TURBOMACHINE PART ADJUSTMENT
An automated method and system for turbine engine parts address the inefficiencies of manual material removal by generating tailored trajectories for robotic adjustment, reducing time and musculoskeletal risks while maintaining part geometry integrity.
Patent Information
- Application Number
- FR2023000793
- Authority / Receiving Office
- FR · FR
- Patent Type
- Patents
- Current Assignee / Owner
- Filing Date
- 2023-01-27
- Publication Date
- 2025-12-26
- Estimated Expiration
- 2043-01-27
AI Technical Summary
The manual removal of excess material from turbine engine parts after material build-up is time-consuming and prone to causing musculoskeletal disorders, and existing methods are inefficient in adapting to the variability of part geometries.
A method and system for automatically generating a material removal trajectory based on a three-dimensional image of the part, identifying excess thickness, and controlling an adjustment tool to remove excess material while preserving the part's original geometry, using a robotic system and computer-aided processes.
Reduces the time required for material removal operations and minimizes musculoskeletal disorders by providing an efficient, automated process that adapts to the unique geometry of each part, ensuring precise and safe material adjustment.
Smart Images
Figure 00000013_0000 
Figure 00000013_0001 
Figure 00000014_0000
Abstract
Description
Title of the invention: TURBOMACHINE PART FITTING technical field
[0001] The present invention relates generally to the repair of turbine engine parts. More particularly, it relates to the removal of material, or adjustment, carried out after material build-up to repair a turbine engine part. PREVIOUS STATE OF THE ART
[0002] In particular, high-pressure turbine components are considered, notably the distributors that form part of the turbine stator. These components are made of nickel superalloy and are subjected to very high stresses during flight phases. These components operate at very high temperatures, up to approximately 1100°C, and are subjected to thermal fatigue stresses.
[0003] These stresses cause wear on these parts, resulting in the formation of cracks on the surface of the parts.
[0004] Repairing the cracks involves adding material by diffusion brazing to fill the gaps, followed by removing excess brazing material to restore the part's original aerodynamic profile. Currently, this material removal, or adjustment, operation is performed manually. It takes approximately one and a half hours per part.
[0005] This operation is therefore time-consuming, and it can cause significant musculoskeletal disorders among operators.
[0006] FR 3 116 456 describes a method for repairing a turbomachine part in which a worn area of a part, in particular the worn leading edge of a blade of a one-piece bladed disc, is machined. Description of the invention
[0007] The invention aims to solve the problems of the prior art by providing a method for fitting a turbomachine part to be placed in a gas stream of the turbomachine, the part having previously undergone a material addition operation to fill a defect in at least one area of its surface, resulting in an excess thickness in at least one area of its surface, characterized in that it comprises the steps of:
[0008] - command to form a three-dimensional image of the part,
[0009] - identification of at least one area of the surface of the part including the on thickness, and identification of excess thickness, in the three-dimensional image of the piece,
[0010] - determination of a trajectory of an adjustment tool, as a function of the geometry of the identified excess thickness, and
[0011] - control for adjusting the part according to the determined trajectory.
[0012] Thanks to the invention, it is possible to automatically generate a material removal trajectory adapted to each part based on the excess material to be removed. The invention allows for adaptation to the variability of the input condition of the parts to be adjusted. Indeed, the position of the excess material on the part is random. Furthermore, the invention makes it possible to avoid altering the sound material of the part.
[0013] The invention is particularly suitable for parts with complex geometry such as high-pressure turbine parts for aircraft engines.
[0014] The invention makes it possible to limit the occurrence of significant musculoskeletal disorders in operators and frees them from time-consuming operations in favor of other operations with higher added value.
[0015] According to a preferred feature, the step of identifying Water less an area of the surface of the part including the excess thickness, and identifying the excess thickness, comprises:
[0016] - a pre-detection of irregularities on the surface of the part, then, in the case of the pre- detection of at least one irregularity,
[0017] - a data filtering representing at least one irregularity to produce area data in which data representing at least one irregularity have been removed, referred to as healthy area data,
[0018] - a surface reconstruction performed by a function having a continuous slope, at starting from healthy area data to produce reconstructed surface data corresponding to a surface without excess thickness,
[0019] - a comparison of the actual surface data of the part with the data of the reconstructed surface, and, depending on the result of the comparison, an identification of the excess thickness.
