Overall open type bladed disc blade three-dimensional geometric contour detection device and method

Abstract

The invention discloses a device and a method for detecting the three-dimensional geometric contour of a blade of an integral open type bladed disc, and aims to solve the problems of difficulty in obtaining integral characteristics, insufficient reliability and poor flexibility when the conventional detection device is used for detecting the three-dimensional geometric characteristics of the integral open type bladed disc. In the aspect of hardware, the device comprises a structured light detection system, a linear module, a supporting frame, a rotating mechanism and a leaf disc to be detected, the structured light detection system is fixedly connected with the supporting frame, the linear module is vertically and fixedly connected to the supporting frame, the rotating mechanism is fixedly connected to the supporting frame, and the supporting frame is connected with the ground through four foot cups. In the aspect of software, the optical axis attitude is optimized and adjusted by adopting a view field overlapping rate, a dynamic reprojection error optimization mechanism is introduced, and calibration parameters are corrected in real time; in the self-adaptive angle adjustment, according to the visual field overlapping rate and the point cloud confidence coefficient of the leaf disc fan-shaped area, the collection angle is automatically supplemented, and multi-visual-angle point cloud coverage and high-precision splicing are realized, so that the integrity and reconstruction precision of the overall three-dimensional contour are ensured.

Description

Technical Field

[0001] This invention relates to a three-dimensional detection device and method for blades of an impeller disk, and more specifically to a three-dimensional geometric contour detection device and method for blades of an integral open impeller disk. Background Technology

[0002] Integral bladed disks (IBs) are key aero-engine components that integrate blades and disks into a single unit, and are widely used in high-performance turbomachinery. Compared to traditional separate blade and disk structures, IIBs offer advantages such as compact structure, light weight, and high strength, effectively improving the engine's thrust-to-weight ratio and operating efficiency. However, their integrated structure also presents significant challenges for subsequent geometric dimension inspection, profile measurement, and defect identification.

[0003] Currently, the 3D contour inspection of integral bladed disks mainly relies on methods such as coordinate measuring machines (CMMs), laser scanning systems, or structured light 3D measurement. Among these, CMMs offer high accuracy but have low measurement efficiency and require precise workpiece clamping, making them unsuitable for large-area rapid measurement of complex curved surfaces. While laser scanning equipment possesses a certain degree of automation, it is prone to measurement blind spots when dealing with deep cavity curved surfaces. Structured light systems are widely used due to their non-contact and high-efficiency advantages, but traditional structured light systems are mostly fixed platform structures, which struggle to meet the complex spatial morphology and inspection posture requirements of integral bladed disks. This is especially true in open bladed disks where blades are densely distributed and surface curvature changes drastically, leading to problems such as occlusion, distortion, and data loss.

[0004] In addition, most of the current mainstream detection devices are based on general-purpose platforms and lack special designs for the characteristic structure of integral open bladed disks. This makes it difficult to meet the actual needs of multi-angle, high-precision, and rapid detection. There is an urgent need for a good three-dimensional geometric contour detection device and method for integral open bladed disks to achieve high-quality reconstruction of blades. Summary of the Invention

[0005] In order to overcome the shortcomings of the prior art, the present invention provides a device and method for detecting the three-dimensional geometric contour of an integral open bladed disk, which can achieve efficient detection of blade defects.

[0006] A device and method for detecting the three-dimensional geometric contour of an integral open bladed disk includes: a structured light detection system 1, a linear module 2, a support frame 3, a rotating mechanism 4, and a bladed disk 5 to be tested. The structured light detection system 1 is fixedly connected to the support frame 3, and the linear module 2 is vertically fixed to the support frame 3. The end of the structured light detection system 1 is integrated into the linear module 2, and the camera and lens in the structured light detection system 1 are both facing the bladed disk 5 to be tested. The rotating mechanism 4 drives the bladed disk 5 to rotate. The support frame 3 is connected to the ground through four feet, which can provide reliable support and fixation to ensure the stability of the support frame 3 during use and prevent shaking.

[0007] Furthermore, the structured light detection system 1 includes a projector 6, a lens coupling device 7, a coupling lens 8, and an optical fiber image transmission bundle 9. The projector 6 is placed above the support frame 3 and is fixedly connected to the lens coupling device 7 through a slot structure. The lens coupling device 7 is used to effectively couple the projection beam into the coupling lens 8. Its bottom is fixedly connected to the support frame 3. One end of the coupling lens 8 is located in the lens coupling device 7, and the other end is connected to the optical fiber image transmission bundle 9. The end of the optical fiber image transmission bundle 9 is connected to the projection lens 10.

