An intake passage protection cover geometry reconstruction method based on multi-source data fusion

Abstract

This invention discloses a method for reconstructing the geometry of an air intake protective shield based on multi-source data fusion, aiming to accurately reconstruct the complex rib structure of the air intake protective shield's internal cavity. The method includes: constructing an outer surface reference model and an inner wall base point set in the same coordinate system; performing robust smooth fitting on the base point set to obtain an unribbed inner wall base surface model, and selecting rib structure base points by calculating the normal residuals; obtaining a rib network skeleton diagram based on the rib structure base points; reconstructing parameterized branch rib models for the branch segments of the rib network, and constructing continuous node patch models for the intersection node segments; finally, fusing the unribbed inner wall base surface, branch rib models, and node patch models to obtain a three-dimensional reconstruction model of the internal cavity. This invention can accurately and automatically reconstruct the internal geometry containing a complex rib network from measured data without relying on the original design model, providing a reliable digital model foundation for subsequent processing and analysis.

Description

Technical Field

[0001] This invention relates to the field of complex component geometry measurement technology, specifically to a method for measuring and constructing the geometry of an intake manifold protective cover casting blank. Background Technology

[0002] The air intake shield is a critical aerodynamic and protective structure installed at the very front of an aircraft engine. Its main functions are to guide airflow, protect internal components from damage by foreign objects, and serve as part of the supporting structure. Air intake shields are typically characterized by complex shapes, narrow internal spaces, severe localized obstruction, and ribbed structures on the inner walls. The complex reinforcing rib design within the air intake shield determines whether it meets the requirements for extreme mechanical and thermal environments, as well as lightweight design. Therefore, during the subsequent manufacturing process of the air intake shield, it is necessary to obtain a true three-dimensional model and geometric shape of the cast blank to study and verify its quality and performance.

[0003] Existing methods for acquiring the three-dimensional structure of geometric components, including three-dimensional scanning methods, can acquire geometric data of the outer surface of parts relatively well. However, for deep cavity parts such as the casting blank of the air intake protective cover, the internal area is often difficult to directly acquire complete internal cavity point cloud data due to the inaccessibility of light, the limited field of view, and the obstruction of the rib structure. As a result, the scanning results can only reflect the outer surface and a small amount of visible area, and cannot fully characterize the internal cavity geometry.

[0004] Furthermore, there is often a large deviation between the cast blank of the air intake protective cover and the theoretical design model. If the design model is relied on too much for direct correction, it is easy to weaken the ability of the cast blank of the air intake protective cover to express the true geometric features, and it is also difficult to adapt to application scenarios where there is no design model or the design model has a large deviation. Summary of the Invention

[0005] To improve the accuracy of constructing the actual three-dimensional structural model of the air intake protective cover, this invention provides a method for reconstructing the geometric structure of the air intake protective cover based on multi-source data fusion.

[0006] The technical solution is as follows:

[0007] This invention provides a method for reconstructing the geometry of an air intake protective shield based on multi-source data fusion, which includes the following steps:

[0008] Acquire point cloud data of the outer surface of the air intake protective cover and construct a reference model of the outer surface of the air intake protective cover;

[0009] Obtain the ultrasonic measurement points on the inner wall of the air intake protective cover, and construct the set of base points on the inner wall of the air intake protective cover in the same coordinate system as the outer surface reference model;

[0010] Robust smooth fitting is performed on the inner wall base point set to obtain the unreinforced inner wall base surface model of the air intake protective cover, and the projection point of each inner wall base point in the inner wall base point set on the unreinforced inner wall base surface model is obtained.

[0011] Based on the normal residual of each inner wall base point and the projection point corresponding to each inner wall base point, inner wall base points whose normal residuals meet the preset residual threshold are selected as rib structure base points. Connectivity analysis is performed on the rib structure base points to obtain rib structure candidate regions.

[0012] The skeleton of the candidate rib structure is extracted to obtain a rib network skeleton diagram, and the candidate rib structure is divided into branch segment region and intersection node segment region according to the rib network skeleton diagram.

[0013] Extract the centerline, rib height, and rib width of the branch segment region, reconstruct a parameterized branch rib model based on the centerline, rib height, and rib width, and extract the rib structure base point of the intersection node segment region and the boundary curve of each branch rib connected to the rib structure base point of the intersection node segment region. Construct an intersection node patch model based on the rib structure base point of the intersection node segment region and the boundary curve.

[0014] Geometric Boolean operations are performed on the unreinforced inner wall base model, the branch rib model, and the intersection node patch model to obtain a reconstructed three-dimensional model of the air intake protective cover.

[0015] Preferably, the acquisition of point cloud data of the outer surface of the air intake protective cover includes:

[0016] At least three standard spheres are arranged on the outer surface of the inlet protective cover casting blank, and the inlet protective cover casting blank is three-dimensionally scanned to obtain the outer surface scanning point cloud data.

[0017] The inlet protective cover casting blank is fixed on a machine tool, and the center coordinates of the standard ball are measured by a machine tool probe.