[0020] According to a preferred feature, the three-dimensional image of the part includes a mesh that represents the real part composed of mesh elements and the pre-detection of irregularity on the surface of the part includes the identification of first mesh elements whose curvature exceeds a curvature threshold as irregularities.
[0021] According to a preferred feature, the pre-detection of irregularity on the surface of the part further includes the identification of second mesh elements close to the first mesh elements over a predetermined number of rows.
[0022] According to a preferred feature, the function having a continuous slope is defined by a polynomial equation.
[0023] According to a preferred feature, the step of identifying at least one area of the part's surface, including the excess thickness, and identifying the excess thickness, involves a fit between a model of the part and the three-dimensional image of the part. This feature is implemented in the case of a highly curved area, in addition to the features described above.
[0024] According to a preferred feature, determining a trajectory of an adjustment tool involves comparing a predetermined trajectory with the coordinates of the points of the identified over-thickness, so as to retain only the points of the programmed trajectory whose distance with the points of the defect is less than a comparison threshold.
[0025] According to a preferred feature, the part adjustment control along the determined trajectory includes a detection of a first contact of the tool with the over-thickness and the control of the tool so that it performs a progressive descent by successive passes.
[0026] The invention also relates to a system for adjusting a turbomachine part to be placed in a gas stream of the turbomachine, the part having previously undergone a material addition operation to fill a defect in at least one area of its surface, resulting in an excess thickness in at least one area of its surface, the adjustment system comprising a three-dimensional measuring device and an adjustment machine, the adjustment system being characterized in that it comprises a control module capable of:
[0027] - to control the three-dimensional measuring device to form a three-dimensional image dimension of the piece,
[0028] - identify at least one area of the surface of the part including the excess thickness, and identify the excess thickness in the three-dimensional image of the part,
[0029] - determine a trajectory of an adjustment tool, as a function of the geometry of the excess thickness identified, and
[0030] - to control the fitting machine to perform the fitting of the part according to the tra determined jectory.
[0031] The system has advantages similar to those previously presented.
[0032] In a particular embodiment, the steps of the process according to the invention are implemented by computer program instructions.
[0033] Consequently, the invention also relates to a computer program on an information medium, this program being capable of being implemented in a computer, this program comprising instructions adapted to the implementation of the steps of a process as described above.
[0034] The invention also relates to a computer-readable information carrier comprising computer program instructions adapted for implementation steps in a process such as the one described above. Brief description of the drawings
[0035] Other features and advantages will become apparent from the following description of a preferred embodiment given by way of non-limiting example, described with reference to the figures in which:
[0036] [Fig-1] illustrates a mechanical part adjustment system according to a method of preferred interpretation of the invention,
[0037] [Fig.2] illustrates the method of adjusting a mechanical part according to a preferred embodiment of the invention,
[0038] [Fig.3] illustrates a step of identifying at least one area of the surface of the part including an excess thickness, and identification of the excess thickness, included in the process of [Fig.2],
[0039] [Fig.4a]
[0040] [Fig.4b]
[0041] [Fig.4c]
[0042] [Fig.4d] illustrate a surface reconstruction step, included in the process of [Fig.2], and
[0043] [Fig.5] illustrates an example of a trajectory generated according to a preferred embodiment of the invention.
[0044] Identical, similar or equivalent parts of the different figures bear the same numerical references so as to facilitate the transition from one figure to another.
[0045] The different parts represented in the figures are not necessarily shown on a uniform scale, in order to make the figures more legible.
[0046] The different possibilities (variants and embodiments) should be understood as not being mutually exclusive and can be combined with each other.
[0047] DETAILED DESCRIPTION OF SPECIFIC EMBODIMENTS
[0048] According to a preferred embodiment shown in [Fig. 1], a mechanical part fitting system is, for example, intended to fit parts to be placed in a gas stream of the turbomachine. The mechanical part is a high-pressure turbine component, in particular a distributor that is part of the turbine stator.
[0049] Consider a mechanical part 1 that has previously undergone a material addition operation due to the formation of cracks on its surface. Adding material to the part results in an excess thickness compared to the part's initial geometry. The purpose of the adjustment is to eliminate this excess thickness and restore the part to its initial geometry, or at least to give it a suitable aerodynamic profile, without localized excess thickness.
[0050] The adjustment system includes a support 2 adapted to receive and hold the Part to be adjusted 1. The support can be mobile so as to be able to provide different positions for the part to be adjusted.
[0051] The fitting system also includes a fitting machine 3 which itself includes a robotic arm 31 configured to be able to move along several axes.
