Method and system for improving wall thickness measurement precision of high-temperature alloy hollow blade casting

Abstract

The invention discloses a method and system for improving the wall thickness measurement precision of a high-temperature alloy hollow blade casting, relates to the technical field of high-temperature alloy hollow blade detection, and can ensure the accuracy and stability of scanning data by designing a CT detection tool so as to improve the wall thickness measurement precision. By carrying out dead-angle-free three-dimensional scanning on the casting, geometrical shape information of the casting can be obtained in all directions, and comprehensive data support is provided for subsequent modeling and measurement. Scanning data are integrated to obtain three-dimensional scanning point cloud data of the forging, so that the actual form of the casting can be better presented, and a basis is provided for subsequent analysis and processing; according to the method, effective feature data are obtained based on the casting type, the modeling mode is determined, then the three-dimensional modeling result of the casting is obtained, the accurate wall thickness measurement result can be obtained through fitting with a solid model, the blade wall thickness detection can be rapidly completed, and the detection precision error is small.

Description

Technical Field

[0001] The invention relates to the technical field of high-temperature alloy hollow blade detection, and more particularly to a method and a system for improving the wall thickness measurement accuracy of a high-temperature alloy hollow blade casting. Background Art

[0002] At present, with the development of civil aviation technology and the unremitting pursuit of high-performance military aircraft by various countries, the performance indicators of aircraft engines are constantly improving, and the temperature at the front inlet of the turbine is constantly increasing. In order to improve the temperature bearing capacity of turbine blades and extend their service life, most advanced aircraft engine turbine blades adopt thin-walled hollow structures, and use the inner cavity as an air cooling channel to achieve composite air cooling of the blades. Therefore, wall thickness is an extremely important geometric dimension indicator of hollow turbine blades, and the accuracy of wall thickness detection directly affects the safety and life of the engine.

[0003] The widely used method for measuring the wall thickness of high-temperature alloy hollow blade castings is the ultrasonic thickness gauge, which is easy to operate and low in cost. By obtaining the elastic constant and density of the hollow blade to be tested; calculating the theoretical sound velocity of the ultrasonic wave at each wall thickness measurement point; using the ultrasonic thickness gauge to measure the wall thickness of each measurement point at a fixed sound velocity, so as to obtain the echo time of the corresponding measurement point; according to the theoretical sound velocity and echo time, the wall thickness value of the corresponding measurement point is obtained. However, the above method has poor positioning accuracy of the measurement position. The coordinates of the measurement position of the wall thickness measurement position of the hollow blade are defined on the drawing, and the positioning accuracy of ultrasonic manual measurement is only about 3mm. Even with the use of auxiliary tooling, it is impossible to achieve a high positioning accuracy.

[0004] Therefore, how to improve the measurement accuracy of the wall thickness of high-temperature alloy hollow blade castings is an urgent problem that technical personnel in this field need to solve. Summary of the invention

[0005] In view of this, the present invention provides a method and system for improving the wall thickness measurement accuracy of a high-temperature alloy hollow blade casting, so as to solve the problems existing in the background technology.

[0006] In order to achieve the above object, the present invention adopts the following technical solution:

[0007] On the one hand, a method for improving the wall thickness measurement accuracy of a high-temperature alloy hollow blade casting is provided, comprising:

[0008] Design CT inspection fixtures and use a 6-point coordinate system to clamp the blades and fix the castings so that the stacking axis of the parts is perpendicular to the CT turntable during scanning;

[0009] Perform three-dimensional scanning of the casting without blind spots to obtain scanning data;

[0010] Integrate the scanning data to obtain the 3D scanning point cloud data of the forging;

[0011] Based on the type of the casting, obtaining effective feature data of the three-dimensional scanning point cloud data;

[0012] Determining a modeling method of the casting according to the type of the effective feature data;

[0013] Acquire a three-dimensional modeling result of the casting based on the modeling method and the effective feature data;

[0014] The 3D modeling results are fitted with the solid model, and the measurement results are obtained based on the wall thickness measurement position.

[0015] Optionally, the CT inspection tooling includes a blade back, a clamping surface, and a base;

[0016] The blade back is completely in contact with the convex surface of the blade casting; the blade back also includes a detection hole;

[0017] The clamping surface positions the casting to ensure that the casting is perpendicular to the base;

[0018] The base provides a stable supporting function.

