Inertial Friction Welding Simulation

Abstract

The invention relates to a method and system for simulating inertial friction welding of parts to be welded, comprising: - an acquisition module (3) configured to acquire an input configuration (9) consisting of a set of physical parameters relating to the parts to be welded and the welding process, - a machine learning module (5) comprising a predictive neural network (51) configured to establish a friction curve (11) representative of the weld between said parts, taking as input said acquired input configuration, - a simulation module (7) configured to simulate the evolving behavior of the welding operation using the friction curve (11) associated with said input configuration (9). Figure for the abstract: Fig. 2.

Description

Title of the invention: SIMULATION OF INERTIAL FRICTION WELDING Technical field

[0001] The present invention generally relates to a method and a system for simulating inertial friction welding and in particular to its modeling using a friction law determined by machine learning. STATE OF THE PRIOR ART

[0002] The reliability of welds is a major concern in several industrial fields. In particular, inertial friction welding has very good mechanical characteristics, avoiding the formation of cracks, porosity or any reduction in strength in the weld zone.

[0003] The evolutionary behavior of the friction welding operation depends on several physical parameters and therefore, it is advantageous to model the welding process in advance in order to choose the physical parameters generating the most optimal and reliable welding process.

[0004] Currently, models of the friction welding process are mainly carried out in descriptive mode. This means that it is necessary to obtain input data of the weld to be modeled before carrying out the numerical simulation. In other words, it is currently not possible to simulate unrealized welds.

[0005] Initial tests to predict the results of inertial friction welding have been carried out. These tests make it possible to predict an inertial friction weld with a welding configuration consisting of a set of well-defined input parameters. However, the results become incorrect as soon as one of the input parameters is modified. Thus, it is necessary to identify a friction law for each welding configuration encountered. This identification is based on a trial-and-error method by changing the pressure in stages until success or until the tests are stopped. This identification procedure is not very precise while being very long and repetitive.

[0006] The object of the present invention is therefore to remedy the aforementioned drawbacks by proposing a method and a system for predictive inertial friction welding simulation making it possible to anticipate the welding results on each configuration while being precise, efficient and rapid. Statement of the invention

[0007] The present invention is defined by a welding simulation system by inertial friction of parts to be welded, comprising:

[0008] - an acquisition module configured to acquire a formed input configuration of a set of physical parameters relating to the parts to be welded and the welding process,

[0009] - a machine learning module comprising a predictive neural network configured to establish a friction curve representative of the weld between said parts by taking said acquired input configuration as input,

[0010] - a simulation module configured to simulate the evolutionary behavior of the welding operation using the friction curve associated with said input configuration.

[0011] This makes it possible to predict the friction law most representative of the evolution of the welding parameters and consequently, to simulate the welding procedure in a predictive manner in order to anticipate the results on each configuration while being precise, efficient and rapid.

[0012] Advantageously, the simulation module is configured to determine the evolutionary behavior of the welding operation by simulating the evolution of the temperature, and / or the deformation and / or the stress in the parts to be welded.

[0013] Advantageously, the physical parameters relating to the parts include parameters relating to the materials of the parts to be welded, to the thicknesses of the parts to be welded, and to the average diameters of the parts to be welded. The physical parameters relating to the welding process are point values ​​of surface welding energy, torque of the parts to be welded, rotation speed, and pressure.

[0014] Each input configuration is selected from a set of input parameters, thus making it possible to predict the most optimal weld. Apart from the material and geometry of the parts, all the physical quantities used as input to the learning module come from current acquisitions during welding.

[0015] Advantageously, the predictive neural network is configured to take as input a point abscissa z of a point considered on the friction curve to be predicted and to give as output the friction coefficient H corresponding to said point abscissa z used as input, the point abscissa z being a dimensionless number determined by the values ​​of said input physical parameters.

[0016] This makes it possible to obtain by iterative prediction the friction curve p=f(z) most representative of the weld.

[0017] Advantageously, the predictive neural network comprises a succession of layers fully connected to each other, the neurons having the same logistic activation function of the sigmoid function type.

[0018] Advantageously, the system comprises a learning database configured to train the neural network, the learning database comprising a set of uniformly weighted learning friction curves, the learning friction curves being preprocessed experimental friction curves.

[0019] Advantageously, the system comprises a preprocessing module comprising preprocessing neural networks having experimental friction curves as input and learning friction curves as output, the preprocessing neural networks being configured to carry out the following steps:

[0020] - smoothing and resampling of each experimental curve to form smoothed and resampled experimental curves, and

[0021] - extrapolation of said smoothed and resampled experimental curves for form the learning friction curves with identical abscissa ranges.

