Progress in reliability methodology for passive systems

1

• Progress in reliability methodology for passive
  systems

C. Bassi, N Devictor, M. Marquès CEA France
A.K. Nayak, D. Saha BARC India
2
Advances

• Applications of RMPS on Gen IV Reactors at CEA
• APSRA Methodology and application at BARC
• Passive system reliability analyses at MIT
• Reliability Analysis of CAREM like PRHRS
  (presented by Mr Gimenez)

3
Applications on Gen IV Reactors at CEA

• Gas-cooled Fast Reactor (GFR) application to
  the Decay Heat Removal system (DHR)
• Very High Temperature Reactor (VHTR)
  application to decay heat removal by radiation
  and conductivity

4
Decay Heat Removal system (DHR) of GFR

• The DHR system consists on
• a metallic guard containment enclosing the
  primary system, not pressurized in normal
  operation and maintaining a backup pressure of
  1.0 MPa, in case of primary circuit
  depressurisation
• three dedicated DHR loops in helium (3100
  redundancy) with secondary loops with pressurised
  water connected to an external water pool
  (ultimate heat sink).
• Each DHR loops can remove the decay heat
• few minutes after the reactor SCRAM by natural
  circulation in case of blackout transient (in
  this case the primary pressure is nearly
  constant, close to the nominal pressure 7 MPa)
• few minutes after the SCRAM by forced circulation
  in case of large break LOCA (in this case the
  pressure is equal to the chosen back-up pressure
  1 MPa) DHR forced circulation is obtained by
  one blower driven by an electrical motor which
  can be supplied by batteries during one day,
• one day after the SCRAM by natural circulation
  and with a pressure of 1.0 MPa.

5
RMPS application to the DHR system of GFR

• Objective Quantify the reliability of the DHR
  in pressurised situation
• Scenario
• Initiating event short duration Loss of Offsite
  Power (10-2/reactor.year)
• complete loss of Forced Circulation
  (emergency diesel generators fail to start or
  circulator failure)
• Failure criteria
• Tfuel_max gt 1600C or Tupper_max gt 1250C between
  0 and 24 hours
• Critical parameter identification
• by deductive approach

6
RMPS application to the DHR system of GFR
DHR nodalization in the CATHARE 2 code
Probabilistic distribution of the critical
parameters

• 100 simulations
• Latin Hypercube Sampling of these 9 random
  variables
• Evaluation by CATHARE code of core max
  temperature

7
RMPS application to the DHR system of GFR
100 CATHARE evaluations, no failure case
Sensitivity analysis

• FRPLAQ (multiplicative factor for laminar NC
  pressure drop in the core region)
• REPLAQ (Reynolds number for turbulent to laminar
  transition in the core region)
• T2DHR (ternary pool and secondary circuit
  temperature level at DHR initiation)

Linear correlation between Tmax and 6 parameters
(R2 0,97)
TMAX 1016.5 (T2DHR0.30211) - (FUITE449.71)
(FRPLAQ63.027) - (ECPLAQ18.222)
- (REPLAQ2.415710-3) (ECDHX15.5575)
Failure probability lt 5.10-6
106 simulations
8
RMPS application to the DHR system of GFR
Sensitivity study upon the DHR isolating valve
opening fraction
(1) Valve modelling in CATHARE Cv ? DP f(Pu)
(3) Criteria f(DP loop)
(2) Prametric study for ? Pu i.e. DP in NC loop
Propagation of uncertainty in CATHARE
calculations and reliability evanuation of the
DHR
Critical parameters and sampling identical to the
reference case
100 CATHARE evaluations
106 simulations
2272 cases gt 1600C (max. 1609C !)
Failure probability ? 2,3.10-3
9
The European RAPHAEL Project RMPS exercise
to the Decay Heat Removal of an HTR
SP Safety - WP4 Safety approach and licensing
issues Nicolas DEVICTOR (CEA)

10
RAPHAEL - SP Safety - WP4Exercise - Decay Heat
Removal

• Aim of the study
• to illustrate the application of method RMPS
  (Reliability Method for Passive System)
• to illustrate the different types of results and
  their possible use in help of design and safety
  analysis
• Support for the exercise
• Simplified modelling of the GT-MHR
• See CRP 3 (on Heat Transport and Afterheat
  Removal for Gas Cooled Reactors under accident
  conditions) and report IAEA RECDOC-1163
• Content of the exercise
• Transient Loss of Fuel Coolant with
  depressurization
• Taking into account some uncertainty sources (see
  slide 3), computation of the probability than
  Tmax exceed 1600C and assessment of the most
  influential uncertainty sources.
• Note Tmax maximal value of the averaged
  temperatures in the compacts

11
RAPHAEL - SP Safety - WP4Exercise - Decay Heat
Removal
Simplified model

• Modelling developed under Cast3M environment
  (ARCTURUS)
• Components core, reflectors, core barrel,
  baffle, reactor vessel, RCCS, concrete
• Physical phenomena
• conduction,
• cavity radiation,
• exchange RCCS
• Boundary conditions
• Fixed heat exchange at the external concrete wall
• Fixed RCCS water temperature
• Model provided by CEA/DM2S/SFME

12
RAPHAEL - SP Safety - WP4Exercise - Decay Heat
Removal
Uncertainty sources taken into account (without
correlation)
Result for nominal values Maximal temperature
1340.5C at 93 hours
13
RAPHAEL - SP Safety - WP4Exercise - Decay Heat
Removal

• Conditional exceedance probability 0,0214
• Conditional to the scenario
• Conditional to the uncertainty model
• Reliability index (?) 2,0258 
• Co-ordinates of the most probable failure point
  

• FORM method used
• 97 calls
• 1 call ?10mn

Comment the result should be confirmed by
Importance Sampling
14
RAPHAEL - SP Safety - WP4Exercise - Decay Heat
Removal

• Importance factors

15
RAPHAEL - SP Safety - WP4Exercise - Decay Heat
Removal

• Sensitivity and elasticities on the ?

