SYSTEM AND METHOD FOR DYNAMICALLY LOCATING A FAULT
OBSERVED ON A COMPONENT
DESCRIPTION
5 TECHNICAL FIELD
The present invention relates to the general field of forging and concerns the
dynamic locating of a fault found in a defective component resulting from the forging
operation. It finds application in all industrial fields and in particular in the field of
aeronautics in which forged components are subjected to maximum constraints of quality
10 and safety.
STATE OF THE PRIOR ART
At the time of shaping forged components, fluctuation in production may lead to
the forming of defects potentially causing the rejection of these components. The
detection of defects in forged components can be performed using inspection means of
ultrasound, magnetic detection or visual type.
The cause of a defect may be related to forging parameters or to faulty
configuration of forming tools. It is therefore important to identify or locate the origin of
a defect to improve the forging operation or tooling.
At the current time, to identify the origin of a defect in a forged component, the
defect is approximately related to the original part before forging. This relating of defects
is imprecise and is performed solely for defects present on the surface and in regions
having easy visual identification.
In addition, with this type of comparison it is not possible to locate a defect
25 generated during an intermediate forging step and it is not possible to analyse
propagation of the defect, which is detrimental to good diagnosis of defects.
The objective of the present invention is therefore to propose a system and method
for the dynamic locating of a defect or fault observed in a defective component, which
overcome the aforementioned drawbacks and provide prospective or retrospective
30 Itnowledge of the propagation of the defect.
DESCRIPTION OF THE INVENTION
The present invention is defined by a system for the dynamic locating of a fault
observed in a defective component related to the forging operation, comprising:
- processing means to model a component shaping operation by forging, using a set
of successive models of the said component;
-processing means to add a fault plotter to a first model of said set of models in a
zone corresponding to the region of the fault of the said defective component to obtain a
first plotted model; and
-processing means to track the said fault plotter over time during the said
modelling from the said first plotted model to locate the origin of the said fault.
It is thus possible prospectively or retrospectively to diagnose the propagation of
the fault in the medium of the component.
Advantageously, the dimensions and position of the zone associated with the said
fault plotter in the first model are substantially similar to the dimensions and position of
the fault region in the defective component.
According to one particular embodiment of the present invention, the modelling of
the shaping operation is dynamic modelling using finite elements forming at each
modelling time step a polygonal mesh representing the component at the corresponding
forging step.
Advantageously, the processing means are configured to define at each modelling
time step the dimensions and position of the said fault plotter as a function of the
elementary elements of the said mesh at the said step in time.
According to a first embodiment of the present invention, the said set of successive
models comprises an initial model corresponding to the component before forging,
intermediate models corresponding to intermediate forging steps, and a final model
corresponding to the forged component, the said first model corresponding to the said
final model and the first plotted model corresponding to a final plotted model, the
processing means being configured to track the said fault plotter in time by reversing the
sequence of the said modelling starting from the said final plotted model.
The processing means are configured to locate the said fault plotter in the initial
1 model to identify the region in which the fault would have been in the component before
forging.
Advantageously, the processing means are configured to locate the said fault
I 5 plotter in intermediate models comprising particular configurations to verify whether the
said particular configurations are likely to induce the said fault.
According to a second embodiment of the present invention, the said set of
successive models comprises an initial model corresponding to the component before
forging and a final model corresponding to the forged component, the said first model
10 corresponding to the said initial model and the said first plotted model corresponding to
an initial plotted model, the processing means being configured to locate the said fault
plotter in the final model to verify whether the fault in the forged component lies outside
a machining region of the said forged component.
Advantageously, the said fault plotter is a contrast element associated with the said
15 polygonal mesh.
The invention also concerns a method for the dynamic locating of a fault observed
in a defective component related to a forging operation, comprising the following steps:
- modelling an operation to shape a component by forging as per a set of successive
models of the said component;
2 0 -adding a fault plotter, to a first model of said set of models, in a zone
corresponding to the region of the fault of the said defective component, to obtain a first
plotted model; and
-tracking the said fault plotter over time during the said modelling starting from the
said first plotted model to locate the origin of the said fault.