[0019] Optionally, after integrating the scanning data to obtain the three-dimensional scanning point cloud data of the forging, the three-dimensional scanning point cloud data is processed, including filtering and gridding the three-dimensional scanning point cloud data, removing the point cloud of the detection tooling, and obtaining the casting point cloud data.

[0020] Optionally, before integrating the scanned data, the scanned data is enhanced, specifically: graying the scanned data to obtain a grayscale image;

[0021] A window is constructed with any pixel point in the grayscale image, and the degree of disorder in the window is calculated and assigned to the pixel point; a disordered image is obtained according to the disorder degrees of all pixels in the grayscale image, and the skewness of the disordered image is calculated, and the grayscale image is divided into a plurality of sub-blocks based on the skewness;

[0022] Taking any sub-block in the grayscale image as a target block, obtaining the gradient amplitude of each pixel in the target block and the mean gradient amplitude of the target block; obtaining a label segmentation threshold based on the gradient amplitudes of all pixels in the target block and the mean gradient amplitude of the target block; the pixel points in the target block whose grayscale values ​​are greater than the label segmentation threshold are reserved points, and the pixel points in the target block whose grayscale values ​​are less than the label segmentation threshold are processed points;

[0023] The processing points in each sub-block are obtained, and the richness of the sub-block is obtained based on the processing points in each sub-block; the sub-blocks are classified according to the richness of each sub-block to obtain suitable sub-blocks, merged sub-blocks and split sub-blocks, and all the merged sub-blocks and split sub-blocks are updated to suitable sub-blocks by splitting and merging operations; the processing points in all the suitable sub-blocks are enhanced to obtain an enhanced image.

[0024] Optionally, obtaining effective feature data of the three-dimensional scanning point cloud data based on the type of the casting specifically includes:

[0025] Acquire the structural complexity of the casting; when the structural complexity of the casting is less than or equal to the preset complexity, the effective feature data is feature point data;

[0026] When the structural complexity of the casting is greater than the preset complexity, the effective characteristic data is characteristic line data.

[0027] Optionally, determining the modeling method of the casting according to the type of the valid feature data includes:

[0028] In the case where the effective feature data is the feature point data, determining the casting modeling method as point-to-line-to-surface method according to the curvature information of the feature points; or,

[0029] When the effective feature data is irregular feature line data and the amount of the effective feature data is small, it is determined that the modeling method of the casting is triangulated network-based modeling.

[0030] Optionally, the specific process of fitting the three-dimensional modeling result with the solid model is:

[0031] According to the rotation matrix and translation matrix obtained in the previous iteration, all the triangular facets in the solid model of the previous iteration are rigidly transformed to obtain the solid model of this iteration;

[0032] Matching the triangular facets that meet the fitting conditions with each triangular facet in the solid model one by one from the 3D modeling result, and forming a facet pair from the two successfully matched triangular facets;

[0033] The rotation matrix and translation matrix of this iteration are obtained based on the least square sum of the fitting parameters between the reference patch and the corresponding target patch in all the patch pairs of this iteration;

[0034] According to the judgment that the absolute value of the difference between the least squares sum of this iteration and the least squares sum of the previous iteration is less than the threshold, it is known to stop the iteration process, and the fitting of the three-dimensional modeling result and the solid model is completed.

[0035] Optionally, the patch pair specifically includes:

[0036] The triangular facet in the solid model of this iteration is used as a reference facet, and a triangular facet with the smallest fitting parameter with the reference facet is selected from the three-dimensional modeling result as a pre-selected facet;

[0037] According to whether the fitting parameters of the pre-selected patch and the reference patch meet a threshold, it is known that the pre-selected patch is a target patch of the reference patch, and the reference patch and the target patch form a patch pair;

[0038] The fitting parameter is one of the facet spacing or the facet angle.

[0039] In another aspect, a system for improving the wall thickness measurement accuracy of a high-temperature alloy hollow blade casting is provided, comprising:

[0040] Prepare the module, design the CT inspection tooling, and use a 6-point coordinate system to clamp the blade to fix the casting so that the stacking axis of the part is perpendicular to the CT turntable during the scanning process;

[0041] The data scanning module performs a three-dimensional scan of the casting without blind spots to obtain scanning data;

[0042] The data integration module integrates the scanning data to obtain the 3D scanning point cloud data of the forging;

[0043] A feature data extraction module, which obtains effective feature data of the three-dimensional scanning point cloud data based on the type of the casting;

[0044] A modeling method determining module, which determines the modeling method of the casting according to the type of the effective feature data;

[0045] A modeling result acquisition module, which acquires a three-dimensional modeling result of the casting based on the modeling method and the effective feature data;

[0046] The measurement result output module fits the 3D modeling results with the solid model and obtains the measurement results based on the wall thickness measurement position.

