Hellenic Navy, Hellenic Naval Academy, Mechanics & Materials Division, Marine Materials Laboratory, Hazjikyriakou Avenue, Piraeus 185 39, Greece
Shipbuilding Technology Laboratory, School of Naval Architecture and Marine Engineering, National Technical University of Athens, 9 Heroon Polytechniou Avenue, Zografou,
Athens 157 73, Greece
b
a r t i c l e
i n f o
Article history:
Received 13 November 2013
Received in revised form 30 May 2014
Accepted 27 July 2014
Keywords:
Thermo-mechanical modelling
Finite element analysis
Convergence ow diagram
Repair welding
a b s t r a c t
In this paper the authors propose a practical ow diagram for the systematic development and solution of
complex thermo-mechanical nite element analysis models. The proposed diagram consists of three different phases and provides a step-by-step guide for the development of the nal thermo-mechanical model,
taking into account convergence issues, mesh density and estimation of time step magnitude. In phase I, a
preliminary thermo-mechanical analysis is carried out in order to get an idea of the model behaviour, the
required resources and the feasibility of the overall analysis. In phase II the nal thermal model is developed
in full, taking into account the mechanical results obtained at the end of phase I, whereas in phase III the
nal mechanical model is generated on the basis of a continuously modied thermal model. The proposed
procedure presented herein in the form of a ow diagram provides the capability for gradual output of the
numerical results (preliminary results, thermal results, mechanical results), while paying attention to the
time-consuming problem of results convergence required for a numerically accurate analysis. The former is
an important issue for large-scale complex simulation projects, whereas the latter provides evidence that
the development of the numerical model has been realized on the basis of the modelling laws. For better
presentation and understanding, the proposed procedure is applied to the study of a nite element analysis
thermo-mechanical model, where increased intricacy generally exists.
2014 Elsevier B.V. All rights reserved.
change effects on the structural response of the structure. This is performed by importing to the mechanical model the nodal temperatures at each time increment and calculating the thermal strain.
From the aforementioned staggered approach it is deduced that
both thermal and mechanical models must normally run with the
same analysis parameters, namely time step magnitude and mesh
density. If, for example, the material undergoes phase transformation accompanied by volume change during a specic short temperature range, a small constant time step and a ne mesh are required
in the areas of transformation for both the thermal and the mechanical analysis. This allows for the accurate monitoring of the transient
stress developed during the transformation temperature range [5].
Thus, for the entire analysis there should be an exact correspondence between the mesh density and the time step magnitude
between the two models. This requirement renders the whole procedure of model development very complex and time-consuming.
1.2. The three major problems: mesh density, time step and
convergence of results
The rst problem arising during the thermo-mechanical modelling is that the thermal and the mechanical models are completely
289
290
entire simulation. Taking into account the variety of different elements and analysis types that exist nowadays in most commercial
nite element platforms, the development of an efcient set of
equations between the reduction of the time step and the gradual
mesh renement that leads to results convergence is a very
difcult and triggering task. In the current proposed diagram, this
spatial and temporal renement is based on the repetitive execution of the thermal and the post-mechanical model. This execution
provides feedback pertaining to the appropriateness of the spatial
and temporal renement that was applied. The latter methodology
has the advantage of applicability in most thermomechanical
simulations except of casting simulation where it is not directly
applicable due to material ow.
The gradual reduction of the initial time step that is applied by
the analyst is strongly affected by all the temperature and time
dependent phenomena that take place during the entire simulation. It is well known, that in a typical transient non-linear thermomechanical analysis, temperature and time dependent magnitudes
exist. Temperature dependent magnitudes can refer, for example,
to the material properties, coefcient of heat convection and convection heat, radiation; whereas time dependent magnitudes can
refer to the moving heat source, heat convection, operational loads,
pressures, existence of gaps, etc.
As a basis for the discussion of the next paragraphs, Fig. 2
depicts typical examples of the temperature dependent heat
capacity, the thermal conductivity, the convection lm coefcient,
the yield stress and the thermal dilatation of a typical mild steel
that undergoes several microstructural transformations depending
on the peak austenitization temperature (Tpeak) [6,7]. In the same
gure, the time dependent moving heat source of a welding arc
is also presented [6]. Furthermore, in Fig. 2f, the temperature
depended axial stress response of an axially xed steel specimen
that undergoes phase change transformation is shown [5,6].
In nite element simulations, the temperatures in the thermal
analysis and the displacements in the mechanical analysis are calculated for every node of the model and are exported at every time
step. Therefore the applied gradual reduction of the initial time
step value should nally:
i. Provide small temperature differences at every node of the
model between all successive analysis steps, so that the temperature dependent phenomena are accurately modelled.
For example, a very small time step can result in very small
nodal temperature differences between all successive analysis steps. It is up to the researcher to decide, whether the
latter temperature difference can accurately model the
non-linear material properties at the areas of solid state
transformations (see Fig. 2a, c and d) or whether it is enough
to accurately account for the convection heat losses (see
Fig. 2b). On the other hand, a relatively bigger time step
can provide larger nodal temperature differences between
successive analysis steps and thus hiding or articially
minimizing the effects of phase change on the transient
mechanical response of the structure. Taking into account
that the steel phase transformation temperature range is
approximately DTtr = 300 C, a practical guideline is to select
the maximum allowable time step that provides temperature differences of maximum 30 C, or 10% of DTtr. This small
temperature difference will later provide the basis for an
accurate mechanical analysis where small thermally
induced stress differences are also required. Therefore the
selected maximum allowable time step should additionally
keep the thermally-induced stress differences between successive analysis steps in a stress analysis smaller than a
small percentage of the material yield stress. In current
paper the value of 5% of the material yield stress is
291
Fig. 2. An example of temperature and time dependent magnitudes in a typical thermomechanical analysis, reprinted from [6,7]. (a) Material heat capacity and thermal
conductivity, (b) surface convection lm coefcient, (c) material yield stress, (d) material thermal dilatation, (e) three dimensional moving heat source, (f) axial stress
response of an axially xed steel specimen that undergoes phase change transformation.
