Coupled Thermal-hydraulic and Neutronic Model for the Asc

1
Coupled Thermal-hydraulic and Neutronic Model for
the Ascó NPP using RELAP5-3D/NESTLE

• L. Batet, R. Pericas, E. Morales and F. Reventós
• Technical University of Catalonia (UPC)
• Department of Physics and Nuclear Engineering

2
Contents

• Introduction
• Ascó NPP description
• Ascó model description
• Thermal-hydraulic
• Neutronics
• Model testing
• Steady State
• Load Rejection transient
• Main Steam line Break transient
• Sensitivities
• Conclusions

3
Introduction

• The Dept. of Physics and Nuclear Engineering of
  the (UPC) holds a large background in the use of
  TH codes for the Safety Analysis of Nuclear Power
  Plants (NPP). The Thermal Hydraulic Studies Group
  has been cooperating for 15 years with the
  operators of the Catalan nuclear plants, Ascó (2
  units) and Vandellós II (all of them 3 loop PWR
  Westinghouse design).
• Ascó-1 NPP started commercial operation on
  December 1984.
• RELAP5 model developed by its analysts.
• Extensively qualified and validated.
• The presentation represents a continuation of the
  presentation in 2006 seminar L. Batet et al.
  Status of the activities related to the use of
  RELAP5-3D at the Technical University of
  Catalonia

4
Introduction

• A 3D vessel component and neutronic data have
  been added to the existing model allow full 3D
  NKTH coupling.
• Neutronic data taken from previous model RELAP5
  (NRC)/PARCS. We have faced problems in adapting
  the data.
• What we are presenting is a set of preliminary
  calculations.
• The work has been basically done by undergraduate
  students (following a previous work presented in
  the 2006 seminar).
• This is a first step done with the aim of
  achieving full neutronic-TH simulation
  capabilities in the UPC.

5
Ascó NPP description

• Ascó-I is owned by ENDESA (100).
• Ascó-II is owned by ENDESA (85) and IBERDORLA
  (15).
• Units are located close to Tarragona, in the
  north east of Spain, and they use the Ebro River
  as a final heat sink.

6
Ascó NPP description

• The actual nominal power of each unit is 2952.3
  MWt and 1028 MWe.
• The reactor vessel is cold head type.
• 3 Siemens (type SG 61 W/D3) steam generators. FW
  fed directly to the upper part of the downcomer
  via J-tubes. The circulation ratio on the
  secondary side of the steam generators is 3.65 at
  rated power.
• The auxiliary feed water system is impulsed by
  one turbo pump and two motor pumps. In the plant
  there are, among others, control systems for the
  reactivity (rods and boron), primary pressure,
  pressurizer level, steam dump and steam generator
  level. The reactor protection system includes
  safety valves in the pressurizer and the steam
  generator.

7
Ascó NPP description

• Summary of the plant main features

8
Plant thermal-hydraulic model

• The RELAP5 model of Ascó NPP is prepared to
  simulate both units of the plant.
• Only slight changes are needed, concerning mainly
  to the fuel load, to switch from one to another.
• The model includes
• Hydrodynamic elements (primary, secondary, safety
  systems and auxiliary systems)
• Heat structures, and
• Control and protection systems.
• The model has been prepared for RELAP5/MOD3.2 and
  has been subjected to a thoroughly validation and
  qualification process, which includes the
  simulation of transients occurred in the plant
  itself.

9
Plant thermal-hydraulic model

• Summary of the model degree of detail

10
Plant thermal-hydraulic model
11
Plant thermal-hydraulic model
AFW
FW modeling
MFW
12
Plant thermal-hydraulic model
Logic diagram (partial) of the MFW control system
13
Vessel thermal-hydraulic model

• In order to perform 3D NK calculations, in the
  Ascó model, the vessel has been converted in a
  set of RELAP5-3D multid components.
• The 3D vessel model had been previously developed
  for Vandellòs-II by an undergraduate student (X.
  Sabaté).
• A previous R5/Parcs model existed (used in the
  MSLB benchmark)

Pseudo 3D Model RELAP5/PARCS
1D Vessel Model Relap5/3.2
3D Vessel Model Relap5-3D
14
Vessel thermal-hydraulic model

• One of the nodalization proposals
• RPV consists of 5 multid (cartesian
  cylindrical)
• 500 hydraulic nodi
• Some of them are disabled (by applying small
  volume factors) in the corners

21 core channels
15
Vessel thermal-hydraulic model

• Model adopted for the vessel

• Vessel inlets and downcomers.
• 3 azimuthal sectors .
• 1 radial sector.
• 5 axial levels.
• Primary circuit to vessel inlets (120º).
• Bypass.
• 8 azimuthal sectors .
• 1 radial sector.
• 1 axial level.
• Core.
• 21 thermal-hydraulic nodes
• 9 axial levels including Core entrance, Core
  exit, Core auxiliary exit and Core.
• 5x5 Cartesian matrix.