[0008] Furthermore, the linear module 2 includes a projection lens 10, a miniature camera 11, a cable chain 12, a drive motor 13, a lead screw 14, a servo motor 15, and an end fixing device 16. The cable chain 12 and the drive motor 13 are vertically installed and fixed on the support frame 3. The fiber optic image bundle 9 and the miniature camera 11 are placed in the cable chain and protected by the cable chain. The lead screw 14 is fixed to the output shaft of the drive motor 13 and vertically fixed to the support frame 3. The servo motor 15 is sleeved on the lead screw 14 and can slide vertically along the lead screw 14. The end fixing device 16 is fixed to the output shaft of the servo motor 15 and is clearance-fitted with the projection lens 10 and the miniature camera 11 to play a fixing role, and can rotate under the drive of the servo motor 15.

[0009] Furthermore, the rotating mechanism 4 includes a rotary motor 17, a rotary motor drive gear 18, a rotary driven gear 19, and a servo cylinder 20. The rotary motor 17 is horizontally fixed to the support frame 3, the rotary motor drive gear 18 is fixed to the output shaft of the rotary motor 17, the servo cylinder 20 is disposed on one side of the rotary motor 17 and is horizontally fixed to the support frame 3, the rotary driven gear 19 is fixed to the output shaft of the servo cylinder 20, and the rotary driven gear 19 meshes with the rotary motor drive gear 18. The impeller 5 is fixedly connected to the rotary driven gear 19 through a fixed column.

[0010] Furthermore, the device sets the following parameters for key hardware structures: the Z-axis stroke of the linear module is not less than the envelope height H of the blade from the root to the tip, the minimum displacement resolution is not greater than 0.01 mm, the repeatability is not greater than ±0.02 mm, and the end-effector rotation fine-tuning range is ±10° with a resolution not greater than 0.01°; the rotation range of the rotating mechanism is 0°-360°, the angle step is not greater than 0.05°, and the repeatability is not greater than ±0.02°; the optical path adjustment mechanism at θ, The adjustment range of the direction is not less than ±3° and the resolution is not greater than 0.01°. The adjustment range of the distance d direction is not less than ±10mm and the resolution is not greater than 0.02mm.

[0011] A method for detecting the three-dimensional geometric contour of blades in an integral open bladed disk, the method comprising the following steps:

[0012] S1: Adjust the detection positions of structured light detection system 1, linear module 2 and bladed disk 5;

[0013] S2: Perform optical path adjustment and system calibration;

[0014] S3: The rotating mechanism 4 drives the bladed disk 5 to be tested to rotate at multiple angles according to the initially set rotation angle. At the same time, the structured light detection system 1 simultaneously acquires point cloud data at each angle and uses the field of view overlap rate and point cloud confidence as evaluation indicators to adaptively adjust the rotation angle.

[0015] S4: 3D reconstruction and defect identification.

[0016] Furthermore, the optical path adjustment and system calibration in step two include the following processes:

[0017] (1) Construct optical models for the camera and projector, and establish their intrinsic parameter matrices respectively:

[0018]

[0019] Where f x f y c is the camera focal length. x c y The parameter with the superscript 'p' indicates the principal point position of the camera, and the parameter with the superscript 'p' indicates the corresponding parameter of the projector.

[0020] (2) Adjust the optical path. Because this device is used in a complex, integrated open-disc inspection scenario with limited space, it is essential to precisely align the projector and camera within their effective field of view. (Introduction) This serves as an evaluation index for field-of-view overlap. If this index does not reach the set threshold η... min Then the servo adjustment mechanism is executed to adjust the optical axis attitude parameters. To optimize the optical path matching between the camera and the projector.

[0021] (3) Joint calibration and dynamic error optimization. After the optical path adjustment is completed, in order to ensure that the camera and projection system have a unified coordinate reference system in 3D reconstruction, Zhang Zhengyou's checkerboard calibration method is used to obtain the camera's extrinsic parameter matrix, and a mapping relationship from the world coordinate system to the image coordinate system is established, thereby obtaining the initial estimated parameter set:

[0022] Θ0={K c ,R pc ,t pc}

[0023] However, in actual deployment, due to factors such as lens distortion, optical axis jitter, and lens thermal deformation, the initial calibration has certain errors. Therefore, a dynamic reprojection error optimization mechanism is introduced. A reprojection error objective function is constructed, and the Levenberg-Marquardt optimization algorithm is used for iterative updates to correct the calibration parameters in real time.

