertainty band and sometimes does not have correct trends);

U the code does not predict the parameter and the reason is not understood (Unqualified - calculation does not show correct trend and behavior, reasons are unknown and unpredictable).

In the tables 5 and 6 the comparison between the PSB and the VVER1000 calculated system and experimental results are summarized. Going in to details, the time of occurrence of the first dry-out is well foreseen in both calculations but the peack cladding temperature is higher in the PSB calculation while in the “Kv calculation” is in good agreement with the experimental value and in both cases are quenched by the loop seal clearing.

039.11


Phase 2 is characterized by the second dry-out occurrence in the facility, the onset of SIT injection and the SIT emptying. The timing of the second dry-out is in good agreement (about 70 s before in the ‘Kv-scaled calculation) and the SIT activation 90 second (set point 4.0 MPa, see section 4), this is due to the primary pressure that in the plant calculation is lower than in the experiment because of the heat of the structures realised to the fluid in the facility. Moreover the peak cladding temperature, during dry-out, is well foreseen in the ‘Kv scaled’ calculation, but at various level the range of dryout occurrence is shorter because of the earlier intervention time of the SITs. In the third phase, regarding the mass depletion in primary loop, the primary mass inventory is smaller in the ‘Kv-scaled’ calculation as confirmed by the earlier occurrence of the final dry-out. This is due to the amount of mass delivered by the hydro-accumulators smaller in the calculation than in the test. In conclusion all the RTA observed during the experiment are reproduced in the Kv scaled calculation, the causes originating the discrepancies are also understood (no U mark) and the demonstration of similarity requested by the UMAE process is fulfilled.

039.12


Table 4: PSB CL-4.1-03 test, Kv-scaled calculation for similarity demonstration: relevant boundary and initial conditions (part 1)

N. Parameter/System Unit PSB CL-4.1-03 test

VVER1000Dem. of

Similarity

Note

1. Primary system volume m3 1.23 370 (1) (1) VVER1000/320 (generic)

2. Break: - area - location - Ar/V

m2

- m-1

7.85E-5

CL with PRZ6.38E-5

2.355E-2

CL with PRZ 6.36E-5

3. Primary system: - HL temperature - CL temperature - mass flow rate - GLoop/Core power - isolation valve closure

°C °C

kg/s kg/s/MW

-

308-311 282-283

7.9 6.99

not applicable

309 283 2416 7.12

not applicable

4. PRZ - pressure - level

MPa

m

15.6 4.94

15.7 4.95

5. Core - initial power - decay power - ∆T - average linear power - core power/vol.

MW

- °C

KW/m MW/m3

1.130

- 27

1.905 0.919

339

- 26

1.878 0.916

6. SG SS: - pressure - SL flow rate - DC level - FW temperature - GFW/Core power - MSIV closure - FW stop - SRV set point

MPa Kg/s

m °C

kg/s/MW- -

MPa

6.93-6.88

0.11 (1) 2.48-2.52

170 -(2) -

13MPa(3) 7.4

6.88 34 (1) 2.40 170 -(2) -

13MPa(3) 7.4

(1) each SL (2) FW in this test has been regulated in order to keep the level constant (3) Primary pressure

7. Accumulators: - no. - position - total volume - pressure - liquid mass - liquid mass/PS vol. - total vol./PS vol. - temperature - isolation

- -

m3 MPa kg

kg/m3 -

°C -

2

DC 0.2

4.08-4.14 157

127.6 0.162 26-32

Lev.=1.3m

2

DC 60 4.1

47000 127.0 0.162

25 Tot. M=36tons

8. LPIS: - no. - position - fluid temperature - delivered flow rate - flow rate/PS volume - actuation set point

- -

°C kg/s

kg/s/m3

-

3

CL 1, 3, 4 -

0.248 0.20

CL T = 500°C

3

CL 1, 3, 4 50 74. -

CL T = 500°C

9. HPIS and AFW - not active not active 10. Reactor coolant pump:

- trip - coastdown

- -

13MPa(1)

4 s

13MPa(1)

4 s

(1) Primary pressure

11. SCRAM - 13MPa(1) 13MPa(1) (1) Primary pressure 12. CMT, PRHR and RHR - - - 13. TEST END s 2591 2850

039.13


Table 5: Judgment of code calculation on the basis of RTA (part 1) UNIT EXP PSB CALC VVER

CALC Judg PSB

Judg VVER

RTA: Pressurizer empting TSE emptying time s 10 12 15 E R

scram time s 57.6 57.5 56.6 E E IPA integrated flow from SL (from 0

up to emptying) kg - - - - -

RTA: Steam generators secondary side behaviour TSE main steam line valve closure s 17.5 23 13.0 R R

difference between PS and SS pressure at 100 s

MPa 0.33 0.44 0.45 R R

SVP SG level: - at end of subcooled blowdown - when PS pres. equals SS pres. - when ACC starts - when LPIS starts

m 2.44;2.44; 2.49;2.38 2.43;2.44; 2.49;2.37 2.40;2.41; 2.47;2.33 2.35;2.26; 2.34;2.30

