Method of shaping a piece of equipment comprising a blade

Abstract

Method for shaping a piece of equipment comprising a blade The present invention relates to a method for shaping a piece (12) of equipment of the turbine blade type, comprising the following steps: - providing a piece (12) comprising a blade (14) in an initial conformation; - providing a nominal definition representing the piece in a nominal conformation; - comparing the initial conformation with the nominal definition in order to determine a conforming or non-conforming character; - for a non-conforming data item, determining a force to be applied to the piece to deform said piece; - applying the force so as to obtain the piece in a deformed conformation, - comparing the deformed conformation with the nominal definition in order to determine a conforming or non-conforming character; - training a self-learning algorithm (82). Figure for the abstract: Figure 1

Description

Title of the invention: Method for shaping a piece of equipment comprising a blade

[0001] The present invention relates to a method for shaping a piece of equipment comprising a blade. The invention applies particularly to static (variable or not) or rotor turbine or compressor blades comprising a blade.

[0002] A blade is typically shaped like a twisted wing, the aerodynamic profile of which is designed to optimize a fluid flow and compress it.

[0003] Conventional methods of manufacturing blades by machining, forging or extrusion do not always allow the required twist or deflection to be obtained. It is therefore known to subject blades to a straightening process, aimed at bringing the shape of the blade into the tolerance intervals around the nominal profile. Such a process is described in particular in document EPI494003.

[0004] The aim of the present invention is to propose a method allowing the automation of the straightening of the blades, as well as a continuous improvement of the result by automatic learning.

[0005] To this end, the invention relates to a shaping method of the aforementioned type, comprising the following steps:

[0006] - supply of a part comprising a blade in an initial conformation, said blade extending along an axis, and comprising at least a first gripping zone arranged at an axial end of the blade;

[0007] - providing a nominal definition representing the part in a conformation nominal, said nominal definition comprising a set of nominal dimensions and associated tolerances;

[0008] - three-dimensional acquisition of a first set of data representing the part in the initial conformation;

[0009] - comparison of the first data set to the nominal definition in order to determine terminate a conforming or non-conforming character of at least one data of the first data set;

[0010] - for non-compliant data, determination by a self-learning algorithm trained on a plurality of parts of the same type of a force to be applied to the part to deform said part with a sufficient probability so that at least the non-compliant data is subsequently determined as being compliant;

[0011] - application of the force determined in the previous step on the part so as to obtain the part in a deformed conformation,

[0012] - three-dimensional acquisition of a second set of data of the part in the deformed conformation, ;

[0013] - comparison of the second data set to the nominal definition in order to determine terminate a conforming or non-conforming character of at least one data of the second data set;

[0014] - measurement and recording of parameters of the first applied force, of the first and second data sets as well as conformity and non-conformity characteristics determined during the comparison steps; and

[0015] - training of the self-learning algorithm based on the measurements and recordings events of the previous step.

[0016] According to other advantageous aspects of the invention, the method comprises one or more of the following characteristics, taken individually or in all technically possible combinations:

[0017] - the force is chosen between: a bending force, perpendicular to the axis of the blade; a torsional torque around said blade axis; and a combined bending and torsional force;

[0018] - at least one three-dimensional acquisition is carried out by means of equipment to measure with or without contact;

[0019] - each of the nominal definition, the first and the second data set includes at least one of: a set of dimensions of a section of the blade, a twist angle, a thickness of a leading edge of the blade, a thickness of a trailing edge of the blade, a position of a key point on an exterior surface of the part, or a combination of any of these data;

[0020] - during the application of the force on the blade, a grip is exerted at least on the first gripping area;

[0021] - during the application of the force on the blade, a second gripping zone of the part is held fixed relative to a frame;

[0022] - the force is a bending force, perpendicular to the axis of the blade; said force in bending being applied by exerting support on a support zone of the blade, said support zone being arranged between the first and second gripping zone of said part;