[0018] Based on the coordinates of the center of the standard sphere, a unified coordinate relationship is established between the scanning coordinate system containing the outer surface scan point cloud data and the machine tool measurement coordinate system.

[0019] Preferably, the step of obtaining ultrasonic thickness measurement data of the inner wall of the air intake protective cover includes:

[0020] The surface coordinates of each ultrasonic measuring point on the outer surface of the blank are obtained by using a machine tool probe.

[0021] The thickness of each ultrasonic measuring point is measured using an ultrasonic thickness gauge to obtain the thickness value corresponding to each ultrasonic measuring point.

[0022] Further preferably, the set of base points for constructing the inner wall of the air intake protective cover includes:

[0023] Based on the aforementioned outer surface reference model, calculate the unit normal vector pointing inwards from each ultrasonic measuring point, and calculate the inner wall base point. coordinates The calculation formula is:

[0024] ;

[0025] in, Indicates the first Coordinates of the inner wall base points Indicates the first Surface coordinates of each ultrasonic measuring point Indicates the first Thickness values ​​corresponding to each ultrasonic measuring point;

[0026] The set of inner wall base points is composed of all the inner wall base points. .

[0027] Preferably, the step of selecting inner wall base points whose normal residuals meet a preset residual threshold as rib structure base points based on the normal residuals of each inner wall base point and the projection points corresponding to each inner wall base point includes:

[0028] For each inner wall base point Determine the inner wall base point on the unreinforced inner wall base surface. Corresponding projection points and the unit normal at the projection point Calculate the inner wall base point Corresponding normal residual The normal residual The calculation formula is:

[0029] ;

[0030] Preset residual threshold If the normal residual satisfy:

[0031] ;

[0032] Then the normal residual The corresponding inner wall base points were selected as rib structure base points.

[0033] Preferably, connected component analysis is performed on the base points of the rib structure to obtain candidate regions for the rib structure, including:

[0034] A measurement point adjacency graph is constructed for the rib structure base points, and a connected component analysis is performed on the measurement point adjacency graph to obtain high residual connected regions;

[0035] Candidate regions for rib structures are selected based on the area, number of points, and aspect ratio characteristics of the high residual connected regions.

[0036] Preferably, the skeleton of the candidate rib structure is extracted to obtain a rib network skeleton diagram, including:

[0037] The candidate regions of the rib structure are subjected to skeletonization processing, the central skeleton line of the candidate regions is extracted, and a rib network skeleton diagram representing the rib connection relationship is generated based on the topology of the skeleton line. :

[0038] ;

[0039] in, A set of skeleton nodes. It is the set of skeleton edges.

[0040] Preferably, the reconstructed parameterized branch reinforcement model includes:

[0041] For each branch segment region corresponding to each edge in the rib network skeleton diagram, extract its branch centerline;

[0042] Construct a cross section along the centerline of the branch, at the parameter position. At a given cross-section, the local reinforcement height of the cross-section is defined as the peak value of the residuals of the reinforcement structural base points within the cross-section neighborhood or the average value of the peak value neighborhood. ;

[0043] The distance between the left and right boundaries corresponding to the point where the residual distribution within the cross section decays to a preset proportion of the peak value is used as the local stiffener width of the cross section. Based on the branch centerline and local rib height Local rib width And predefined root transition parameters, and reconstruct the parameterized branch bar model using a sweep algorithm.

[0044] Preferably, the construction of the cross-node patch model includes:

[0045] A smooth surface that satisfies both positional and tangential continuity with the boundary curve is constructed as the patch model for the intersection node.

[0046] Preferably, after obtaining the reconstructed three-dimensional model of the air intake protective cover, the intersection of the normal directions at each ultrasonic measuring point on the three-dimensional model is obtained to obtain the predicted thickness of the model.

[0047] The model-predicted thickness is compared with the measured thickness. If the thickness error in a local area exceeds a preset threshold, ultrasonic measuring points are densified in that local area, and the three-dimensional model of the intake duct protective cover is reconstructed based on the densified ultrasonic measuring points for iterative correction.

[0048] Compared with the prior art, the beneficial effects of the technical solution of the present invention include at least the following:

[0049] (1) Through a multi-source data fusion strategy, this method combines three-dimensional scanning that can acquire the complete shape with ultrasonic thickness measurement that can penetrate the measurement, integrates multi-source data using a unified coordinate method, and thus acquires complete point cloud data of the internal cavity space; and through robust smooth fitting, it realizes the reconstruction of discrete point data into a continuous model of the outer surface, automatically identifies the topology of the rib network through connected domain analysis and skeleton extraction, and generates continuous and smooth internal cavity geometry through parametric sweep (branch ribs) and boundary constraint surface fitting (cross nodes), and finally realizes the reconstruction of the actual three-dimensional model of the air intake protective cover.