[0047] It can be seen from the above technical solutions that the present invention discloses a method and system for improving the wall thickness measurement accuracy of high-temperature alloy hollow blade castings, which has the following beneficial effects compared with the prior art:

[0048] 1. Improve measurement accuracy: By designing CT inspection tooling and using a 6-point coordinate system to clamp the blade to fix the casting, the stacking axis of the part is perpendicular to the CT turntable during the scanning process, which can ensure the accuracy and stability of the scanning data, thereby improving the accuracy of wall thickness measurement.

[0049] 2. 3D scanning without blind spots: 3D scanning of castings without blind spots can obtain all-round geometric shape information of castings, including internal structure and external surface, providing comprehensive data support for subsequent modeling and measurement.

[0050] 3. Integrate scanning data: Integrate the scanning data to obtain the 3D scanning point cloud data of the forging, which can better present the actual shape of the casting and provide a basis for subsequent analysis and processing.

[0051] 4. Modeling and measurement results: Based on the casting type, effective feature data is obtained, and the modeling method is determined, and then the three-dimensional modeling result of the casting is obtained. By fitting with the solid model, accurate wall thickness measurement results can be obtained. The method of the present invention can quickly complete the blade wall thickness detection with small detection accuracy error. BRIEF DESCRIPTION OF THE DRAWINGS

[0052] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on the provided drawings without paying creative work.

[0053] Figure 1 A flow chart of the method provided by the present invention;

[0054] Figure 2 A schematic diagram of the system structure provided by the present invention;

[0055] Figure 3 This is a schematic diagram of the detection tooling structure provided by the present invention. DETAILED DESCRIPTION

[0056] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0057] The embodiment of the present invention discloses a method for improving the wall thickness measurement accuracy of a high-temperature alloy hollow blade casting. Figure 1 As shown, including:

[0058] Design CT inspection fixtures and use a 6-point coordinate system to clamp the blades and fix the castings so that the stacking axis of the parts is perpendicular to the CT turntable during scanning;

[0059] Perform three-dimensional scanning of the casting without blind spots to obtain scanning data;

[0060] Integrate the scanning data to obtain the 3D scanning point cloud data of the forging;

[0061] Based on the type of casting, obtain effective feature data of 3D scanning point cloud data;

[0062] Determine the casting modeling method based on the type of valid feature data;

[0063] Obtain the three-dimensional modeling results of castings based on the modeling method and effective feature data;

[0064] The 3D modeling results are fitted with the solid model, and the measurement results are obtained based on the wall thickness measurement position.

[0065] In a specific embodiment, Figure 3 As shown, the CT inspection tooling includes a blade back 3, a clamping surface 1, and a base 4;

[0066] The blade back 3 is completely in contact with the convex surface of the blade casting; the blade back also includes a detection hole 2;

[0067] Clamping surface 1, positioning the casting to ensure that the casting is perpendicular to the base;

[0068] The base 4 provides a stable support.

[0069] In a specific embodiment, after the scanning data is integrated to obtain the three-dimensional scanning point cloud data of the forging, the three-dimensional scanning point cloud data is processed, including filtering and gridding the three-dimensional scanning point cloud data, removing the point cloud of the detection tooling, and obtaining the casting point cloud data.

[0070] In a specific embodiment, before integrating the scan data, the scan data is further enhanced, specifically: graying the scan data to obtain a grayscale image;

[0071] A window is constructed with any pixel point in the grayscale image, and the degree of disorder in the window is calculated and assigned to the pixel point; a disordered image is obtained according to the disorder degree of all pixels in the grayscale image, and the skewness of the disordered image is calculated, and the grayscale image is divided into multiple sub-blocks based on the skewness;

[0072] Taking any sub-block in the grayscale image as the target block, obtain the gradient amplitude of each pixel in the target block and the mean gradient amplitude of the target block; obtain the marking segmentation threshold based on the gradient amplitude of all pixels in the target block and the mean gradient amplitude of the target block; the pixel points in the target block whose grayscale value is greater than the marking segmentation threshold are reserved points, and the pixel points in the target block whose grayscale value is less than the marking segmentation threshold are processed points;

[0073] The processing points in each sub-block are obtained, and the richness of the sub-block is obtained based on the processing points in each sub-block; the sub-blocks are classified according to the richness of each sub-block to obtain suitable sub-blocks, merged sub-blocks and split sub-blocks, and all merged sub-blocks and split sub-blocks are updated to suitable sub-blocks by splitting and merging operations; the processing points in all suitable sub-blocks are enhanced to obtain an enhanced image.

