43
Safety Design and Analysis Hiroyuki Sato Japan Atomic Energy Agency Training Course on High Temperature Gas-cooled Reactor Technology October 19-23, Serpong, Indonesia

Safety Design and Analysis - International Atomic Energy … ·  · 2015-11-04Safety Design and Analysis Hiroyuki Sato Japan Atomic Energy Agency Training Course on High Temperature

Embed Size (px)

Citation preview

Safety Design and Analysis

Hiroyuki Sato

Japan Atomic Energy Agency

Training Course on

High Temperature Gas-cooled Reactor Technology

October 19-23, Serpong, Indonesia

p.2

Safety Design Objective

protect people and the environment from harmful effects of ionizing radiation

Safety principles

Safety Requirements

Safety Guides

The fundamental safety objective is to

The basis for requirements and measures for the protection of people and the environment against radiation risks and for the safety of facilities and activities that give rise to radiation risks.

Design requirements for the structures, systems and components of a nuclear power plant, as well as for procedures and organizational processes that are required to meet safety principles

Recommendations and guidance on how to comply with the safety requirements, indicating an international consensus that it is necessary to take the measures recommended

p.3

Safety Requirements

Chapter Title Requirement #

Contents

Management of Safety in Design

1-3 Responsibilities, Plant design management, Safety throughout the lifetime of the plant

Principal Technical Requirements

4-12 Fundamental safety function, Defense in depth, Safety assessment

General Plant Design 13-42 Design basis, Safety Classification, Design considerations, Safety analysis

Design Specific Plant Systems 43-82

Reactor core and associated features, Reactor coolant systems, Containment structure and confinement, Instrumentation and control systems, Backup power supply, Supporting systems and auxiliary systems, other power conversion systems, radiation protection

*Safety Requirements for Modular HTGRs proposed by JAEA

p.4

Fundamental Safety Function

Confinement of Radioactivity

Control of Heat Removal

Control of Heat Generation

Control Chemical Attack

• Low core power density

• Un-insulated RPV • Slow heat up due

to massive graphite components

• Heat removal by passive conduction & radiation

• Large negative temperature coefficient intrinsically reduce reactor power

• Two independent and diverse systems for reactor shutdown

• Non-reacting coolant

• Limited source of water

• Graphite and coated layer protects fuel

p.5

Inherent Features on Radioactivity Confinement

1.0

0.8

0.6

0.4

0.2

0

Temperature [oC]

Failu

re f

raction [

-]

1600 1800 2000 2200 2400 2600 2800

• Primary radionuclide retention barrier in the HTGR consists of the three ceramic coating layers surrounding the fissionable fuel kernel to form a fuel particle.

• Fuel particles can withstand extremely high temperature without losing their ability to retain radionuclides under accident conditions.

* INL, INL/EXT-10-18610 (2010).

p.6

Inherent Features on Control of Heat Removal

Air

Reactor cavity cooling system (RCCS

Co

re

Inner reflector

Side reflector

Core barrel

RPV RCCS

Cavity wall

Conduction

Radiation

Natural convection

HTGR can remove core heat even engineered safety system does not work due to large thermal capacity, high thermal conductivity and low power density in core

Conduction

p.7

Inherent Features on Control of Heat Generation

Control rod

Circulator

VCS

Radiation

Natural convection

Water

Primary system

Reject to atmosphere

RPV

Heat removal

Heat removal

HTTR LOFC Test

• In order to demonstrated the characteristics, LOFC test was conducted by using the HTTR.

• All circulator was stopped without scram.

• Reactor reached stable condition by itself, i.e., only by physical phenomena without active forced cooling and control rod insertion.

Stop of circulator

Core flow rate Test result

Reactor power Test result

100

50

0

Flo

w r

ate

(%)

30

15

0 P

ow

er (

%)

Elapsed time (hr) 0 1 2 3 4 5 6

HTGR can intrinsically shut down reactor without relying on the control rod insertion because of large negative temperature coefficient

p.8

Inherent Features on Control of Chemical Attack

Helium Gas Turbine

Reduce risk of water ingress by utilizing helium gas turbine

Enable to eliminate water cooling system by employing dry cooling tower due to high temperature waste heat

SiC Oxidation

SiC (s) + 3/2O2 (g)→SiO2 (s) + CO (g)

SiC (s) + CO (g) →SiO (g) + 2C (s)

105

1

Temperature (K)

SiC

SiC + C

Protective oxide layer region

Gasification region

1800 1400 1000

Protective oxide layer may formed due to oxidation of SiC layer with oxygen in air.