[0050] (2) This method divides the identified rib network of the air intake protective cover into regular branch segments and irregular intersection segments. For branch segments, parametric sweep modeling based on centerline, rib height, and rib width is used, which is efficient and regular. For intersection segments, surface patches that satisfy G1 continuity are constructed using the original measurement points in the region and the boundaries of adjacent ribs. Based on the partitioning modeling strategy, this method can automatically and accurately handle single ribs, cross ribs, and complex node connection areas without manual intervention. Attached Figure Description

[0051] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0052] Figure 1 This is a flowchart of a method for reconstructing the geometry of an air intake protective cover based on multi-source data fusion, provided in an embodiment of the present invention.

[0053] Figure 2 This is a schematic diagram of the standard ball pasting position provided as an example in an embodiment of the present invention;

[0054] Figure 3 This is an exemplary schematic diagram of the missing point cloud data region provided in an embodiment of the present invention;

[0055] Figure 4 This is an exemplary mapping diagram of the machine tool coordinate system and the scanning coordinate system provided in an embodiment of the present invention;

[0056] Figure 5 This is a schematic diagram of the location of ultrasonic measuring points provided in an embodiment of the present invention;

[0057] Figure 6 This is an exemplary schematic diagram of ultrasonic measuring point offset provided in an embodiment of the present invention;

[0058] Figure 7 This is an exemplary schematic diagram of branch segment region extraction provided in an embodiment of the present invention;

[0059] Figure 8 This is an exemplary schematic diagram of the extraction of the intersection node segment region provided in an embodiment of the present invention;

[0060] Figure 9 This is a schematic diagram of the cross-section of a branch reinforcement provided as an example in an embodiment of the present invention. Detailed Implementation

[0061] The technical solutions in the embodiments of this specification will be clearly and completely described below with reference to the accompanying drawings. In the context of this invention, the term "protective cover" means "intake duct protective cover".

[0062] The following description provides examples and does not limit the scope, applicability, or examples set forth in the claims. Changes may be made to the function and arrangement of the described elements without departing from the scope of this specification. Various processes or components may be appropriately omitted, substituted, or added to the examples. For example, the described methods may be performed in a different order than described, and various steps may be added, omitted, or combined. Furthermore, features described with respect to some examples may be combined into other examples.

[0063] Traditionally, the three-dimensional structure of geometric components is obtained through three-dimensional scanning methods. However, for deep cavity components such as air intake protective cover casting blanks, the internal area is often difficult to obtain complete internal cavity point cloud data directly due to the inaccessibility of light, limited field of view, and rib structure obstruction, making it impossible to accurately characterize its internal cavity geometry.

[0064] In view of this, existing technologies still have the problem of difficulty in accurately reconstructing the complex geometry of the air intake shield. Therefore, this application is proposed to improve the accuracy of the three-dimensional structure reconstruction of the air intake shield.

[0065] This embodiment uses a cast blank of an air intake protective cover as the object. The air intake protective cover has a complex external curved surface, a narrow internal cavity, localized rib structures, and partially invisible internal cavity areas. This cast blank of the air intake protective cover is used as the object of the three-dimensional structural digital reconstruction in this embodiment, referring to... Figure 1 As shown, Figure 1This embodiment of the specification provides a method for reconstructing the geometry of an air intake shield based on multi-source data fusion, which includes the following steps:

[0066] Step S1: Obtain point cloud data of the outer surface of the air intake protective cover and construct a reference model of the outer surface of the air intake protective cover;

[0067] Step S2: Obtain the ultrasonic measurement points of the inner wall of the air intake protective cover, and construct the inner wall base point set of the air intake protective cover in the same coordinate system as the outer surface reference model;

[0068] Step S3: Perform robust smooth fitting on the inner wall base point set to obtain the unreinforced inner wall base surface model of the air intake protective cover; and obtain the projection point of each inner wall base point in the inner wall base point set on the unreinforced inner wall base surface model.

[0069] Step S4: Based on the normal residual of each inner wall base point and the projection point corresponding to each inner wall base point, select the inner wall base points whose normal residual meets the preset residual threshold as rib structure base points, and perform connected component analysis on the rib structure base points to obtain rib structure candidate regions.

[0070] Step S5: Extract the skeleton of the candidate rib structure to obtain the rib network skeleton diagram, and divide the candidate rib structure into branch segment region and intersection node segment region according to the rib network skeleton diagram.

[0071] Step S6: Extract the centerline, rib height, and rib width of the branch segment region, and reconstruct the parameterized branch rib model based on the centerline, rib height, and rib width; and extract the rib structure base point of the intersection node segment region and the boundary curve of each branch rib connected to the rib structure base point of the intersection node segment region, and construct the intersection node patch model based on the rib structure base point of the intersection node segment region and the boundary curve.

[0072] Step S7: Perform geometric Boolean operations on the unreinforced inner wall base model, the branch rib model, and the cross node patch model to obtain the reconstructed three-dimensional model of the air intake protective cover.

[0073] In one embodiment, acquiring point cloud data of the outer surface of the air intake protective cover includes:

[0074] At least three standard spheres are arranged on the outer surface of the inlet protective cover casting blank, and the inlet protective cover casting blank is three-dimensionally scanned to obtain the outer surface scanning point cloud data.