[0074] Specifically, the steps of calculating the skewness of the chaotic image include:

[0075] Construct a chaos histogram of the chaos image, obtain the probability of each chaos degree based on the chaos histogram, and calculate the skewness of the chaos image based on the probability:

[0076]

[0077] Where t represents the skewness of the chaotic image; N represents the number of all pixels in the chaotic image; p i Represents the frequency corresponding to the i-th degree of chaos in the chaos histogram.

[0078] The step of dividing the degree image into a plurality of sub-blocks based on the skewness comprises:

[0079] Based on big data, a commonly used block size is obtained as the initial size. When the skewness of the chaotic image is less than 0, the initial size is increased; when the skewness of the chaotic image is greater than 0, the initial size is reduced; based on the final determined size, the grayscale image is divided into multiple sub-blocks.

[0080] The step of obtaining a marking segmentation threshold based on the gradient amplitudes of all pixels in the target block and the average value of the gradient amplitudes of the target block includes: obtaining a threshold correlation coefficient of the target block based on the gradient amplitudes of all pixels in the target block and the average value of the gradient amplitudes of the target block:

[0081]

[0082] Where D represents the threshold correlation coefficient; f i Represents the gradient amplitude of the i-th pixel; represents the mean value of the gradient amplitude corresponding to the target block; c′ represents the size of the target block; β represents a hyperparameter; e represents a natural constant;

[0083] The label segmentation threshold of the target block is obtained based on the threshold correlation coefficient of the target block.

[0084] The calculation method of the marker segmentation threshold is:

[0085]

[0086] Where F represents the marker segmentation threshold; represents the mean value of the gradient amplitude corresponding to the target block; D represents the threshold correlation coefficient; f max Indicates the maximum value of the gradient amplitude of the pixels in the target block.

[0087] The step of obtaining a marking segmentation threshold based on the gradient magnitudes of all pixels in the target block and the average value of the gradient magnitudes of the target block also includes:

[0088] The necessity of correction of the target block's mark segmentation threshold is obtained based on the target block and its neighboring sub-blocks. When the necessity of correction is greater than the preset threshold, the target block's mark segmentation threshold is corrected. The calculation method of the necessity of correction is:

[0089]

[0090] Where L represents the necessity of modifying the target block marker segmentation threshold; F k represents the label segmentation threshold of the kth neighborhood sub-block corresponding to the target block; F k+1 Indicates the marking segmentation threshold of the k+1th neighboring sub-block corresponding to the target block; represents a hyperparameter;

[0091] Obtain the contrast sub-block of the target block, and calculate the correction of the marker segmentation threshold of the target block according to the contrast sub-block:

[0092]

[0093] Among them, F′ represents the corrected marker segmentation threshold of the target block; represents the mean value of the gradient amplitude of the contrast sub-block; f max ′ represents the maximum value of the gradient amplitude of all pixels in the contrast sub-block; f i Represents the gradient magnitude corresponding to the i-th pixel.

[0094] The step of obtaining the richness of the sub-block based on the processing points in each sub-block comprises:

[0095] According to the processing points in each sub-block, the corresponding richness of the sub-block is obtained as follows:

[0096]

[0097] Among them, u represents the richness corresponding to the sub-block; M v Indicates the frequency of occurrence of the vth gray level in the sub-block; M v+1 represents the frequency of occurrence of the v+1th gray level in the sub-block; V represents the number of all gray levels in the sub-block; p(v) represents the probability of occurrence of the vth gray level in the sub-block.

[0098] In a specific embodiment, based on the type of casting, obtaining effective feature data of the three-dimensional scanning point cloud data specifically includes:

[0099] Obtaining the structural complexity of the casting; when the structural complexity of the casting is less than or equal to the preset complexity, the effective feature data is the feature point data;

[0100] When the structural complexity of the casting is greater than the preset complexity, the effective characteristic data is the characteristic line data.