• Lower plenum.
• 8 azimuthal sectors.
• 2 radial sectors.
• 1 axial level.
• Upper head.
• 8 azimuthal sectors .
• 2 radial sectors.
• 2 axial levels.

16
Vessel thermal-hydraulic model
Core bypass
Core
Downcomer
Core
Upper plenum and Upper head
Lower plenum
17
Vessel neutronic model

• Neutronic data taken from previous model
  R5/PARCS.
• At that time (2001) Cross Section libraries were
  taken from SEANAP (Univ. Polit. Madrid)
• Additional work has been done to transfer from
  PARCS format values to NESTLE format values.
• We have faced problems in adapting the data

18
Vessel neutronic model

• 17x17 Matrix 157 elements/level.
• 13 axial levels.
• 2 reflector levels.
• 11 power levels.
• 190 different compositions (XS).
• Mapping from 13 levels (NK) ? 8 levels (TH).

Dont ask me why
19
Vessel neutronic model
157 neutronic nodes
20
Control Rod core distribution
Control Rod group A
Control Rod group B
Control Rod group C
Control Rod group D
21
Model Testing

• Steady state
• Load rejection transient.
• Load rejection (50) corresponding to the testing
  after fuel reload (Ascó I, 1999, cycle 13)
• MSLB
• UPC participated in the OECD-CSNI PWR MSLB
  benchmark with RELAP5/PARCS. As a culmination, a
  MSLB was calculated for the Ascó NPP (A. Cuadra,
  JL Gago, F. Reventós, Analysis of a
  Main-Steam-Line Break in Ascó NPP, ANS Technical
  Paper - Thermal Hydraulics, Volume 146, Number
  1, April 2004, Pages 41-48)

22
Steady State

• Steady state calculation at Beginning Of Life
  conditions, for the Ascó-1 NPP cycle 13.

23
Steady State
Something to be improved in the kinetics data or
in mapping
24
Steady State
25
Load rejection

• Transient from 100 to 50 steady state power in
  12 seconds.
• Control Rod group D insertion.
• Intervention of kinetics, TH, control systems
  the actuation of one system affecting the others
• We usually perform this test whenever we make an
  improvement to the model.
• We performed this test with the 1D TH (point
  kinetics) model using R5-3D in 2006

26
Load rejection
27
Load rejection
Control Rod group D
28
Load Rejection
Power axial profiles
29
Load rejection
Axial top level temperature evolution
Power evolution
30
Main Steam Line Break

• Objective
• Try to get the good results obtained after the
  MSLB benchmark using R5/PARCS
• Loop 2 pipe break at 10,060 seconds (loop 2)
  inside containment (double-guillotine between
  nodes 750-752).
• Stop AFW Turbopumps, 3 min after the break.
• Flow control to 15 of AFW motorpumps, 4 min
  after the break.
• Base case 1 sensitivity no safety injection
  (no credit for boron injection)

31
MSLB Base Case
32
MSLB Base Case
Isnt the primary a closed system?
33
MSLB Base Case
Axial top level temperature evolution
Power evolution
34
MSLB Sens. No safety injection
35
MSLB Sens. No safety injection
36
Ongoing work and conclusions

• The well validated Ascó model for R5 has been
  converted into a 3D model by substituting the
  vessel volumes by multid components.
• 3D NK using Nestle
• XS library adapted from previous Parcs model
  (facing problems)
• On going try to refine the conversion process
  and use the gt600 compositions in the original
  model
• At present we depend on borrowed XS
• On going initiating a research line in
  neutronics (try to use SCALE to produce 2 group
  XS)

37
Ongoing work and conclusions

• The goal is to have fully 3D NKTH computational
  capabilities at the THSG.
• Plant data available to (partially) validate the
  future model (e.g. temperature and power maps for
  the 50 load rejection transient)
• Comments and suggestions are welcome.