Helium gas turbine

Reactor

To dry cooling tower

He

Heat exchangers

Part

ial p

ress

ure

of

oxy

gen

(Pa

) IAEA, IAEA-TECDOC-784 (1995).

p.9

Defense in Depth

Normal operation

AOO DBA DEC Operational states Accident conditions

Control, limiting and protection

systems

Conservative design, high

quality in construction

and operation

Engineered safety features, safety systems and

procedures

Complementary measures and accident management (AM)

No core melt

Conditions practically eliminated

Level 1 Level 2 Level 3

Plants states (considered in design)

LWR (IAEA SSR-2/1, DS462)

Modular HTGRs

AOO Multiple failures

AM

Conditions practically eliminated

HTGR - Japan’s proposal

Level 4

Severe accidents [including core melt]

Normal operation

Operational states

Level 1 Level 2

Level 1 Level 2 Future

modular HTGRs

Level 3 Level 4

Level 3

Inherent safety features

Engineered safety features and inherent safety features

Plants states (considered in design)

Accident conditions (DBA)

Level 3a Level 3b

[No large FP release] Single failure criterion

p.10

Physical Barrier

“Set of barriers”

Coated fuel particle (CFP)

CFP

Core graphite

Reactor coolant pressure boundary

Confinement

Modular HTGRs Future modular HTGRs

“A barrier”

Level of defense in depth and physical barriers shall be distinguished.

Regardless of the redundancy or diversity of physical barriers, the defense in depth to assure the reactor safety is applied.

p.11

Safety Classification

In compliance to defense in depth, Structures, Systems, & Components (SSC) which are indispensable to achieve TRLC are classified into the following categories with 3 classes;

Prevention System (PS) SSCs whose failure or malfunction have the possibility to cause a excessive amount of radiation exposure to the public outside the site boundary e.g. Reactor coolant pressure boundary, etc.

Mitigation System (MS) SSCs whose function are to prevent an escalation of conditions or put under control immediately, thereby prevent or mitigate possible undue radiation exposure to the public or site personnel e.g. Engineered safety features, etc.

Safety Evaluation Summary

p.12

• Safety evaluation is made for the purpose of demonstrating that safety requirements are met for all postulated initiating events that could occur over a broad range of operational states, including different levels of availability of the safety systems.

• There are two basic types of safety evaluation - Deterministic safety analysis Used to verify that design can met with acceptance criteria. - Probabilistic risk assessment Used to determine the probability of damage for each physical barrier, and evaluate the risk that arises from low frequency sequences

Acceptance Criteria

Event Selection

Safety Analysis

The following slide will show brief procedure of safety evaluation taking the HTTR safety evaluation result as an example

HTTR Outline

p.13

Major specification

Thermal power 30 MW

Fuel Coated fuel particle / Prismatic block type

Core material Graphite

Coolant Helium

Inlet temperature 395C

Outlet temperature 950C

Pressure 4 MPa

Containment vessel

Reactor pressure vessel

Intermediate heat

exchanger (IHX)

Hot- gas duct

HTTR Graphite-moderated and helium-cooled VHTR

Fuel Rods Graphite

Block

First criticality : 1998 Full power operation : 2001 50 days continuous 950oC operation : 2010 Loss of forced cooling test at 9MW : 2010

HTTR Configuration

p.14

TLRC and Plant States

p.15

Dose Limits Criteria

Normal Operation 1 mSv of annual radiation exposure outside the site boundary

Accident No significant risk of radiation exposure to the public* * effective dose equivalent shall not exceed 5 mSv

Anticipated Operational Occurrence (AOO)

A postulated event does not result in core damage & the event can be put under control with the condition which allows the resumption of normal operation