[0052] The robotic arm 31 is adapted to be equipped with a fitting tool 32. Preferably, several fitting tools can be fitted to the robotic arm, for example, tools with abrasives whose surface has grains of different particle sizes, so as to perform different tasks such as cutting, grinding, sanding, and smoothing. Preferably, an automatic tool changer 33 is fitted to the fitting machine 3 so as to equip the robotic arm 31 with the most suitable fitting tool 32.
[0053] The adjustment system also includes a three-dimensional measuring device 4, for example by photogrammetry. The three-dimensional measuring device 4 is capable of performing a three-dimensional measurement of the part to be adjusted 1.
[0054] The adjustment system also includes a control module 10 that controls the various elements of the adjustment system described above. The control module 10 has the general structure of a computer. In particular, it includes a processor 100 executing a computer program implementing the method according to the invention, a memory 101, input interfaces 102, and output interfaces 103.
[0055] These different elements are conventionally connected by a bus 105.
[0056] The input interfaces 102 are connected to the various elements described above and are intended to receive information from them.
[0057] Processor 100 executes the processes that will be described below. These processes are carried out in the form of computer program code instructions that are stored in memory 101 before being executed by processor 100.
[0058] Memory 101 can also store the results of the treatments performed.
[0059] The output interfaces 103 are connected to the various elements described above and are intended to send instructions to them.
[0060] To simplify the representation, the various mechanical and electrical connections of the elements described above have not been shown in [Fig.1].
[0061] The operation of the adjustment system is ensured by the control module 10 and is detailed in the form of a process described below.
[0062] Figure 2 illustrates the mechanical part fitting process according to an embodiment of the invention. The fitting process comprises steps E1 to E5 implemented by the control module 10 described above.
[0063] It is assumed that a part 1, such as a high-pressure turbine stator distributor of an aircraft engine, has previously undergone a material addition operation due to the formation of cracks on its surface. The material addition results in an increased thickness compared to the surface of the original part. This increased thickness constitutes a defect in the part.
[0064] It is also assumed that part 1 has been fixed to support 2. This is the part to be adjusted.
[0065] Step El is a command to the three-dimensional measuring device 4 to measure the part to be fitted 1, for example by photogrammetry, and form a three-dimensional image of it.
[0066] To achieve this, the part is scanned and then polygonized. This produces a mesh that represents the actual part composed of basic elements, here of the triangular type. The theoretical profile of the surface considered is continuous, which implies that the mesh elements composing it have essentially the same characteristics.
[0067] The three-dimensional image data thus produced is stored.
[0068] The next step E2 is a three-dimensional image processing to identify the area(s) of part 1 that contain added material, i.e., excess thickness. The excess thickness itself is also identified. Correspondingly, areas without excess thickness are also identified. In the subsequent processing, only the areas with excess thickness are processed, and the areas without excess thickness are excluded from the processing and left untouched. In other words, the processing is limited to the areas with excess thickness.
[0069] Step E2 is described with reference to Figures 3 and 4a to 4d and comprises substeps E21 to E24.
[0070] Figures 4a to 4d represent a search area for a sound zone containing an excess thickness. The zone has, for example, a diameter between 10 and 20 mm, for example 15 mm. Indeed, among all the types of solder excess thickness, none has dimensions greater than 15 mm in all directions. Such a search area size therefore ensures that sound zones are systematically found. This size also allows for a balance to be struck between the following factors: processing time (non-linear), overlap between different zones to ensure no zone is missed, and the resulting duplicate analyses (time-consuming redundancy).
[0071] Substep E21 is a pre-detection of irregularities on the surface of the mechanical part. An irregularity is shown in [Fig. 4a]. This may be an excess thickness to be treated or another irregularity not addressed within the scope of the invention, such as a cooling hole present in the air channel.
[0072] To this end, the curvature of the mesh elements is analyzed. During a In the first selection, mesh elements whose curvature exceeds a configurable curvature threshold are identified as irregularities. The curvature threshold varies depending on the areas of the analyzed part.
[0073] Preferably, a second selection is made to also select the mesh elements near the mesh elements identified as irregularities, over a number of ranks which is for example 5. The number of ranks is determined empirically and can be parameterized.
[0074] The set of selected mesh elements is the set of pre-detected irregularities. It is assumed in the following that at least one irregularity has been detected.
[0075] The next substep E22 is a filtering of pre-detected irregularities with the aim of removing them from the three-dimensional image data.