[0101] For example, when the effective feature data is feature point data, the feature point data may be obtained by a curvature estimation method or a principal component analysis method. When the effective feature data is feature line data, the feature line data may be obtained by a feature line fitting method.

[0102] In a specific embodiment, determining the modeling method of the casting according to the type of effective feature data includes:

[0103] When the effective feature data is feature point data, the casting modeling method is determined to be point-to-line-to-surface method according to the curvature information of the feature points; or,

[0104] When the effective feature data is irregular feature line data and the amount of effective feature data is small, it is determined that the modeling method of the casting is triangulated network-based modeling.

[0105] In a specific embodiment, the specific process of fitting the three-dimensional modeling result with the solid model is:

[0106] According to the rotation matrix and translation matrix obtained in the previous iteration, all the triangular facets in the solid model of the previous iteration are rigidly transformed to obtain the solid model of this iteration;

[0107] Matching the triangular facets that meet the fitting conditions with each triangular facet in the solid model one by one from the 3D modeling result, and forming a facet pair from the two successfully matched triangular facets;

[0108] The rotation matrix and translation matrix of this iteration are obtained based on the least square sum of the fitting parameters between the reference patch and the corresponding target patch in all the patch pairs of this iteration;

[0109] According to the judgment that the absolute value of the difference between the least squares sum of this iteration and the least squares sum of the previous iteration is less than the threshold, it is known to stop the iteration process, and the fitting of the three-dimensional modeling result and the solid model is completed.

[0110] In a specific embodiment, forming a patch pair specifically includes:

[0111] The triangular facet in the solid model of this iteration is used as the reference facet, and a triangular facet with the smallest fitting parameter with the reference facet is selected from the 3D modeling result as the pre-selected facet;

[0112] According to the fitting parameters of the pre-selected patch and the reference patch meeting the threshold, it is known that the pre-selected patch is the target patch of the reference patch, and the reference patch and the target patch form a patch pair;

[0113] The fitting parameter is one of the patch spacing or the patch angle.

[0114] When the fitting parameter is the patch spacing, the barycentric coordinates of each reference patch are translated according to the barycentric coordinates of the solid model of this iteration to obtain the barycentric coordinates of each reference patch after translation; the barycentric coordinates of each target patch are translated according to the barycentric coordinates of the 3D modeling result to obtain the barycentric coordinates of each target patch after translation;

[0115] According to the barycentric coordinates of each reference face patch after translation, the barycentric coordinates of each target face patch after translation, the number of face patch pairs, and the rotation matrix, a least square sum function of the distances between the reference face patches in all face patch pairs and the corresponding target face patches is constructed;

[0116] Solve the rotation matrix that minimizes the least squares sum function as the rotation matrix of this iteration; obtain the translation matrix of this iteration based on the centroid coordinates of the entity, the centroid coordinates of the 3D modeling result, and the rotation matrix of this iteration.

[0117] The method for obtaining the centroid coordinates of the solid model is as follows: obtaining the centroid coordinates of each reference surface piece, and taking the average value of the centroid coordinates of each reference surface piece as the centroid coordinates of the solid model;

[0118] The method for obtaining the barycentric coordinates of the three-dimensional modeling result is: obtaining the barycentric coordinates of each target facet, and taking the average value of the barycentric coordinates of each target facet as the barycentric coordinates of the three-dimensional modeling result.

[0119] On the other hand, a system for improving the wall thickness measurement accuracy of high temperature alloy hollow blade castings is provided. Figure 2 As shown, including:

[0120] Prepare the module, design the CT inspection tooling, and use a 6-point coordinate system to clamp the blade to fix the casting so that the stacking axis of the part is perpendicular to the CT turntable during the scanning process;

[0121] The data scanning module performs a three-dimensional scan of the casting without blind spots to obtain scanning data;

[0122] The data integration module integrates the scanning data to obtain the 3D scanning point cloud data of the forging;

[0123] Feature data extraction module, which obtains effective feature data of 3D scanning point cloud data based on the type of casting;

[0124] Modeling method determination module, which determines the modeling method of the casting according to the type of effective feature data;

[0125] Modeling result acquisition module, which acquires the three-dimensional modeling result of the casting based on the modeling method and effective feature data;

[0126] The measurement result output module fits the 3D modeling results with the solid model and obtains the measurement results based on the wall thickness measurement position.

