FUSION WELDING OPTIMIZATION

Abstract

The invention relates to a method for determining an optimized welding direction (112) for a parting line (106), a welding beam (108) being intended to be moved to travel along an edge of the parting line (106) according to the optimized welding direction (112) while being maintained at a constant predefined angle with respect to the optimized welding direction (112), comprising: - for each of several potential welding directions (112) passing through a center of the parting line (106), a determination of a thickness of the parting line (106) as a function of a position on the welding direction (112), this thickness being taken along a straight line making the predefined angle with the welding direction (112) and passing through the position considered, and a calculation (304) of a weldability score from the thickness as a function of the position on the welding direction (112);and - a selection as the optimized welding direction (112), of the potential welding direction (112) with the best weldability score. Figure for the abstract: Fig. 1;

Description

Title of the invention: OPTIMIZATION OF FUSION WELDING Technical field of the invention

[0001] The present invention relates to a method for determining an optimized welding direction for a joint plane, a method for determining an optimized joint plane, as well as a corresponding computer program and system.

[0002] The invention acts as a technological solution dedicated to the industrialization of mechanically welded parts forming part of a mechanical system. It focuses on the concept of assembling metal parts by fusion welding process. Technological background

[0003] To date, mechanically welded parts are defined from mechanical strength and dimensioning studies carried out by design offices, with a significant modification of certain interfaces for assembly, in particular the joint planes. When a joint plane profile is a candidate for welding, tests on representative specimens and similar or close production means are carried out. From the measurements carried out on these tests, a welding range is established to define the optimal parameterization to be applied for the weld, and a partial or total redefinition of the joint plane is carried out. Once a joint plane has been selected, the main method for determining the welding direction to be used is trial and error on the specimen as well as feedback on the welding of parts that are similar from a topological point of view.

[0004] The current method of defining welding direction is limited by its manual and case-specific nature. The cost of testing representative welded specimens can be high and does not guarantee rapid convergence towards optimal joint planes. In addition, feedback can trap designers into historical profiles derived from obsolete or unjustified specifications and technological know-how.

[0005] It may thus be desirable to provide a method for determining an optimized welding direction in a joint plane which makes it possible to overcome at least some of the aforementioned problems and constraints. Summary of the invention

[0006] A method is therefore proposed for determining an optimized welding direction for a joint plane, a welding beam being intended to be moved to travel along an edge of the joint plane in the optimized welding direction while being maintained at a constant predefined angle with respect to the optimized welding direction, characterized in that it comprises: - for each of several potential welding directions: • a determination of a thickness of the joint plane as a function of a position in the welding direction, this thickness being taken along a straight line making the predefined angle with the welding direction and passing through the position considered, and • a calculation of a weldability score from the thickness as a function of the position in the welding direction; and - a selection as optimized welding direction, of the potential welding direction with the best weldability score.

[0007] The invention may further comprise one or more of the following optional features, in any technically possible combination.

[0008] Optionally, the calculation of the weldability score includes a calculation of an evaluation of a cumulative thickness to be welded and the weldability score is calculated from this evaluation of the cumulative thickness to be welded, to be better when the evaluation of the cumulative thickness to be welded decreases.

[0009] Also optionally, the thickness being determined for several discrete positions on the welding direction, the evaluation of the cumulative thickness to be welded is the sum of the squares of these discrete thicknesses, the weldability score being calculated from this sum of the squares.

[0010] Also optionally, the determination of the thickness as a function of the position in the welding direction comprises, for each of several discrete positions in the welding direction separated by a constant pitch, a counting of a number of points of a mesh superimposed on the joint plane, the points counted being those strictly included in a strip inclined relative to the welding direction, this strip being centered on the discrete position considered and of width such that the strips are contiguous, this number of points being taken as the thickness of the joint plane at the discrete position considered.

[0011] Also optionally, the potential welding directions pass through a center of the joint plane.

[0012] Also optionally, the center of the joint plane is a center of gravity of points of a mesh superimposed on the joint plane, these points being located in the joint plane.