[0024]

[0025] in These are the actual coordinates of the image points. Estimate the predicted coordinates of image points using the current parameters.

[0026] (4) Optical path stability evaluation and adjustment termination criterion. To determine whether the adjustment has reached a stable state, an optical path stability index S is introduced. opt This parameter, taking into account factors such as field-of-view overlap stability and calibration error convergence speed, is defined as follows:

[0027]

[0028] Among them, A overlap A represents the area of ​​the effective overlap between the camera and the projected pattern. target E represents the expected detection area; E is the current reprojection error; E max α is the upper limit of the reprojection error; α and β are weighting coefficients, satisfying α + β = 1.

[0029] The optical path adjustment is considered to have reached a stable state and the adjustment process is terminated when the following conditions are met:

[0030]

[0031] Where τ stable η is the optical path stability threshold. min E represents the minimum allowable value for field-of-view overlap. tol This is the upper limit of the allowable calibration error.

[0032] During the detection process, a coarse-fine adjustment method is adopted. The coarse adjustment uses a large step size to achieve rapid field coverage (η≥0.75), while the fine adjustment uses a small step size and combines reprojection error and field overlap rate to achieve precise matching. (pixels, η≥0.85).

[0033] Furthermore, the adaptive angle adjustment method in step three includes the following process:

[0034] Let the initial rotation angle sequence be: Θ0={θ1,θ2,…θ n Divide the circumference of the bladed disk into n sector regions: B = {b1, b2, ... b} nIn structured light 3D reconstruction, overlapping regions contribute to the quality of multi-view point cloud stitching and the accuracy of feature matching. Let V be the field of view acquired from two adjacent angles. i V i+1 Then its field-of-view overlap rate is defined as:

[0035]

[0036] Determine whether the minimum overlap requirement ω is met. i ≥ω min Otherwise, add an angle:

[0037] For each reconstructed region b i Let the corresponding point cloud confidence distribution be C. i ={c1,c2,…,c n The confidence level equilibrium index for this region is defined as the standard deviation.

[0038]

[0039] when And the maximum confidence level max(C) i )<τ max If this occurs, it indicates that the point cloud quality in that area is uneven, and an auxiliary angle needs to be introduced.

[0040] Based on the above two indicators, construct an overall evaluation function:

[0041]

[0042] Where α1+α2=1 is an adjustable weighting factor.

[0043] When Ψ(θ) i )>τ Ψ Automatically add new angles: Then it enters a loop for iterative detection until all regions meet the angular sufficiency requirement: Ψ(θ) i )≤τ Ψ .

[0044] The advantages of this invention compared to the prior art are:

[0045] 1. This invention provides a device and method for detecting the three-dimensional geometric contour of blades in an integral open bladed disk, addressing the lack of dedicated detection equipment for integral open bladed disks. This invention employs an optical fiber image transmission bundle and a miniature camera to transmit structured light images from a projector to the integral open bladed disk with high fidelity, achieving stable imaging in confined and complex spaces. This overcomes the limitations of traditional equipment being too large to access, and, combined with rotational scanning and multi-angle viewpoint planning, enables full-coverage detection of all blades in the bladed disk.

[0046] 2. By adjusting the optical path and calibrating the system, rapid alignment and high-precision calibration between the projector and camera can be achieved in detection environments with limited space and frequent attitude changes, significantly improving the imaging consistency of the structured light system and the accuracy and robustness of 3D reconstruction.

[0047] 3. By constructing an adaptive angle adjustment strategy based on the overlap rate of the field of view and the confidence of the point cloud, we can ensure that information from multiple perspectives is fully acquired, effectively improve the coverage integrity of the point cloud and the quality balance of the reconstructed area, and overcome the limitations of the fixed angle scheme in the detection of complex structures.

[0048] 4. This invention fully considers the testing requirements of different specifications and sizes of integral open impellers during the design process. By setting adjustable linear module stroke, rotating mechanism bearing diameter, and load capacity, the device can adapt to impeller structures of different diameters and heights. Simultaneously, a dynamic error optimization mechanism is introduced during calibration to ensure calibration accuracy for impellers of different sizes. An adaptive angle adjustment mechanism is introduced during angle division, allowing the system to automatically correct the detection step size and adjust the rotation angle according to the actual size of the impeller. Therefore, this invention not only meets the testing requirements of specific impeller models but also adapts to application scenarios of impellers of different sizes, possessing good versatility and flexibility.

