2.1.2.4
Probabilistic safety assessment
Level 1 PSA
for power operations
The level 1 PSA study for NPP V-1 was prepared in the framework of the
PHARE Programme by the UK company Electrowatt Engineering Services in cooperation with
RELKO and VÚJE in 1992. The objectives of the study were as follows:
to determine core damage frequency (CDF)
based on the failure trees method and event trees method,
to identify initiating events and
accident sequences with a predominating contribution to core damage,
to assess the effects of various
modifications of safety systems on CDF,
to specify recommendations for the
updating of emergency operating procedures based on predominant accident sequences.
The study assessed NPP V-1 unit 1 as a
pilot unit, being on power, for the status quo of in July 1992. The study contemplated
initiating events (transients and LOCA), internal fires and internal floods. In a majority
of cases, own specific data for the individual intensities of component failures were used
in the PSA. Published generic data were used whenever specific data were not available.
Event trees and failure trees were modeled in the Risk Spectrum software environment. A
very detailed model of the unit was set up for the individual technological systems,
electric switchboards and I&C systems with the corresponding controls down to the
sensor and relay level. Also, this programme was used to finally quantify the model.
Results of the first Level 1 PSA study:
CDF = 1.70E-03/year
Since „small reconstruction “ of NPP V-1 was under preparation during July 1992, the
obtained PSA model was used to evaluate and to supplement suggested changes into the
project. Under the „small reconstruction“, improvements of I&C systems, electric
and technological circuits were made. Also emergency operating procedures were
supplemented by adding critical safety functions restoration guidelines.
The changes made in the configurations of both NPP V-1 units by the end of 1993, and the
completion of symptoms-based emergency procedures resulted in a new model of Level 1 PSA.
Level 1 PSA study after „small reconstruction“ was drafted in
December 1993. The new model incorporated, in addition to unit hardware modifications,
also remedial operator actions during accidents. The modeling of operator's interventions
enabled the completion of symptoms-based emergency procedures. Values obtained from a
detailed analysis of the human factor reliability, using the THERP method were used for
the probability analysis of operator's error.
Results of Level 1 PSA study after „small reconstruction “:
CDF = 8.9E-04/year
Based on the results obtained from PSA study some objectives were set for the NPP V-1
„gradual upgrading“ under preparation.
Preliminary suggestions for changes in unit systems were used to set up Level 1 PSA model
”Basic Engineering”. The model was subsequently quantified, and based on sensitivity
analyses, the most appropriate project of change for the individual unit systems was
selected.
The PSA model ”Basic Engineering” was used to determine the
preliminary core damage frequency - CDF of 2.57E-04/year, without symptom-oriented
emergency operating procedures.
Having completed the NPP V-1 “gradual upgrading“, a new Level 1 PSA model was
developed to reflect the actual status of the unit in 2000. The Level 1 PSA study after
„gradual reconstruction “ of NPP V-1 was developed by Slovak engineering and research
companies in cooperation with VÚJE.
|
Fig. 2.1.1 Comparison of the outcome of NPP V-1 upgrading: PSA aspect
|
Main results of the recent Level 1
PSA study:
CDF = 2.56E-05 /year
Table 2.1.1 documents the results of the most recent NPP V-1 PSA study from the aspect of
the contributions of the individual initiating events.
Table 2.1.1
Contributions
of initiating events to RC melting frequencies |
Initiating
event |
Description |
Frequency
[1/year] |
%
of overall CDF |
LL-LOOP2,5 |
Large
leak from loop 2 or 5 |
7.00E-6 |
23.0 |
ML(32-100) |
Medium
leak (32-100 mm) |
4.30E-3 |
19.8 |
SL |
Small
leak (0-32 mm) |
2.80E-2 |
15.6 |
LL-LOOP1,4 |
Large
leak from loop 1 or 4 |
7.00E-6 |
12.8 |
SGTM |
SG
manifold rupture |
5.00E-6 |
3.8 |
IFL |
Interfacing
LOCA |
5.00E-4 |
3.5 |
PSL |
Leak
from pressurizer |
6.90E-4 |
2.8 |
SGTR |
SG
tube rupture |
8.90E-4 |
2.8 |
LL-LOOP3,6 |
Large
leak from loop 3 or 6 |
7.00E-6 |
2.6 |
FIRE-TGHALL |
Fire
at the machine room |
6.20E-2 |
2.4 |
ML(100-200) |
Medium
leak (100-200 mm) |
1.50E-4 |
2.3 |
2TG |
Trip
of both TG |
5.50E-1 |
2.3 |
LOP |
Complete
failure of house power supply |
2.50E-2 |
2.2 |
SHB |
Main
steam header rupture |
2.20E-3 |
1.3 |
|
Other
events |
|
2.8 |
Main recommendation of the most recent PSA study
To reduce CDF, the PSA study has recommended to develop and implement symptom-oriented
emergency operating procedures - SBEOP. The development of SBEOP is in the course and
implementation is expected by the year 2003.
Plans and ongoing activities related to NPP V-1 PSA
Based on the most recent PSA model for NPP V-1 the system of real-time risk monitoring
system ”EOOS - Risk Monitor” is implemented for power operation, for which NPP
Bohunice have been licensed.
Development of Level 1 PSA study for low-power and shut-down reactor (shut down PSA) is in
progress.
The PSA model will have to be updated after the development of SBEOP.
The power plant has systematically used PSA as a tool for the development and setting of
priorities with respect to improvements that were implemented during the upgrading
programme aimed at achieving the best safety improvements possible. Moreover, the
conclusions and the results of the PSA have been incorporated into the training programme
for NPP V-1 operators, and selected emergency scenarios have been implemented into the new
unit simulator at VÚJE.
    
|