[0013] A method for determining an optimized joint plane between two parts to be welded is also proposed, comprising: - a reception of a definition of an initial joint plane between the two parts; - starting from the initial joint plane, an iteration of the following steps: • a determination of an optimized welding direction in the plane joint, in accordance with a method according to the invention, the weldability score of the optimized welding direction found being associated with the joint plane, and • as long as the weldability score does not satisfy an exit condition of the iteration, a modification of the joint plane and a continuation of the iteration, otherwise the joint plane with the best weldability score is taken as the optimized joint plane.

[0014] Optionally, the definition of the initial joint plane comprises positions of pilot points and a geometric shape consecutively connecting the consecutive pilot points to form a closed curve, and the modification of the joint plane comprises a modification of the position of at least a portion of the pilot points.

[0015] There is also provided a computer program downloadable from a communication network and / or recorded on a computer-readable medium, characterized in that it comprises instructions for executing the steps of a method according to the invention, when said program is executed on a computer.

[0016] A method for welding two parts is also proposed, comprising: - determining an optimized welding direction for a joint plane between the two parts according to the invention, or determining an optimized joint plane between the two parts according to the invention; and - welding the two parts by moving the welding beam to travel along the edge of the joint plane in the optimized welding direction while being maintained at the constant predefined angle with respect to the optimized welding direction.

[0017] Also provided is a data processing system designed to implement a method according to the invention. Brief description of the figures

[0018] The invention will be better understood with the aid of the following description, given solely by way of example and with reference to the appended drawings in which: - [Fig. 1] is a three-dimensional view of a welding of two parts, illustrating in particular a joint plane between these two parts, - [Fig.2] is a simplified view of a data processing system for welding optimization, - [Fig.3] is a block diagram of a method for determining an optimized welding direction for a joint plane, - [Fig.4] is a block diagram of a method for determining an optimized joint plane, using the method of [Fig.3], - [Fig.5] is a block diagram of a two-part welding process using the process of [Fig.4], and - [Fig.6] is a view of a joint plane and a potential welding direction, illustrating an example of determining a joint plane thickness along this welding direction. Detailed description of the invention

[0019] With reference to [Fig.l], an example of an assembly 100 obtained by welding will now be described.

[0020] The assembly 100 comprises two parts 102, 104 designed to be welded to each other. More precisely, each of the two parts 102, 104 has a flat welding face. These flat welding faces have geometries that mirror each other and are designed to be joined and aligned with each other in order to be welded. These flat faces thus form a joint plane 106 of the two parts 102, 104.

[0021] To carry out the welding, a welding beam 108 is emitted in the plane of the joint plane 106 by a suitable welding apparatus 110, to reach an edge of the joint plane 106. This welding beam 108 is moved to travel along the edge of the joint plane 106 to fuse the two parts 102, 104 at the joint plane 106. More precisely, the welding beam 108 is moved in a welding direction 112, that is to say parallel to this welding direction 112. The latter extends in the plane of the joint plane 106 and passes, for example, through a center of the joint plane 106, while being maintained at a constant predefined angle with respect to this welding direction 112 during its movement, for example perpendicular as in the example illustrated in [Fig.l].

[0022] In [Fig.l], the welding apparatus 110 and the welding beam 108 are illustrated at three successive positions along the welding direction 112. Fusion welding is for example one of: TIG / MIG (from the English "Tungsten Inert Gas / Metal Inert Gas") welding, laser welding, plasma welding, and electron beam welding (also designated by the acronym FE).

[0023] It is possible to define, at each position s on the welding direction 112, a thickness E(s) of the joint plane 106, this thickness E(s) being taken along the welding beam 108 passing through the position s considered on the welding direction 112. In the example of [Fig.l], the thickness E(s) is thus taken along the perpendicular to the welding direction 112 passing through the position s considered. This is the thickness that the welding beam 108 must pass through at each position s to carry out welding over the entire joint plane 106.

[0024] The thickness curve E(s) represents the history (or gradient) of the thickness of the joint plane 106 crossed by the welding beam 108 during its movement along the welding direction 112.

[0025] Referring to [Fig.2], an example of a data processing system 200 designed to optimize the joint plane 106 and the welding direction 112, will now be described.

