Method for determining damage of a mechanical component by displacement measurements using stereoscopic
By measuring the displacement and deformation of mechanical parts using image stereo correlation technology, the problem of accuracy in identifying damage to mechanical parts was solved, enabling the determination of the priority direction of damage and the differentiation of healthy areas.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- SAFRAN AIRCRAFT ENGINES SAS
- Filing Date
- 2024-08-26
- Publication Date
- 2026-04-17
AI Technical Summary
Existing technologies struggle to accurately identify damage to mechanical components and their preferred direction during mechanical testing, and also have difficulty distinguishing between healthy areas and damaged states of materials.
The displacement of mechanical components in space is measured using stereo correlation techniques of images, the deformation distribution is calculated, and the preferred orientation is determined. This is then compared with reference elements to determine damage.
It enables precise location and priority direction identification of damage to mechanical components under mechanical stress, improving the accuracy and reliability of mechanical testing.
Smart Images
Figure CN121889639A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to a method for determining damage to a mechanical component subjected to mechanical stress based on data obtained through stereo correlation of digital images. The invention also relates to an apparatus that enables the implementation of this method. Background Technology
[0002] Stereo correlation of digital images is a known technique that enables the measurement of motion fields, such as displacement or deformation fields. This technique involves matching two digital images of an observed surface in two different deformation states (the so-called reference state and the so-called deformed state). Each pixel in the first image is compared with the corresponding pixel in the second image, and their similarity is measured. This is a non-destructive surface inspection technique.
[0003] Documents US-A1-2013 / 06570, FR-A1-3 127 042, and WO-A1-2016 / 076975 are examples of image stereo correlation applications.
[0004] When conducting mechanical testing, it is known to apply black and white patches to the surface of the part under study to provide a sufficiently random texture that can be tracked by the camera used. These cameras are typically equipped with charge-coupled device (CCD) sensors, the acquisition frequency of which depends on the phenomenon being studied. For cases with fast or cyclic dynamics, this can vary from a few frames per second to thousands of frames per second.
[0005] Digital image stereo correlation techniques can be used during mechanical testing, such as static testing. This testing can be used to characterize material properties or check material quality. It can also be used to test and verify the mechanical strength of mechanical components.
[0006] During these static mechanical tests, it is necessary to determine the conditions under which static damage occurs. Damage refers to any form of deterioration in the properties of a material, which may be related to, for example, the appearance of microcracks at the level of fibers, strands, braided planes, multiple braided planes, or the entire material; delamination of the material at different scales; or, in the case of composite materials, fiber delamination from the matrix or fiber breakage. For example, in the case of metallic materials, this could be microcracks between grain boundaries or cracks within a single crystal.
[0007] Typically, deformation along the direction of mechanical stress is analyzed to derive the limit value of deformation at the surface of a mechanical component.
[0008] However, the results obtained remain to be interpreted because there are no indicators to distinguish the end of the so-called healthy zone of a material from the transition to a "damaged" state. Furthermore, even when damage is known to exist in the material of a component, it is sometimes difficult to clearly identify the preferred direction of stress, whether it is the fiber direction of a composite material or the necking direction of a metallic material.
[0009] This direction is an important factor to consider in order to verify that the tests performed do indeed apply stress in the direction or area we actually want to study during the testing period.
[0010] Therefore, one object of the present invention is to provide an improved method for determining damage during improved mechanical testing.
[0011] Another object of the present invention is to provide a method for identifying one or more preferred directions of damage during mechanical testing. Summary of the Invention
[0012] A method for determining damage to a mechanical component subjected to mechanical stress is proposed, the method comprising the following steps: a) Use stereo correlation of images to measure the displacement of the mechanical component in space along the x, y, and z directions within a given area of the component; then, the following steps are performed by a computer. b) Based on the displacement measured in step a), calculate the deformation of the component in the given area; c) Show the distribution of deformation defined along the y direction as a function of deformation defined along the x direction in the reference frame; d) Determine the region of the deformation distribution shown in step c); e) Determine the preferred orientation of the distribution of the shown deformations; and f) Compare the determined distribution area and / or preferred orientation with a reference element to determine if the component is damaged.