[0022] Advantageously, the simulation module is configured to use each friction curve as input to simulate the evolution during welding of the temperature of the parts, and / or of the deformation of the parts and / or of the stress on the parts, for each input configuration associated with said each friction curve.

[0023] Thus, the simulation becomes predictive and frees itself from the physical realization of the weld. Indeed, the simulation uses the curve predicted by the neural diagram and no longer requires welding a configuration to obtain the friction curve by inverse identification.

[0024] Advantageously, the simulation module is configured to carry out the simulation using a finite element calculation technique.

[0025] The invention also relates to a method for simulating inertial friction welding of parts to be welded, comprising the following steps implemented by computer:

[0026] - acquire an input configuration formed from a set of physical parameters relating to the parts to be welded and the welding process,

[0027] - establish by a predictive neural network a friction curve representative of the welding between said parts for said input configuration,

[0028] - simulate the evolutionary behavior of the welding operation using the curve of friction associated with said input configuration.

[0029] Advantageously, the simulation of the evolving behavior of the welding operation comprises the simulation of the evolution of the temperature, and / or of the deformation and / or of the stress in the parts to be welded.

[0030] Advantageously, the predictive neural network is previously trained using a mini-batch gradient descent type optimization technique, using a learning database comprising a set of learning friction curves corresponding to pre-processed experimental friction curves.

[0031] The invention also relates to a computer program product comprising ins- code instructions for implementing the steps of an inertial friction welding simulation method according to any of the above characteristics, when said code instructions are executed by a processing unit. Brief description of the drawings

[0032] Other characteristics and advantages of the invention will appear on reading preferred embodiments of the invention made with reference to the attached figures among which:

[0033] [Fig.l] schematically illustrates an inertial friction welding process between two parts to be welded;

[0034] [Fig.2] schematically illustrates a system for simulating inertial friction welding of parts to be welded, according to a preferred embodiment of the invention;

[0035] [Fig.3] is a graph schematically illustrating a friction curve determined by the machine learning module, according to the invention;

[0036] [Fig.4] schematically illustrates a simulation system comprising a training device, according to a preferred embodiment of the invention;

[0037] [Fig.5] schematically illustrates the steps of a simulation method that can be implemented by a system according to the embodiment of [Fig.4];

[0038] The [Fig.6A],

[0039] [Fig.6B] and [Fig.6C] are graphs illustrating the steps performed on an experimental friction curve to obtain the corresponding learning friction curve, according to the invention; and

[0040] [Fig.7] is a graph illustrating several friction curves predicted by the machine learning module, according to the invention. DETAILED DESCRIPTION OF THE INVENTION

[0041] The principle of the invention consists of using a neural diagram in order to predict a friction curve representative of the weld to be simulated, without needing to actually carry out a friction weld.

[0042] [Fig.l] schematically illustrates an inertial friction welding process between two parts pl, p2 to be welded.

[0043] Inertial friction welding is a solid-state welding process consisting of rotating one of the parts before bringing them into contact under the effect of pressure. This process is described by time evolution curves of the speed V, the pressure P and the material consumption M.

[0044] More particularly, this welding process comprises three phases. The first phase (Phase 1) is a launch phase during which one of the parts pl is set into rotation R and then disengaged. After disengaging, the part pl continues to rotate by inertia with decreasing rotational speed R.

[0045] The second phase (Phase 2) is a welding phase which comprises two periods. During the first period, the two parts pl, p2 are brought into contact under the effect of a pressure force E which increases rapidly before stabilizing at a constant pressure. The rotation speed R continues to decrease freely. During this braking period the parts pl, p2 heat up locally by friction. During the second period, the parts pl, p2 deform under the effect of the heat and the pressure force. The assembly of two parts shortens due to the effect of an increase in material consumption. The rotation speed R continues to decrease.

[0046] The third phase (Phase 3) is a forging phase. This phase begins after the complete deceleration of the part pl, i.e. after its rotation speed has reached zero. During this third phase, there is no more material consumption and the parts pl, p2 cool while the pressure force E is kept constant or possibly increased.

[0047] [Fig. 2] schematically illustrates a system for simulating inertial friction welding of parts to be welded, according to a preferred embodiment of the invention. [Fig. 2] is also used to illustrate a method for simulating inertial friction welding, according to an embodiment of the invention.

[0048] This simulation system 1 comprises an acquisition module 3, a machine learning module 5, and a simulation module 7.

[0049] The acquisition module 3 is configured to acquire an input configuration 9 from a set of input configurations. Each input configuration 9 comprises a first set 91 of physical parameters relating to the parts to be welded and a second set 93 of physical parameters relating to the welding process.