292
293
294
295
Fig. 4. The results at the end of the linear loop (NE = 6079 elements, Dt = 0.1 s). (a) Transient temperature at the mid-length of the treated zone (Tmax = 864 C, Tmin = 25 C),
(b) temperature distribution during cooling (Tmax = 33 C, Tmin = 25 C), (c) von Mises static stress at the mid-length of the treated zone (rmax = 1063 MPa, rmin = 0 MPa), (d)
von Mises static stress due to the applied operational pressure only (rmax = 205 MPa, rmin = 0.5 MPa), (e) nal displacement magnitudes due to the applied operational
pressure only (dmin = 0 mm, dmax = 0.515 mm, scale factor 10).
4.2. Phase II: nalizing the thermal model and obtaining the thermal
results
Phase II pertains to the main analysis of the thermal problem. It
aims at calculating the transient and residual temperature distribution of the structure under investigation. The ow diagram of
phase II is shown in Fig. 6.
At the beginning of this phase, the thermal model is executed
using the mesh density and the time step derived from phase I.
Depending on the convergence criteria, the models mesh is progressively rened until the desired level of accuracy in the areas
of interest is attained [33]. For example, the gradual increase of
mesh density as we approach the weld line AB will allow better
estimation of the size of the heat affected zones where phase
transformations occur during the overall treatment. Note that
the mesh of the thermal model must also be rened but to
a lesser degree at areas of high stress gradients of the mechanical model, as observed at the end of phase I. Notwithstanding
296
Fig. 5. The results at the end of phase I (NE = 7587 elements, Dt = 0.08 s). (a) Transient temperature at the mid-length of the treated zone (Tmax = 954 oC, Tmin = 25 C), (b)
temperature distribution during cooling (Tmax = 34 C, Tmin = 25 C), (c) transient von Mises stress at the mid-length of the treated zone (rmax = 360 MPa, rmin = 3 MPa), (d)
von Mises residual stress (rmax = 393 MPa, rmin = 0.5 MPa), (e) nal von Mises stress (rmax = 382 MPa, rmin = 0.6 MPa), (f) nal displacement magnitudes (dmin = 0 mm,
dmax = 0.522 mm, scale factor 10).
that the latter local mesh renement does not necessarily contribute to the accuracy of the thermal analysis results, it will contribute to the faster solution and convergence of the mechanical
model carried out in phase III. As long as thermal results convergence has been attained for specic mesh density, the thermal
model is re-executed by reducing the time step in order to conrm that changes of the time step do not signicantly affect the
temperature results. This procedure may require a few more
loops to complete in order to provide a better estimation of
the maximum temperatures reached in the heat affected zones
and to calculate the cooling rates in the transformation areas,
necessary for the post thermal-stress analysis (phase III). At this
stage of phase II, further remeshing or time step reduction are
not expected to strongly affect the results, thus the nal thermal
results, such as maximum temperatures and cooling rates, can
be obtained. Notice that mesh renement and step time reduction can be performed simultaneously for the case of analysts
297
was equal to 30 C (or 10% of DTtr) which is also within the desirable range. From the results depicted in Fig. 7 it is also deduced
that the maximum temperature calculated by the software
(Tmax = 1390 C) was not above the melting point of the material
which is equal to 1537 C [42]. Thus no melted zone was created.
If the actual treatment had resulted in the formation of a melted
zone then the thermal model developed during this phase should
have been further adapted with respect to the experimental or
actual results (if any), as indicated in Fig. 6 in order to compensate
for the unknown parameters that are involved in the numerical
analysis. The duration of the analysis of the nal converged
thermal model (single run) was 3.5 CPU-hours2 whereas its size
was equal to 3 GB.
2
CPU Intel Core i3 M350 @ 2.27 GHz (dual core, dual thread), RAM 4 GB, HDD,
Windows 64 bit OS.
298
Fig. 7. The nal mesh density of the thermal model as obtained at the end of phase II (NE = 20,080 elements, Dt = 0.037 s). (a) Transient temperature at the mid-length of the
treated zone (Tmax = 1390 C, Tmin = 25 C), (b) temperature distribution during cooling (Tmax = 39 C, Tmin = 25 C).
299
Fig. 9. The nal mesh density of the mechanical model as obtained at the end of phase III (NE = 42,773 elements, Dt = 0.025 s). (a) Transient temperature at the mid-length of
the treated zone (Tmax = 1516 C, Tmin = 25 C), (b) temperature distribution during cooling (Tmax = 39 C, Tmin = 25 C), (c) transient von Mises stress at the mid-length of the
treated zone (rmax = 385 MPa, rmin = 0 MPa), (d) von Mises residual stress (rmax = 423 MPa, rmin = 0.2 MPa), (e) nal von Mises stress (rmax = 407 MPa, rmin = 0.4 MPa), (f)
nal displacement magnitudes (dmin = 0 mm, dmax = 0.55 mm, scale factor 10).
300
[7]
[8]
[9]
[10]
[11]
[12]
[13]
[30]
[31]
301