7/28/2019 Failure Assessment of Piping
1/1
Failure assessment of piping under severe accident loadsJens
Arndt, Jrgen Sievers (GRS)
In face of severe accident scenarios with moltencore material
which occurred in 2011 at Fukushima
Daiichi and in 1979 at Three Mile Island-2 theintegrity
assessment of primary circuit devicesrequires a special
concern.
Method ASTOR
For the accomplishment of a simplified analysisconcerning
integrity of the components during asevere accident and especially
the question whichcomponent fails first in framework of
thermo-hydraulic analysis with system codes like ASTEC orATHLET an
efficient method called ASTOR(Approximated Structural Time Of
Rupture) hasbeen further developed. This method employs the
hypothesis of linear damage accumulation formodeling damage
progression. A failure time sur-face which is generated by
structural finite ele-ment (FE) analysis of varying pressure and
tem-perature loads serves as a basis for estimations offailure
times. The time of failure is strongly depend-ent on the changing
stress level during the transientloading and the temperature
dependent materialproperties characterizing plastification and
creep.
ASTOR calculates foreach point of time whichis characterized by
a
temperature and a pres-sure value a damageincrement. The result
ofthe summation of damageincrements is a damagevalue D(t). The
failurecan be assumed whenD(t) reaches a value of 1or a smaller
value ifsafety factors are inclu-ded.
Simulation of a severe accident scenario
In following the results of a FE-based failure
assessment and an ASTOR failure analysis for areactor coolant
line (RCL) of a PWR loaded duringan assumed severe accident
scenario (stationblackout) with molten core material in the
lowerhead of the reactor pressure vessel (RPV-LH) arecompared. The
time of failure of the RCL is ofspecial concern because a failure
before the RPV-LHs failure may enable a significant
pressuredecrease. The figure shows calculated failure timesbased on
FE-analyses (Run A and B with differentassumptions concerning
failure criteria) andASTOR results with different damage values.
Theinvestigation shows that ASTOR results for damage
values of about 0.4 - 0.5 are close to the FEresults.
(1) (2)(3)(4)(5)
D-Damagefactor (ASTOR)(1) D = 0.4 @ 46760 s
(4) D = 0.5 @ 47750 s(5) D = 1.0 @ 49118 s
(3) Run B-Failure @ 47656 s
(2) Run A-Failure @ 47471 s
ConclusionThe method ASTOR enables a fast estimation offailure
times and can be integrated into the frame-work of thermohydraulic
system analysis programseasily. The application of ASTOR is limited
to theboundary conditions concerning pipe geometry,material data
and type of loading used for genera-tion of the failure surface. An
uncertainty of thecalculated failure times exists but can be
con-strained by a decrease of the assumed damagelimit value.
Dependent on the required accuracy ofthe time of failure of a pipe
three failure assess-ment methods are accomplishable:- ASTOR
(useful for implementation in systemcodes, limited applicability,
limited accuracy, exten-sive concerning generation of failure
surface);- FE-analysis with simplified FE-model (flexibleconcerning
application, limited applicability con-cerning complexity, high
accuracy);- Complex FE-analysis model with considera-tion of
interaction between components (exten-sive concerning generation of
analysis model, flexi-ble concerning application, high
accuracy).
Gesellschaft fr Anlagen- und Reaktorsicherheit (GRS) mbH,
Schwertnergasse 1, 50667 Cologne, GermanyJens Arndt, Tel. +49 221
2068 679 / [email protected]; Dr. Jrgen Sievers, +49 221 2068 747
/[email protected]
-
tF
Time of failure
Pressure P(MPa)
Temperature
T(C)