[0075] The inlet protective cover casting blank is fixed on a machine tool, and the center coordinates of the standard ball are measured by a machine tool probe.

[0076] Based on the coordinates of the center of the standard sphere, a unified coordinate relationship is established between the scanning coordinate system containing the outer surface scan point cloud data and the machine tool measurement coordinate system.

[0077] Explained, in this embodiment, the multi-source data includes scanned point cloud coordinate data and ultrasonic testing of the protective cover's wall thickness data. Using the center of a standard sphere as a reference, a unified spatial reference is established between the scanned point cloud coordinate system and the machine tool's wall thickness measurement coordinate system. This multi-source data is then fused to ultimately obtain an accurate fit to the actual geometry of the air intake protective cover.

[0078] Illustratively, the establishment of a unified spatial reference between the scanned point cloud coordinate system and the machine tool wall thickness measurement coordinate system, such as... Figure 2 As shown, the intake manifold protective cover casting blank is fixed on the machine tool worktable, and the standard spheres are measured by the machine tool probe to obtain the center coordinates of each standard sphere. Using the center coordinates of the standard spheres, a spatial correspondence between the machine tool measurement coordinate system and the scanning point cloud coordinate system can be established.

[0079] After arranging the standard spheres, a 3D scan of the cast blank of the air intake duct protective cover was performed to obtain the scanned point cloud data of the outer surface of the blank. Illustratively, as... Figure 3 As shown, due to the inaccessible and ribbed structures within the inner cavity of the intake manifold protective cover casting blank, and the inherent limitations of optical measurement methods, the obtained scanning point cloud data mainly includes the outer surface of the blank and a small portion of the visible inner surface area, while data is missing from the deep cavity and areas obscured by the ribbed structures. After completing the standard sphere scanning point cloud measurement, ultrasonic measuring points are arranged on the outer surface of the blank. In another specific embodiment, the ultrasonic measuring points are preferably arranged along a preset measuring line, which can be a regular grid measuring line or a curved measuring line generated based on the curved shape of the outer surface of the blank, with the latter being preferred; furthermore... Figure 4 As shown, the machine tool probe touches and measures each ultrasonic measuring point one by one, recording the surface coordinates of each ultrasonic measuring point to form a set of ultrasonic measuring points on the outer surface of the blank. Through this method, the precise surface coordinates of each ultrasonic measuring point under a unified measurement system can be obtained, providing a foundation for the fusion of ultrasonic thickness measurement data and scanning point cloud data.

[0080] In one embodiment, to improve the stability and accuracy of spatial registration, six standard spheres are used for reference. Figure 3 As shown.

[0081] In one embodiment, acquiring ultrasonic thickness measurement data of the inner wall of the air intake protective cover includes:

[0082] The surface coordinates of each ultrasonic measuring point on the outer surface of the blank are obtained by using a machine tool probe.

[0083] The thickness of each ultrasonic measuring point is measured using an ultrasonic thickness gauge to obtain the thickness value corresponding to each ultrasonic measuring point.

[0084] For example, let the first The surface coordinates of each ultrasonic measuring point are: The corresponding thickness value is Then the thickness values ​​of all measuring points constitute the ultrasonic thickness measurement dataset.

[0085] In this step, the thickness value corresponding to each ultrasonic measuring point is calculated based on the ultrasonic echo time of flight, satisfying the following:

[0086]

[0087] in, The speed at which ultrasound waves propagate in the material being tested. For the first Echo flight time at each ultrasonic measurement point.

[0088] In another specific embodiment, after thickness measurement, the outer surface scanning point cloud data is stitched, filtered, and fitted to construct a reference model of the blank's outer surface. Further, based on the coordinates of the center of the standard sphere measured by the machine tool probe and the coordinates of the center of the standard sphere fitted from the scanning point cloud, a unified coordinate relationship is established between the scanning point cloud coordinate system and the machine tool measurement coordinate system, accurately mapping the surface coordinates of each ultrasonic measuring point to the coordinate system of the outer surface reference model.

[0089] In another embodiment, the set of base points for constructing the inner wall of the air intake protective cover includes:

[0090] Based on the aforementioned outer surface reference model, calculate the unit normal vector pointing inwards from each ultrasonic measuring point, and calculate the inner wall base point. coordinates The calculation formula is:

[0091] ;

[0092] in, Indicates the first Coordinates of the inner wall base points Indicates the first Surface coordinates of each ultrasonic measuring point Indicates the first Thickness values ​​corresponding to each ultrasonic measuring point;

[0093] The set of inner wall base points is composed of all the inner wall base points. .

[0094] Each ultrasonic measuring point can be understood as a location where the thickness of the inner wall is measured using ultrasound. Explained in this embodiment, for each ultrasonic measuring point, the unit normal along the inward direction at that point is calculated based on the outer surface reference model. Let the... The unit normal along the inward direction at each ultrasonic measuring point is Then it can be combined with its thickness value Offset along the normal direction (see reference) Figure 6 The inner wall base points are constructed. It can be understood that the set of inner wall base points reflects the spatial response position of the inner wall in the direction normal to the outer surface reference model based on the ultrasonic thickness measurement information, which includes both the geometric information of the inner wall base surface and the geometric information of the protruding parts of the rib structure.