[0023] - the effort is a torsional torque around the blade axis; one of the elements of gripping and the second portion of the part is held fixed relative to a frame (30); and the torsional force is applied by pivoting, relative to the frame, the other of said first and second gripping zones;

[0024] - the force is a torsional torque around the blade axis; and the torsional force is applied by rotating the first and second gripping areas relative to each other;

[0025] - the part further comprises an identifier, the method further comprising the steps of: reading the identifier; identifying the nominal definition corresponding to the piece of equipment from among a plurality of nominal definitions; and identifying the self-learning algorithm corresponding to the piece of equipment from among a plurality of self-learning algorithms.

[0026] The invention further relates to an assembly for implementing a method as described above, comprising: a measuring device, capable of acquiring three-dimensional data of the part, in the initial conformation and / or in the deformed conformation; an installation capable of applying the force to the part; and an electronic device comprising a data memory and at least one self-learning algorithm trained on a plurality of parts of the same type as the part, said measuring device, the installation and the electronic device being connected by at least one communication channel.

[0027] The invention will be better understood on reading the description which follows, given solely by way of non-limiting example and made with reference to the drawings in which:

[0028] [Fig-1] [Fig.l] is a schematic view, from above, of an installation and a assembly for implementing a shaping method according to one embodiment of the invention;

[0029] [Fig.2][Fig.3] Figures 2 and 3 are partial schematic views of the installation of [Fig.l];

[0030] [Fig.4] [Fig.4] is a flowchart of a shaping method according to one embodiment of the invention; and

[0031] [Fig.5] [Fig.5] is a flowchart of a learning method for a shaping method according to an embodiment of the invention.

[0032] [Fig.l] represents an installation 10 for implementing a method for manufacturing a piece of equipment 12.

[0033] The part 12, visible in Figures 1 and 3, is in particular of the turbine or compressor blade type. In particular, the part 12 comprises a metal blade 14, extending along a main axis 16.

[0034] The blade 14 comprises a curved surface. In particular, the blade 14 comprises two opposite curved faces 18, 19, each of said faces extending substantially along the axis 16.

[0035] In the embodiment shown, the part 12 further comprises a first gripping zone 20, arranged at a first axial end of the blade 14, in the form of a blade root. The root 20 and the blade 14 are formed in a single piece from a metallic material such as a base alloy of iron, titanium, aluminum, copper or nickel, or from a plastic material, or are formed in a composite manner. in a metallic material and in a plastic material.

[0036] In a variant not shown, the part 12 further comprises a heel, arranged at a second axial end 22 of the blade 14, opposite the root 20.

[0037] The installation 10 is intended to carry out an operation of straightening the blade 14 of the part 12 in order to give it a desired final shape, in the event of too great a difference between this desired shape - conforming to a definition of the blade - and the shape of the actual blade resulting from standard manufacturing processes, for example forging, extrusion or machining.

[0038] The installation 10 comprises a fixed frame 30 and at least two mobile units 32, 34, 36, 38, capable of exerting a force on the blade 14 in order to deform it. In the embodiment shown, the mobile units comprise: a gripping unit 32; a support unit 34; a bending unit 36; and a torsion unit 38.

[0039] The installation 10 further comprises an electronic module 40, said module comprising in particular a microprocessor. Preferably, the electronic module 40 also comprises: a module for converting analog data - from the sensors - into digital data; a data preprocessing algorithm, in particular for removing aberrant data and standardizing the digital data for subsequent processing by a program; and possibly a display or a sound transmitter.

[0040] We consider an orthonormal base (X, Y, Z) associated with the frame 30, the direction Z representing the vertical.

[0041] The frame 30 comprises: an upper surface extending in a plane (X, Y); and means 44, 45 for guiding the gripping, support, bending and torsion units on said upper surface.

[0042] The guide means 44, 45 are for example rails. In the embodiment shown, first rails 44 are arranged along a first axis 46 parallel to X and second rails 45 are arranged along a second axis 47 parallel to Y.