[0026] For example, the system 200 is a computer system comprising a data processing unit 202 (such as a microprocessor) and a main memory 204 (such as a RAM memory, from the English "Random Access Memory") accessible by the processing unit 202. The computer system 200 further comprises for example a network interface and / or a computer-readable medium, such as for example a local medium 206 (such as a local hard disk) or a remote medium (such as a remote hard disk and accessible via the network interface through a communication network) or even a removable medium (such as a USB key, from the English "Universal Serial Bus", or a CD, from the English "Compact Disc" or a DVD, from the English "Digital Versatile Disc") readable by means of an appropriate reader of the computer system 200 (such as a USB port or a CD and / or DVD disk reader).A computer program 208 containing instructions for the processing unit 202 is recorded on the local medium 206 and / or downloadable via the network interface. This computer program 208 is for example intended to be loaded into the main memory 204, so that the processing unit 202 executes its instructions. To facilitate the description of the computer program 208, the instructions will be described subsequently as organized into software modules. However, this presentation does not prejudge the form of the computer program, which may be any.

[0027] Alternatively, all or part of these modules could be implemented in the form of hardware modules, that is to say in the form of an electronic circuit, for example micro-wired, not involving a computer program.

[0028] The computer program 208 thus comprises a module 210 for determining an optimized welding direction in a joint plane. The module 210 is in particular designed to implement a method 300 for determining a welding direction, such as that which will be described in more detail later with reference to [Fig. 3],

[0029] The computer program 208 further comprises a module 212 for determining an optimized joint plane, using the module 210. The module 212 is in particular designed to implement a method 400 for determining an optimized joint plane, such as that which will be described in more detail later with reference to [Fig.4].

[0030] With reference to [Fig. 3], an example of the method 300 for determining an optimized welding direction in a joint plane 106 will now be described.

[0031] For each of several potential welding directions passing through a center of the joint plane 106, the following steps 302, 304, 306 are implemented.

[0032] For example, the potential welding directions are angularly spaced by a fixed angular pitch, for example between 1° and 20°. Preferably, the potential welding directions only cover 180° of the joint plane at most.

[0033] For example, a mesh being superimposed on the joint plane, the center of the joint plane is a center of gravity of the points of the mesh located in the joint plane. The points of the mesh have for example the same weight for the determination of the center of gravity. To do this, the mesh is preferably made up of elements of surface (plane) sufficiently small to discretize the geometry of the joint plane. For example, a maximum element surface of one twenty-fifth of the surface of the joint plane is recommended. Preferably, the elements must be of the same geometric type (for example, triangles or quadrilaterals).

[0034] During a step 302, the thickness E(s) of the joint plane is determined as a function of the position s on the welding direction considered. Thus, in the present invention, the thickness E(s) is a function giving the thickness of the joint plane perpendicular to the welding direction considered, that is to say the length of the portion of the perpendicular to the welding direction considered extending on the joint plane, as explained previously with reference to [Fig.l].

[0035] For example, the thickness E(s) of the joint plane is determined for several discrete positions on the welding direction. The center of the joint plane is for example one of these positions and the other positions are for example separated from each other by a fixed pitch. This pitch is preferably of the order of magnitude of the average distance between two neighboring nodes of the mesh, that is to say between 0.5 and 1.5 times this distance. Preferably, the discrete positions extend on either side of the center of the joint plane over at least 1.1 times the distance between the center and the node of the mesh in the joint plane furthest from the center. This condition makes it possible to ensure that the discrete positions extend over the entire length of the joint plane.

[0036] In this case, for each position s on the welding direction, we count for example the number of points of the mesh strictly included in a band inclined relative to the welding direction 112 by the predefined angle of the welding beam 108, centered on the position s considered and of width such that the bands are contiguous, for example equal to the pitch between the positions when the predefined angle is a right angle (band perpendicular to the welding direction). This number of points is taken as the thickness E(s) of the joint plane at the position s considered.

