An online measurement method for micro-deformation of a special-shaped thin-walled part

Abstract

This invention relates to the field of online measurement technology and discloses an online measurement method for micro-deformation of irregularly shaped thin-walled parts. The method includes: setting measurement nodes while maintaining the clamping and contact relationship of the irregularly shaped thin-walled part; collecting initial and current distance measurements using an online ranging device, and collecting the wall thickness of the measurement point and the surface distance to the nearest support point to form an initial reference point cloud and a current measurement point cloud; establishing a data correspondence relationship between corresponding measurement nodes after rigid registration; filtering valid measurement nodes; determining the rescan order based on the difference between the initial and current distance measurements, the wall thickness of the measurement point, and the surface distance to the nearest support point; identifying the local maximum node within the set of directly adjacent points and connecting them to form a setback ridge; performing a second online scan along the setback ridge, and updating the micro-deformation results of the setback ridge and its directly adjacent areas. This invention is used for on-site micro-deformation detection of irregularly shaped thin-walled parts.

Description

Technical Field

[0001] This invention relates to the field of online measurement technology, and more specifically, to an online measurement method for micro-deformation of irregularly shaped thin-walled parts. Background Technology

[0002] In the machining or forming of irregularly shaped thin-walled parts in aerospace, mold making, and precision equipment industries, it is often necessary to complete staged measurements under the constraints of fixtures. These parts have significant surface undulations, marked local thickness variations, and dispersed support contact points. After processing, minor surface retractions are often related to the clamping condition, support distance, and wall thickness distribution. On-site measurement personnel typically aim to obtain verifiable micro-deformation results using machine tool axes, measuring slides, or online ranging equipment without disassembling the part or realigning the fixture, while also recording obstructions, invalid echoes, and over-range areas.

[0003] Current field measurement procedures mostly employ fixed-density line laser scanning, offline coordinate measuring machine (CMM) re-measurement, or uniform densification of the point cloud across the entire surface. Fixed-density scanning is convenient for programming and cycle control, but it is not sensitive to narrow allowances in thin-walled, far-support areas; offline re-measurement requires disassembling or moving parts, which can easily introduce new positioning discrepancies; and full-surface densified scanning increases measurement downtime, and manual data cleanup is still required in areas with curved surface obstruction and unstable echoes.

[0004] In point cloud processing, a common practice is to rigidly register the initial point cloud with the current point cloud and then find abnormal areas based on point position deviations. This approach can obtain an overall deviation map, but when the deviation magnitudes of two locations are similar, it is difficult to distinguish whether the deviation is located near the support of a thick wall or far from the support of a thin wall simply by looking at the magnitude of the deviation. If the threshold is repeatedly adjusted to find the peak value, the measurement results will be affected by the part size, the scanning path, and the operator's experience. Summary of the Invention

[0005] This invention provides an online measurement method for micro-deformation of irregularly shaped thin-walled parts, which solves the technical problems mentioned in the background art.

[0006] To achieve the above-mentioned objectives, the present invention provides the following technical solution.

[0007] An online measurement method for micro-deformation of irregularly shaped thin-walled parts, applied to irregularly shaped thin-walled parts clamped by fixtures; While maintaining the contact relationship between the irregularly shaped thin-walled part and the fixture, measurement nodes are set. The online ranging device collects the initial distance measurement at the same measurement node in the initial state, and collects the current distance measurement in the same measurement posture after the process node; Collect the wall thickness at the measurement point and the surface distance to the nearest support point to form an initial reference point cloud and a current measurement point cloud; Rigidly register the initial reference point cloud and the current measurement point cloud, establish the correspondence between the initial distance measurement, the current distance measurement, the wall thickness of the measurement point and the surface distance to the nearest support point on the same measurement node, and take the measurement node that completes the correspondence and is not marked as unmeasured, invalid echo or beyond the effective range of the online distance measurement device as the effective measurement node. The priority of rescanning for effective measurement nodes is determined based on the difference between the initial and current distance measurements, the wall thickness of the measurement point, and the surface distance to the nearest support point. This priority increases with the increase of the difference and the surface distance to the nearest support point, and decreases with the increase of the wall thickness of the measurement point. Without setting an artificial threshold within the set of directly adjacent points determined by the scanning order, the effective measurement nodes whose rescanning values ​​satisfy the local maximum relationship are determined as local maximum nodes, and the local maximum nodes with the relationship of directly adjacent point sets are connected according to the scanning order to form a retreat ridge. A second online scan is performed along the yielding ridge line, and the micro-deformation results of the yielding ridge line and its directly adjacent regions are updated using the results of the second online scan.