[0043] Each of the gripping, support, bending and torsion units comprises: a structure 50; and a means 52 for motorized movement of said structure, along at least one of the means 44, 45 for guiding the frame 30.

[0044] More precisely, in the embodiment shown, the structures 50 of the gripping unit 32 and of the support unit 34 are able to move along X, along the first rails 44; and the structures 50 of the bending unit 36 ​​and of the torsion unit 38 are able to move along Y, along the second rails 45. In a variant not shown, at least one mobile unit of the installation 10 is able to move in two directions in a plane (X, Y).

[0045] Preferably, the means 52 for motorized movement of the units are controlled by the electronic module 40; and each of the means 52 is coupled to a sensor 53 of a position of the corresponding unit on the frame 30. Each position sensor 53 is connected to the electronic module 40. The connections between the electronic module 40 and the displacement means 52 and each sensor 53 are preferably wired.

[0046] The gripping unit 32, more precisely visible in [Fig. 2], will now be described. In addition to its structure 50, the gripping unit 32 comprises a rotary gripping member 54.

[0047] The member 54 is capable of exerting a grip on the part 12, in particular on the foot 20, so as to keep said foot fixed relative to said member 54. In particular, the member 54 comprises a housing 55 ([Fig.2]), capable of receiving and holding the foot 20.

[0048] Furthermore, the member 54 is rotatable relative to the structure 50 of the unit 32, along a third axis 56 parallel to X. The housing 55 preferably extends along the third axis 56.

[0049] More specifically, in cooperation with the torsion unit 38, the rotary member 54 is capable of applying a torsional torque to the part 12, as will be detailed below.

[0050] The gripping unit 32 further comprises: a sensor 58 of the angular position of the member 54 relative to the structure 50; and a sensor 59 of the torque exerted by an angular displacement of said member relative to said structure, this torque can be applied to the part in both clockwise and counterclockwise directions.

[0051] According to one embodiment, the gripping unit 32 comprises a device for detecting a mark 60 ([Fig.l]) arranged on the part 12, for example on the foot 20. Thus, when the part 12 is assembled to the housing 55, an angular position of the part 12 is associated with the angular position of the member 54, detected by the sensor 58.

[0052] Preferably, the member 54 and the sensors 58, 59 are connected to the electronic module 40 by a wired connection.

[0053] The support unit 34 comprises an upper surface 61. Said upper surface is capable of being arranged under a second gripping zone of the part 12.

[0054] Preferably, the upper surface 61 of the support unit 34 has a shape suitable for contact with said second gripping zone of the part 12. The second gripping zone may be formed by the heel, or a portion of blade at an axial distance from the root and the heel.

[0055] More specifically, in the embodiment shown, the upper surface 61 of the unit 34 has a shape suitable for contact with the heel, and preferably has a shape complementary to that of the heel.

[0056] Optionally, the support unit 34 further comprises means (not shown) for adjusting a height along Z of the upper surface 61 relative to its structure 50. Said means are connected to the electronic module 40.

[0057] The bending unit 36, more precisely visible in [Fig. 2], will now be described. In addition to its structure 50, the bending unit 36 ​​comprises a press 62. In particular, the press 62 comprises an application head 64, movable along Z relative to the structure 50 of the bending unit 36. The application head 64 is capable of exerting a downward vertical force on the part 12, more precisely on a specific zone of the blade 14, as will be described below. One end of the head 64 comprises in particular a bearing surface 65, intended to come into contact with the blade 14 or the heel.

[0058] The flexion unit 36 ​​further comprises: a sensor 66 of the travel of the head 64 along Z; and a sensor 68 of the force exerted by or on said head.

[0059] Preferably, the press 62 and the sensors 66, 68 are connected to the electronic module 40 by wired connection.