2 5
BRIEF DESCRIPTION OF THE DRAWINGS
A description will now be given of non-limiting examples of embodiments of the
invention with reference to the appended Figures in which:
Fig. 1 schematically illustrates a forging process;
Fig. 2 schematically illustrates a system 12 for the dynamic locating of a fault
observed in a defective component 1, according to the invention;
Fig. 3 schematically illustrates a set of successive models of the shaping of the
component, according to the invention;
5 Figs. 4 and 4A-4D illustrate a method for dynamically locating a fault observed in a
defective component, according to one preferred embodiment of the invention;
Figs. 5A-5E illustrate modelling of the forging of a blank offset from the tooling; and
Figs. 6 and 6A-6C illustrate a method for the dynamic locating of a fault observed in
a defective component, according to another embodiment of the invention.
DETAILED DESCRIPTION OF PREFERRED EMBODIMENTS
The concept at the base of the invention is the use of modelling of component
shaping to track propagation of the fault.
Fig. 1 schematically illustrates a forging process.
15 The components to be forged which are generally of cylindrical shape called ingots
or blanks l a are hot or cold worked between two dies or forging tools 3 by hammering or
pressure using a forging press, to form forged components. The final component i f often
resembles a large disc or dish of particular geometry.
For hot working, the component 1 is placed in a furnace 7 before the flattening step
20 and in some cases the forging and flattening steps are repeated several times before
obtaining the final component i f .
On completion of forging, inspection is carried out using ultrasound, magnetic
detection or visual means 9 to verify that a forged component i f does not comprise any
faults before starting a machining operation for example on this component i f .
25 The present invention proposes locating the fault dynamically to identify the origin
thereof. This makes it possible for example to determine whether the fault is isolated or
reproducible on a series of components and in this case corrective action can be
performed to avoid the reproduction of this fault.
Fig. 2 schematically illustrates a system 12 for the dynamic locating of a fault
30 observed in a defective component 1 according to the invention.
By dynamic locating of a fault is meant the time-space identification of the fault in
relation to the space taken up by the component 1 and at successive times of its shaping.
In other words, it is the spatial locating of the fault in the component 1 at every moment
of its transformation.
5 The locating system 12 comprises data input means 13, processing means 15,
storage means 17 and output means 19 comprising viewing means 20. The processing
means 15 allow the execution of one or more computer programmes comprising
programme code instructions stored in the storage means 17 and designed to implement
the method for dynamically locating a fault.
10 More particularly, the processing means 15 are configured to model the shaping
operation of a component 1 by forging in accordance with a set of successive models of
the component.
Fig. 3 schematically illustrates a set of successive models 21a-21t of the shaping of
the component 1 comprising an initial model 21a representing the blank l a (i.e. the
15 component before forging), intermediate models 21b, 21c representing intermediate
forged components lb, lc, and a final model 21f representing the forged component i f
(i.e. the final component after forging).
It will be noted that modelling can be conducted in three dimensions (3D) or
optionally in two dimensions (2D) for axisymmetric components.
2 0 Therefore, the processing means 15 are used for the numerical solving of equations
modelling the working of the component 1 under the action of forging tools 3 in relation
to forging parameters e.g. comprising a temperature range of the component 1 and of
the tooling 3, a pressure range, heat transfer coefficients, the density of the component
1, a range of working speed, etc. These parameters allow numerical modelling to best
25 represent actual forging operations in the workshop. For example, the heat transfer
coefficients allow heat dissipated by the component 1 into the surrounding medium
during the forging operation due to radiation and/or convection to be taken into account,
and in particular when the temperature of the component 1 is high (e.g. in the order of
1000°C).
Numerical solving is conducted iteratively and for example uses meshes 23 to
discretise the continuous geometric domain of the component using finite elements
described by vertices or nodes 25 and edges 27. Therefore dynamic modelling via finite
elements, at each iteration or modelling step in time, forms a polygon mesh 23 (e.g.
5 triangular) which represents the component 1 at the corresponding forging step.
