Nuclear Power Plant Emergency

Power Supply System Diesel

Generator Reliability Modeling
and Analysis

Presented by
Muneer. M.P
(Enrolment No. ENGG02201201014)

Guided by Technology Advisor

Dr. J. Jayapandian Smt. N.M Vijaya
INTRODUCTION
 Electrical power supply system of Nuclear Power Plant (NPP)
comprise of
 Normal Power Supply: derived from grid or plant generator
 Emergency Power Supply: derived from normal power supply
and is backed up by onsite Emergency Diesel Generators

 If normal power supply is not available, Emergency Diesel

Generator (EDG) supply essential loads of the plant

 A total loss of AC power is referred “Station Blackout (SBO)”

occurs as a result of complete failure of both onsite and offsite
AC power supplies

2
SCOPE & OBJECTIVE
 Studies shows that SBO is a significant contributor to the
risk associated with operation of NPP (NUREG-6890)
 Unreliability of EDG can significantly increase the SBO
risk and good reliability is to be ensured
 EDGs qualified for NPPs are complex system consisting
many subsystems working in tandem
 Current Probabilistic Safety Assessment (PSA) uses
generic modelling of EDG
 The objective of this work is to develop detailed reliability
models for EDG subsystems and evaluate reliability

3
OVERVIEW OF THE WORK

 Analysis carried out on typical DG set taking PFBR

DG set as reference

 Failure modes considered: DG fail to run and DG fail

to start
 Method used: Fault Tree Analysis

 Fault trees were developed and analyzed using

ISOGRAPH software

4
 EDG SYSTEM

Generator
side

View of PFBR
DG Set

Engine Side

5
EDG SYSTEM
Electrical Power Output
Cooling Lub Oil Speed Sensing Generator
Governor
Water System Excitation
System System Sensing

DIESEL ELECTRICAL
ENGINE GENERATOR

Coupling

Fuel Oil Starting Air Charge air &

System Exhaust gas Control Panel
System
system

6
EDG RELIABILITY MODELING

Reliability modeling involves

• System study and function flow diagram

preparation

• Fault tree development

• Basic event modeling

• Consideration of Common Cause Failure (CCF)

7
Function Flow Diagram Preparation

 The EDG system is composed of various

subsystems, and its performance is depending on
the performance of the individual subsystems.

 Function flow diagram for each subsystem is

prepared which show the various components
forming the subsystem

8
Starting Air System

9
Fuel Oil System

10
Cooling water system

11
Lube oil system

12
Charge Air &
Exhaust Gas System

13
Governor system

14
Excitation system

15
Fault Tree Development

 Fault tree analysis is an analytical technique, whereby an

undesired state of the system is specified, and the system is
then analyzed to find all credible ways in which the undesired
event can occur

 DG fail to function as top event and fail to start and fail to run
are sub events forms main tree

 Main fault tree divided into sub trees each representing failure
of each subsystem

 Each sub tree is developed further to represent subsystem

failures in terms of component failures

16
 Fault Tree Development

Fault tree development DG failed to

function

DG

DG failed to
start

DG FS

Fuel oil sys Starting air cooling Charge air & Lube oil DG I&C
fail to start system FS water sys FS exhaust system FS logic system DG fail to
system FS FS run

FUEL SYS FS ST.AIR FS COOLING FS CHARGE FS LUB SYS FS CNTRL FS

DG FR

fuel oil sys Cooling Lube oil sys Engine fail Governor Generator Excitation Control Charge &
fail to run water sys FR fail to run to run fail to run fail to run sys fail to logic sys exhaust sys
run fail to run FR

FUEL FR COOLING FR LUB SYS FR ENGINE FR GOV. FR GEN. FR EXC.FR CNTRL FR CHARGE FR

17
 Basic Event Modeling

Model Generic form Example

Fixed Model Fixed failure Valve fail to open on
probability demand
Dormant model  . Power supply unit
Q   . MTTR
2 failure
Constant failure 
Pump fail to run
(    )t
and repair rate Q(t )  (1  e )

model

  Component Failure rate   Inspection interval

MTTR=Mean Time to Repair   Component Repair rate

18
Basic Event Modeling

Component reliability data source includes

 IAEA TECDOC-478, IAEA TECDOC-508

 US. NRC Component reliability data sheet, US NRC CCF

Parameter estimation

 OREDA hand book

 IEEE 500, IEEE 493

 Literatures and event reports

19
Incorporation of Common Cause
Failure (CCF)

• In CCF, two or more components fail in same mode

due to the shared cause

• CCF modeling methods available in ISOGRAPH

include Alpha factor, beta factor, Multiple Greek
Letter (MGL)