[0008] Compared with the prior art, the present invention has the following substantial features and significant progress: This invention, under the on-site condition of maintaining the clamping and contact relationship of irregular thin-walled parts, uses the initial distance measurement, current distance measurement, wall thickness at the measurement point, and surface distance to the nearest support point to jointly determine the order of rescanning values. It also extracts the set of directly adjacent points, so that the densified measurement position is simultaneously affected by the current displacement change, local thickness difference, and actual support distance. After the second online scan along the set of ... Attached Figure Description

[0009] Figure 1 This is a diagram illustrating the online measurement and ridge rescanning mechanism of an online measurement method for micro-deformation of irregularly shaped thin-walled parts according to the present invention. Detailed Implementation

[0010] The following description is provided in conjunction with the accompanying drawings. It should be understood that the following embodiments are for illustrative purposes only and are not intended to limit the invention; those skilled in the art can make equivalent substitutions or combinations for the specific implementations.

[0011] See Figure 1 , Figure 1The irregular thin-walled component 100, clamp 200, online ranging device 300, thickness measuring device 400, measurement control unit 500, point cloud registration unit 600, ridge line determination unit 700, and micro-deformation result output unit 800 in this embodiment correspond to the measured object, clamping and holding structure, distance acquisition structure, thickness acquisition structure, measurement control structure, point cloud processing structure, ridge line determination structure, and result output structure, respectively.

[0012] It should be noted that the initial distance measurement refers to the distance collected by the online distance measuring device 300 at the same measurement node in the initial state; the current distance measurement refers to the distance collected by the online distance measuring device 300 in the same measurement posture after the process node; the wall thickness at the measurement point refers to the thickness collected by the thickness measuring device 400 at the same measurement node; and the surface distance to the nearest support point refers to the surface path distance from the measurement node along the surface of the part to the nearest actual support contact center.

[0013] This embodiment is applied to irregularly shaped thin-walled parts clamped by a fixture. The irregularly shaped thin-walled part 100 has a free-form surface, local thickness variation, boundary openings, holes or obstruction areas, and the fixture 200 maintains the clamped state of the irregularly shaped thin-walled part 100 through multiple support blocks, clamping blocks or equivalent contact parts.

[0014] Specifically, the wall thickness and support constraints of the irregular thin-walled component 100 are not consistent at different measurement nodes. If only uniform density scanning is performed, areas far from the actual support contact center and with smaller wall thickness at the measurement point may be missed by low-density scanning due to narrower local peak values. If only the difference between the current measurement and the initial measurement is sorted, it is difficult to distinguish whether the same difference occurs in the thick-walled near-support area or the thin-walled far-support area.

[0015] Furthermore, the settings for measurement nodes, import of nominal CAD models, and mapping of actual support contact centers are as follows: In one embodiment, the measurement control unit 500 first sets measurement nodes. When setting measurement nodes, the nominal CAD model of the irregular thin-walled part 100 is imported, the actual support contact center of the multi-point support fixture is obtained, the actual support contact center is mapped onto the surface of the part, and a scanning trajectory and measurement nodes intersecting the scanning trajectory are generated based on the surface topography and measurable area distribution of the irregular thin-walled part 100. The nominal CAD model is used to provide the nominal surfaces, boundaries, holes, narrow slots, protrusions, recesses, and turning points of the irregular thin-walled part 100, and the actual support contact center is used to determine the actual constraint position of the fixture 200 on the irregular thin-walled part 100.