[0060] The torsion unit 38, more precisely visible in [Fig. 3], will now be described. In addition to its structure 50, the torsion unit 38 comprises an upper flange 70 and a lower flange 72. At least one of the upper 70 and lower 72 flanges is movable along Z relative to the structure 50 of the unit 38. Optionally, each of the upper 70 and lower 72 flanges is movable along Z relative to said structure 50.

[0061] According to one embodiment, at least one of the upper 70 and lower 72 flanges is also movable along the X axis relative to the structure 50.

[0062] Each of the upper 70 and lower 72 flanges comprises at least one finger 74, 76. The upper 70 and lower 72 flanges are configured so as to grip the second gripping zone 22 of the part 12 between the fingers 74, 76. Preferably, the end of each finger 74, 76 comprises a surface 78 configured for contact with the second gripping zone. In an embodiment not shown, the surfaces 78 are configured for contact with the heel.

[0063] In the embodiment shown, each of the upper 70 and lower 72 flanges comprises two fingers 74, 76, each finger 74 of the upper flange being arranged vertically above a finger 76 of the lower flange.

[0064] Preferably, the torsion unit 38 further comprises a sensor 79 of force exerted by or on the upper flange and / or the lower flange.

[0065] Preferably, the upper flange 70 and / or the lower flange 72 and / or the sensor 79 are connected to the electronic module 40 by wire connection.

[0066] A first operation of a method for shaping the part 12 will now be described. Said first operation, called the deflecting operation, takes place as follows:

[0067] It is initially considered that the foot 20 is assembled to the housing 55 of the member 54 of the gripping unit 32, so that the axis 16 of the part and the third axis 56 of rotation of the member 54 are merged, as shown in Figures 1 and 3.

[0068] First of all, a support zone 80 on the blade 14 is determined by a self-learning algorithm 82. Said algorithm 82 is stored in the electronic module 40, or in a control device 94, as will be described later.

[0069] The support zone 80 has, for example, dimensions similar to the support surface 65 of the application head 64 of the bending unit 36.

[0070] Based on the location of the support zone 80 on the blade 14, the algorithm 82 then determines: an angular position of the member 54 relative to the corresponding structure of the gripping unit 32; a position along X of said gripping unit 32; and a position along Y of the bending unit 36. Said positions are configured to allow the application head 64 to come into contact with the support zone 80 during a downward vertical movement of said head.

[0071] The algorithm 82 also determines a position along X of the support unit 34, so that the upper surface 61 is in contact with the part 12 during the vertical movement of the application head 64. The position of the support unit 34 is notably chosen so that, during its vertical movement, the head 64 is arranged along X between the gripping unit 32 and the support unit 34.

[0072] Based on the positions determined above, the displacement means 52 are actuated by the electronic module 40 to arrange the gripping units 32, 34 for support and 36 for bending at the corresponding locations on the frame 30; and the member 54 is placed in the corresponding angular position.

[0073] Then, the press 62 is actuated by the electronic module 40, so as to move the application head 64 downwards. The bearing surface 65 thus comes into contact with the bearing zone 80. The downward movement of the head 64 is continued so as to apply a force to the bearing zone 80. The part 12, supported axially on either side of said bearing zone, undergoes a first flexural deformation, called deflection, perpendicular to the main axis 16.

[0074] The stroke of the head 64, since contact with the part 12, is detected by the stroke sensor 66 and the force sensor, to detect the contact. Similarly, the force exerted between the head 64 and the part 12 is detected by the force sensor 68.

[0075] When a desired stroke and / or force, determined by the algorithm 82, is reached, the downward movement of the head 64 is stopped. The position of the support zone 80, the stroke and force measurements are stored in a data memory 96 described later. The head 64 is moved upward to release the part 12.

[0076] A second operation of a method for shaping the part 12 will now be described. Said second operation, called the untwisting operation, takes place as follows:

[0077] As in the deflection operation described above, it is initially considered that the foot 20 is assembled to the housing 55 of the member 54, so that the axes 16 and 56 are merged.