• Beta factor modeling is used since

Q  (1  β ) Q
I t
- It is simple QCCF  β . Qt
- Experience from using this model shows that it gives reasonably
accurate results for redundancy level up to 3 or 4 (mosleh et al.)

20
Cooling water system

21
Component failure data

SL No Component Failure rate Source

(hr-1)
1 Engine driven pump 6.68E-5 OREDA
2 Heat exchanger 5.7E-6 Ever green
3 Thermostat valve 1.00E-7 IEEE 493
4 Non return valve 2.00E-7 Ever green
5 Butterfly valve 7.00E-6 IAEA TECDOC 478
6 Level switch 7.20E-7 IAEA TECDOC 478
7 Temperature switch 5.00E-7 IAEA TECDOC 478
US NRC component
8 Tank 1.82E-8
reliability data sheet

22
 Cooling
water system
failed to run

COOLING FR

Fault tree
JW high DG S W
E T lev el temp. trip
low trip sys failed

JW T TR SW SYS FAIL
ET LOW TR

S O of JW
S O of E T E T failed S W P failed NRV -3 SWC failed
high temp.
low lev el to run failed to extract
switch
switch JW temp. heat
high
HTS SO SWP FR NRV-3
LS SO ET RP SWC
JW T HIGH r=5.5e-007
r=7.2e-007 r=1.82e-008 r=6.68e-005 r=2e-007

SW C No sec.
failed water
No JW JW C failed circulation
circulation to extract
heat
SWC BL SEC WAT
JW CIRC JWC
r=5.4e-006

JW P fail to JW loop TS V failed JW C failed

run NRV No sec.
failures water
circulation
JWP FO JW NRV F TSV TC JCC LH
SEC WAT
r=6.68e-005 r=1e-007 r=5.4e-006

NRV -1 NRV -2
failed failed S econdary Secondary
water inlet water outlet
closed closed
NRV-1 NRV-2
SEC W. IN SEC.W OU
r=2e-007 r=2e-007

BTV -1 BTV -2 BTV -3 BTV -4

failed failed failed failed

BTV-1 BTV-2 BTV-3 BTV-4

r=7e-006 r=7e-006 r=7e-006 r=7e-006

23
RESULTS

 Failure rate of overall DG system is

– Fail to start : 2.27E-3 per demand
– Fail to run : 7.93E-4 per hour

24
 RESULTS
SL No Failure rate
Subsystem
Fail to start (d-1) Fail to run (hr-1)
1 Fuel oil system 1.06E-05 4.14E-04
2 Starting air system 5.51E-04 --
3 Cooling water system 3.00E-05 1.15E-04
4 Charge and exhaust gas system 1.63E-05 1.17E-05
5 Lub oil system 2.00E-05 7.92E-05
6 Engine 2.02E-04* 1.14E-04
7 Governor 2.86E-04* 1.29E-05
8 Electrical generator 2.18E-05
5.38E-04*
9 Generator excitation system 1.86E-05
10 Control logics 6.15E-04 5.90E-06
Total 2.27E-03 7.93E-04

* Fail to start analysis of engine, governor and generator is not carried out in
component level. However overall subsystem level data is adopted from NRC
report
25
 RESULTS – PIECE PART CONTRIBUTION

DG fail to start
Engine Governor Generator & excitation
9% 13% system
DG ‘fail to start’ is Lub oil 24%
mainly contributed by 1%
control logic system, Charge air
1%
starting air system, and Cooling
generator & excitation 1%
system

Control logics
Starting air 27%
24%
Fuel oil
26
 RESULTS- COMPARISON

Fail to start

1.0E-03
Failure rate ( demand-1)