2.41;2.40; 2.40;2.34 2.41;2.40; 2.40;2.34 2.37;2.36; 2.35;2.30 2.18;2.16; 2.28;2.21

2.38;2.38; 2.39;2.32 2.35;2.35; 2.33;2.28 2.33;2.32; 2.32;2.25 2.30;2.30; 2.30;2.25

E

E

E

R

E

E

E

E

SVP SG pressure - at end of subcooled blowdown - when PS pres. equals SS pres. - when ACC starts - when LPIS starts

MPa 7.31;7.34; 7.36;7.31 7.30;7.34; 7.37;7.30 6.79;6.81; 6.83;6.79 5.36;5.39; 5.41;5.37

7.26;7.26; 7.26;7.26 7.15;7.15; 7.15;7.15 6.30;6.30; 6.30;6.30 4.35;4.35; 4.35;5.35

7.28;7.28; 7.28;7.28 7.34;7.34; 7.33;7.3 7.2;7.2; 7.2;7.2 7.0;7.0; 7.0;7.0

E

E

E

R

E

E

R

R

RTA: Subcooled blowdown TSE upper plenum in sat. conditions s 16 17 24 E R

break two phase flow s 113 120 150 E R IPA break flow up to 30 s kg 183.7 157 55110 R E

RTA: First dryout occurrence TSE time of dry out s 97 97 108 E E

range of dry out occurrence at various core levels

s 97 - 102 97 – 124 108 - 112 R E

peak cladding temperature K 589 720 582 R E SVP average linear power kW/m 1.416 1.35 1.3 R E

maximum linear power kW/m 1.416 1.24 1.3 R E core power / primary mass kW/kg 2.01 1.85 2.1 R E

IPA integral of dry out at 2/3 of core height

K s - - - - -

NDP primary mass / initial mass % 47.6 47.5 42 E R time of loop seal clearing s 109 – 105

loop 1&4 113 - 109 loop 3&4

108 - 123 loop 3&4

E/E

E/R

RTA: Rewet by loop seal clearing TSE range of rewet occurrence s 102 - 107 120 - 163 110 - 123 R E

time when rewet is completed s 109 163 116 R E TSE PS pressure equal to SS pressure s 150 178 173 R R SVP break flow at 200 s

break flow at 1000 s kg/s -

- 3.45 0.12

- -

- -

- -

IPA integrated flow (200 - 1000 s) kg 236.52 219 62924 E R

039.14


Table 6: Judgment of code calculation on the basis of RTA (part 2) UNIT EXP PSB CALC VVER

CALC Judg PSB

Judg VVER

RTA: Mass distribution in primary side TSE time of minimum mass

occurrence s 430

2430 411 2780

390 2650

E R

R R

SVP minimum primary side mass kg 171.3 140.6

143 101

33500 17500

R R

M M

av. linear power at min. mass

kW/m 0.304 0.304 0.43 E R

RTA: Second dryout occurrence TSE time of dry out s 405 - 330 - R


s 401-478 - 330-360 - R

peak cladding temperature K 590 - 560 - R SVP average linear power kW/m 0.425 - 0.44 - E

core power / primary mass kW/kg 1.41 - 2.0 - M IPA integral of dry out at 2/3 of

core height K - - - - -

NDP primary mass / initial mass % 21.6 - 18.0 - R RTA: Accumulators behavior

TSE accumulators injection starts s 406 – 414 412 360 E R accumulators injection stops s 1365 – 1452 1418 1000 E R IPA total mass delivered by

accumulators kg - - - - -

NDP minimum mass/initial mass % 20.7 16.7 23.0 M R primary mass (acc. start)/initial mass

% 21.3 25.8 15.2 R R

RTA: Final dryout occurrence TSE time of dry out s 2077 2380 1600 R R


s 2077-2313 2380 - 2800 1610-1920 M R

peak cladding temperature K 783 828 780 R E SVP average linear power kW/m 0.304 0.304 0.3 E E

rate of rod temperature increase

K/s 0.8 0.7 2.1 R M

core power / primary mass kW/kg 1.06 1.3 1.8 R M IPA integral of dry out at 2/3 of

core height K s - - - - -

NDP primary mass / initial mass % 20.6 15.7 13.0 R M RTA: LPIS intervention

TSE LPIS start s 2432 2785 2699 R R range of rewet occurrence s 2482-2518 2910-3010 2700-2780 R R final rewetting s 2559 3056 2720 M R IPA integrated flow from start to

end of rewet kg 96.7 77 33000 R R

NDP primary mass (LPIS start)/initial mass

% 16.8 12 7.0 R R

039.15


6 CONCLUSION

This paper deals with the problem of scaling complex scenario measured in experimental facilities and replicated by the use of Cathare2 v1.5b computer code. More in details it is focused on the ‘demonstration of similarity’ study following the requirements of the UMAE methodology for the prediction of the uncertainty bands.

The reference data base for the activity is the SBLOCA transient, counterpart of the LOBI facility experiment, carried out in the framework of the OECD PSB-VVER Project.