[0016] The actual support contact center can be obtained from the encoder position of the support head in fixture 200, the position obtained by the machine tool probe, or the fixture coordinate record. Regardless of the conventional position acquisition method used, the actual support contact center must be mapped to the surface of the part. This mapping converts the actual contact position of fixture 200 into a position that can establish a surface path relationship with the surface measurement nodes of the irregular thin-walled part 100.

[0017] When generating scanning trajectories based on surface morphology and the distribution of measurable areas, parallel scanning trajectories can be used in areas with gentle changes in freeform surface, while denser conventional scanning trajectories can be used in areas with significant curvature changes, near boundaries, narrow grooves, or obstructed edges. The settings for scanning trajectories and measurement nodes should ensure that measurement nodes with the same name can be associated with the initial distance measurement, the current distance measurement, the wall thickness at the measurement point, and the surface distance to the nearest support point, respectively.

[0018] In one implementation, each measurement node is assigned a node number, and this node number is used to store the correspondence between the initial distance measurement, the current distance measurement, the wall thickness at the measurement point, and the surface distance to the nearest support point on the same measurement node. The node number is used to maintain a one-to-one correspondence between measurement nodes of the same name and is not used as physical input for the rescanning order value. For a measurement node, if the initial distance measurement, the current distance measurement, the wall thickness at the measurement point, or the surface distance to the nearest support point corresponding to the node number is missing, then that measurement node cannot be used as a valid measurement node to participate in the determination of the rescanning order value.

[0019] The process of obtaining the surface distance to the nearest support point can be accomplished using the initial reference point cloud and the actual support contact center. Specifically, after the initial state scan and point cloud stitching are completed, the surface path distance to the nearest actual support contact center is calculated along the surface of the irregular thin-walled part 100, starting from the measurement node, and this surface path distance is used as the surface distance to the nearest support point. This calculation can use the adjacency relationship of the initial reference point cloud, the surface approximation path of the nominal CAD model, or the surface path of the triangular mesh.

[0020] Furthermore, the methods for initial state acquisition, wall thickness acquisition at measurement points, and current state acquisition are as follows: The initial state is the state when the irregular thin-walled part 100 is clamped but has not yet experienced the process node to be monitored. In the initial state, the measurement control unit 500 controls the online ranging device 300 to move along the pre-set scanning trajectory. The online ranging device 300 acquires the initial distance at each corresponding measurement node in a predetermined measurement posture. The initial distance is the distance from the measurement origin of the online ranging device 300 to the surface of the irregular thin-walled part 100. The initial distance includes both the nominal shape information of the irregular thin-walled part 100 and the fixed distance information resulting from the fixed installation relationship.

[0021] The thickness measuring device 400 collects the wall thickness at the corresponding measurement node. The wall thickness at the measurement point is the actual thickness of the irregular thin-walled component 100 at the measurement node. The acquisition of the wall thickness at the measurement point can be performed after the initial scanning. During acquisition, the thickness measuring device 400 contacts or approaches the corresponding measurement node. After acquisition, the thickness measuring device 400 withdraws from the measurement position while maintaining the contact relationship between the irregular thin-walled component 100 and the fixture 200. The wall thickness at the measurement point is used as the denominator in determining the values ​​before and after rescanning, reflecting the influence of local thickness variations of the thin wall at the measurement node on the collapsibility sensitivity.

[0022] Process nodes can be the completion of a processing layer, a forming pass, an unloading operation, or other nodes requiring monitoring. Upon reaching a process node, the irregularly shaped thin-walled part 100 remains on the fixture 200. The actuator retracts to a position that does not obstruct the measurement optical path of the online ranging device 300. After the device stops moving, the measurement control unit 500 calls the same measurement program as the initial state. The online ranging device 300 acquires the current distance measurement according to the same scanning trajectory, the same movement direction, and the same measurement posture. The same scanning trajectory, the same movement direction, and the same measurement posture ensure that the current distance measurement can be reliably compared with the initial distance measurement at the same measurement node.