Accident (ACD) A postulated event does not lead to considerable core damage, that the event does not cause, in its process, a secondary damage which would lead to another abnormal condition, & that the function of the barriers against the release of radioactive materials in the event is adequate.

p.16

Acceptance Criteria for Fuel during AOOs

The coated layer of fuel shall maintain integrity during AOOs

• Failure of coated layer observed above 2000oC is due to degradation of mechanical strength by thermal decomposition of SiC & internal pressure increase because of CO & FP production

• It is estimated that impact of burnup increase on temperature is not significant

• Integrity of coated layer under 1600oC is assured

The peak fuel temperature shall be less than 1600 oC

1.0

0.8

0.6

0.4

0.2

0

Temperature [oC]

Failu

re f

raction [

-]

1600 1800 2000 2200 2400 2600 2800

* INL, INL/EXT-10-18610 (2010).

p.17

Acceptance Criteria for Core during ACDs

The core shall maintain subcriticality & coolability during ACDs a. The fuels shall be maintained in the

graphite fuel block or sleeve b. The structural integrity of the

graphite support structures e.g. support posts shall prevent the core from collapsing

Residual effective thickness at bottom plate of graphite sleeve shall be larger than 5 mm

Residual effective diameter of support posts shall be larger than 80 mm

Graphite sleeve

Fuel compact

Bottom plate

Support post

Coaxial piping

High temperature plenum block

p.18

Acceptance Criteria for RCPB

• The material composing RCPB shall have stable strength during normal operation & abnormal conditions

• The pressure load imposed on RCPB shall not exceed the allowable limit • In case of the HTTR, acceptance criteria are determined in accordance with “High-

Temperature Metallic Component Design Guideline”

Allowable Pressure AOO: Allowable pressure for service condition II in MITI Notice No.501 ACD: Allowable pressure for service condition in III in MITI Notice No.501 (High temperature component refers to “High-Temperature Metallic Component Design Guideline”)

Pressure on the RCPB shall not exceed 110% of the maximum allowable working pressure

(a) Pressure on the RCPB (except the IHX heat transfer tube) shall not exceed 120% of the maximum allowable working pressure (b) IHX heat transfer tube shall avoid creep buckling

p.19

Acceptance Criteria for RCPB

Allowable Temperature • 2 ¼ Cr-Mo steel The degradation of tension strength by thermal aging under 550oC for short time is negligible • Austenite steel Allowable temperature for the material in FBR component design guideline is 650oC • Hastelloy XR Characteristic of creep rupture under 1000oC is stable , Accumulated creep damage under 980oC is below the allowable limit

ACD: 550oC, AOO: 500oC (50oC margin to ACD)

ACD: 650oC, AOO: 550oC (50oC margin to ACD)

ACD: 1000oC, AOO: 980oC

p.20

Event Selection

• Event selection is performed based on deterministic approach • Abnormal events to be postulated as AOOs & ACDs are selected based on the

investigation of main causes which affect each item of the acceptance criteria • The initiating events are identified by FMEA & classified into similar event groups • The most severe events with respect to the acceptance criteria within each similar event

group are selected as representative events • Occurrence frequency of representative events are examined in order to confirm the

adequacy of above event selection

Core damage

Increase of fuel temp.

Graphite corrosion

Air ingress

Abnormality of reactivity control system

Loss of core support

Reactivity addition

Evaluation item Event Sequence Abnormal Events

Water ingress

p.21

General Procedure

Design Database Safety Design Core Design

Safety Analysis

Acceptance Criteria

Safety Analysis Report

Output

Input

SSC Design

p.22

Safety Analysis Code

• Evaluation item related to acceptance criteria, e.g. peak fuel temperature, RCPB temperatures, etc. are analyzed by safety evaluation codes

• In case of the HTTR, following codes were used

Codes Objective

BLOOST-J2 Obtain transient response of fuel during abnormal events related to RI

THYDE-HTGR Obtain transient response of plant during abnormal events

TAC-NC Obtain transient response of reactor temperature during DLOFC accidents

RATSAM6 Obtain shear force stress during DLOFC accidents

COMPARE-MOD1 Obtain transient response of pressure & temperature in CV during DLOFC accidents

GRACE/OXIDE-3F Obtain oxidation rate of graphite materials by intruded air or water during related accidents e.g. DLOFC accident or PPWC tube rupture, etc.