When a fault 33 is observed in a defective component it, data is recorded relating
to the fault 33 (e.g, the dimensions and position of the fault in the component i t ) and
these data are imported into the dynamic locating system 12.
The input means 13 of the dynamic locating system 12 are used to input data
10 relating to the fault 33 and enable the processing means 15 to insert an equivalent of the
fault 33 into the model 21t corresponding to the defective component it.
More specifically the processing means 15 are configured so that, to a first model
2 l t belonging to the set of models 21a-21t of the forging operation, they add a fault
plotter 43 in a zone corresponding to the region of the fault in the defective component
15 i t to obtain a first plotted model 21t. In other words, the first plotted model 21t
represents the defective component i t at the time the fault is detected.
Advantageously, the zone associated with the fault plotter 43 in the first model 21t
has substantially similar dimensions and positioning to the dimensions and position of the
fault region 33 in the defective component it.
2 0 The processing means 15 together with the viewing means 20 then allow the
tracking in time of the fault plotter 43 during modelling starting from the first plotted
model 21t to diagnose the dynamics of the fault 33. In this manner the propagation of the
fault 33 in the component 1 can be tracked prospectively (i.e. moving forward in time) or
retrospectively (i.e. going back in time) starting from the first plotted model 21t.
25 In particular, at each modelling time step the dimensions and position of the fault
plotter 43 can be defined as a function of the elementary elements (i.e. nodes 25 and/or
edges 27) of the mesh 23 at the current time step.
For example, the fault plotter 43 is a contrast element which represents the fault 33
in terms of dimension and position and which can be integrated into the corresponding
30 polygon mesh 23 of the model 21 using a known technique of CAD type. For example, the
fault plotter 43 can be represented by a coloured contour contrasting with the mesh 23,
enclosing a surface substantially equal to the surface of the actual fault 33, and defined in
relation to nodes 25 in the vicinity of the fault. It will be noted that it is not necessary to
have knowledge either of the type or of the precise shape of the fault 33.
5 Figs. 4 and 4A-4D illustrate a method for dynamically locating a fault detected in a
defective component, according to one preferred embodiment of the invention.
According to this embodiment, the first model to which the fault plotter is added
corresponds to the final model 21f representing the forged component if, so that the first
plotted model 21t corresponds to a final plotted model 21tf representing a defective
10 forged component l t f .
At step E l the processing means 15 model the shaping operation of a component
by forging 1 in accordance with successive models 21a-21f comprising an initial model
21a corresponding to the component before forging (blank) l a and a final model 21f
corresponding to the forged component i f .
15 At step E2, after a fault has been detected in a defective forged component (see Fig.
4A), the processing means 15 add a fault plotter 43 to the final model 2lf in a zone
corresponding to the fault region of the defective forged component i f to obtain a final
plotted model 21tf (see Fig. 48).
Fig. 4A is an example illustrating part or more specifically one half of a defective
20 forged component l t f having a forging lap 33 (e.g. a small crack) seen in a micrographic
cross-section of the component 1. In addition, Fig. 48 gives a 2D illustration of- the
corresponding part of the final plotted model 21tf integrating the fault plotter 43 in a
zone 44 corresponding to the fault region 34 of the defective forged component l t f in Fig.
4A.
2 5 At step E3, the processing means 15 together with the viewing means 20 allow the
tracking in time of the fault plotter 43 by reversing the kinetics (i.e. by reversing the
sequence of modelling) starting from the final plotted model 21tf to identify the origin of
the fault in the defective forged component l t f (see Figs. 4C and 4D). It is thus possible to
track the history of the region which suffered the defect.
When modelling the forging operation, the polygon mesh 23 of the model 21
representing the component 1 becomes deformed over time. In other words, the relative
positions of the nodes 25 are modified in relation to one another during the
transformation. This generates variations in the expanse and location of the fault plotter
5 43 which is defined in relation to neighbouring nodes 25 whose coordinates are known at
each modelling step. Therefore, by going back in time, it is possible to identify the
morphology, geometry and position of the fault 33 before flattening of the component 1.