[0050] The first set 91 corresponding to the physical input parameters relating to the parts, preferably comprises parameters relating to the materials, thicknesses, and average diameters of the parts to be welded. The second set 93 corresponding to the physical input parameters relating to the welding process, preferably comprises point values ​​from the welding process comprising magnitudes of surface welding energy, inertia or torque of the parts to be welded, rotation speed, and pressure force.

[0051] The automatic learning module 5 is configured to establish a friction law (or curve) 11 representative of the welding between the parts to be welded pl, p2 for each input configuration.

[0052] It will be noted that the friction law is a so-called Stribeck curve, which defines the evolution of the friction coefficient M as a function of a dimensionless variable noted z characterizing the parameters of the process studied. This law or curve is known to those skilled in the art.

[0053] More particularly, the automatic learning module 5 comprises a predictive neural network 51 having as input a point abscissa z of a point considered on the friction curve 11 to be predicted, and as output the friction coefficient M corresponding to said abscissa z used as input. This makes it possible to obtain by iterative prediction, either point by point, the friction curve p=f(z) characterizing the weld according to the input configuration.

[0054] Indeed, [Fig.3] is a graph schematically illustrating a friction curve determined by the automatic learning module, according to the invention.

[0055] The point abscissa z is a dimensionless number determined by the point values ​​of the physical input parameters. Thus, the friction law p=f(z) integrates the influence of all the physical input parameters relating to the parts to be welded pl, p2 and to the welding process. Apart from the material and geometry of the parts, all the physical quantities used as input to the learning module 5 come from machine acquisitions during the welding process.

[0056] For example, the dimensionless variable z is equal to the viscosity of the material multiplied by the linear velocity divided by the pressure.

[0057] Thus, the neural network 51 makes it possible to obtain a friction curve 11 for each weld integrating the influence of different parameters such as the material, the geometry of the parts and the point parameters of the welding process. The friction curve 11 is subsequently used as input to a numerical simulation to model the welding process.

[0058] The predictive neural network 51 comprises a succession of layers that are fully connected to each other. Each neuron of the previous layer is connected to all the neurons of the next layer. For example, the total number of layers is between 5 and 25 and the number of neurons per layer is between 20 and 50.

[0059] Advantageously, the neurons all have the same non-linear activation function such as the sigmoid-type logistic function.

[0060] The training of the predictive neural network 51 is carried out using a previously constructed learning database comprising a set of learning friction curves (see Figs. 4 and 5).

[0061] The simulation module 7 is configured to numerically simulate the evolving behavior of the welding operation using the friction curve 11 associated with each input configuration 9.

[0062] The simulation module 7 comprises a calculation unit 71, a memory 73 and an output unit 75 comprising for example a screen.

[0063] More particularly, the simulation module 7 preferably has as input data the rotation speed of the rotating part, the pressure on the rotating part, the inertia of the plates of the inertial friction machine, and the curve of friction 11 obtained via machine learning module 5.

[0064] The calculation unit 71 in association with the memory 73 uses the friction curve 11 as input to a finite element simulation to determine the evolution of the temperature, the deformation and the stress of the simulated welding for the input configuration 9 associated with the friction curve. This makes it possible to access areas of the part that are inaccessible experimentally. For example, the simulation makes it possible to model the rise in temperature at the interface of the parts to be welded p1, p2 as well as the deformation of the parts and in particular, the formation of the welding bead created by heating and pressure. This simulation can be displayed on the screen or printed by the output unit 75.

[0065] Numerical simulation by finite elements no longer requires the physical realization of a weld according to a given configuration. Indeed, it is no longer necessary to obtain a friction curve corresponding to a physical weld according to the given configuration to carry out a descriptive simulation by inverse identification. The simulation according to the invention does away with the physical realization of the weld. Indeed, it simply uses the curve 11 predicted by the neural network 51 and thus becomes predictive.

[0066] The method according to the invention is thus defined in two complementary steps. The first step involves the use of a neural diagram in order to predict the friction curve p=f(z) representative of the weld to be simulated. The objective of the first step is therefore to obtain a friction curve or law integrating the influence of different parameters associated with each input configuration of the weld.

[0067] The second step involves using the friction law predicted by the neural diagram as input to a numerical simulation, for example by finite elements, to predict the welding behavior.

[0068] [Fig.4] schematically illustrates a simulation system comprising a training device, according to a preferred embodiment of the invention.

[0069] The simulation system 1 comprises an acquisition module 3, a machine learning module 5, a simulation module 7, and a training device 13.