FLOWNET/TRUMP Obtain fuel & block temperature during channel blockage accidents

HTCORE Obtain release rate of FP from core after the occurrence of accident

PLAIN Obtain distribution of FP plateout in primary loop during nominal operation

Y. Shiina et al., JAERI-M 90-034 (1990). H. Mikami et al., JAERI-M 88-256 (1988).

K. Sawa et al., JAERI-M 91-198 (1991).

p.23

Calculation Condition (1/2)

• Initial process values are set based on rated condition of high temperature operation mode with errors estimated in the operation

Rated condition Safety Design Evaluation

Reactor power 100% (30MW) 102.5% (30.75MW)

Reactor outlet temp. 950oC 967oC

Reactor inlet temp. 395oC 397oC

Primary loop pressure 41 kg/cm2 42.5 kg/cm2

∆Tout 2= (Stability of Tin) 2 + (Tout – Tin)2 x {(Error of P) 2 + (Error of F) 2 }

• Reactor outlet temperature variation during operation is determined using the following expression

Tout : Reactor outlet temp., Tin: Reactor inlet temp., P: Reactor power, F: Reactor flow rate

p.24

Calculation Condition (2/2)

• The parameters for the safety design evaluation shall be specified such that they give conservative result to a reasonable extent in view of the objective of the analysis

- Thermal conductivity of graphite, emissivity, temperature coefficients , CR insertion curve, reactor kinetic parameters, flux skewing • The SSC belongs to MS-1 & MS-2 specified in safety design are allowed to be taken into

account in the safety design evaluation - MS-1: CR, ACS, VCS, CV, EG, Emergency purification system, etc. - MS-2: Stack, post-ACD instrumentation, etc • A single failure of a component or within a system which is designed to cope with an

accident shall be assumed in addition to postulating an initiating event for assessment - Stop of one of two AGC, Malfunction of one of two system in VCS, Malfunction of one of two system in emergency purification system, etc • The analysis of an accident shall take into account unavailability of off-site power if

functions of the engineered safety features are expected - ACS starts after 60 s elapsed from event initiation

p.25

Representative Events

• Challenges to core heat removal – Pressurized loss-of-forced circulation (PLOFC) accident – Depressurization loss-of-forced circulation (DLOFC) accident, etc.

• Challenges to control heat generation – Accidental control rod withdrawal – Station blackout without trip, etc.

• Challenges to control chemical attack – Air ingress due to helium coolant pressure boundary leak/break, etc. – Water ingress due to boundary leak/break in water cooler

• External hazards – Earthquake, Hurricane, etc.

p.26

DLOFC Accident Initiating Event

Rupture of concentric pipe in primary cooling system

p.27

DLOFC Accident Sequence

• Evaluation items related to acceptance criteria, e.g. peak fuel temperature, RCPB temperatures, etc. during Depressurized Loss-of-Forced Circulation (DLOFC) accidents are analyzed by codes corresponding to their capability

• Dose evaluation for DLOFC is conducted by using temperature response of core & shear force ratio in primary loop obtained in the above analysis

Rupture of primary co-axial piping

Release of primary coolant into CV

Depressurization of primary loop

Reactor scram

Equalization of pressure between

reactor and CV

Core heat removal by VCS

NC in reactor

Oxidation of graphite

component

Release of coolant mass & energy into CV

Pressure & temperature increase

in CV

Oxidation becomes negligible due to core cooldown

CV pressure & temperature

decrease Core cooldown

CV cooldown

RATSAM6

COMPARE-MOD1

GRACE

THYDE-HTGR

TAC-NC

p.28

DLOFC Analysis Model

• 2D (R-Z) finite element temperature analysis code TAC-NC is used to obtain temperature response of fuel & RPV

• Decay heat is considered in the core (CR insertion is assumed)

• NC from core to the flow path between RPV and permanent reflector is considered