[0049] The technical solution of the present invention will be further described below with reference to the accompanying drawings and embodiments: Attached Figure Description

[0050] Figure 1 This is a schematic diagram of the overall structure of the three-dimensional geometric contour detection device for an integral open bladed disk according to the present invention;

[0051] Figure 2 This is a schematic diagram of the structured light projection system of the present invention;

[0052] Figure 3 This is a schematic diagram of the linear module of the present invention;

[0053] Figure 4 This is a schematic diagram of the rotating mechanism of the present invention;

[0054] Figure 5 This is a flowchart of a method for detecting the three-dimensional geometric contour of an integral open bladed disk blade according to the present invention. Detailed Implementation

[0055] The embodiments of the technical solution of the present invention will now be described in detail with reference to the accompanying drawings. The following embodiments are only used to more clearly illustrate the technical solution of the present invention. Unless otherwise stated, the technical or scientific terms used in this application should have the ordinary meaning understood by those skilled in the art.

[0056] Specific implementation method one: Combining Figures 1 to 4 This embodiment describes a device and method for detecting the three-dimensional geometric contour of an integral open bladed disk, comprising: a structured light detection system 1, a linear module 2, a support frame 3, a rotating mechanism 4, and a bladed disk 5 to be tested. The structured light detection system 1 is fixedly connected to the support frame 3, and the linear module 2 is vertically fixedly connected to the support frame 3. The end of the structured light detection system 1 is integrated into the linear module 2, and the camera and lens in the structured light detection system 1 are both facing the bladed disk 5 to be tested. The rotating mechanism 4 drives the bladed disk 5 to rotate. The support frame 3 is connected to the ground through four feet, which provide reliable support and fixation, ensuring the stability of the support frame 3 during use and preventing shaking.

[0057] In this embodiment, the support frame 3 is used to fix the entire device and provide a stable installation base. All other mechanisms are set on the support frame 3 and integrated with the support frame 3. The support frame 3 is supported on the ground by four evenly distributed feet at the bottom.

[0058] The structured light detection system 1 is used to project and transmit structured light stripe patterns;

[0059] The linear module 2 is used to realize the vertical linear motion and rotational motion of the end fixing device, ensuring that the projection lens and the fiber optic image transmission bundle can scan the area from the leaf root to the leaf tip.

[0060] The rotating mechanism 4 is used to rotate the impeller to a specified angle and to move the impeller back and forth.

[0061] Specific Implementation Method Two: Combining Figures 1 to 4 This embodiment describes a structured light detection system 1 comprising a projector 6, a lens coupling device 7, a coupling lens 8, and an optical fiber image bundle 9. The projector 6 is positioned above a support frame 3 and is fixedly connected to the lens coupling device 7 via a slot structure. The lens coupling device 7 is used to effectively couple the projected light beam into the coupling lens 8, and its bottom is fixedly connected to the support frame 3. One end of the coupling lens 8 is located in the lens coupling device 7, and the other end is connected to the optical fiber image bundle 9. The end of the optical fiber image bundle 9 is connected to the projection lens 10.

[0062] In this embodiment, the projector 6 is located above the support frame 3, providing a light source for the entire mechanism;

[0063] The lens coupling device 7 is fixed above the support frame 3 and is connected to the light output port of the projector 6 through a slot, coupling the structured light pattern from the projector 6 into the fiber optic image transmission bundle 9.

[0064] The coupling lens 8 is integrated inside the lens coupling device 7. Its optical axis is coaxial with the output optical axis of the projector 6. Its end is connected to the fiber optic image bundle 9, which can accurately image the structured light pattern emitted from the projector 6 onto the incident end face of the fiber optic image bundle 9.

[0065] One end of the fiber optic image bundle 9 is connected to the coupling lens 8 to receive the structured light pattern after it has been focused by the lens, and the other end is connected to the projection lens 10. The projection lens 10 projects the structured light stripe pattern onto the blade disk under test to achieve accurate projection of the stripe pattern.

[0066] Before testing, the position of the structured light detection system 1 needs to be adjusted, and the projector 6 needs to be turned on to adjust the light path to ensure that a clear striped pattern can be formed on the blade disk. The light path transmission path is: projector - coupling lens - fiber optic image bundle - projection lens - blade disk.

[0067] Specific implementation method three: Combining Figures 1-4 This embodiment describes a linear module 2 comprising a projection lens 10, a miniature camera 11, a cable chain 12, a drive motor 13, a lead screw 14, a servo motor 15, and an end-fixing device 16. The cable chain 12 and the drive motor 13 are vertically mounted and fixed on a support frame 3. The fiber optic image bundle 9 and the miniature camera 11 are placed in the cable chain and protected by the cable chain's traction. The lead screw 14 is fixed to the output shaft of the drive motor 13 and vertically fixed to the support frame 3. The servo motor 15 is sleeved on the lead screw 14 and can slide vertically along the lead screw 14. The end-fixing device 16 is fixed to the output shaft of the servo motor 15 and has a clearance fit with the projection lens 10 and the miniature camera 11, serving a fixing function and being able to rotate under the drive of the servo motor 15.