[0023] In areas with irregular concave surfaces, sidewalls, or unstable reflections, the online ranging device 300 can acquire data according to a pre-set supplementary viewing angle. The supplementary viewing angle is only used to obtain the actual measurable surface points in the current measurement point cloud and does not infer or fill in unmeasured areas. For measurement nodes that cannot be reliably acquired due to occlusion, invalid echoes, or exceeding the effective range of the online ranging device, the point cloud registration unit 600 or the measurement control unit 500 marks them as unmeasured, having invalid echoes, or exceeding the effective range of the online ranging device.

[0024] Furthermore, the initial reference point cloud, the current measured point cloud, the unified coordinates, and the rigid registration method are as follows: The initial reference point cloud is formed by multiple local point clouds acquired in the initial state. The current measurement point cloud is formed by multiple local point clouds acquired after the process node. In order to enable the initial reference point cloud and the current measurement point cloud to be compared on the same coordinate basis, the point cloud registration unit 600 converts the point clouds from different measurement perspectives into a unified coordinate system. The unified coordinate system can be established through the calibration results of the online ranging device 300, the machine tool axis position, the reference surface of the fixture 200, or the existing benchmark of the irregular thin-walled part 100.

[0025] Before switching to a unified coordinate system, the camera intrinsic parameters, laser plane, and sensor measurement coordinates can be calibrated according to the standard calibration procedure of the online ranging device 300. The rigid conversion relationship between the measurement coordinates of the online ranging device 300 and the machine tool coordinates can be obtained through calibration plates, standard spheres, or gauge blocks.

[0026] After point clouds from different measurement perspectives are transferred to a unified coordinate system, the point cloud registration unit 600 can perform rigid stitching within the overlap zone. Rigid stitching only involves overall translation and rotation. Through this constraint, the current measurement point cloud retains the true surface changes of the irregular thin-walled part 100 after the process node, avoiding the deformation introduced by the registration process from being mistakenly taken as a micro-deformation result.

[0027] When performing rigid registration between the initial reference point cloud and the current measurement point cloud, initial alignment can be achieved first using the existing benchmark of the irregular thin-walled component 100, the reference surface of the fixture 200, or a region close to the actual support contact center that does not participate in the deformation to be measured. Then, rigid registration is performed using iterative nearest point or least squares methods. Rigid registration also only involves overall translation and rotation. After completing rigid registration, the point cloud registration unit 600 establishes a correspondence between the initial reference point cloud and the current measurement point cloud, forming a correspondence between the initial distance measurement, the current distance measurement, the wall thickness of the measurement point, and the surface distance to the nearest support point on the corresponding measurement nodes.

[0028] For a given measurement node, it is considered a valid measurement node only if the initial distance measurement, current distance measurement, wall thickness at the measurement point, and surface distance to the nearest support point have all established correspondences, and the node is not marked as unmeasured, having an invalid echo, or exceeding the effective range of the online ranging equipment. Valid measurement nodes are the objects of determination for rescan order and local maximum node determination. Measurement nodes marked as unmeasured, having an invalid echo, or exceeding the effective range of the online ranging equipment remain invalid and are not involved in determining rescan order or local maximum node determination.

[0029] Furthermore, the determination of the order of rescan values ​​and the formation method of the retreat ridge line are as follows: After the effective measurement nodes are determined, the ridgeline determination unit 700 determines the rescan order based on the difference between the initial and current distance measurements, the wall thickness of the measurement point, and the surface distance to the nearest support point. The difference is the absolute magnitude of the difference between the initial and current distance measurements, i.e. The greater the difference, the more significant the change in the surface position of the effective measurement node along the same measurement posture; therefore, the rescanning order value should be increased. The greater the surface distance to the nearest support point, the farther the effective measurement node is from the actual support contact center on the surface span; therefore, the rescanning order value should be increased. The greater the wall thickness of the measurement point, the higher the local bending resistance at the effective measurement node; under the same difference amplitude and the same surface distance to the nearest support point, the rescanning order value should be decreased.