It will be noted that the fault 33 may result from an original flaw in the material of
the blank la, or from a defect related to forging parameters (working speed,
10 temperature, etc.) or from a geometric defect of the component 1 and/or of the shaping
tools 3.
Fig. 4C illustrates the locating of the fault plotter 43 in the initial model 21a. This
allows identification of the region 44a in which the fault would have been if it had been
contained in the blank la.
15 To identify other sources of the fault, the dynamic locating system 12 is used to
locate the fault plotter 43 in intermediate models likely to induce the fault.
For example, Fig. 4D illustrates the locating of the fault plotter 43 in an intermediate
model 21b having a particular configuration.
This intermediate model 21b shows that the fault plotter 43 is positioned in the
20 vicinity of an area 44b of concavity normally generated by penetration of tooling 3 into
the material of the component 1 to be forged.
Therefore an abnormally marked concavity or inflection in this zone 44b may be the
cause of the lap or fault 33 in the defective component ltf. It is noted that accentuation
of the concavity may result for example from a parallelism defect between the dies or
25 forging tools 3 and/or from poor centring of the blank l a in relation to the centre of the
tooling 3.
It is therefore of interest to model the different scenarios which generate
accentuated concavity to diagnose or verify whether it is indeed the concavity which led
to this fault 33.
For example, Figs. 5A-5E illustrate the modelling of forging using a blank 21a that is
deliberately offset from the tooling 3.
Fig. 5A shows a blank 21a whose central axis A1 is offset by a few millimetres from
the axis of symmetry A2 of the tooling 3.
5 Fig. 5B clearly shows an increase in concavity in the zone 44b of original concavity in
the vicinity of which the fault has been located. This clearly evidences a high risk of a lap
formation in the concavity zone as illustrated in Figs 5C-5E. In particular, Fig. 5E shows
that the final model 21f of the forged component is highly dissymmetric.
It is therefore possible to modify the configuration or geometry of the forging tools
10 3 so that the penetration of the two tool cores 3 into the component 1 no longer
generates any concavity or that the flow of material occurs in different manner so that
the concavity is reduced or eliminated.
In general, the locating of the origin of the fault 33 according to the invention allows
the modelling of different effects or scenarios which may lead to the cause of the fault so
15 that it can be remedied.
Figs. 6 and 6A-6C illustrate a method for the dynamic locating of a fault detected in
a defective component according to another embodiment of the invention.
According to this embodiment, the first model to which the fault plotter 44 is added
corresponds to the initial model 21a representing the component before forging (the
20 blank) l a so that the first plotted model corresponds to an initial plotted model 21ta
representing a defective blank lta.
At step E l l , the processing means 15 model the shaping operation of a component
by forging according to successive models comprising an initial model 2la corresponding
to the component l a before forging and a final model 21f corresponding to the forged
25 component i f .
At step E12, after a fault 33 has been detected in a defective blank 1ta (see Fig. 6A),
the processing means 15 add a fault plotter 44 to the initial model 21a in a zone
corresponding to the region of the fault 33 in the defective blank lta, to obtain an initial
plotted model 21ta (see Fig. 68).
Fig. 6A is an example illustrating a defective blank l t a having a small fault 33 on its
surface and Fig. 6B gives a 20 illustration of the initial plotted model 21ta integrating the
fault plotter 44 in a zone corresponding to the region of the fault 33 in the defective blank
lta in Fig. 6A.
5 At step E13, the processing means 15 together with the viewing means 20 allow the
locating of the fault plotter 44 in the final model 2lf to verify whether the fault in the
forged component lies outside a machining area 46 of this forged component. This allows
economy of components by checking whether or not the forged component after
machining will be impacted by the initial fault 33 of the component lta before forging.
10 Fig. 6C illustrates the locating of the fault plotter 44 in the final model and the
machining contour 46. This example shows that the fault does not affect the finished
component after machining and therefore the starting blank can be used.