[0078] First, the algorithm 82 determines one or more areas of the part 12 intended to come into contact with the fingers 74, 76 and said positions are determined accordingly. The algorithm 82 thus determines: an initial angular position of the member 54 relative to the corresponding structure of the gripping unit 32; a position along X of said gripping unit 32; a position along Y of the torsion unit 38; and a position along Z of the upper flange 70 and / or the lower flange 72 of said torsion unit, relative to the corresponding structure. Said positions are configured so that a third gripping area of ​​the part 12 is maintained between the upper and lower flanges, as described above. The third gripping area may be any section of the blade, or the heel.

[0079] On the basis of the positions determined above, the displacement means 52 are actuated by the electronic module 40 to arrange the gripping units 32 and 38 of torsion at the corresponding locations on the frame 30; the member 54 is placed in the corresponding initial angular position; then the upper flange 70 and / or the lower flange 72 is / are moved along Z so as to grip the third gripping zone.

[0080] Preferably, the force exerted by the flanges 70, 72 on the part 12 is controlled by the installation 10, in particular at the end of the stroke.

[0081] Then, the rotary gripping member 54 is rotated about the third axis 56 relative to the structure 50 of the gripping unit 32. The blade 14 undergoes a torsional torque and consequently a second deformation, called untwisting. The angular displacement of the member 54 is measured by the angular position sensor 58; and the torsional torque exerted by the member 54, as well as its clockwise or counterclockwise direction, are measured by the torque sensor 59.

[0082] When a desired angular displacement and / or torque, determined by the program 82, is reached, the angular displacement of the member 54 is stopped at a final angular position. The angular displacement and torque measurements are stored in the data memory 96.

[0083] The deflecting and untwisting operations described above are implemented in a method 100 for shaping parts 12 as described above. The method 100 is represented in the form of a flowchart in [Fig.4].

[0084] For the implementation of the method 100, the installation 10 described above is incorporated into an assembly 90, in which said installation 10 is associated with the following elements, represented schematically in [Fig.l]: an imaging device 91; an automaton 92; and a control device 94.

[0085] The imaging device 91 is capable of acquiring data representative of the part 12 in three dimensions, and in particular of the blade 14. The imaging device 91 notably comprises a reader 95 of two-dimensional traceability and identification data.

[0086] The imaging device 91 is for example a contact measuring machine, for example a three-dimensional measuring machine (CMM) or a multi-dimension measuring bench. Alternatively, the imaging device 91 may be a non-contact measuring equipment such as a three-dimensional scanner, or a one-dimensional or two-dimensional scanner, used in several directions with respect to the part.

[0087] The automaton 92, for example of the robot arm type, is capable of moving the part 12 between the imaging device 91 and the installation 10, as well as assembling and / dissociating said part 12 and the housing 55 of the gripping unit 32.

[0088] The control device 94 comprises a data memory 96 and a program memory 98. The program memory 98 comprises as many self-learning algorithms as there are blade definitions. Alternatively, the program memory 98 comprises a self-learning algorithm 82 and a parameter set for each part definition 12. The control device 94 can be implemented in the form of a computer, an embedded system or a control automaton, said computer, system or automaton comprising the data memory 96 and the program memory 98.Alternatively, the control device 94 is implemented in a dissociated form in which the data memory 96 is stored on a remote server and the program memory 98 is stored in a computer, an embedded system or a PLC, or vice versa, or in a completely remote form, in which the database and the program memory 98 are each stored on a server or in a cloud.

[0089] The control device 94 is connected to the electronic module 40 of the installation 10, to the imaging device 91 and to the automaton 92.

[0090] Preferably, the imaging device 91, the automaton 92 and the electronic module 40 each communicate with the control device 94 via a wireless link. The control device 94 can thus be remote from the assembly 90, either in a server room or in a cloud. When the control device 94 is a computer or an automaton, the imaging device 91, the automaton 92 and the electronic module 40 each communicate with the control device 94 preferably via a wired link.