[0095] Since the set of inner wall base points contains both inner wall base surface information and rib structure protrusion information, if a uniform surface fitting is directly applied to all inner wall base points, the local rib structures will exert a pulling effect on the overall fitting result, causing the fitted surface to deviate from the true inner wall base surface. Therefore, in the above embodiment, to reduce the influence of local protruding base points on the fitting result, a robust smoothing fitting method is used to fit the inner wall base point set to obtain an unribbed inner wall base surface. The robust smoothing fitting method can be effectively applied to surface fitting of multi-source data in the coordinate system constructed in this method.

[0096] In one embodiment, after obtaining the unreinforced inner wall base surface, based on the normal residual of each inner wall base point and the projection point corresponding to each inner wall base point, inner wall base points whose normal residual meets a preset residual threshold are selected as ribbed structure base points, including:

[0097] For each inner wall base point Determine the inner wall base point on the unreinforced inner wall base surface. Corresponding projection points and the unit normal at the projection point Calculate the inner wall base point Corresponding normal residual The normal residual The calculation formula is:

[0098] ;in, For the first The normal residual of each inner wall base point relative to the unreinforced inner wall base surface; For projection point The unit outward normal vector of the unreinforced inner wall base surface is determined by the tangent plane of the unreinforced inner wall base surface at that point, and its direction points to one side of the inside of the part.

[0099] Preset residual threshold If the normal residual satisfy:

[0100] ;

[0101] Then the normal residual The corresponding inner wall base points were selected as rib structure base points.

[0102] satisfy The measuring points corresponding to the inner wall base points are high residual measuring points, and the inner wall base points corresponding to the high residual measuring points are rib structure base points.

[0103] The threshold It can be determined according to different needs. For example, it can be determined based on a combination of the measurement uncertainty of ultrasonic thickness measurement and the minimum protrusion height of the rib structure. The principle for determining the value is as follows: The error must be greater than the overall measurement error of ultrasonic thickness measurement (including the accuracy of the ultrasonic thickness gauge itself, the error in normal offset calculation, and the coordinate registration error) to avoid misjudging measurement noise as stiffening structure; at the same time, it should be less than the minimum design protrusion height of the stiffening structure to ensure that the stiffening structure's base point can be effectively identified. In engineering practice, Typically, the overall ultrasonic thickness measurement error can be taken as 2 to 3 times, and not greater than 1 / 3 of the minimum rib protrusion height. In this embodiment, for the specimen being measured, the overall ultrasonic thickness measurement error is approximately 0.1 to 0.2 mm, and the minimum rib protrusion height is approximately 1.5 mm. Take 0.5mm.

[0104] Understandably, in the above method, for the inner wall base area of ​​the protective cover that is actually unreinforced, the residual of the inner wall base point relative to the unreinforced inner wall base surface is usually close to zero; for the area of ​​the protective cover that actually has a ribbed structure, the residual of the inner wall base point relative to the unreinforced inner wall base surface is usually significantly greater than zero.

[0105] In one embodiment, connected component analysis is performed on the rib structure base points to obtain candidate regions for the rib structure, including:

[0106] A measurement point adjacency graph is constructed for the rib structure base points, and a connected component analysis is performed on the measurement point adjacency graph to obtain high residual connected regions;

[0107] Candidate regions for rib structures are selected based on the area, number of points, and aspect ratio characteristics of the high residual connected regions.

[0108] In one specific embodiment, a measurement point adjacency graph is constructed based on the measurement line arrangement order of each ultrasonic measurement point, and a connected component analysis is performed on the measurement point adjacency graph for measurement points with high residuals to obtain several high residual connected regions:

[0109]

[0110] in, Indicates the first A high residual connected region.

[0111] To remove local noise points and small protrusions in non-reinforced structures, each high residual connected region is filtered based on the number of points, area, and aspect ratio. Only high residual connected regions that meet preset conditions are retained as candidate regions for reinforced structures. The filtering conditions include:

[0112]

[0113] and:

[0114]

[0115] in, For the first The number of measurement points contained in a high residual connected region. and These represent the scales of the high residual connected region along the principal and secondary directions, respectively. The threshold number of points. The aspect ratio threshold is used. In engineering practice, the point count threshold and aspect ratio threshold can be set based on the initial information of the test part (before casting) of the intake manifold protective cover casting blank.

[0116] In one embodiment, after obtaining candidate regions for rib structures, the skeleton of the candidate regions is extracted to construct a rib network skeleton diagram:

[0117]

[0118] in, A set of skeleton nodes. This is the skeleton edge set. Based on the topological relationships of the skeleton diagram, the candidate rib structure region is divided into branch segment regions and intersection node segment regions. The branch segment region corresponds to the portion of a single rib extending along its length, while the intersection node segment region corresponds to the location where two or more ribs intersect, branch, or connect. The branch segment region extraction process is described in [reference needed]. Figure 7 Reference for the process of extracting intersection node segments Figure 8 .