[0013] Therefore, due to the method described in this invention, damage caused by mechanical stress on a component can be studied, particularly the orientation that the damage tends to follow during stress application. In fact, using stereo correlation of images and subsequent processing to measure displacement makes it possible to track deformation during stress application to the component, thereby obtaining the evolution of the damage orientation.
[0014] The method according to the invention may include one or more of the following features, which may be employed individually or in combination with each other: -The distribution area is polygonal; - In step d), the distribution area is determined in real time; -In step d), the distribution area is determined at a predetermined time; - In step e), the preferred orientation is determined based on the principal deformation; - In step e), the preferred orientation of the distribution of deformations is performed by linear regression; - In step e), the preferred orientation of the deformation distribution is determined by statistical methods; - Mechanical components are aircraft parts, particularly turbine housings, and preferably turbine fan housings.
[0015] The present invention also relates to an apparatus for carrying out the method for determining damage to a mechanical component as described above, the apparatus comprising: - A device for measuring the displacement of a mechanical component over a given region using stereo correlation of images, and - A means for performing at least steps b) to f) by means of a computer.
[0016] The present invention also relates to an aircraft comprising at least one device for implementing the method described above for determining damage to mechanical components. Attached Figure Description
[0017] The invention will be better understood through the following description, given only by way of example and with reference to the accompanying drawings, in which: Figure 1 A schematic diagram of a test bench according to the present invention is shown, the test bench including mechanical components and equipment for implementing a method for determining damage to the mechanical components.
[0018] Figure 2 A schematic diagram of a mechanical component (particularly a mechanical component made of woven composite material) is shown.
[0019] Figure 3 This is a block diagram of a process for determining damage to mechanical parts according to the present invention.
[0020] Figure 4 It is a graphical representation of the distribution of deformation along the y direction as a function of deformation along the x-direction at different times during mechanical testing.
[0021] Figure 5 It is a graphical representation of the distribution of deformation along the y direction as a function of deformation along the x-direction, and its preferred orientation is determined. Detailed Implementation
[0022] Figure 1 A test stand D for mechanical component 10 is shown, which, among other things, includes a camera C for acquiring digital images by stereo correlation (enabling displacement measurement) and a device 20 equipped with a computer device 20 for implementing the method for determining damage to the mechanical component as described below.
[0023] Patches M are also applied to the surface of mechanical component 10 to make it easier for camera C to track.
[0024] The camera C used for image acquisition is a CCD camera with a capture frequency of a few frames per second. The acquisition area is illuminated by LED lights. Furthermore, the illumination can be continuous or synchronized with camera C.
[0025] The following examples relate to mechanical parts, particularly those made of composite materials (more specifically, 3D braided composite materials). However, these mechanical parts can also be 2D braided composite parts, unidirectional composite parts, or metal parts.
[0026] Figure 2 A mechanical component 10 made of a 3D woven composite material is shown. The composite material includes weft fibers 11 and warp fibers 12 woven together on a woven plane P or multiple woven planes stacked along the thickness direction. The weft fibers 11 and warp fibers 12 can be, for example, carbon, glass, or a mixture of both, or more generally other types of fibers. These fibers are embedded in a matrix 13, which contains one or more polymers capable of being cured to form the final mechanical component 10. The matrix 13 can be, for example, selected from thermosetting resins or thermoplastic resins.
[0027] More complex weave structures can also be used in these tests. For example, the warp fibers 12 are no longer limited to a single weave plane P, but instead pass through multiple planes of the material.
[0028] In fact, component 10 can be a mechanical part of an aircraft. In particular, the component can be a turbine housing, and for example, the housing of a turbine fan.
[0029] During mechanical testing, mechanical component 10 is subjected to stresses that may cause deformation or even damage to component 10.