[0068] A miniature camera 11 and a projection lens 10 are arranged opposite to each other in an end fixing device 16, with the lens facing the surface of the bladed disk 5 to be measured, for collecting structured light patterns.

[0069] The drag chain 12 is fixedly connected to the support frame 3, providing flexible traction and protection for the miniature camera 12 and the fiber optic image transmission bundle 9, preventing entanglement, bending or damage caused by movement.

[0070] The drive motor 13 is vertically mounted on the support frame 3 and controls the movement of the lead screw 14;

[0071] The lead screw 14 is set vertically and connected to the support frame 3 through a fixed seat to form a vertical lifting transmission channel;

[0072] When the drive motor 13 rotates and drives the lead screw 14 to rotate, the servo motor 15 generates lifting displacement due to the helical motion of the lead screw, thereby achieving height adjustment;

[0073] The end-fixing device 16 is fixedly installed at the output end of the servo motor 15 to fix the miniature camera 11 and the projection lens 10. The detection end can move up and down as a whole with the linear module 2 to ensure that the projection and the miniature camera are aligned with the area to be measured.

[0074] Before testing, the position of the linear module 2 needs to be pre-adjusted. At this time, the drive motor 13 drives the lead screw to move, and the lead screw 14 drives the servo motor 15 to perform linear reciprocating motion in the vertical direction. The end fixing device 16 moves accordingly and can rotate under the drive of the servo motor 15, which can drive the miniature camera 11 and the projection lens 10 to the designated position, so as to realize the layer-by-layer scanning and detection of the blade from the root to the tip.

[0075] Specific implementation method four: Combination Figures 1 to 4 In this embodiment, the rotating mechanism 4 includes a rotary motor 17, a rotary motor drive gear 18, a rotary driven gear 19, and a servo cylinder 20. The rotary motor 17 is horizontally fixed to the support frame 3. The rotary motor drive gear 18 is fixed to the output shaft of the rotary motor 17. The servo cylinder 20 is located on one side of the rotary motor 17 and is horizontally fixed to the support frame 3. The rotary driven gear 19 is fixed to the output shaft of the servo cylinder 20 and meshes with the rotary motor drive gear 18. The impeller 5 is fixedly connected to the rotary driven gear 19 through a fixed column.

[0076] The rotary motor 17 is horizontally mounted and fixed to the support frame 3. Its output shaft is connected to the rotary motor drive gear 18 to provide rotational power and drive the driven gear 19 meshing with it to rotate.

[0077] The servo electric cylinder 20 is located on one side of the rotary motor 17 and is horizontally fixed to the support frame 3. Its output shaft is connected to the rotary driven gear 19. The extension and retraction of the servo electric cylinder 20 can drive the rotary driven gear 19 to move back and forth along the axial direction, thereby realizing the translational adjustment of the impeller 5.

[0078] Driven by the rotary motor 17, the drive gear 18 drives the driven gear 19 to rotate synchronously, thereby achieving the angle adjustment of the impeller.

[0079] The impeller 5 is fixedly connected to the rotating driven gear 19 via a fixed column, so that it can be adjusted in angle and horizontal position as the gear rotates.

[0080] Before testing, the position of the impeller 5 needs to be pre-adjusted. At this time, the rotary motor 17 outputs power to control the drive gear 18 to rotate, so that the driven gear 19 meshing with it transmits power to the impeller. The blades to be tested can rotate into the detection field of view in sequence. Then, the servo cylinder 20 is controlled to output power to drive the driven gear 19 to move along the axis. By fine-tuning back and forth, the blades are ensured to be at the optimal imaging distance, thereby improving the detection accuracy and system adaptability.

[0081] Specific Implementation Method Five: Combining Figures 1 to 4This embodiment describes the following parameters for the key hardware structure of the device: the Z-axis travel of the linear module is not less than the envelope height H of the blade from the root to the tip; the minimum displacement resolution is not greater than 0.01 mm; the repeatability is not greater than ±0.02 mm; the end-effector rotation fine-tuning range is ±10° with a resolution not greater than 0.01°; the rotation range of the rotation mechanism is 0°-360°; the angle step is not greater than 0.05°; and the repeatability is not greater than ±0.02°; the optical path adjustment mechanism at θ... The adjustment range of the direction is not less than ±3° and the resolution is not greater than 0.01°. The adjustment range of the distance d direction is not less than ±10mm and the resolution is not greater than 0.02mm.