The steps envisaged for the ‘Kv-scaled’ calculation have been execute (as in section 5) and the acceptability of the results has been evaluated by the application of the qualitative accuracy evaluation. This involves the identification of Ph. W., of RTA onside the Ph. W and the parameter ranges that characterize the RTA themselves. The main obtained results are summarized below.

• The code is adequate for predicting the phenomena and all the RTA of interest in the transient as demonstrated by the calculation carried out by the PSB nodalization;

• The Kv scaled calculation with VVER1000 nodalization has been performed. • The differences between experiment and Kv calculation are not relevant and

understood. In conclusion the ‘demonstration of similarity’ has been achieved, an Analytical

Simulation Model is available to perform the ‘nominal conditions calculation’ and to the application of the uncertainty band derived from the accuracy database, as foreseen in the UMAE methodology.

REFERENCES

[1] D’Auria F., et al. Scaling of Natural Circulation in PWR System. North-Holland Physics Publishing. July, 1990.

[2] D’Auria F., N. Debrecin, G.M. Galassi. Outline of the Uncertainty Methodology based on Accuracy Extrapostion (UMAE). J. Nuclear Technology, Vol. 106. January, 1995.

[3] D’Auria F., et al. UMAE Application: Contribution to the OECD/CSNI UMS Vol II. J. DCMN NT 307(97) Rev.1. June, 1998.

[4] Ambrosini W., R. Bovalini, F. D’Auria. Evaluation of Accuracy of Thermal-Hydraulic Code Calcualtions. J. Energia Nucleare, Vol.7 N.2. May 1990.

[5] D’Auria F., M. Leonardi, R. Pochard. Methodology for the Evaluation of Thermal-Hydraulic Codes Accuracy. International Conference on New Trends in Nuclear System Thermohydraulics. Pisa, Italy, May 30 – June 2, 1994.

[6] Melikhov O. I., et al., Pre-test Experiment Proposal Report (Test 3). OECD PSB-facility Project, EREC. 2003.

[7] D’Auria F., et al. Addressing the Scaling Issue by Thermalhydraulic System Codes: Recent Results. Fifth International Conference of Yugoslav Nuclear Society. 2004.

039.16


[8] Araneo D., “Nodalization of the VVER 1000 for the CATHARE2 V1.5B/MOD5.1 code: STEADY STATE QUALIFICATION” WD.A.1.6.2-TAR/08, Pisa, November, 2004.

[9] Muellner N., D. Araneo. Calculation of the 12 tests in the reference VVER-1000 NPP (Balakovo3) using Relap5 mod 3.2.2 gamma and Cathare 2 mod 1.5. TACIS Project 30303, DIMNP, University of Pisa; WD.A.1.6.5-TAR/2, Rev. 0. 2005.

[10] Del Nevo A., D. Araneo, 2004, “Nodalization Qualification Process of the PSB-VVER Facility for Cathare2 Thermal-Hydraulic Code” W.D.A.1.6.1-TAR/01, Pisa

[11] Elkin I.V., et. Al., “Report about PSB-VVER heat losses ”, PSB-VVER report PSB 04, OECD PSB-VVER project, 01 April 2003

[12] Elkin I.V, et al., 2004, “Single-phase performance characteristics of the PSB-VVER pump”, OECD PSB-VVER project, Electrogorsk

[13] Del Nevo A., 2005, “Cathare2V1.5b and RELAP5/MOD3.3 Post Test Analyses and Accuracy Quantification of PSB Test CL-0.5-03”, W.D.A.2.2.1-TAR/01, Pisa, March 2005

[14] D’Auria F., G. M. Galassi, P. Gatta, L. Mastrantonio. Development of a Nodalization for WWER 1000 NPP for Relap5mod3.1 and Relap5mod3.2. Pisa University, DIMNP -NT 263 (95) Rev1. Pisa, Italy, Sept. 1996

[15] Aprile G. A. Analisi di Sicurezza nei Reattori VVER 1000. Master Degree Thesis in Nuclear Engineering. Tutor: F D’Auria., G. M. Galassi. Pisa, 1996

[16] Aprile G. A., F. D’auria, M. Frogheri, G. M. Galassi. Application of a Qualified WWER-1000 Plant Nodalization for Relap5/m3.2 Computer Code. International conference: Nuclear option in countries with small and medium electricity grid. 7-9 October 1996 – Opatija, (HR).

[17] D’Auria F., G. M. Galassi, W. Giannotti, D. Araneo. Nodalization of the VVER 1000 for the Relap5 Code. DIMNP, Università di Pisa; NT 482(02), Rev. 0. October 2002

[18] Melikhov O. I., et al., Pre-test Experiment Proposal Report (Test 3). OECD PSB-facility Project, EREC. 2003.

[19] Cherubini M. Scaling Analysis for Russian WWER Type Reactor. Master Degree Thesis in Nuclear Engineering. Tutor: F D’Auria., O. Melikhov, G. M. Galassi. Pisa, 2004

SCALING OF SMALL BREAK LOCA IN VVER 1000 SYSTEM · International Conference Nuclear Energy for New Europe 2005 Bled, Slovenia, September 5-8, 2005 039.1 SCALING OF SMALL BREAK LOCA

Documents