[0091] The method 100 for shaping parts 12 will now be described. Said method 100 is implemented by the self-learning algorithm 82 of the control device 94.

[0092] A production batch is provided, consisting of parts 12 in an initial state. Said parts 12 are substantially identical to each other and for example obtained by forging.

[0093] Each of the parts 12 of said production batch comprises an identifier, in particular a visual identifier 104 such as a serial number, a one-dimensional or two-dimensional barcode, an RFID device or a combination of two or all three types of identifier. The visual identifier makes it possible to link each part 12 of a production batch to a nominal definition 112. A nominal definition is in particular defined by a material, a definition plane defined by a set of nominal dimensions, and a set of associated tolerances, defining for example the minimum and maximum dimensions admissible in the three dimensions of the blade.The nominal definition 112 can for example be described by means of discrete information representative of the geometry of the part 12, such as for example a set of dimensions in an orthonormal basis (X, Y, Z) of a section of the blade, for example defined at an axial distance from a surface of the root 20, a twist angle around the axis 16, for example defined at an axial distance from a surface of the root 20, a thickness of a leading edge of the blade, a thickness of a picking edge of the blade, a position of one or more particular points on an outer surface of the blade, or a combination of any of these data.

[0094] The visual identifier 104 also makes it possible to link each part 12 to a self-learning algorithm 82, as will be described later.

[0095] Each of the parts 12 of the production batch is subjected to the following steps:

[0096] First of all, a first three-dimensional acquisition 106 of the part 12, in particular of the blade 14, is carried out by the imaging device 91. The first three-dimensional acquisition generates a first set of data 114 representative of the geometry of the part 12 in the initial conformation.

[0097] According to one embodiment, this first set of data 114 comprises at least the same information as that of the nominal definition 112, or only part of said information.

[0098] According to one embodiment, the imaging device 91 scans the identifier 104 of the part 12. The identifier 104 is sent to the control device 94, which compares the identifier to a set of nominal definitions stored in the data memory 96.

[0099] The comparison makes it possible to identify the nominal definition 112 as well as its definition characteristics, corresponding to the manufactured part 12. The comparison also makes it possible to identify the self-learning algorithm 82 to be used, said algorithm being associated with the nominal definition of the part 12.

[0100] Then, each data of the first data set 114 is compared (step 110) to the corresponding data of the nominal definition 112 of the part 12. The identifier of the part 12, said first set of data 114 and the conformity or non-conformity of each data item of the first set 114 of data are stored in the data memory 96.

[0101] The comparison makes it possible to determine whether the identified part 12, in the initial state, conforms to the nominal profile of the blade.

[0102] If at least one data item of the first data set 114 does not conform to the corresponding data item of the nominal definition 112, for example a twist angle measured at a particular distance along the axis is greater than the maximum permissible definition angle at that same distance, the algorithm 82 performs the following operations (step 116):

[0103] - determination of the deformation / straightening operation most suitable for deforming the part to the desired profile, between the deflection and untwisting operations; then,

[0104] - if the deflection operation is chosen, determination of the position of the zone support 80, and the force or stroke to be applied to the part 12 to obtain the desired deformation; or

[0105] - if the untwisting operation is chosen, determination of the angular displacement or of the torsion torque to be applied to part 12 to obtain the desired deformation.

[0106] Then (step 118), the automaton 92 places the part 12 in the housing 55 of the installation 10; and the operation defined in the previous step by the algorithm 82 is carried out by said installation 10, as described previously.

[0107] In particular, in the embodiment considered, the information on the support zone 80, determined by the algorithm 82, is transmitted to the electronic module 40 of the installation 10.

[0108] Then, the automaton 92 transfers the part 12 into the imaging device 91; and a second three-dimensional acquisition of the part 12 in a deformed state, in particular of the blade 14, is carried out, so as to generate a second set of data 120 representative of the geometry of the blade 14 in the deformed conformation.