[0030] The method according to the invention enables the determination of damage to a component 10 subjected to mechanical stress, and the method includes the following steps: a) Using stereo correlation of the image, measure the displacement of the component in space along the x, y, and z directions over a given area of component 10; then, the following steps are performed by a computer. b) Calculate the deformation of the component in the given area based on the displacement measured in step a); c) Show the distribution of deformation defined along the y direction as a function of deformation defined along the x direction in the reference frame; d) Determine the region of distribution of the deformation shown in step b); e) Determine the preferred orientation of the distribution of the shown deformations; and f) Compare the obtained preferred orientation and / or distribution area with a reference element to determine whether part 10 is damaged.
[0031] The various stages of the method 100 according to the present invention are as follows: Figure 3 As shown in the figure, and described in more detail below.
[0032] In step a) Figure 3 In section 102, stereo correlation of the image is used to measure the displacement of component 10 in a given area of component 10 along the x, y, and z directions in space. These directions are orthogonal. The operator is free to choose the area to be observed. In practice, these measurements are performed using a camera C equipped with a CCD sensor.
[0033] After step a), the measurement results are sent to a computer device 20 capable of performing the following steps.
[0034] In step b) Figure 3 In step 104 (marked as 104), the deformation of component 10 in a given region is calculated based on the displacement measured in step 104. Specifically, the deformation is calculated based on the x, y, and z displacement fields obtained using stereo correlation of the image.
[0035] These deformations are typically calculated using the following equations: [Mathematical Expression 1]
[0036] Where ε is a tensor, U i and U j It is a displacement field.
[0037] In step c) Figure 3 In the reference frame (marked as 106), the calculated deformation is shown, such that the deformation (ε) is obtained as defined along the x-direction. xx The y-direction-bounded deformation of the function of ε yy The distribution of ). Note that deformation (ε) defined along the z-direction can also be obtained. xy These deformations represent shearing.
[0038] In step d) Figure 3 In step c), the region of the distribution of the deformation shown in step 108 is determined. This determination of the distribution region can be performed in real time or at a predetermined time.
[0039] For example, Figure 4 It is a graphical representation of the distribution of deformation defined in the y-direction as a function of the deformation defined in the x-direction of component 10 at different times during mechanical stress. Figure 4A shows the distribution 30 of the deformation before the test began. All points are concentrated, and the area of distribution 30 is relatively small. Figure 4 B shows the deformed distribution 30' during the test. These points remain concentrated, but they are more dispersed than before the experiment began. Furthermore, the area of distribution 30' increases but remains relatively small. Figure 4 C shows the distribution of deformation 30'' after the application of mechanical stress. At this stage, the points appear more dispersed, and the area distributed 30'' is larger than in the previous stage. Figure 4 A and Figure 4 A larger value in B indicates that damage to the component has already occurred.
[0040] These deviations in the distribution can be represented by an estimated region of polygon 31 that covers the entire distribution of the deformed points, such as... Figure 4 C or Figure 5 As shown.
[0041] Alternatively, not shown, the deviation in the distribution can be represented by an estimated region of an elliptical structure (particularly as part of an approximation of the distribution).
[0042] In step e) Figure 3 In step d), the preferred orientation is determined based on the deformation distribution shown in step d). This preferred orientation can be determined by the deformation (ε) defined along the x-direction. xx The y-direction-bounded deformation of the function of ε yy The distribution of the deformation points is determined by linear regression.
[0043] Alternatively, the preferred orientation can be determined based on multiple principal deformations (e.g., a first principal deformation and a second principal deformation). These principal deformations can be determined, for example, by the Mohr circle. The use of principal deformations particularly facilitates changing the reference frame and simplifys calculations. Linear regression can also be performed to determine the preferred orientation.
[0044] Typically, using linear regression to determine priority allows the results to be summarized using the slope as the sole parameter. This also simplifies the method and reduces computation time.
[0045] For the two variants mentioned, preference can also be determined using statistical methods (especially based on standard deviation).