• NC in cavity is neglected • VCS panel temperature is set as

boundary condition (90oC) • Oxidation heat is neglected • Conservativeness are taken into

account for initial conditions, thermal properties & emissivity

• Air intrudes into core by NC

RP

V

VC

S

Perm

anen

t re

flec

tor

Rem

ova

ble

ref

lect

or

Core

Removable reflector

Carbon block

R-Z Finite Element Model for Transient Simulation

p.29

Radiation Path in DLOFC Accident

Containment Vessel

Reactor building

RPV Circulating activity in PCS

Primary cooling system (PCS)

Lift off of plated-out FP

FP release from fuel matrix

Radiation cloud

Emergency ventilation

system

Leakage

FP release (Immediate release)

Skyshine gamma-ray exposure

External and internal exposure from radiation cloud

Direct gamma-ray exposure

Radiation exposure

FP release (Delayed release)

p.30

DLOFC Simulation Results (1/2)

• Peak fuel temperature does not exceed initial temperature due to small power density

• Peak RPV temp is close to limit due to NC

• Oxidation of support post is larger than graphite sleeve since the air contact with the support post during the early stage

K. Kunitomi et al., Nucl. Eng. Des., 233, 235-249 (2004).

p.31

DLOFC Simulation Results (2/2)

• Peak CV pressure is close to limit • However, increase in CV volume needs careful

consideration since CO concentration is close to limit

Reactor scram

Elapsed time (s)

Pre

ssu

re (

kg/c

m2)

Acceptance criteria

Flammable range

Detonation range

JAEA, JAEA Oarai R&D center Reactor installment licensing application document [Supplementary volume 3 HTTR].

p.32

Water Ingress Accident Initiating Event

Rupture of heat transfer tube in water cooler

p.33

Water Ingress Accident Sequence

• Though the pressure of pressurized water loop (3.5MPa) is lower than primary loop (4.0MPa), it is assumed that the ingress of water immediately occurs after the rupture of heat transfer tube

• Amount of water ingress into primary loop is performed by hand calculation considering pumping by feed water pump & gravity

• Amount of water ingress into core is calculated considering accumulated heat in SG structure & profile of primary loop flow rate

Rupture of PPWC heat transfer tube

Reactor scram, PPWC pump stop

Oxidation of graphite

component

Oxidation becomes negligible due to core cooldown

OXIDE-3F

ACS startup

Water ingress into primary loop

Water ingress into core

Core cooldown

THYDE-HTGR

p.34

Water Ingress Simulation Results

Water ingress into primary loop 175 kg

Oxidized graphite 44 kg

• Reacted water with graphite is considerably smaller than amount of water ingress into primary loop because of IVs actuation & GC stop

• Reactor power does not increase due to large shutdown margin

• Safety valve of primary loop does not open

→ No FP release into CV Elapsed time [min]

Pre

ssu

re [

kg/c

m2]

Tem

per

atu

re [

oC

]

S. Saito et al., JAERI-1332 (1994).

Primary coolant pressure

Peak fuel temperature

p.35

Major Results of Abnormal Events

Evaluation items Event Results

Fuel temp. Leakage in inner pipe of concentric pipe 1715oC

RPV temp. DLOFC events 530oC

RCPB pressure Rupture of inner pipe of concentric pipe in primary cooling system

45.9 kg/cm2

PPWC tube temp. Rupture of pipe in pressurized water cooling system 368oC

IHX tube temp. Rupture of inner pipe of concentric pipe in secondary cooling system

956oC

CV pressure Rupture of concentric pipe in primary cooling system 4.7 kg/cm2

Graphite oxidation Rupture of concentric pipe in primary cooling system 3.6 mm (Sleeve)

Effective dose rate Rupture of concentric pipe in primary cooling system 1.5 mSv

JAEA, JAEA Oarai R&D center Reactor installment licensing application document [Supplementary volume 3 HTTR].

p.36

Radionuclide Transportation in HTGR

RCPB

Reactor building (RB)

Fuel element

Removal by Purification system

Dust sorption

Plate-out on RCPB

Plate-out on RB

Release to building

Lift off

Release to environment

Normal operation

ACD

p.37

Plate-out on RCPB

Coolant flow

Mass transfer region

Mass transfer

Diffusion

Sublimation

Adsorption desorption

FP concentration

RCPB

Circulating FP in primary loop deposits on RCPB “Plate-out”.