[0076] Filtering irregularities results in surface area data in which irregularity data have been removed ([Fig.4b]).
[0077] The surface area data after defect filtering thus identifies a so-called healthy area.
[0078] The next substep E23 is a surface reconstruction at the level of the area of the part's surface that has been removed, using the surrounding healthy area on either side. The reconstruction is performed by a function having a continuous slope, for example, defined by a polynomial equation. The polynomial equation is, for example, of order 5 or higher, which in practice allows for obtaining good quality results.
[0079] Thus, the reconstruction uses the healthy area surface data determined in the previous step and results in reconstructed surface data corresponding to a defect-free surface ([Fig.4c]).
[0080] The next substep E24 is a comparison of the actual surface data of the part with the reconstructed surface data in the area considered ([Fig.4d]).
[0081] The comparison makes it possible to isolate the portion of the mesh corresponding to the excess thickness. This step allows for the identification and characterization of the excess thickness located in the area under consideration.
[0082] For a given overthickness, the result of step E2 is a data set representing the overthickness, called overthickness data, and a data set representing the area including the overthickness. The data are typically point coordinates. For example, the overthickness data are the point coordinates of the given overthickness. It is possible to associate an overthickness value with each point considered.
[0083] It should be noted that in certain highly curved areas, and for patch-type repairs, the curvature of the basic elements (elementary triangles) is already significant even in the absence of irregularities. In this case, the use of the curvature of the elements The basic mesh used for pre-detecting irregularities on the part surface (substep E21) may not be optimal. For these areas, an alternative method for identifying and characterizing excess thickness (step E2) can be a fit between a part model and the three-dimensional image of the part. This fit can be performed between a portion of the part model (corresponding to the theoretical geometry of the part, as given by its CAD representation) and a portion of the actual measured surface. Such a local (rather than global) fit eliminates potential distortions of the part relative to its theoretical CAD model (in particular, relative distortions of one part of the part with respect to another, for example, the driving blade with respect to the driven blade). However, this method is less effective on welds and prominent defects in air channels.In the example process described here, for the most curved areas of the part (leading edge for example), this alternative method is used instead of the surface reconstruction method presented above with reference to figures 4a to 4b.
[0084] Step E2 is followed by step E3, which is the determination of a trajectory for the fitting tool 32. Step E3 is described for an over-thickness detected, or defect, in a given area, and is repeated for all over-thicknesses detected on the part considered.
[0085] The part was previously processed by CAD / CAM software to produce a programmed trajectory. The programmed trajectory is a machining trajectory that, after machining, would lead to a surface corresponding to the surface of the theoretical CAD model of the part.
[0086] The programmed trajectory is compared with the coordinates of the defect points in the area under consideration. Each point on the programmed trajectory and each point on the defect are compared, using a comparison threshold, so as to retain only those points on the programmed trajectory whose distance from the defect points is less than the comparison threshold. The threshold value is, for example, between 0.1 and 0.3 mm.
[0087] It is possible to perform local alignments between a portion of the CAD geometry and the actual geometry in order to compensate for any deformations that might cause discrepancies between the theoretical CAD geometry and the actual geometry of the part, in particular relative deformations of one part of the part with respect to another part, and thus reposition the toolpath. This improves the accuracy of the tool's positioning relative to the part.
[0088] Alternatively, the CAM calculation of the tool trajectory is performed on the basis of the reconstructed surface (without excess thickness), obtained at the end of sub-step E23 and which is based on the actual surveyed geometry rather than on the theoretical CAD basis. The calculation complexity is greater, but the accuracy is increased since this makes it possible to eliminate any possible deformations of the part.
[0089] The points thus selected form a trajectory of the adjusting tool, as illustrated in [Fig. 5]. [Fig. 5] represents a portion of the surface of a mechanical part containing defects. For one of the defects, the generated trajectory is represented by dashed lines.
[0090] The next step E4 is a command to the fitting machine 3 to perform the fitting of the part according to the determined trajectory of the fitting tool 32.
[0091] Optionally, particularly when the programmed toolpath is obtained based on the theoretical CAD model of the part and without local registration, the initial contact of the tool with the excess material can be detected, for example, with a wattmeter. The tool is then commanded to perform a gradual descent, in successive passes, for example with a vertical increment of between 0.01 and 0.05 mm, or even between 0.02 and 0.03 mm. This contact detection followed by gradual descent eliminates errors in the programmed toolpath due to deformations of the actual part compared to its theoretical CAD model.