[0082] Specific Implementation Method Six: Combination Figures 1 to 5 This embodiment describes a device and method for detecting the three-dimensional geometric contour of an integral open bladed disk blade. The method includes the following steps:

[0083] S1: Adjust the detection positions of structured light detection system 1, linear module 2 and bladed disk 5;

[0084] S2: Perform optical path adjustment and system calibration;

[0085] S3: The rotating mechanism 4 drives the bladed disk 5 to be tested to rotate at multiple angles according to the initially set rotation angle. At the same time, the structured light detection system 1 simultaneously acquires point cloud data at each angle and uses the field of view overlap rate and point cloud confidence as evaluation indicators to adaptively adjust the rotation angle.

[0086] S4: 3D reconstruction and defect identification.

[0087] Specific implementation method seven: Combination Figures 1 to 5 This embodiment describes a detection process in step two, which includes the following steps:

[0088] (1) Construct optical models for the camera and projector, and establish their intrinsic parameter matrices respectively:

[0089]

[0090] Where f x f y c is the camera focal length. x c y The parameter with the superscript 'p' indicates the principal point position of the camera, and the parameter with the superscript 'p' indicates the corresponding parameter of the projector.

[0091] (2) Adjust the optical path. Because this device is used in a complex, integrated open-disc inspection scenario with limited space, it is essential to precisely align the projector and camera within their effective field of view. (Introduction) This serves as an evaluation index for field-of-view overlap. If this index does not reach the set threshold η...min Then the servo adjustment mechanism is executed to adjust the optical axis attitude parameters. To optimize the optical path matching between the camera and the projector.

[0092] (3) Joint calibration and dynamic error optimization. After the optical path adjustment is completed, in order to ensure that the camera and projection system have a unified coordinate reference system in 3D reconstruction, Zhang Zhengyou's checkerboard calibration method is used to obtain the camera's extrinsic parameter matrix, and a mapping relationship from the world coordinate system to the image coordinate system is established, thereby obtaining the initial estimated parameter set:

[0093] Θ0={K c ,R pc ,t pc}

[0094] However, in actual deployment, due to factors such as lens distortion, optical axis jitter, and lens thermal deformation, the initial calibration has a certain error. Therefore, a dynamic reprojection error optimization mechanism is introduced, and a reprojection error objective function is constructed:

[0095]

[0096] in These are the actual coordinates of the image points. Estimate the predicted coordinates of image points using the current parameters.

[0097] (4) Optical path stability evaluation and adjustment termination criterion. To determine whether the adjustment has reached a stable state, an optical path stability index S is introduced. opt It can comprehensively consider factors such as the stability of field of view overlap and the convergence speed of calibration error, and is defined as follows:

[0098]

[0099] Among them, A overlap A represents the area of ​​the effective overlap between the camera and the projected pattern. target E represents the expected detection area; E is the current reprojection error; E max α is the upper limit of the reprojection error; α and β are weighting coefficients, satisfying α + β = 1.

[0100] The optical path adjustment is considered to have reached a stable state and the adjustment process is terminated when the following conditions are met:

[0101]

[0102] Where, τ stable η is the optical path stability threshold. min E represents the minimum allowable value for field-of-view overlap. tol This is the upper limit of the allowable calibration error.

[0103] During the detection process, a graded adjustment method of coarse adjustment and fine adjustment is adopted. Coarse adjustment achieves rapid field coverage with a larger step size (η≥0.75), while fine adjustment achieves precise matching by using a smaller step size and combining reprojection error and field overlap rate. (pixels, η≥0.85).

[0104] Specific implementation method eight: Combination Figures 1 to 5 This embodiment describes a detection process in step three, which includes the following steps:

[0105] Let the initial rotation angle sequence be: Θ0={θ1,θ2,…θ n Divide the circumference of the bladed disk into n sector regions: B = {b1, b2, ... b} n In structured light 3D reconstruction, overlapping regions contribute to the quality of multi-view point cloud stitching and the accuracy of feature matching. Let V be the field of view acquired from two adjacent angles. i V i+1 Then its field-of-view overlap rate is defined as:

[0106]

[0107] Determine whether the minimum overlap requirement ω is met. i ≥ω min Otherwise, add an angle:

[0108] For each reconstructed region b i Let the corresponding point cloud confidence distribution be C. i ={c1,c2,…,c n The confidence level equilibrium index for this region is defined as the standard deviation.