[0037] This example is illustrated in [Fig.6], where the references of [Fig.l] are repeated for the joint plane and the welding direction, where the center of the joint plane 106 bears the reference 602, where the current position on the welding direction 112 bears the reference 604, where the strip bears the reference 606 and is perpendicular to the welding direction 112, like the welding beam 108, where the mesh bears the reference 608, and where the distance between the center 602 and the node of the mesh in the joint plane further from the center 602 bears the reference D. In the example of [Fig.6], the number of mesh points in the band 606 is forty-five. Thus, the thickness E(604) at position 604 is forty-five.

[0038] Returning to [Fig.3], during a step 304, a weldability score S is calculated from the thickness E(s).

[0039] Preferably, the calculation of the weldability score S comprises a calculation of an evaluation of a cumulative thickness to be welded and the weldability score S is calculated from this evaluation of the cumulative thickness, to be better when the evaluation of the cumulative thickness decreases. The weldability score S is for example taken equal to this evaluation of the cumulative thickness. In the calculation of the evaluation of the cumulative thickness, the thickness at each point contributes all the more strongly as the thickness is high at this point. The contributions (i.e. the thicknesses at the points considered) are thus accumulated in the evaluation, with each thickness weighted by itself.Thus, the use of the evaluation of the cumulative thickness to be welded degrades the weldability score S (for example by making it high) for welding directions having large thicknesses to cross, cumulatively or in peak (i.e. when the thickness is high punctually or for many consecutive points), and, conversely, improves the weldability score S (for example by making it low) for welding directions having small thicknesses to cross. Indeed, the welding operation is all the easier when the cumulative thickness to be welded is low.

[0040]

[0041] For example, when the thickness E(s) is determined for several discrete positions on the welding direction 112, the evaluation of the cumulative thickness can be the sum of the squares of these discrete thicknesses, the weldability score S being calculated from this sum of the squares, for example taken equal to this sum: g _ y .E(sy • More generally, the weldability score S can be calculated from a sum of polynomials of degree greater than or equal to 2, having the thickness as a variable, for example taken equal to this sum: S = £.P( E(S|) ), with P Ie polynomial of degree greater than or equal to 2. Alternatively, the weldability score S can be calculated from the product of the squares of the discrete thicknesses, for example taken equal to this product: $ „ yjE(sp

[0042] During a step 306, the potential welding direction with the best weldability score S is selected as the optimized welding direction. For example, the best weldability score S is the smallest, as in the case where the weldability score S is equal to the sum of the squares of the discrete thicknesses.

[0043] With reference to [Fig.4], an example of a method 400 for determining a plane of optimized joint, will now be described.

[0044] During a step 402, operating data of the method are received, for example in a main control file.

[0045] The operating data includes a definition of an initial joint plane. This definition includes, for example, the positions of pilot points, as well as a geometric shape consecutively connecting the consecutive pilot points to form a closed curve delimiting a closed surface. An example of a geometric shape is a succession of straight line segments, each segment being delimited by two consecutive pilot points. Alternatively, the pilot points can be connected by splines, each spline being able to consist of one or more other points, called intermediate points.

[0046] The operating data further includes a definition of a mesh making it possible to discretize the joint plane in the form of nodes and finite elements.

[0047] Generally, two degrees of freedom are assigned to each pilot point and define the position of the pilot point in the plane of the joint plane, the values ​​of these degrees of freedom being initialized to initial values. The positions of the pilot points are for example located in an orthogonal reference frame. The position of each pilot point can thus be governed by an Xp component and a Yp component.

[0048] During a step 404, a mesh is superimposed on the initial joint plane. This mesh allows a regular discretization of the joint plane making it possible to define the topological density in the plane of the joint plane (i.e. for example the number of nodes per unit area). For example, the mesh is produced from triangles and / or quadrilaterals. Preferably, the mesh must be sufficiently fine so that the thickness of the joint plane in any direction passing through the center of gravity of this mesh is crossed by at least four points of the mesh. The number of elements of the mesh is thus for example increased until this condition is satisfied.

[0049] Starting from the initial joint plane, the following steps 406, 408, 410 are iterated.

[0050] During a step 406, the center of the current joint plane is determined. This is for example from the center of gravity of the mesh points included in the current joint plane, as explained previously.