[0070] The training device 13 comprises a preprocessing module 15 configured to construct a learning database 17 intended to be used for training the automatic learning module 5.

[0071] The preprocessing module 13 comprises preprocessing neural networks 151 comprising a succession of layers whose number of neurons is for example between 20 and 50 per layer. The total number of layers can be between 5 and 25. The layers are preferably fully connected to each other. The neurons preferably have the same activation function which can be a sigmoid function.

[0072] It will be noted that as a variant, the training device 13 may not be integrated into the simulation system.

[0073] [Fig.5] schematically illustrates a simulation method that can be implemented by a system according to the embodiment of [Fig.4].

[0074] In step E1, raw experimental friction curves 110 are constructed from the physical welding processes.

[0075] In step E2, the preprocessing module 13 is configured to determine learning friction curves from the experimental friction curves.

[0076] More particularly, the preprocessing neural networks 151 perform preprocessing operations on the experimental friction curves 110 to provide the learning friction curves 111 as output. These preprocessing operations include smoothing, resampling and extrapolation operations of the experimental curves 110.

[0077] Figs. 6A-6C are graphs illustrating the steps performed on an experimental friction curve to obtain the corresponding learning friction curve, according to the invention.

[0078] Each graph represents the friction coefficient h as a function of the dimensionless variable z and more precisely of the logarithm of z.

[0079] In particular, [Fig.6A] illustrates the smoothing and resampling effects performed by the preprocessing neural networks 151 on an experimental raw curve. The star-shaped points 110a represent the raw curve while the single points 110b represent the smoothed and resampled curve according to a rough estimate in 50 points.

[0080] Fig. 6B illustrates the extrapolation of the experimental curve by extending the abscissa range over the low 120a and high 120b z values. This allows the extension of the abscissa ranges of the smoothed and resampled experimental curves, thus forming learning friction curves with extended abscissa ranges.

[0081] Indeed, the extension to low 120a and high 120b abscissas z is carried out so that the discretization of the points of each of the curves is on an identical z range for each of the curves of the learning database 17.

[0082] For example, for the extension at low 120a and high 120b abscissa z, the preprocessing neural networks 151 use at most 40% of the points located on either side of the ends of the smoothed curve sampled at 50 points in order to deduce the extension locally.

[0083] [Fig.6C] illustrates the points of the experimental curve and the corresponding learning friction curve 111. The points of the experimental curve are represented by stars 110a. According to this example, the learning friction curve 111 is the result of smoothing and resampling operations to 200 points after an extrapolation operation.

[0084] In total, according to this example, each learning friction curve 111 represents 200 data points. All these learning friction curves 111 defined on the same abscissa range z form a relevant learning database 17.

[0085] Advantageously, the learning friction curves 111 are uniformly weighted. Thus, the weightings of each point of a curve depend on the total number of points for the curve. For example, for a curve described by 50 points, each of the 50 points will have a weighting of 1 / 50 during training.

[0086] The training database thus comprises all the points of each of these uniformly weighted friction curves. It will be noted that the training database 17 can comprise several hundred training friction curves with different input configurations.

[0087] Advantageously, an inverse identification of the friction curve is carried out by the simulation module 7 for each of the welds in the learning database 17.

[0088] In step E3, the learning database 17 is used to carry out the training of the machine learning module 5.

[0089] Advantageously, a mini-batch gradient descent type optimization algorithm is used for training the predictive neural network 51 of the machine learning module 5. The weights of the neural network are updated progressively. During an entire learning cycle, all of the mini-batches are used. For example, the size of the mini-batches is between 32 and 128 points. During training, a common technique called “momentum” can be used each time the weights of the neural network are updated. The factor used when updating the weights is, for example, between 10% and 90%. An example of the “momentum” technique is described by P. Kim in MATLAB Deep Learning with Machine Learning, Neural Networks and Artificial Intelligence, 2017.

[0090] Advantageously, once the training has been carried out, blind tests are carried out in order to validate the relevance of the friction curve predicted during simulation calculations. The success criterion is the correct prediction of the consumption of the material at the end of the weld and of the temperature at the interface during the weld.

[0091] After training, the machine learning module 5 is ready to predict friction curves from different configurations.

[0092] Indeed, in step E4, for each input configuration 9 acquired by the acquisition module 3, the automatic learning module 5 uses this configuration input to establish a corresponding friction curve 11 representative of the weld.