The deposited FP becomes radiation source for worker does in maintenance.

A part of deposited FP departures from RCPB in case of ACD

The mechanism of plate-out can be classified in to the following:

- Mass transfer from coolant flow to wall proximity region of RCPB

- Adsorption and desorption equilibrium between wall proximity region and RCPB surface

- Diffusion in RCPB

- Sublimation from RCPB to coolant

p.38

Plate-out Experiments

Country Facility FP Material Condition

GBR Dragon Cs, Sr, I、Ag ― FP plate-out

DEU AVR/VAMPYR-I Cs, Sr, Ag, I Ti, 15Mo3, 4541, 4961, ST35.8, 10CrMo910

FP plate-out considering dust effect, 850-900oC, Laminar

DEU AVR/VAMPYR-II Cs, Ag, I Incoloy800, Inconel617 FP plate-out considering dust effect 400-850oC, Turbulent

DEU SCAFEX Cs, Sr, I, Ag ― FP plate-out, 100oC-350oC, Re:1500-4000

DEU LAMINAR Cs, I Incoloy800, Inconel617, 10CrMo910, 15Mo3

FP plate-out, 300-900oC, Re:6000-10000

FRA PEGASE/SAPHIR Cs 15Mo3, 4541, Nimocast713LC, Inconel625

FP plate-out, 585-970oC, 11-13g/s

FRA PEGASE/CPL-2 Cs, I Incoloy800, Hastelloy B, SS347, SS410, T22

FP plate-out, 350-750oC,

FRA SILOE/COMEDIE Cs, Co, I, Ag, Te,Cr

Incoloy800, Hastelloy X, SS AISI 347

FP plate-out, 600-835oC, 16-45g/s

JPN JMTR/OGL-I Cs, I SUS, Hastelloy X FP plate-out, RT-950oC, 10-60 m/s

USA GAIL Cs, Sr, I 1/2Cr 1/2Mo FP plate-out, 90-500oC, Re:12000

USA Peach Bottom Unit1 Cs, Sr, I ― FP plate-out

USA Fort St. Vrain Cs, Sr ― FP plate-out

p.39

Evaluation for Plate-out FP

IAEA, IAEA-TECDOC-978 (1997).

Plate-out activity on the circulator in Fort. St. Vrain was radiochemically examined at General Atomics.

Generally, the comparison between prediction and test data are in good agreement.

(Over prediction in Cs-134 by factor of 2, Underprediction in Cs-137 by factor of 1.4)

p.40

Removal in Purification System

Pre charcoal trap (PCT)

Oxide copper bed

Cold charcoal trap (CCT)

Molecular sieve trap

Filter

Filter

Heater

Heater

A part of circulating FP is removed by purification system CCT and PCT removes Kr and Xe, as well as I, Br, respectively

N. Sakaba, Nucl. Eng. Des., 233, 147-154 (2004).

p.41

Lift off from RCPB

Chemical desorption

FP departure from RCPB

Increase in velocity Temperature increase

Moisture increase

Failure in piping Failure in heat transfer tube Coolant leakage

Physical departure

FP concentration

decrease

RCPB

Coolant flow

Lift force

Drag force

O H H

O H H

O

H H H

p.42

Release to Confinement

RCPB leakage or RCPB failure • The following FP is released to confinement

on reactor coolant a. Circulating FP in primary loop b. FP detached from RCPB c. FP adsorbed in dust Tube rupture in SG • According to the amount of water and

steam intruded in primary loop, FP is released to confinement through safety valve.

SG

Reactor

Safety valve

FP

Confinement

p.43

Release to Environment

Released FP from primary loop

FP accumulated in core

Confinement A part of FP is deposited on the

confinement surface Velocity is the dominant parameter

because of the large velocity in case of rupture in RCPB

Environment

Released from stack

Leakage in confinement