[0051] During a step 408, an optimized welding direction for the current joint plane is determined by implementing the method 300. The method 300 is preferably implemented each time with the same parameters, in particular: same mesh, same inter-position pitch on the welding direction, same formula for the weldability score, same angular pitch for the potential welding directions. Thus, it is possible to compare the results between the joint planes. The weldability score associated with the optimized welding direction found is taken as the weldability score of the joint plane considered.

[0052] During a step 410, as long as the weldability score of the current joint plane does not satisfy an exit condition of the iteration, the current joint plane is modified to obtain a new joint plane and the iteration is continued by returning to step 406, with this new joint plane as the current joint plane.

[0053] For example, the modification of the current joint plane involves a modification of the positions of at least some of the pilot points. Indeed, when the position of one or more pilot points is modified, the shape of the current joint plane is itself modified, thus generating a variant of the current joint plane. This parameterization thus makes it possible to define and modify the joint plane from a numerical point of view.

[0054] Preferably, areas of the joint plane called bulk areas are provided, in which the pilot points cannot be located. These areas make it possible to guide the optimization of the shape of the joint plane and to maintain the desired general shape of the assembly.

[0055] This modification is for example carried out by means of a parametric automated optimization algorithm, seeking to optimize the weldability score. The implemented parametric automated optimization algorithm may for example be a differential algorithm based on an iterative strategy for constructing and enriching an interpolating response surface of the radial basis function type (from the English “Radial Basis Function”). The enrichment criterion between each iteration may for example be the theoretical optimum identified by the differential evolution algorithm (known per se) on the basis of the response surface constructed at said iteration. A variant may consist of using an interpolating response surface of the polynomial regression function type, and provided with an iterative enrichment strategy by a local SQP (“Sequential Quadratic Programming”) algorithm.

[0056] If the weldability score of the current joint plane satisfies the exit condition of the iteration, the joint plane with the best weldability score is taken as the optimized joint plane.

[0057] For example, the exit condition of the iteration may be that the weldability score S of the current joint plane is lower than a predefined threshold. In this case, the last joint plane, i.e. the one whose weldability score is lower than the predefined threshold, is taken as the optimized joint plane. Alternatively, the exit condition of the iteration may be that the quality criterion R2 or Q2 of the RBF response surface or polynomial regression surface considered reaches a significant threshold value, for example 0.9. In another variant, the exit condition of the iteration is that a predefined number of iterations is reached, for example one hundred iterations.

[0058] During a step 412, the optimized joint plan is provided (in particular the positions of its pilot points and the indication of the connection curve used (straight line segment, spline, etc.), with the optimal welding direction found (for example, expressed by an angle in degrees relative to a predefined axis of the welding plan).

[0059] At least one of: - for each welding direction of the optimized joint plane, the list of contributions of the thicknesses used to calculate the weldability score S along the considered welding direction (for example, in the case where the weldability score is the sum of the squares of the thicknesses, the list of the squares of the thicknesses is provided); and - the weldability criterion S for each of the welding directions evaluated for the optimized joint plane, which makes it possible to prioritize the different possible welding directions according to the weldability score S, and therefore to identify the optimal welding direction, as well as to quantify the gain on the weldability score S.

[0060] The preceding output data are for example provided in a readable and editable results file (for example in .txt format) or not (specific formatting).

[0061] With reference to [Fig.5], an example of a method 500 for welding the two parts 102, 104 will now be described.

[0062] During a step 502, an optimized joint plane is determined, with its associated optimized welding direction, by implementing the method 400.

[0063] During a step 504, the two parts 102, 104 are obtained, each with a flat welding face according to the determined optimized welding plan.

[0064] During a step 506, the flat welding faces of the two parts 102, 104 are joined to each other.

[0065] During a step 508, the two parts 102, 104 are welded to each other, by moving a welding beam according to the determined optimized welding direction, so that this welding beam travels along an edge of the joint plane.

[0066] It may happen that the parts 102, 104 are provided with reinforcements intended to facilitate welding, but which must not be kept for the final assembly. Thus, during an optional step 510, planing of the two parts 102, 104 around the weld may be provided to remove this reinforcement.

[0067] In conclusion, it will be noted that the invention is not limited to the embodiments described above. It will indeed appear to those skilled in the art that various modifications can be made to the embodiments described above, in light of the teaching which has just been disclosed to them.