[0030] In one implementation, the rescan order value for each valid measurement node is determined according to the following formula:

[0031] In the above formula, Indicates valid measurement nodes The order of rescans, Indicates valid measurement nodes The initial distance measurement, Indicates valid measurement nodes The current ranging, Indicates valid measurement nodes The wall thickness at the measuring point Indicates valid measurement nodes The surface distance to the nearest support point. The initial distance measurement, current distance measurement, wall thickness at the measurement point, and surface distance to the nearest support point are all converted to the same unit of length before being substituted into the above formula, and this same unit of length is maintained consistently within the same process node; after this processing, and The product and sum They have the same cubic dimension of length, and the values ​​after the rescan are dimensionless.

[0032] The method for determining the rescanning priority value described above increases the rescanning priority value as the difference amplitude increases, as the surface distance to the nearest support point increases, and decreases as the wall thickness of the measuring point increases. This relationship corresponds to the measurement scenario of the irregular thin-walled part 100: when the difference amplitude of two effective measuring nodes is the same, the effective measuring node with a smaller wall thickness and a larger surface distance to the nearest support point is more likely to correspond to the local retreat of the thin-walled far support area; when the wall thickness and surface distance to the nearest support point of two effective measuring nodes are the same, the effective measuring node with a larger difference amplitude should be rescanned first.

[0033] The ridge line determination unit 700 does not set a manual threshold within the set of directly adjacent points determined by the scanning order. The set of directly adjacent points consists of measurement nodes that are directly adjacent to the effective measurement nodes in the scanning order. For an effective measurement node, if its rescanning order value satisfies a local maximum relationship, that is, its rescanning order value is not less than the rescanning order values ​​of all effective adjacent measurement nodes in the set of directly adjacent points, then the effective measurement node is determined as a local maximum node. This process avoids repeatedly setting manual thresholds for different sizes of the irregular thin-walled parts 100, and also avoids missing narrow peaks due to excessively high manual thresholds or introducing too many non-critical areas due to excessively low manual thresholds.

[0034] In one implementation, the set of locally largest nodes can be represented by the following formula:

[0035] In the above formula, This represents the receding ridge formed by connecting the local maximum nodes in the scan order. Indicates valid measurement nodes The set of directly adjacent points, Indicates valid measurement nodes The order of rescans, Indicates the effective adjacent measurement nodes within the set of directly adjacent points. The order of rescanning values.

[0036] After determining the local maximum node, the ridge line determination unit 700 connects the local maximum nodes with directly adjacent point sets in the scanning order to form the ridge line. The ridge line is formed by connecting the local maximum nodes with directly adjacent point sets in the scanning order and is attached to the actual measured surface of the irregular thin-walled part 100.

[0037] The ridge line is not a new structure formed on the irregular thin-walled part 100, nor is it a solid part obtained by machining the irregular thin-walled part 100. The ridge line is a set of surface positions determined based on the rescanning sequence values ​​of the effective measurement nodes, the set of directly adjacent points, and the local maximum node. Since the ridge line is attached to the actual measured surface of the irregular thin-walled part 100, the online ranging device 300 can perform a second online scan along the ridge line, thereby increasing the point density in areas where local peaks are more likely to occur.

[0038] Furthermore, the method for updating the second online scan and micro-deformation results is as follows: After the ridge line is formed, the measurement control unit 500 determines the execution path for the second online scan based on the ridge line and controls the online ranging device 300 to perform the second online scan along the ridge line. The second online scan is performed under the condition that the contact relationship between the irregular thin-walled part 100 and the fixture 200 remains unchanged, and the measurement posture of the online ranging device 300 still corresponds to the same measurement posture in the initial state acquisition and the current state acquisition. The second online scan can use the original minimum sampling interval within the rated allowable range of the online ranging device 300 to increase the point density near the ridge line.