[0109]

[0110] when And the maximum confidence level max(C) i )<τ max If this occurs, it indicates that the point cloud quality in that area is uneven, and an auxiliary angle needs to be introduced.

[0111] Based on the above three indicators, an overall evaluation function is constructed:

[0112]

[0113] Where α1+α2=1 is an adjustable weighting factor.

[0114] When Ψ(θ) i )>τ Ψ Automatically add new angles: Then it enters a loop for iterative detection until all regions meet the angular sufficiency requirement: Ψ(θ)i )≤τ Ψ .

[0115] The above description is merely a preferred embodiment of the present invention and is not intended to limit the invention. Various modifications and variations can be made to the present invention by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. A device for detecting the three-dimensional geometric contour of an integral open bladed disk blade, characterized in that: The system includes a structured light detection system 1, a linear module 2, a support frame 3, a rotating mechanism 4, and a test impeller 5. The structured light detection system 1 is fixedly connected to the support frame 3, and the linear module 2 is vertically fixed to the support frame 3. The end of the structured light detection system 1 is integrated into the linear module 2, and the camera and lens in the structured light detection system 1 are both facing the test impeller 5. The rotating mechanism 4 drives the impeller 5 to rotate. The support frame 3 is connected to the ground through four feet, which provide reliable support and fixation to ensure the stability of the support frame 3 during use and prevent shaking.

2. The device for detecting the three-dimensional geometric contour of an integral open bladed disk blade according to claim 1, characterized in that: The structured light detection system 1 includes a projector 6, a lens coupling device 7, a coupling lens 8, and an optical fiber image transmission bundle 9. The projector 6 is placed above the support frame 3 and is fixedly connected to the lens coupling device 7 through a slot structure. The lens coupling device 7 is used to effectively couple the projection beam into the coupling lens 8. Its bottom is fixedly connected to the support frame 3. One end of the coupling lens 8 is located in the lens coupling device 7, and the other end is connected to the optical fiber image bundle 9. The end of the optical fiber image bundle 9 is connected to the projection lens 10.

3. The three-dimensional geometric contour detection device for an integral open bladed disk according to claim 2, characterized in that: The linear module 2 includes a projection lens 10, a miniature camera 11, a cable chain 12, a drive motor 13, a lead screw 14, a servo motor 15, and an end-fixing device 16. The cable chain 12 and the drive motor 13 are vertically mounted and fixed on the support frame 3. The fiber optic image bundle 9 and the miniature camera 11 are placed in the cable chain and protected by the cable chain. The lead screw 14 is fixed to the output shaft of the drive motor 13 and vertically fixed to the support frame 3. The servo motor 15 is mounted on the lead screw 14 and can slide vertically along the lead screw 14. The end-fixing device 16 is fixed to the output shaft of the servo motor 15 and is in clearance fit with the projection lens 10 and the miniature camera 11 to fix them. It can also rotate under the drive of the servo motor 15.

4. The three-dimensional geometric contour detection device for an integral open bladed disk according to claim 3, characterized in that: The rotating mechanism 4 includes a rotary motor 17, a rotary motor drive gear 18, a rotary driven gear 19, and a servo cylinder 20. The rotary motor 17 is horizontally fixed to the support frame 3. The rotary motor drive gear 18 is fixed to the output shaft of the rotary motor 17. The servo cylinder 20 is located on one side of the rotary motor 17 and is horizontally fixed to the support frame 3. The rotary driven gear 19 is fixed to the output shaft of the servo cylinder 20 and meshes with the rotary motor drive gear 18. The impeller 5 is fixedly connected to the rotary driven gear 19 through a fixed column.

5. The three-dimensional geometric contour detection device for an integral open bladed disk according to claim 4, wherein the key hardware structure of the device is configured with the following parameters: the Z-axis stroke of the linear module is not less than the envelope height H of the blade from the root to the tip; the minimum displacement resolution is not greater than 0.01 mm; and the repeatability is not greater than [missing value]. ±0.02mm, the end-effector rotation fine-tuning range is ±10°, and the resolution is no greater than 0.01°; the rotation range of the rotation mechanism is 0°-360°, the angle step is no greater than 0.05°, and the repeatability is no greater than ±0.02°; the optical path adjustment mechanism at θ, The adjustment range of the direction is not less than ±3° and the resolution is not greater than 0.01°. The adjustment range of the distance d direction is not less than ±10mm and the resolution is not greater than 0.02mm.