[0068] In particular, the operating data may comprise at least one of: the length of the welding directions, the coordinates of the center of the welding plane ... joint used as a pivot point for welding directions, the discretization step of welding directions, the angular step between two welding directions, clutter zones of the welding plane.

[0069] In the detailed presentation of the invention which is made above, the terms used must not be interpreted as limiting the invention to the embodiments set out in the present description, but must be interpreted to include all equivalents whose prediction is within the reach of those skilled in the art by applying their general knowledge to the implementation of the teaching which has just been disclosed to them.

Claims

Claims

1. Method (300) for determining an optimized welding direction (112) for a joint plane (106), a welding beam (108) being intended to be moved to travel along an edge of the joint plane (106) in the optimized welding direction (112) while being maintained at a constant predefined angle with respect to the optimized welding direction (112), characterized in that it comprises: - for each of several potential welding directions (112): • a determination (302) of a thickness of the joint plane (106) as a function of a position on the welding direction (112), this thickness being taken along a straight line making the predefined angle with the welding direction (112) and passing through the position considered, and • a calculation (304) of a weldability score from the thickness as a function of the position on the welding direction (112);and - a selection (306) as optimized welding direction (112), of the potential welding direction (112) with the best weldability score.;

2. The method of claim 1, wherein the calculation (304) of the weldability score comprises a calculation of an evaluation of a cumulative thickness to be welded and the weldability score is calculated from this evaluation of the cumulative thickness to be welded, to be better when the evaluation of the cumulative thickness to be welded decreases.

3. Method according to claim 2, in which, the thickness being determined for several discrete positions on the welding direction (112), the evaluation of the cumulative thickness to be welded is the sum of the squares of these discrete thicknesses, the weldability score being calculated from this sum of the squares.

4. Method according to any one of claims 1 to 3, in which the determination (302) of the thickness as a function of the position on the welding direction (112) comprises, for each of several discrete positions on the welding direction separated by a step constant, a count of a number of points of a mesh (608) superimposed on the joint plane (106), the points counted being those strictly included in a strip (606) inclined relative to the welding direction (112) by the predefined angle, this strip (606) being centered on the discrete position (604) considered and of width such that the strips are contiguous, this number of points being taken as the thickness of the joint plane (106) at the discrete position (604) considered.

5. A method according to any one of claims 1 to 4, wherein the potential welding directions pass through a center of the joint plane.

6. A method according to claim 5, wherein the center of the joint plane is a center of gravity of points of a mesh superimposed on the joint plane, these points being located in the joint plane.

7. Method (400) for determining an optimized joint plane between two parts (102, 104) to be welded, comprising: - receiving a definition of an initial joint plane between the two parts; - starting from the initial joint plane, an iteration of the following steps: • determining an optimized welding direction in the joint plane, in accordance with a method according to any one of claims 1 to 6, the weldability score of the optimized welding direction found being associated with the joint plane, and • as long as the weldability score does not satisfy an exit condition of the iteration, modifying the joint plane and continuing the iteration, otherwise the joint plane with the best weldability score is taken as the optimized joint plane.

8. The method of claim 7, wherein defining the initial parting plane includes pilot point positions and a geometric shape consecutively connecting the pilot points to form a closed curve, and wherein modifying the parting plane includes modifying the position of at least some of the pilot points.

9. Computer program (208) downloadable from a communications network and / or recorded on a computer-readable medium, characterized in that it comprises instructions for the execution of the steps of a method according to any one of claims 1 to 8, when said program is executed on a computer.

10. A method of welding two parts (102, 104), comprising: - determining an optimized welding direction (112) for a joint plane (106) between the two parts (102, 104) according to any one of claims 1 to 6, or determining an optimized joint plane between the two parts (102, 104) according to claim 7 or 8; and - welding the two parts (102, 104) by moving the welding beam (108) to travel along the edge of the joint plane (106) in the optimized welding direction (112) while being maintained at the constant predefined angle with respect to the optimized welding direction (112).

11. A data processing system (200) adapted to implement a method according to any one of claims 1 to 8.

Images