[0039] The rescan points obtained from the second online scan also need to undergo range checks, invalid echo checks, and coordinate transformations. If a rescan point exceeds the effective range of the online ranging device, has no valid echo, or cannot be reliably acquired due to viewpoint obstruction, it is marked as invalid. For valid second online scan results, the point cloud registration unit 600 converts them to unified coordinates and establishes a correspondence with the initial reference point cloud.

[0040] In one implementation, after the second online scan, the directed distance between the current measurement surface corresponding to the current measurement point cloud and the initial reference surface corresponding to the initial reference point cloud is calculated along the unit local normal of the initial reference surface corresponding to the initial reference point cloud. For a rescan point or a valid measurement node, the calculation relationship of the directed distance can be expressed by the following formula:

[0041] In the above formula, Indicates position The directed distance, Indicates the initial reference surface at position point, Indicates the current position of the measurement surface. point, Indicates the initial reference surface at position The unit local normal. The unit local normal is normalized before participating in the dot product, so that... The modulus is The unit local normal can be obtained by fitting the neighborhood plane of the initial reference point cloud, interpolating the triangular mesh surface normals of the initial reference point cloud, or the surface normals of the nominal CAD model. Therefore, It is a length vector. It is a dimensionless unit direction vector. Maintain the dimension of length.

[0042] When updating the micro-deformation results, the yield ridge line and its directly adjacent areas are updated using the results of the second online scan, while other areas retain the micro-deformation results obtained by comparing the current measured point cloud with the initial reference point cloud. This avoids excessively long online measurement times caused by high-density repeated scanning of the entire surface, while ensuring higher point density for the yield ridge line position and peak values ​​on the yield ridge line. The micro-deformation result output unit 800 outputs the micro-deformation distribution of the entire surface, the position of the yield ridge line, the peak values ​​on the yield ridge line, the key section curves, and an explanation of the invalid areas.

[0043] The full-surface micro-deformation distribution is used to display the overall surface deviation of the irregular thin-walled part 100 after the process node; the yield ridge position is used to identify the continuous surface area guided by the second online scan; the peak value on the yield ridge is used to characterize the maximum local micro-deformation at the yield ridge position; the key section curve is used to show the trend of micro-deformation results changing with the surface path on a specified section; the invalid area description is used to record areas that are not measured, have invalid echoes, or exceed the effective range of the online ranging device.

[0044] Furthermore, the methods for handling repeated scans, point deviations, and the dispersion of repeated measurements are as follows: To verify the stability of the micro-deformation results under the current state, a repeat scan is performed on the current state without changing the irregular thin-walled part 100, the fixture 200, or the measurement posture. The repeat scan can reuse the same scan trajectory, the same direction of movement, and the same measurement posture acquired under the current state, or it can execute the same path as the second online scan for the yielding ridge line and its directly adjacent areas.

[0045] In one implementation, for the same location The positional deviation between the repeated scan results and the current measurement surface in the current measurement point cloud can be expressed by the following formula:

[0046] In the above formula, Indicates position Positional deviation, Indicates the current location of the measured point cloud. point, Indicates the location of repeated scan results. The formula is used to illustrate that the positional deviation can be determined by the difference in spatial points at the same location; specifically, the degree of dispersion of repeated measurements can be output in the form of maximum positional deviation, average positional deviation, or standard deviation.

[0047] If the positional deviation between the repeated scan result and the current measured point cloud is within the nominal uncertainty range of the online ranging device 300, it indicates that the current micro-deformation result has stable repeatability. If the positional deviation increases significantly, it is necessary to combine the standard part remeasurement results to distinguish between online ranging device 300 drift, actual changes in the irregular thin-walled part 100, changes in the contact relationship of the fixture 200, or local invalid echoes. For positions with inconsistent repeated scans, obstructed viewing angles, or exceeding the effective range of the online ranging device, the invalid marking should continue to be maintained.