[0046] Similar to Figure 4 C, Figure 5 The deformation (ε) defined along the y direction is shown as a function of the deformation defined along the x-direction. yy An example of distribution 30'' is given, where the region of distribution has been determined and represented by polygon 31. In this example, statistical methods are used to determine the preference orientation 32 of distribution 30''.
[0047] This preference orientation indicates the direction in which damage to mechanical component 10 tends to propagate.
[0048] In the specific case where component 10 is made of 3D woven composite material, as described above, determining the preferred orientation allows for checking whether the material is already under stress along the warp or weft direction. This information is essential for the design of the composite material. In practice, the orientation direction is selected based on future stress.
[0049] As mentioned earlier, as an alternative, multiple principal deformations can be used, particularly a first principal deformation that is a function of the second principal deformation, the first and second principal deformations being determined, for example, by the Mohr circle. In this case, the same method can be used to obtain and process [the results]. Figure 4 and Figure 5 The graphic shown represents a similar graphic representation (not shown here).
[0050] In step f) Figure 3 In section 112), the previously determined distribution area and / or preferred orientation are compared with reference elements to determine whether component 10 is damaged. These reference elements may be study pieces that undergo additional testing to enable verification of the quality of the reference elements or intentional quality defects.
[0051] This comparison validates that the applied mechanical stress corresponds to the actual expected deformation and / or damage of the tested mechanical component.
[0052] The methods described above enable the study of damage caused by mechanical stress on a component. In fact, using stereo correlation of images and subsequent processing to measure displacement allows for tracking deformation during loading of the component, thereby obtaining the evolution of the orientation that the damage tends to follow during stress.
[0053] The present invention also provides an apparatus D for implementing the method 100 described above. The apparatus D includes means C for measuring the displacement of mechanical component 10 by stereo correlation of images over a given area, and means 20 for implementing at least steps b) to f) by a computer.
[0054] The present invention also provides an aircraft comprising at least one device D for implementing the method 100 described above. Thus, device D is mounted in the aircraft on at least one component to be inspected. In this way, the physical condition and / or damage of one or more components to be inspected can be directly assessed during flight of the aircraft, thereby detecting any deformation, cracks, and / or damage as soon as they occur and taking appropriate action.
Claims
1. A method (100) for determining damage to a mechanical component (10) subjected to mechanical stress, the method comprising the steps of: a) The displacement of the component (10) in a given region of the component along the x, y, and z directions in space is measured by image stereo correlation (102); then, the following steps are performed by a computer. b) Based on the displacement measured in step a), calculate (104) the deformation of the component in the given region; c) In the reference frame, (106) is shown as a deformation (ε) defined along the x-direction. xx The deformation (ε) of the function of ) along the y-direction yy The distribution of ); d) Determine the region of the distribution of deformation shown in step c) in (108); e) Determine the preferred orientation of the distribution of the deformation shown in (110); and f) Compare the previously determined distribution area and / or the preferred orientation with a reference element (112) to determine whether the component is damaged.
2. The method according to claim 1, wherein, The distribution area is polygonal.
3. The method (100) according to claim 1, wherein, In step d), the distribution area is determined in real time.
4. The method (100) according to any one of claims 1 to 3, wherein, In step d), the distribution area is determined at a predetermined time.
5. The method (100) according to any one of claims 1 to 4, wherein, In step e), the preferred orientation is determined based on the principal deformation.
6. The method (100) according to any one of claims 1 to 5, wherein, In step e), the preferred orientation of the distribution of the deformation is determined by linear regression.
7. The method (100) according to any one of claims 1 to 5, wherein, In step e), the determination of the preferred orientation of the distribution of the deformation is performed by statistical methods.
8. The method (100) according to any one of claims 1 to 7, wherein, The mechanical component (10) is an aircraft component, particularly a turbine housing, preferably a turbine fan housing.
9. An apparatus (D) for carrying out the method (100) according to any one of claims 1 to 8, comprising: - A device (C) for measuring the displacement of the mechanical component (10) in the defined area using stereo correlation of an image, and -A device (20) for performing at least steps b) to f) by means of a computer.
10. An aircraft comprising at least one device (D) according to claim 9.