6. A device and method for detecting the three-dimensional geometric contour of an integral open bladed disk blade as described in any one of claims 1-4, characterized in that: The method includes the following steps: S1: Adjust the detection positions of structured light detection system 1, linear module 2 and bladed disk 5; S2: Perform optical path adjustment and system calibration; S3: The rotating mechanism 4 drives the bladed disk 5 to be tested to rotate at multiple angles according to the initially set rotation angle. At the same time, the structured light detection system 1 simultaneously acquires point cloud data at each angle and uses the field of view overlap rate and point cloud confidence as evaluation indicators to adaptively adjust the rotation angle. S4: 3D reconstruction and defect identification.

7. The method for detecting the three-dimensional geometric contour of an integral open bladed disk blade according to claim 5, characterized in that: The optical path adjustment and system calibration method in step two includes the following process: (1) Construct optical models for the camera and projector, and establish their intrinsic parameter matrices respectively: Where f x f y c is the camera focal length. x c y The parameter with the superscript 'p' indicates the principal point position of the camera, and the parameter with the superscript 'p' indicates the corresponding parameter of the projector. (2) Adjust the optical path. Because this device is used in a complex, integrated open-disc inspection scenario with limited space, it is essential to precisely align the projector and camera within their effective field of view. (Introduction) This serves as an evaluation index for field-of-view overlap. If this index does not reach the set threshold η... min Then the servo adjustment mechanism is executed to adjust the optical axis attitude parameters. To optimize the optical path matching between the camera and the projector. (3) Joint calibration and dynamic error optimization. After the optical path adjustment is completed, in order to ensure that the camera and projection system have a unified coordinate reference system in 3D reconstruction, Zhang Zhengyou's checkerboard calibration method is used to obtain the camera's extrinsic parameter matrix, and a mapping relationship from the world coordinate system to the image coordinate system is established, thereby obtaining the initial estimated parameter set: Θ0={K c ,R pc ,t pc } However, in actual deployment, due to factors such as lens distortion, optical axis jitter, and lens thermal deformation, the initial calibration has a certain error. Therefore, a dynamic reprojection error optimization mechanism is introduced, and a reprojection error objective function is constructed: in These are the actual coordinates of the image points. Estimate the predicted coordinates of image points using the current parameters. (4) Optical path stability evaluation and adjustment termination criterion. To determine whether the adjustment has reached a stable state, an optical path stability index S is introduced. opt It can comprehensively consider factors such as the stability of field of view overlap and the convergence speed of calibration error, and is defined as follows: Among them, A overlap A represents the area of ​​the effective overlap between the camera and the projected pattern. target E represents the expected detection area; E is the current reprojection error; E max α is the upper limit of the reprojection error; α and β are weighting coefficients, satisfying α + β = 1. The optical path adjustment is considered to have reached a stable state and the adjustment process is terminated when the following conditions are met: Where τ stable η is the optical path stability threshold. min E represents the minimum allowable value for field-of-view overlap. tol This is the upper limit of the allowable calibration error. During the detection process, a graded adjustment method of coarse adjustment and fine adjustment is adopted. Coarse adjustment achieves rapid field coverage with a larger step size (η≥0.75), while fine adjustment achieves precise matching by using a smaller step size and combining reprojection error with field overlap rate. (pixels, η≥0.85).

8. The method for detecting the three-dimensional geometric contour of an integral open bladed disk blade according to claim 5, characterized in that: The optical path adjustment and system calibration method in step three includes the following process: Let the initial rotation angle sequence be: Θ0={θ1,θ2,…θ n Divide the circumference of the bladed disk into n sector regions: B = {b1, b2, ... b} n In structured light 3D reconstruction, overlapping regions contribute to the quality of multi-view point cloud stitching and the accuracy of feature matching. Let V be the field of view acquired from two adjacent angles. i V i+1 Then its field-of-view overlap rate is defined as: Determine whether the minimum overlap requirement ω is met. i ≥ω min Otherwise, add an angle: For each reconstructed region b i Let the corresponding point cloud confidence distribution be C. i ={c1,c2,…,c n The confidence level equilibrium index for this region is defined as the standard deviation. when And the maximum confidence level max(C) i )<τ max If this occurs, it indicates that the point cloud quality in that area is uneven, and an auxiliary angle needs to be introduced. Based on the above two indicators, construct an overall evaluation function: Where α1+α2=1 is an adjustable weighting factor. When Ψ(θ) i )>τ Ψ Automatically add new angles: Then it enters a loop for iterative detection until all regions meet the angular sufficiency requirement: Ψ(θ) i )≤τ Ψ .

Images