• Information for steering RD
• Average value of the conductivities
• in the reflector and the core
• Variability of conductivity
• in the reflector
• Bounds of the thermal inertia
• in the graphite

16
APSRA methodology and application

• Assesment of Passive System ReliAbility (ASPRA)
  developed by BARC (India)
• A.K Nayak, M.R. Gartia, A. Antony, G. Vinod and
  R.K.Sinha
• Reliability analysis of a boiling two-phase
  natural circulation system using the ASPRA
  methodology.
• Proceedings of ICAPP 2007, Nice, France

17
APSRA methodology and application
APSRA methodology
18
APSRA methodology and application

• Application to the MHT system of the AHWR
• Evaluation of the reliability of the boiling
  two-phase system

The Main Heat Transport (MHT) system consists of
a common reactor inlet header from which 452
inlet feeders branch out to an equal number of
fuel channels in the core. The two-phase mixture
leaving the core is separated into steam and
water in the steam drum, by gravity. The feed
water issues in the form of jets through J tubes
allowing proper mixing of feed water with the
saturated water
19
APSRA methodology and application

• Identification of natural circulation failure
  (steps II and III)
• For normal operational states, natural
  circulation failure in AHWR occurs if
• Clad surface temperature rises above 400C (673K)
  or/and
• CHF occurs with or without flow induced
  instability
• Key parameters causing the failure (step IV)
• Fission heat generation rate
• Level in steam drum
• Pressure in the system
• Feed water temperature/subcooling
• Calculation with best estimate code RELAP5/mod
  3.2

20
APSRA methodology and application

Parametric studies to determine the failure
points and generation of the failure surface
(step V)
21
APSRA methodology and application

• Identify the causes for the deviation of key
  parameters (step VI)
• The deviation of key parameters caused by
• Failure of active components valves, pumps,
  control systems
• Failure of passive components passive valve

Fault tree /event tree analyses
22
APSRA methodology and application

• Evaluation of failure probability for the system
  (step VII)

the probability of failure for each point of the
failure surface can be calculated by considering
the probability of deviations of all those
parameters which are responsible for causing the
failure
23
Passive system reliability analyses at MIT
The impact of uncertainties on the performance of
passive systems. L.P. Pagani, G.E. Apostolakis
P. Hejzlar Nuclear Technology Vol. 149 Feb. 2005
1

• Reliability analysis of the DHR system (2x50
  loops) of a 600 MW GFR cooled by helium
• Scenario LOCA transient
• Uncertain parameters power, pressure, cooler
  wall temperature, Nusselts number and friction
  factor, in forced, mixed and free convection.
• Failure criteria Tcore gt 1600C in the hot
  channel or Tcoregt 850C in the average channel
• Simulation of the steady-state behavior
  simplified T-H code
• Reliability evaluation 10000 Monte Carlo
  simulations

24
Passive system reliability analyses at MIT

• Comparison passive (DHR in natural circulation)
  VS active (DHR in forced circulation with
  blowers)
• Functional failure (due to uncertainties)
• ? Functional failure important for passive
  system, negligible for active system
• Functional failure hardware (components)
  failures

? Active system failure dominated by the common
cause failure of the blowers. ? Active system
more reliable then passive for 2 and 3 loop
design ? Increase of redundancy more effective
for functional reliability
25
Passive system reliability analyses at MIT
Incorporating reliability analysis into the
design of passive cooling systems with an
application to a gas-cooled reactor F.J. Mackay,
G.E. Apostolakis P. Hejzlar Nuclear Engineering
Design (Article in press)

• Reliability analysis of the DHR system (2x50
  loops) of a 600 MW GFR cooled by helium
• Scenario LOCA
• Failure criteria Tcladgt1600C or TDHR pipe
  wallsgt850C
• Simulation transient analysis with RELAP5-3D
• (the 2 DHR loop are modeled)
• Uncertain parameters core roughness, DHR
  12check valve leakage, Containment structure
  and core heat transfer coefs, shaft inertia
• Reliability evaluation 128 LHS Monte Carlo
  simulations
• Failure probability estimation 0.305

26
Passive system reliability analyses at MIT

• An important finding was the discovery that the
  smaller pressure loss through the DHR heat
  exchanger than through the core would make the
  flow to bypass the core through one DHR loop, if
  two loops operated in parallel.
• This finding is a warning against modelling only
  one lumped DHR loop and assuming that n of them
  will remove n times the decay power.
• Failure probability too high.
• Risk-based design changes insulation of DHR
  pipes inner surface, minimise heat transfer of
  the structure (increase back up pressure), use of
  more reliable valve.
• Proposal of use of a combination of active and
  passive systems.
• This study is an example of the kind of insights
  that can be obtained by including a reliability
  assessment of design process.