[0048] As an optional implementation, the surface reflection state of the irregular thin-walled part 100 is stable, the measurement optical path of the online ranging device 300 is unobstructed, the thickness measuring device 400 can collect the wall thickness of the measuring point at the same measurement node, and the actual support contact center of the fixture 200 can be mapped onto the surface of the part. The measurement control unit 500 collects the initial distance measurement in the initial state, and collects the current distance measurement after the process node. The point cloud registration unit 600 forms the initial reference point cloud and the current measurement point cloud and completes rigid registration. Since there are no unmeasured, invalid echoes, or measurement nodes that exceed the effective range of the online ranging device, most of the same measurement nodes can be used as valid measurement nodes to participate in the determination of the rescanning order value. At this time, the ridge line determination unit 700 calculates the rescanning order value for the valid measurement nodes. If a local thin-walled area of ​​the irregular thin-walled part 100 is far from the actual support contact center, and the difference between the current distance and the initial distance is large, then the valid measurement nodes in this area obtain a higher rescanning order value. The ridge line determination unit 700 identifies the local maximum node within the set of directly adjacent points and connects them according to the scanning sequence to form the ridge line. The online ranging device 300 performs a second online scan along the ridge line, and the micro-deformation result output unit 800 outputs the full-surface micro-deformation distribution and the peak value on the ridge line.

[0049] In one alternative implementation, two valid measurement nodes may have similar difference amplitudes, but one valid measurement node has a smaller measuring point wall thickness and a larger surface distance to the nearest support point, while the other valid measurement node has a larger measuring point wall thickness and a smaller surface distance to the nearest support point. If only the difference amplitude is compared, the two valid measurement nodes may have similar importance; however, this implementation prioritizes the valid measurement node with a smaller measuring point wall thickness and a larger surface distance to the nearest support point by using the rescan priority value. The ridge line is not simply selected as an isolated point with the largest difference amplitude, but rather a local maximum node is found within the set of directly adjacent points, and these nodes are connected in the scanning order to form a continuous surface region. Thus, the second online scan of the online ranging device 300 focuses on the ridge line and its directly adjacent region, where micro-deformation peaks are more likely to occur. The micro-deformation result output unit 800 updates this region using the second online scan result, giving the ridge line position and the peak value on the ridge line a more reliable spatial resolution.

[0050] As an optional implementation, holes, outer boundaries, sidewalls, or unstable reflection areas in the irregularly shaped thin-walled component 100 may cause some measurement nodes to be marked as unmeasured, invalid echoes, or exceeding the effective range of the online ranging device. For these measurement nodes, this implementation retains the invalid marking and excludes them from the determination of rescan order values ​​and local maximum node determination. This process avoids mistaking missing, saturated, or unreliable data as local maximum nodes. If the obstruction or invalid echo is located near the yield ridge, the online ranging device 300 can acquire data within a pre-set supplementary viewing angle, but the supplementary viewing angle must still meet the same measurement posture correspondence and rigid registration requirements. For locations that still cannot be reliably acquired, the micro-deformation result output unit 800 retains a record in the invalid area description and does not supplement with speculative data.

[0051] Although embodiments of this application have been described in conjunction with the accompanying drawings, those skilled in the art can make various modifications and variations without departing from the spirit and scope of this application, and all such modifications and variations fall within the scope defined by the appended claims.

Claims

1. An online measurement method for micro-deformation of irregularly shaped thin-walled parts, applied to irregularly shaped thin-walled parts clamped by a fixture, characterized in that: While maintaining the contact relationship between the irregularly shaped thin-walled part and the fixture, measurement nodes are set. The online ranging device collects the initial distance measurement at the same measurement node in the initial state, and collects the current distance measurement in the same measurement posture after the process node; Collect the wall thickness at the measurement point and the surface distance to the nearest support point to form an initial reference point cloud and a current measurement point cloud; Rigidly register the initial reference point cloud and the current measurement point cloud, establish the correspondence between the initial distance measurement, the current distance measurement, the wall thickness of the measurement point and the surface distance to the nearest support point on the same measurement node, and take the measurement node that completes the correspondence and is not marked as unmeasured, invalid echo or beyond the effective range of the online distance measurement device as the effective measurement node. The priority of rescanning for effective measurement nodes is determined based on the difference between the initial and current distance measurements, the wall thickness of the measurement point, and the surface distance to the nearest support point. This priority increases with the increase of the difference and the surface distance to the nearest support point, and decreases with the increase of the wall thickness of the measurement point. Without setting an artificial threshold within the set of directly adjacent points determined by the scanning order, the effective measurement nodes whose rescanning values ​​satisfy the local maximum relationship are determined as local maximum nodes, and the local maximum nodes with the relationship of directly adjacent point sets are connected according to the scanning order to form a retreat ridge. A second online scan is performed along the yielding ridge line, and the micro-deformation results of the yielding ridge line and its directly adjacent regions are updated using the results of the second online scan.