[0119] In one embodiment, the reconstructed parameterized branch reinforcement model includes:

[0120] For each branch segment region corresponding to each edge in the rib network skeleton diagram, extract its branch centerline;

[0121] Construct a cross section along the centerline of the branch, at the parameter position. At a given cross-section, the local reinforcement height of the cross-section is defined as the peak value of the residuals of the reinforcement structural base points within the cross-section neighborhood or the average value of the peak value neighborhood. ;

[0122] The distance between the left and right boundaries corresponding to the point where the residual distribution within the cross section decays to a preset proportion of the peak value is used as the local stiffener width of the cross section. Based on the branch centerline and local rib height Local rib width And predefined root transition parameters, and reconstruct the parameterized branch bar model using a sweep algorithm.

[0123] For example, for each branch segment region, its branch centerline is extracted, and a cross-sectional plane is constructed along the branch centerline. In the... Branch reinforcement parameter location At, in the cross-sectional neighborhood The peak value of the internal residual or the average value of the peak neighborhood is taken as the local height of the cross section, and is denoted as:

[0124]

[0125] Meanwhile, the distance between the left and right boundaries corresponding to the point where the residual distribution within the cross section decays to a preset threshold value is taken as the local width of the cross section, denoted as:

[0126]

[0127] in, and The first Branch reinforcement at parameter location The left and right boundary points on the cross section.

[0128] The peak neighborhood average value refers to: in the cross-sectional neighborhood First, locate the rib structure base point with the largest residual (i.e., the peak point), and then take the arithmetic mean of the residuals of several rib structure base points within a preset radius centered on the peak point.

[0129] The two value selection methods described above have different applicable scenarios: when the distribution of reinforcement structural base points within the cross-sectional neighborhood is dense and the measurement noise is low, using the residual peak value as the local reinforcement height can more sensitively reflect the true height of the reinforcement top; when there are individual outlier high residual points within the cross-sectional neighborhood or the measurement noise is high, using the peak neighborhood average value can effectively suppress the influence of outliers, making the modeling results more robust. In engineering practice, one method can be selected based on the density of ultrasonic measurement points and the measurement noise level. In this embodiment, using the peak neighborhood average value can obtain a smoother and more stable branch reinforcement model.

[0130] The root transition parameters describe the transition morphology at the connection between the branch reinforcement and the unreinforced inner wall base surface. These parameters can be obtained based on casting process requirements or by fitting the decay gradient of the reinforcement structure base point residual near the reinforcement width boundary. This ensures a smooth transition of the reconstructed branch reinforcement model at the connection with the unreinforced inner wall base surface, avoiding sharp edges. In a specific embodiment, predefined root transition parameters include, for example, the root transition fillet radius. and the root transition angle Among them, the root transition fillet radius The radius of the transition arc between the sidewall of the branch reinforcement and the base surface of the unreinforced inner wall, and the root transition angle. The angle of inclination of the sidewall of the branch reinforcement relative to the normal of the cross section.

[0131] Furthermore, based on the relationship between the branch reinforcement cross-sectional shape and the base surface, the root transition parameters of the branch reinforcement at that location can be extracted; based on the branch centerline and local reinforcement height... Local width And the root transition parameters, using a sweep algorithm, are used to reconstruct the branch reinforcement model from this data, for reference. Figure 9 As shown.

[0132] For the intersection node segment region, the inner wall base points within the node's neighborhood and the boundary curves of each branch reinforcement connected to the intersection node segment region are extracted. The intersection node patch model is then reconstructed based on these inner wall base points and branch reinforcement boundary curves. To ensure a smooth transition between the node region and each branch reinforcement, a smooth surface is constructed that satisfies both positional continuity (C0 continuity) and tangential continuity (G1 continuity) with the boundary curves of each adjacent branch reinforcement model, serving as the intersection node patch model.

[0133] Finally, geometric Boolean operations are performed on the unreinforced inner wall base model, the branch rib model, and the intersection node patch model to obtain a reconstructed 3D model of the air intake protective shield. This step, based on geometric Boolean operations, forms a single, seamless, and geometrically complete 3D model of the reconstructed air intake protective shield from the unreinforced inner wall base model, the branch rib model, and the intersection node patch model. The geometric Boolean operation process is as follows: First, the branch rib model is merged with the unreinforced inner wall base model to form a unified whole with the base surface; then, the intersection node patch model is merged again with the assembly obtained in the previous step to fill or reinforce the node areas. Compared to the initial modeling, this model contains all the structural features actually possessed by the air intake protective shield and can be directly used for subsequent engineering analysis.

[0134] In one embodiment, after obtaining the reconstructed three-dimensional model of the air intake protective cover, the intersection of the normal directions at each ultrasonic measuring point on the three-dimensional model is obtained to obtain the predicted thickness of the model.

[0135] The model-predicted thickness is compared with the measured thickness. If the thickness error in a local area exceeds a preset threshold, ultrasonic measuring points are densified in that local area, and the three-dimensional model of the intake duct protective cover is reconstructed based on the densified ultrasonic measuring points for iterative correction.