[0093] More particularly, the automatic learning module 5 uses the input configuration comprising the physical parameters relating to the parts to be welded (materials, thicknesses, and average diameters) as well as the point values ​​resulting from the welding process (welding surface energy, inertia, rotation speed, and pressure) to determine the point abscissas z and to predict the friction curve p=f(z). This friction curve representative of the weld associated with the input configuration is predicted iteratively by the already trained predictive neural network.

[0094] [Fig.7] is a graph illustrating several friction curves predicted by the machine learning module, according to an example of the invention.

[0095] This example shows three curves 11a, 11b, 11e for three different inlet configurations 9, namely three different pressures, 48 ​​MPa, 68 MPa and 88 MPa respectively, all other physical parameters (geometry, materials, surface energy and speed) being almost constant between these three curves.

[0096] For each configuration 9 of a geometry, a material pair and / or a welding parameter, the automatic learning module 5 provides a friction curve 11.

[0097] Then, in step E5, the friction curve 11 representative of the weld of interest is used to supply the simulation module 7.

[0098] Having as input data the rotation speed of the rotating part, the pressure on the rotating part, the inertia of the plates of the inertial friction machine, and the friction curve 11, the simulation module 7 determines the evolutionary behavior of the welding operation.

[0099] More particularly, the simulation module 7 preferably has as input data the rotation speed of the rotating part, the pressure on the rotating part, the inertia of the plates of the inertial friction machine, and the friction curve 11 obtained via the automatic learning module 5.

[0100] The calculation unit 71 of the simulation module 7 uses finite element calculation to determine the evolutionary behavior of the welding operation by preferably simulating the evolution of the temperature, the deformation and the stress in the modeled parts.

Claims

Claims

1. System for simulating inertial friction welding of parts to be welded, characterized in that it comprises: - an acquisition module (3) configured to acquire an input configuration (9) formed from a set of physical parameters relating to the parts to be welded and to the welding process, - a machine learning module (5) comprising a predictive neural network (51) configured to establish a friction curve (11) representative of the weld between said parts by taking as input said acquired input configuration, - a simulation module (7) configured to simulate the evolving behavior of the welding operation by using said friction curve (11) associated with said input configuration (9).

2. System according to claim 1, characterized in that the simulation module (7) is configured to determine the evolutionary behavior of the welding operation by simulating the evolution of the temperature, the deformation and the stress in the parts to be welded.

3. System according to any one of the preceding claims, characterized in that the predictive neural network (51) is configured to take as input a point abscissa z of a considered point of the friction curve (11) to be predicted, and to give as output the friction coefficient h corresponding to said point abscissa z used as input, the point abscissa z being a dimensionless number determined by the values ​​of said physical input parameters.

4. System according to any one of the preceding claims, characterized in that the predictive neural network (51) comprises a succession of layers fully connected to each other, the neurons of said predictive neural network (51) having the same logistic activation function of the sigmoid function type.

5. System according to any one of the preceding claims, characterized in that it comprises a training database (17) configured to train the predictive neural network (51), the training database (17) comprising a set of uniformly weighted training friction curves (111), the training friction curves being pre-processed experimental friction curves.

6. System according to claim 5, characterized in that it comprises a preprocessing module (15) comprising preprocessing neural networks (151) having experimental friction curves (110) as input and learning friction curves (111) as output, the preprocessing neural networks being configured to carry out the following steps: - smoothing and resampling of each experimental curve to form smoothed and resampled experimental curves, and - extrapolation of said smoothed and resampled experimental curves to form the learning friction curves (111) with identical abscissa ranges.

7. System according to any one of the preceding claims, characterized in that the simulation module (7) is configured to use each friction curve (11) as input to simulate the evolution during welding of the temperature of the parts, and / or the evolution of the deformation of the parts, and / or the evolution of the stress on the parts, for each input configuration (9) associated with said each friction curve (11).

8. Method for simulating inertial friction welding of parts to be welded, characterized in that it comprises the following steps implemented by computer: - acquiring an input configuration (9) formed from a set of physical parameters relating to the parts to be welded and to the welding process, - establishing by a predictive neural network (51) a friction curve (11) representative of the weld between said parts for said input configuration, - simulating the evolving behavior of the welding operation using the friction curve associated with said input configuration, the method preferably being implemented by a simulation system according to any one of claims 1 to 7.

9. Method according to claim 8, characterized in that the predictive neural network (51) is previously trained according to an optimization technique of the mini-batch gradient descent type, using a training database (17) comprising a set of training friction curves (111) corresponding to pre-processed experimental friction curves.

10. Computer program product comprising code instructions for implementing the steps of a welding simulation method by inertial friction according to any one of claims 8 or 9, when said code instructions are executed by a processing unit.

Images