2. The online measurement method for micro-deformation of irregularly shaped thin-walled parts according to claim 1, characterized in that, When setting measurement nodes, import the nominal CAD model of the irregular thin-walled part, obtain the actual support contact center of the multi-point support fixture, map the actual support contact center onto the surface of the part, and generate the scanning trajectory and the measurement nodes intersecting the scanning trajectory according to the surface morphology and measurable area distribution of the irregular thin-walled part.

3. The online measurement method for micro-deformation of irregularly shaped thin-walled parts according to claim 2, characterized in that, Set node numbers for the measurement nodes, and use the node numbers to store the correspondence between the initial distance measurement, the current distance measurement, the wall thickness of the measurement point, and the surface distance to the nearest support point on the same-name measurement node.

4. The online measurement method for micro-deformation of an irregularly shaped thin-walled part according to claim 1, characterized in that, The online ranging device is a line laser profile sensor, which collects the initial ranging and the current ranging data according to the same scanning trajectory, the same direction of movement, and the same measurement posture.

5. The online measurement method for micro-deformation of an irregularly shaped thin-walled part according to claim 1, characterized in that, The surface distance to the nearest support point is the surface path distance from the measurement node along the surface of the part to the nearest actual support contact center, and the surface path distance is determined based on the initial reference point cloud and the actual support contact center.

6. The online measurement method for micro-deformation of an irregularly shaped thin-walled part according to claim 1, characterized in that, Measurement nodes marked as unmeasured, invalid echoes, or exceeding the effective range of the online ranging equipment are kept invalid and measurement nodes with invalid marks are excluded from the determination of rescan order values ​​and local maximum node determination.

7. The online measurement method for micro-deformation of an irregularly shaped thin-walled part according to claim 1, characterized in that, Before rigid registration, point clouds from different measurement perspectives are transferred to a unified coordinate system, and rigid stitching involving only overall translation and rotation is performed within the overlap zone to form the current measurement point cloud while keeping the geometric position of the original measurement points unchanged.

8. The online measurement method for micro-deformation of an irregularly shaped thin-walled part according to claim 1, characterized in that, The set of directly adjacent points consists of measurement nodes that are directly adjacent to the effective measurement nodes in the scanning sequence. The yielding ridge is formed by connecting the local maximum nodes with the relationship of the set of directly adjacent points in the scanning sequence and is attached to the actual measured surface of the irregular thin-walled part.

9. The online measurement method for micro-deformation of an irregularly shaped thin-walled part according to claim 1, characterized in that, After the second online scan, along the local normal of the initial reference surface corresponding to the initial reference point cloud, the directed distance between the current measurement surface corresponding to the current measurement point cloud and the initial reference surface corresponding to the initial reference point cloud is calculated. The yield ridge and its directly adjacent regions are updated using the results of the second online scan. Other regions retain the micro-deformation results obtained by comparing the current measurement point cloud with the initial reference point cloud. The full surface micro-deformation distribution, yield ridge position, peak value on the yield ridge, key section curve, and invalid region description are output.

10. The online measurement method for micro-deformation of an irregularly shaped thin-walled part according to claim 1, characterized in that, Under the condition that the irregular thin-walled part, the fixture and the measurement posture remain unchanged, the current state is repeatedly scanned, the positional deviation between the repeated scan result and the current measurement point cloud is compared, and the degree of dispersion of repeated measurement is output according to the positional deviation.

Images