[0109] The second data set 120 is compared to the nominal definition 112 in order to determine whether the identified part 12, in the deformed state, is compliant.

[0110] The position, angle, stroke, torque and / or force values, measured by the sensors 53, 58, 59, 66, 68, 79 during the deflection and / or untwisting operation, as well as the conformity and non-conformity characteristics of the second set of data are stored in the data memory 96 of the control device 94, in association with the identifier of the part 12.

[0111] By an automatic learning method, these values ​​are taken into account in the self-learning algorithm 82, corresponding to step 116 described above, to determine the optimal deformation / straightening operation to be implemented on each following part 12 of the batch concerned, and for each part 12 of each batch manufactured according to the definition plan of the same nominal part.

[0112] According to one embodiment, if at least one piece of data from the second set of data 120 does not fall within the tolerance interval around the nominal profile, i.e. if at least one piece of data is considered non-compliant, the part 12 is subjected to a new iteration of steps 116 and 118 in order to subject it to a second deformation if this is necessary and authorized on the reference of the part.

[0113] Preferably, the algorithm 82 limits the deformations of the part 12 to a single deflection operation and a single untwisting operation.

[0114] According to one embodiment, the algorithm 82 is configured to bring the part into conformity with the necessary minimum. According to another embodiment, the program 82 is configured to tend towards the nominal profile.

[0115] According to one embodiment, the self-learning algorithm 82, implemented in step 116, is developed at the end of a learning phase 200 taking place as follows ([Fig.5]):

[0116] A test batch of parts 212, in an initial state, is provided. The parts 212 are analogous to the parts 12 of the production batch, described above. For example, each test batch comprises at least one hundred parts, and more preferably at least four hundred parts. Each of the parts 212 of the test batch is subjected to the following steps:

[0117] The identifier 104 of the part 212, as well as the nominal definition 112 of said part are stored in the data memory 96 of the control device 94.

[0118] A first three-dimensional acquisition 216 of the part 212 is carried out by the imaging device 91.

[0119] Next, an operator compares a profile of the part 212 with the nominal definition 112 described above. Based on this comparison, the operator evaluates an appropriate operation, chosen from the deflecting and untwisting operations described above, to which to subject the part 212 considered to deform it.

[0120] The operator then subjects said part 212 to the appropriate operation using the installation 10, by controlling the gripping units 32, 34 of support, 36 of bending and 38 of torsion of said installation. The geometric data of the blade in the initial conformation, the positions, displacements, forces and torques previously described, measured by the sensors of the installation 10, as well as the conformity characteristics of the geometric data are stored (step 218) in the data memory 96.

[0121] Then, a second three-dimensional acquisition 220 of the part 212 thus deformed is carried out by the imaging device 91.

[0122] The geometric data of the deformed blade - generated by the second three-dimensional acquisition 220 - is compared to the nominal definition 112. The comparison makes it possible to determine whether the identified part 212 has been deformed so as to enter into the tolerance interval 114 around the nominal profile of the blade.

[0123] These operations are repeated on a plurality of parts 212 of the test batch. A correlation analysis is carried out between the geometric measurements of each initial part and the geometric measurements of each deformed part in comparison with the theoretical profile, in order to determine the influential parameters for the straightening of the blades corresponding to a nominal blade reference. Then, the self-learning algorithm 82 is developed, on the basis of the influential parameters determined by the correlation analysis. The algorithm 82 is then stored in the program memory 98.

[0124] During the implementation of the method 100 on the parts 12 of the production batch, the information stored in the memory 96 makes it possible to train said algorithm by automatic learning, as indicated above, in order to improve the recommendations for torque / angle and force / travel values ​​to be imparted to each part 12 to make it conform, with sufficient probability, to the nominal geometric profile.

[0125] According to another embodiment, the learning phase 200 can be carried out without human intervention, the self-learning algorithm 82 being able to follow a predefined experimental plan, or an experimental plan whose limits are predefined, the algorithm building the plan within these limits.