CLAIMS
1. A system for the dynamic locating of a fault (33) observed in a component
obtained after a forging operation of a blank, characterized in that it comprises:
5 - processing means (15) adapted for the modelling of said forging operation via the
numerical solving of equations modelling working of the component (1) during the said
forging operation as a function of forging parameters, the said solving being performed
iteratively and leading to a set of time successive models (21a-21f) of the forming of said
component (1) during the said forging operation, the said set comprising an initial model
10 (21a) corresponding to the said blank (la), intermediate models (21b-21e) corresponding
to intermediate forging steps and a final model (21f) corresponding to the said forged
component (1);
input means to provide the said processing means (15) with data relating to a fault
(33) observed in said component (I), a component having a fault being called a defective
15 component (It);
-processing means (15) adapted to add a fault plotter (43) to a first model
belonging to the said set of models (21a-21f), in a zone corresponding to the region of
the fault in the said defective component (It) to obtain a first plotted model (21t)
representing the defective component (It) at the time of detection of the fault; and
2 0 -viewing means (20) linked with the processing means (15) and adapted to track
the said fault plotter (43) over time during the said modelling starting from the said first
plotted model (Zlt), retrospectively or prospectively, to diagnose the dynamics of the
said fault (33).
2 5 2. The system according to claim 1, characterized in that the dimensions and
positioning of the zone associated with the said fault plotter (43) in the first plotted
model (21t) are substantially similar to the dimensions and positioning of the region of
the fault (33) in the defective component (It).
3. The system according to claim 1 or 2, characterized in that the modelling of the
said forging operation is dynamic modelling with finite elements forming a polygon mesh
(23) at each iteration representing the component at the corresponding forging step.
5 4. The system according to claim 3, characterized in that the processing means (15)
are configured to define at each iteration the dimensions and position of the said fault
plotter (43) as a function of the elementary elements of the said polygon mesh (23) at the
said step in time.
10 5. The system according to any of claims 1 to 4, characterized in that the said first
model corresponds to the said final model (21f) and the first plotted model (21t)
corresponds to a final plotted model (21tf), and in that the processing means (15) are
configured to track in time the said fault plotter (43) by reversing the sequence of the said
modelling starting from the said final plotted model (21tf).
6. The system according to claim 5, characterized in that the processing means (15)
are configured to locate the said fault plotter (43) in the initial model (21a) to identify the
region in which the fault would have been in the component before forging.
2 0 7. The system according to claim 5 or 6, characterized in that the processing means
(15) are configured to locate the said fault plotter (43) in intermediate models comprising
particular configurations to verify whether the said particular configurations are likely to
cause the said fault.
2 5 8. The system according to any of claims 1 to 4, characterized in that the said first
model corresponds to the said initial model (2la) and the said first plotted model (21t)
corresponds to an initial plotted model (21ta), and in that the processing means (15) are
configured to locate the said fault plotter (43) in the final model (21f) to verify whether
the fault in the forged component lies outside a machining region (46) of the said forged
30 component.
9. The system according to any of claims 3 to 8, characterized in that the said fault
plotter (43) is a contrast element associated with the said polygon mesh (23).
5 10. A method for dynamically locating a fault (33) observed in a component
obtained after the forging operation of a blank, characterized in that it comprises the
following steps:
-modelling the said forging operation by the numerical solving of equations
modelling the working of the component (1) during the said forging operation as a
10 function of forging parameters, the said solving being performed iteratively and leading
I
i
to a set of time successive models (21a-21f) of the forming of the said component (1)
I
during the forging operation, the said set comprising an initial model @la) corresponding
to the said blank (la), intermediate models (21b-21e) corresponding to intermediate , forging steps and a final model (21f) corresponding to the said forged component (1);
1 15 . -obtaining data relating to a fault (33) observed in said component (I), a
I component having a fault being called a defective component (It);
-adding a fault plotter (43) to the first model belonging to said set of models, in a
zone corresponding to the region of the fault in the said defective component to obtain a
first plotted model (It) representing the defective component (It) at the time of
20 detection of the fault; and
-tracking the said feult plotter (43) in time during the said modelling starting from
the said first plotted model, retrospectively or prospectively, in order to diagnose the
dynamics of the said fault.