1.0E-04 NRC report

Analyzed result
1.0E-05

1.0E-06
Start air Cooling Charge Fuel oil Lub oil Control
water air logics

27
RESULTS – PIECE PART CONTRIBUTION

Piece part contribution Charge air

DG fail to run 1% Lub oil Engine
10% 14%
Cooling
15% Governor
DG ‘fail to run’ is mainly 2%
contributed by fuel oil Generator
system followed by 3%
cooling water system Excitation
and engine 2%
Fuel oil
Control logics
52%
1%

28
 RESULTS - COMPARISON
Fail to run
1.0E-03
Failure rate ( hour-1)

1.0E-04

1.0E-05 Analyzed result

NRC report

1.0E-06

29
DG MISSION TIME RELIABILITY

 The targeted reliability level of EDG is 0.975 per

demand ( US NRC RG 1.155)

 The term demand means an event requiring the

EDG to start and supply power to the safety-related
bus

 Reliability of DG to start and operate successfully up

to different mission time of 24 hours is 0.9789

30
CONCLUSION

 The detailed reliability modelling and analysis of

EDG is carried out
 DG subsystem level failure rate is assessed

 Piece part contribution of subsystems in DG failure is

analyzed

 DG mission time reliability is within targeted reliability

of 0.975 even for scenario demanding continuous
operation of DG for 24 hours.

31
FUTURE SCOPE

 Emergency Power Supply system comprise DG sets,

emergency switchgear and logics associated with
the isolation of class IV system for loss of power
supply and planned sequential pick up of loads. The
overall detailed reliability analysis of Emergency
Power Supply System considering all other
equipment is proposed for future work

32
PUBLICATIONS

 Presented a paper titled “Assessment of Reliability of

Nuclear Power Plant Diesel Generator System by Fault
Tree Analysis” in “National Conference on Reliability and
Safety Engineering -2014 (NCRS-14)”, held at Anna
University BIT Campus, Tirichirapally on 13-15 Feb. 2014
 REFERENCES
1. S.A. Eide, C.D. Gentillon, T.E. Wierman, INL, "Reevaluation of station Blackout Risk
at Nuclear Power Plants", NUREG/CR-6890, U.S. NRC,2005.
2. J.D. Andrews and T.R. Moss “Reliability and Risk Assessment”
3. S. A. Eide, T. E. Wierman and D. M. Rasmuson, “Detailed Study of Emergency
Diesel Generator Performance Using EPIX/RADS Database” Idaho National
Laboratory, USA, Sep. 2008
4. Zbigniew Matuszak, Leszek Chybowski, “Simulation of Unavailability of the Offshore
Unit’s Power Plant System with Use of Selected Algorithms” Technical University of
Szczecin, Poland
5. Congling Dong et al, "Marine Propulsion System Reliability Research Based on Fault
Tree Analysis”, Advanced Shipping and Ocean Engineering, Vol. 2 Iss. 1, PP. 27-33,
Mar. 2013
6. Akira Kubo et al "Analysis of Reliability/Failure of Newly Built Diesel Engines"
International Council on Combustion Engines, Paper No:95, CIMAC Congress, 2004

7. Manual 82340 (Revision C), “EGB - Proportional Governor/Actuator with Hydraulic

Amplifier Systems”, Woodward, Inc.

34
REFERENCES
8. IAEA-TECDOC-478 “Component Reliability Data for Use in Probabilistic Safety
Assessment”, 1988

9. IAEA-TECDOC-508 “Survey of Ranges of Component Reliability Data for Use

in Probabilistic Safety Assessment”, 1989

10. OREDA – Offshore Reliability Data Handbook, 4th edition, 2002

11. Evergreen Safety & Reliability Technologies, LLC “Historical Nuclear Power
Plant Component Failure Rate Data”

12. U.S. NRC, “Component Reliability Data Sheets Update 2010”, January 2012

13. A. Mosleh et al “Procedures for Treating Common Cause Failures in Safety

and Reliability Studies”, NUREG/CR-4780, USNRC, 1987

14. U.S. NRC, “CCF Parameter Estimations Update 2010”, January 2012

35
THANKYOU
36