[0136] The preset threshold is determined based on the machining allowance requirements of the air intake protective cover and the comprehensive measurement error of ultrasonic thickness measurement. The principle for its selection is that the threshold should be greater than the comprehensive ultrasonic thickness measurement error to avoid misinterpreting normal measurement fluctuations as model deviations; simultaneously, it should be less than the minimum machining allowance allowed by subsequent processing steps to ensure that the accuracy of the reconstructed model meets the processing requirements. In engineering practice, this threshold is typically taken as 2 to 3 times the comprehensive ultrasonic thickness measurement error. In this embodiment, the threshold is set to 0.3 mm.

[0137] The inlet duct protective shield geometry reconstruction method described in the above embodiments combines a multi-source data fusion strategy with 3D scanning that yields a complete shape and ultrasonic thickness measurement that allows for penetration measurement. It integrates multi-source data using a unified coordinate system to obtain complete point cloud data of the internal cavity. Robust smooth fitting reconstructs a continuous model of the outer surface from discrete point data. Connected domain analysis and skeleton extraction automatically identify the topology of the rib network. Parametric sweeping (branch ribs) and boundary constraint surface fitting (intersection nodes) generate a continuous and smooth internal cavity geometry, ultimately achieving the reconstruction of the actual 3D model of the inlet duct protective shield. This reconstruction method significantly improves reconstruction accuracy and completeness, more realistically reflecting the actual geometric state of the cast blank, including deviations caused by casting deformation. This provides a reliable digital model foundation for subsequent processing of the aircraft nose section and the inspection and analysis of the inlet duct protective shield.

[0138] The various embodiments in this specification are described in a progressive manner. The same or similar parts between the various embodiments can be referred to each other. Each embodiment focuses on describing the differences from other embodiments.

[0139] Furthermore, embodiments of this specification also provide a computer-readable storage medium storing instructions that, when executed on a computer or processor, cause the computer or processor to perform one or more steps in the embodiments of the above-described method for reconstructing the geometry of the air intake shield based on multi-source data fusion. If the constituent modules of the above-described electronic device are implemented as software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium.

[0140] In the above embodiments, implementation can be achieved, in whole or in part, through software, hardware, firmware, or any combination thereof. When implemented in software, it can be implemented, in whole or in part, as a computer program product. A computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, all or part of the flow or function according to the embodiments of this specification is generated. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in or transmitted through a computer-readable storage medium. The computer instructions can be transmitted from one website, computer, server, or data center to another website, computer, server, or data center via wired (e.g., coaxial cable, fiber optic, Digital Subscriber Line (DSL)) or wireless (e.g., infrared, wireless, microwave, etc.) means. The computer-readable storage medium can be any available medium that a computer can access or a data storage device such as a server or data center that integrates one or more available media. The available media can be magnetic media (e.g., floppy disks, hard disks, magnetic tapes), optical media (e.g., Digital Versatile Discs (DVDs)), or semiconductor media (e.g., Solid State Disks (SSDs)).

[0141] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. This program can be stored in a computer-readable storage medium, and when executed, it can include the processes of the embodiments of the methods described above. The aforementioned storage medium includes various media capable of storing program code, such as ROM, RAM, magnetic disks, or optical disks. Unless otherwise specified, the technical features of this embodiment and its implementation can be combined arbitrarily.

[0142] The above embodiments are merely preferred embodiments described in this specification and are not intended to limit the scope of this specification. Any modifications and improvements made by those skilled in the art to the technical solutions of this specification without departing from the spirit of this specification should fall within the protection scope defined by the claims of this specification.

Claims

1. A method for reconstructing the geometric structure of an air intake protective shield based on multi-source data fusion, characterized in that: Includes the following steps: Acquire point cloud data of the outer surface of the air intake protective cover and construct a reference model of the outer surface of the air intake protective cover; Obtain the ultrasonic measurement points on the inner wall of the air intake protective cover, and construct the set of base points on the inner wall of the air intake protective cover in the same coordinate system as the outer surface reference model; Robust smooth fitting is performed on the inner wall base point set to obtain the unreinforced inner wall base surface model of the air intake protective cover, and the projection point of each inner wall base point in the inner wall base point set on the unreinforced inner wall base surface model is obtained. Based on the normal residual of each inner wall base point and the projection point corresponding to each inner wall base point, inner wall base points whose normal residuals meet the preset residual threshold are selected as rib structure base points. Connectivity analysis is performed on the rib structure base points to obtain rib structure candidate regions. The skeleton of the candidate rib structure is extracted to obtain a rib network skeleton diagram, and the candidate rib structure is divided into branch segment region and intersection node segment region according to the rib network skeleton diagram. Extract the centerline, rib height, and rib width of the branch segment region, reconstruct a parameterized branch rib model based on the centerline, rib height, and rib width, and extract the rib structure base point of the intersection node segment region and the boundary curve of each branch rib connected to the rib structure base point of the intersection node segment region. Construct an intersection node patch model based on the rib structure base point of the intersection node segment region and the boundary curve. Geometric Boolean operations are performed on the unreinforced inner wall base model, the branch rib model, and the intersection node patch model to obtain a reconstructed three-dimensional model of the air intake protective cover.

