TY - GEN
T1 - Dynamic behaviour of pipelines in power plants
AU - Calonius, Kim
PY - 2010
Y1 - 2010
N2 - Dynamic excitation due to a pipe break can cause pipe to
abruptly displace and hit the components, instrumentation
and equipment nearby. In order to minimize the risk of
such damage by pipe whips in a power plant, different
types of restraints and supports are designed for the
pipelines.
The usability of different types of elements provided by
Abaqus, a commercial general-purpose finite element code,
in modelling the dynamic behaviour of pipelines is
tested. A relatively short pipe line section with one
bend and one restraint is chosen as a test case. Simple
and typical nuclear power plant pipeline geometry and
materials are chosen. The stiffness of the restraint as
well as the flexural stiffness of the pipe cross-section
are solved with static compression simulations with a
detailed model using three-dimensional solid and shell
elements. After that, the model is substituted with
couple of simpler models using pipe and/or elbow elements
for the pipe and a spring element for the restraint.
The eigenmodes of models are calculated and compared with
each other. The pipe whip is simulated with nonlinear
dynamic analyses with the most adequate models according
to the preliminary linear analyses. The displacement and
stress results of different models are compared with each
other and the reliability and adequacy of different
element types are discussed. Sensitivity study is made by
varying analysis type, material properties, mesh density,
element properties and inner pressure.
The results of the most adequate simple models with the
right combinations of special-purpose elements provided
by Abaqus corresponded well to the ones of the much
larger three-dimensional solid and shell element models.
The post-yield behaviour of the pipe steel material has
some notable effect on the behaviour of the pipe in a
guillotine pipe break. Obviously, rate dependence makes
the structure slightly dynamically stiffer. The inner
pressure of the pipe and its assumed decay after the pipe
break has a major influence on the results.
It is shown, how the stress distribution along the pipe
axis, around the cross-section and through the thickness
of the wall can be solved with elbow elements on a fairly
detailed level. Long pipe runs with many supports and
restraints should preferably be modelled with simple
structural elements such as pipe and elbow elements and
special-purpose elements in order to save time and
numerical errors due to overly large models.
AB - Dynamic excitation due to a pipe break can cause pipe to
abruptly displace and hit the components, instrumentation
and equipment nearby. In order to minimize the risk of
such damage by pipe whips in a power plant, different
types of restraints and supports are designed for the
pipelines.
The usability of different types of elements provided by
Abaqus, a commercial general-purpose finite element code,
in modelling the dynamic behaviour of pipelines is
tested. A relatively short pipe line section with one
bend and one restraint is chosen as a test case. Simple
and typical nuclear power plant pipeline geometry and
materials are chosen. The stiffness of the restraint as
well as the flexural stiffness of the pipe cross-section
are solved with static compression simulations with a
detailed model using three-dimensional solid and shell
elements. After that, the model is substituted with
couple of simpler models using pipe and/or elbow elements
for the pipe and a spring element for the restraint.
The eigenmodes of models are calculated and compared with
each other. The pipe whip is simulated with nonlinear
dynamic analyses with the most adequate models according
to the preliminary linear analyses. The displacement and
stress results of different models are compared with each
other and the reliability and adequacy of different
element types are discussed. Sensitivity study is made by
varying analysis type, material properties, mesh density,
element properties and inner pressure.
The results of the most adequate simple models with the
right combinations of special-purpose elements provided
by Abaqus corresponded well to the ones of the much
larger three-dimensional solid and shell element models.
The post-yield behaviour of the pipe steel material has
some notable effect on the behaviour of the pipe in a
guillotine pipe break. Obviously, rate dependence makes
the structure slightly dynamically stiffer. The inner
pressure of the pipe and its assumed decay after the pipe
break has a major influence on the results.
It is shown, how the stress distribution along the pipe
axis, around the cross-section and through the thickness
of the wall can be solved with elbow elements on a fairly
detailed level. Long pipe runs with many supports and
restraints should preferably be modelled with simple
structural elements such as pipe and elbow elements and
special-purpose elements in order to save time and
numerical errors due to overly large models.
M3 - Conference article in proceedings
SN - 978-951-38-7591-6
VL - 1
T3 - VTT Symposium
SP - 113
EP - 125
BT - Baltica VIII
PB - VTT Technical Research Centre of Finland
CY - Espoo
T2 - BALTICA VIII - International Conference on Life Management and Maintenance for Power Plants
Y2 - 18 May 2010 through 20 May 2010
ER -