Claims

Claims

1. A method (100) of shaping a piece of equipment (12) of the turbine or compressor blade type, the shaping method comprising the following steps: - provision of a part (12) comprising a blade (14) in an initial conformation, said blade extending along an axis (16), and comprising at least a first gripping zone (20) arranged at an axial end of the blade; - providing a nominal definition (112) representing the part in a nominal conformation, said nominal definition comprising a set of nominal dimensions and associated tolerances; - three-dimensional acquisition (106) of a first set of data (114) representing the part in the initial conformation; - comparing the first data set (114) to the nominal definition (112) in order to determine a compliant or non-compliant character of at least one data item of the first data set (114); - for non-compliant data, determination by a self-learning algorithm (82) trained on a plurality of parts of the same type of a force to be applied to the part to deform said part with sufficient probability so that at least the non-compliant data is subsequently determined as being compliant; - application of the force determined in the previous step on the part (12) so as to obtain the part in a deformed conformation, - three-dimensional acquisition of a second set of data (120) of the part in the deformed conformation; - comparing the second data set (120) to the nominal definition (112) in order to determine whether at least one item of data in the second data set (120) is compliant or not; - measurement and recording of parameters of the applied force, of the first (114) and second (120) sets of data as well as of the conformity and non-conformity characteristics determined during the comparison steps; and - training the self-learning algorithm (82) based on the measurements and recordings from the previous step.

2. A method according to claim 1, wherein the force is chosen from: a bending force, perpendicular to the axis (16) of the blade (14); a torsional torque around said axis (16) of the blade (14); and a combined bending and torsional force.

3. Method according to claim 1 or claim 2, in which at least one three-dimensional acquisition (116) is carried out by means of contact or non-contact measuring equipment.

4. The method of claim 3, wherein each of the nominal definition (112), the first (114) and the second (120) data set comprises at least one of: a set of dimensions of a section of the blade, a twist angle, a thickness of a leading edge of the blade, a thickness of a trailing edge of the blade, a position of a key point on an exterior surface of the part, or a combination of any of these data.

5. Method according to claim 1 or claim 2, in which, during the application of the force on the blade, a grip is exerted at least on the first gripping zone (20).

6. Method according to claim 1 in which, during the application of the force on the blade, a second (22) gripping zone of the part (12) is kept fixed relative to a frame (30);

7. Method according to claim 6 in which the force is a bending force, perpendicular to the axis (16) of the blade (14); said bending force being applied by exerting pressure on a support zone (80) of the blade, said support zone being arranged between the first (20) and the second (22) gripping zone of said part.

8. Method according to claim 6 in which: the force is a torsional torque around the axis (16) of the blade (14); one of the gripping elements and the second portion of the part is held fixed relative to a frame (30); and the torsional force is applied by pivoting, relative to the frame, the other of said first and second gripping zones.

9. A method according to claim 6 wherein: the force is a torsional torque about the axis (16) of the blade (14); and the torsional force is applied by pivoting, relative to each other, the first (20) and the second (22) gripping zones.

10. The method of claim 1 wherein the part further comprises an identifier (104), the method further comprising the steps of: reading the identifier; identifying the nominal definition (112) corresponding to the piece of equipment (12) from among a plurality of de- nominal finishes; and identifying the self-learning algorithm (82) corresponding to the piece of equipment (20) from among a plurality of self-learning algorithms.

11. Assembly (90) for implementing a method according to one of the preceding claims, comprising: - a measuring device (91), capable of acquiring three-dimensional data of the part (12), in the initial conformation and / or in the deformed conformation; - an installation (10) capable of applying the force to the part (12); and - an electronic device (94) comprising a data memory (96) and at least one self-learning algorithm (82) trained on a plurality of parts of the same type as the part (12), said measuring device (91), the installation (10) and the electronic device (94) being connected by at least one communication channel.

Images