[0092] Step E4 involves selecting an adjustment tool 32 and controlling the tool changer 33 to equip the adjustment arm 31 with the selected adjustment tool 32.
[0093] Step E4 may involve the successive use of two fitting tools, a first tool being used to perform a coarse material removal and a second tool being used to perform a gentle abrasion to produce a smooth surface. Of course, the number of successive tools may exceed two, each tool producing a finer abrasion than the previous one.
[0094] The next step E5 is an inspection of the machined part.
[0095] According to a first variant, the control is carried out visually by an operator, for example during a manual finishing operation which follows the machining.
[0096] According to a second embodiment, the three-dimensional measuring device 4 is controlled to measure the part 1 after fitting, for example by photogrammetry, and forms a three-dimensional image of it. This step is similar to step EL. The image formed is analyzed to verify the conformity of the fitting result with an expected result.
Claims
Demands
1. A method for fitting a turbomachine part to be placed in a turbomachine gas stream, the part having previously undergone a material addition operation to fill a defect in at least one area of its surface, resulting in an excess thickness in at least one area of its surface, characterized in that it comprises the steps of: - command (E1) to form a three-dimensional image of the part, - identification (E2) of an area of the surface of the part including the excess thickness, and identification of the excess thickness, in the three-dimensional image of the part, - determination (E3) of a trajectory of a fitting tool, as a function of the geometry of the identified excess thickness, and - command (E4) to fit the part according to the determined trajectory.
2. A method for fitting a part according to claim 1, wherein the step (E2) of identifying water less an area of the part surface including the excess thickness, and identifying the excess thickness, comprises: - A pre-detection (E21) of irregularities on the part surface, then, in the case of pre-detection of at least one irregularity, - a filtering (E22) of data representing the at least one irregularity to produce area data in which the data representing the at least one irregularity have been removed, referred to as healthy area data, - a surface reconstruction (E23) performed by a function having a continuous slope, from the healthy area data to produce reconstructed surface data corresponding to a surface without excess thickness, - a comparison (E24) of the data of the actual part surface with the data of the reconstructed surface, and, depending on the result of the comparison,an identification of the excess thickness.
3. A method for fitting a part according to claim 2, wherein the three-dimensional image of the part includes a mesh that represents the actual part composed of mesh elements and the pre-detection of irregularity on the surface of the part includes the identification of first mesh elements whose curvature exceeds a curvature threshold as irregularities.
4. A method for fitting a part according to claim 2, wherein the pre-detection (E21) of irregularity on the surface of the part further comprises the identification of second mesh elements close to the first mesh elements over a predetermined number of rows.
5. A method for fitting a part according to any one of claims 2 to 4, wherein the function having a continuous slope is defined by a polynomial equation.
6. A method for fitting a part according to claim 1, wherein the step (E2) of identifying at least one area of the surface of the part including the excess thickness, and identifying the excess thickness comprises a fit between a model of the part and the three-dimensional image of the part.
7. A method for fitting a part according to any one of claims 1 to 6, wherein the determination (E3) of a trajectory of a fitting tool involves comparing a predetermined trajectory with the coordinates of the points of the identified overhang, so as to retain only the points of the programmed trajectory whose distance with the points of the defect is less than a comparison threshold.
8. A method for fitting a part according to any one of claims 1 to 7, wherein the control (E4) for fitting the part along the determined trajectory includes a detection of a first contact of the tool with the over-thickness and the control of the tool to make a progressive descent by successive passes.
9. A fitting system for a turbomachine part to be placed in a gas stream of the turbomachine, the part having previously undergone a material addition operation to fill a defect in at least one area of its surface, resulting in an excess thickness in at least one area of its surface, the fitting system comprising a three-dimensional measuring device (4) and a fitting machine (3), the fitting system being characterized in that it comprises a control module (10) capable of: - controlling the three-dimensional measuring device (4) to form a three-dimensional image of the part, - identifying an area of the surface of the part including the excess thickness, and identifying the excess thickness, in the three-dimensional image of the part, - determining a trajectory of a fitting tool, as a function of the geometry of the identified excess thickness, and - order the fitting machine (3) to perform the fitting of the part according to the determined trajectory.
10. A computer program comprising instructions for carrying out the steps of the process according to any one of claims 1 to 8 when said program is executed by a computer.
11. Computer-readable recording medium on which is recorded a computer program comprising instructions for carrying out the steps of the process according to any one of claims 1 to 8.