下面是壹個簡單的例子，妳可以先練習試試看。
1.3.18 Inertia welding simulation using Abaqus/Standard and Abaqus/CAE
Products: Abaqus/Standard Abaqus/CAE
Objectives
This example demonstrates the following Abaqus features:thermal-mechanical coupling for inertia welding simulation,semi-automatic remeshing using Python scripting and output database scripting methods for extracting deformed configurations,defining a complex friction law in a user subroutine,flywheel loading through user subroutine definitions, andcombining and presenting results from a sequence of output database (.odb) files.
Application description
This example examines the inertia friction welding process of the pipes shown in Figure 1.3.18–1. The specific arrangement considered is the resulting as-welded configuration shown in Figure 1.3.18–2.
In this weld process kinetic energy is converted rapidly to thermal
energy at a frictional interface. The resulting rapid rise in interface
temperature is exploited to produce high-quality welds. In this example
the weld process is simulated, and the initial temperature rise and
material plastic flow are observed. An important factor in the process
design is control of the initial speed of the flywheel so that, when the
flywheel stops, the temperature rises to just below the melting point,
which in turn results in significant flow of material in the region of
the weld joint. Understanding the friction, material properties, and
heat transfer environment are important design aspects in an effective
inertia welding process; therefore, simulation is a helpful tool in the
process design.Geometry
The weld process in this example is shown in Figure 1.3.18–1,
where two pipes are positioned for girth-weld joining. The two pipes
are identical, each with a length of 21.0 mm, an inside radius of
42.0 mm, and an outside radius of 48.0 mm. The pipes are adjacent,
touching each other initially at the intended weld interface.Materials
The pipes are made of Astroloy, a high-strength alloy used in gas turbine components. Figure 1.3.18–3
shows flow stress curves as a function of temperature and plastic
strain rate. At temperatures relevant to the welding process, the
material is highly sensitive to plastic strain rate and temperature.
Specific heat is a function of temperature, as shown in Figure 1.3.18–4.Other material properties are defined as follows:Young's modulus:180,000 MPaPoisson's ratio:0.3Density:7.8 × 10–9 Mg/mm3Conductivity:14.7 W/m/C at 20C 28 W/m/C at 1200C
Initial conditions
The pipes are initially set at 20°C, representing room temperature. Boundary conditions and loading
A
pressure of 360 MPa is applied to the top surface of the upper pipe.
The initial rotational velocity of the flywheel is set at 48.17 rad/s,
or 7.7 revolutions per second. The mass moment of inertia of the
flywheel is 102,000 Mg mm2. Interactions
The
principal interaction occurs at the weld interface between the pipes;
however, a secondary concern is the possibility of contact of weld flash
with the side of the pipes. The weld-interface friction behavior is
assumed to follow that described by Moal and Massoni (1995), where the
ratio of shear stress to the prescribed pressure is observed to be a
complex function of interface slip rate. The heat generation from the
frictional sliding, combined with plastic deformation, contributes to
the temperature rise in the pipes.
Abaqus modeling approaches and simulation techniques
Abaqus/CAE
and Abaqus/Standard are used together to affect the weld simulation in a
way that permits extreme deformation of the pipes in the weld region.
This process is automated through the use of Python scripts. Three cases
are studied in this example.Summary of analysis casesCase 1Initial flywheel velocity = 48.17 rad/s. This case produces a successful weld.Case 2Initial
flywheel velocity = 20.0 rad/s. This case illustrates an unsuccessful
weld scenario; the flywheel has insufficient energy to begin the weld
process.Case 3Initial flywheel velocity =
70.0 rad/s. This case illustrates an unsuccessful weld scenario; the
flywheel has excessive energy, resulting in a temperature rise into the
liquidus regime of the pipe material.The
following sections discuss analysis considerations that are applicable
to all the cases. Python scripts that generate the model databases and
Abaqus/Standard input files are provided for Case 1, with instructions
in the scripts for executing the Case 2 and Case 3 simulations.Analysis types
The
analysis is nonlinear, quasi-static with thermal-mechanical coupling. A
fully coupled temperature-displacement procedure is used.Analysis techniques
The
key feature required for successful simulation of this process is
remeshing. In this example, because of the large deformation near the
weld region, multiple analyses are employed to limit element distortion.
These analyses are executed in sequence, with remeshing performed
between executions, and are automated through the use of Python scripts.
At each remesh point the current model configuration represents a
significant change in the pipes' shape and in the current analysis
mesh. Abaqus/CAE is used to extract the outer surface of the pipes,
reseed the surface, and remesh the pipe regions. This process employs
the Abaqus Scripting Interface PartFromOdb command, which is used to extract orphan mesh parts representing the deformed pipes. These parts are then passed to the Part2DGeomFrom2DMesh command. This command creates a geometric Part
object from the orphan mesh imported earlier. Once the profile of the
deformed part has been created, options in the Mesh module are used to
remesh the part. The new mesh results in a new Abaqus/Standard analysis,
and the map solution procedure maps state variables from the previous
analysis (see “Mesh-to-mesh solution mapping,” Section 12.4.1 of the Abaqus Analysis User's Manual).Mesh design
The
pipes are modeled as axisymmetric. The element formulation used is the
fully coupled temperature-displacement axisymmetric elements with twist
degrees of freedom (element types CGAX4HT and CGAX3HT), where the twist
degree of freedom enables modeling of rotation and shear deformation in
the out-of-plane direction. The hybrid formulation is required to handle
the incompressible nature of the material during the plastic flow. The
mesh is divided into two regions for each pipe. In the region near the
weld interface, smaller elements are created (see Figure 1.3.18–5).
During the remeshing process, the region near the weld surface is
recalculated so that the new flash region is also meshed with smaller
elements (see Figure 1.3.18–6).Material model
The
material model defined for this example approximates the
high-temperature behavior of Astroloy, where it is reported by Soucail
et al. (1992) using a Norton-Hoff constitutive law to describe the
temperature and strain-rate viscoplastic behavior. A similar model is
defined in Abaqus as a rate-dependent perfectly plastic material model.
For the loading in this model, these material parameters result in the
onset of local plastic flow only after the interface temperature has
exceeded roughly 1200C, near the material solidus temperature of
1250C. Above this temperature the Mises flow stress is highly sensitive
to variations in temperature and strain rate. A special adjustment in
the flow stress at high strain rates is necessary to avoid divergence
during the iteration procedure of the nonlinear solution. In the
material model definition an extreme case of stress data is defined when
the strain rate is 1.0 × 106 s–1. Stress data when the strain rate equals zero are also defined to be the same as the stress data at strain rate 1.0 × 10–5 s–1. As a result of large deformation, thermal expansion is not considered in the material model. It
is assumed that 90% of the inelastic deformation energy contributes to
the internal heat generation, which is the Abaqus default for inelastic
heat fraction.Initial conditions