2. The method for reconstructing the geometric structure of the air intake protective cover according to claim 1, characterized in that: The acquisition of point cloud data of the outer surface of the air intake protective cover includes: At least three standard spheres are arranged on the outer surface of the inlet protective cover casting blank, and the inlet protective cover casting blank is three-dimensionally scanned to obtain the outer surface scanning point cloud data. The inlet protective cover casting blank is fixed on a machine tool, and the center coordinates of the standard ball are measured by a machine tool probe. Based on the coordinates of the center of the standard sphere, a unified coordinate relationship is established between the scanning coordinate system containing the outer surface scan point cloud data and the machine tool measurement coordinate system.

3. The method for reconstructing the geometric structure of the air intake protective cover according to claim 2, characterized in that: The acquisition of ultrasonic thickness measurement data of the inner wall of the air intake protective cover includes: The surface coordinates of each ultrasonic measuring point on the outer surface of the blank are obtained by using a machine tool probe. The thickness of each ultrasonic measuring point is measured using an ultrasonic thickness gauge to obtain the thickness value corresponding to each ultrasonic measuring point.

4. The method for reconstructing the geometric structure of the air intake protective cover according to claim 3, characterized in that: The set of base points for constructing the inner wall of the air intake protective cover includes: Based on the aforementioned outer surface reference model, calculate the unit normal vector pointing inwards from each ultrasonic measuring point, and calculate the inner wall base point. coordinates The calculation formula is: ； in, Indicates the first Coordinates of the inner wall base points Indicates the first Surface coordinates of each ultrasonic measuring point Indicates the first Thickness values ​​corresponding to each ultrasonic measuring point; The set of inner wall base points is composed of all the inner wall base points. .

5. The method for reconstructing the geometric structure of the air intake protective cover according to claim 1, characterized in that: The step of selecting inner wall base points whose normal residuals meet a preset residual threshold as rib structure base points based on the normal residuals of each inner wall base point and the corresponding projection points includes: For each inner wall base point Determine the inner wall base point on the unreinforced inner wall base surface. Corresponding projection points and the unit normal at the projection point Calculate the inner wall base point Corresponding normal residual The normal residual The calculation formula is: ； Preset residual threshold If the normal residual satisfy: ； Then the normal residual The corresponding inner wall base points were selected as rib structure base points.

6. The method for reconstructing the geometric structure of the air intake protective cover according to claim 1, characterized in that: Connectivity analysis is performed on the base points of the rib structure to obtain candidate regions for the rib structure, including: A measurement point adjacency graph is constructed for the rib structure base points, and a connected component analysis is performed on the measurement point adjacency graph to obtain high residual connected regions; Candidate regions for rib structures are selected based on the area, number of points, and aspect ratio characteristics of the high residual connected regions.

7. The method for reconstructing the geometric structure of the air intake protective cover according to claim 1, characterized in that: The skeleton of the rib structure candidate region is extracted to obtain a rib network skeleton diagram, including: The candidate regions of the rib structure are subjected to skeletonization processing, the central skeleton line of the candidate regions is extracted, and a rib network skeleton diagram representing the rib connection relationship is generated based on the topology of the skeleton line. : ； in, A set of skeleton nodes. It is the set of skeleton edges.

8. The method for reconstructing the geometric structure of the air intake protective cover according to claim 1, characterized in that: The reconstructed parameterized branch rib model includes: For each branch segment region corresponding to each edge in the rib network skeleton diagram, extract its branch centerline; Construct a cross section along the centerline of the branch, at the parameter position. At a given cross-section, the local reinforcement height of the cross-section is defined as the peak value of the residuals of the reinforcement structural base points within the cross-section neighborhood or the average value of the peak value neighborhood. ; The distance between the left and right boundaries corresponding to the point where the residual distribution within the cross section decays to a preset proportion of the peak value is used as the local stiffener width of the cross section. Based on the branch centerline and local rib height Local rib width And predefined root transition parameters, and reconstruct the parameterized branch bar model using a sweep algorithm.

9. The method for reconstructing the geometric structure of the air intake protective cover according to claim 1, characterized in that: The construction of the cross-node patch model includes: A smooth surface that satisfies both positional and tangential continuity with the boundary curve is constructed as the patch model for the intersection node.

10. The method for reconstructing the geometric structure of the air intake protective cover according to claim 1, characterized in that: After obtaining the reconstructed three-dimensional model of the air intake protective cover, the intersection of the normal directions at each ultrasonic measuring point on the three-dimensional model is obtained to obtain the predicted thickness of the model. The model-predicted thickness is compared with the measured thickness. If the thickness error in a local area exceeds a preset threshold, ultrasonic measuring points are densified in that local area, and the three-dimensional model of the intake duct protective cover is reconstructed based on the densified ultrasonic measuring points for iterative correction.

Images