[0127] In this specification, each embodiment is described in a progressive manner, and each embodiment focuses on the differences from other embodiments. The same or similar parts between the embodiments can be referred to each other. For the device disclosed in the embodiment, since it corresponds to the method disclosed in the embodiment, the description is relatively simple, and the relevant parts can be referred to the method part.

[0128] The above description of the disclosed embodiments enables one skilled in the art to implement or use the present invention. Various modifications to these embodiments will be apparent to one skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention will not be limited to the embodiments shown herein, but rather to the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A method for improving the wall thickness measurement accuracy of high-temperature alloy hollow blade castings, characterized in that: include: Design CT inspection fixtures and use a 6-point coordinate system to clamp the blades and fix the castings so that the stacking axis of the parts is perpendicular to the CT turntable during scanning; Perform three-dimensional scanning of the casting without blind spots to obtain scanning data; Integrate the scanning data to obtain the 3D scanning point cloud data of the forging; Based on the type of the casting, obtaining effective feature data of the three-dimensional scanning point cloud data; Determining a modeling method of the casting according to the type of the effective feature data; Acquire a three-dimensional modeling result of the casting based on the modeling method and the effective feature data; The 3D modeling results are fitted with the solid model, and the measurement results are obtained based on the wall thickness measurement position.

2. A method for improving the wall thickness measurement accuracy of a high-temperature alloy hollow blade casting according to claim 1, characterized in that: The CT inspection tooling includes a blade back, a clamping surface, and a base; The blade back is completely in contact with the convex surface of the blade casting; the blade back also includes a detection hole; The clamping surface positions the casting to ensure that the casting is perpendicular to the base; The base provides a stable supporting function.

3. The method for improving the wall thickness measurement accuracy of a high-temperature alloy hollow blade casting according to claim 1, characterized in that: After the scanning data is integrated to obtain the 3D scanning point cloud data of the forging, the 3D scanning point cloud data is processed, including filtering and gridding the 3D scanning point cloud data, removing the point cloud of the detection tooling, and obtaining the casting point cloud data.

4. The method for improving the wall thickness measurement accuracy of a high-temperature alloy hollow blade casting according to claim 1, characterized in that: Before integrating the scanned data, the scanned data is also enhanced, specifically: graying the scanned data to obtain a grayscale image; A window is constructed with any pixel point in the grayscale image, and the degree of disorder in the window is calculated and assigned to the pixel point; a disordered image is obtained according to the disorder degrees of all pixels in the grayscale image, and the skewness of the disordered image is calculated, and the grayscale image is divided into a plurality of sub-blocks based on the skewness; Taking any sub-block in the grayscale image as a target block, obtaining the gradient amplitude of each pixel in the target block and the mean gradient amplitude of the target block; obtaining a label segmentation threshold based on the gradient amplitudes of all pixels in the target block and the mean gradient amplitude of the target block; the pixel points in the target block whose grayscale values ​​are greater than the label segmentation threshold are reserved points, and the pixel points in the target block whose grayscale values ​​are less than the label segmentation threshold are processed points; The processing points in each sub-block are obtained, and the richness of the sub-block is obtained based on the processing points in each sub-block; the sub-blocks are classified according to the richness of each sub-block to obtain suitable sub-blocks, merged sub-blocks and split sub-blocks, and all the merged sub-blocks and split sub-blocks are updated to suitable sub-blocks by splitting and merging operations; the processing points in all the suitable sub-blocks are enhanced to obtain an enhanced image.

5. The method for improving the wall thickness measurement accuracy of high-temperature alloy hollow blade castings according to claim 1, characterized in that: The obtaining of effective feature data of the three-dimensional scanning point cloud data based on the type of the casting specifically includes: Acquire the structural complexity of the casting; when the structural complexity of the casting is less than or equal to the preset complexity, the effective feature data is feature point data; When the structural complexity of the casting is greater than the preset complexity, the effective characteristic data is characteristic line data.

6. The method for improving the wall thickness measurement accuracy of high-temperature alloy hollow blade castings according to claim 1, characterized in that: Determining the modeling method of the casting according to the type of the effective feature data includes: In the case where the effective feature data is the feature point data, determining the casting modeling method as point-to-line-to-surface method according to the curvature information of the feature points; or, When the effective feature data is irregular feature line data and the amount of the effective feature data is small, it is determined that the modeling method of the casting is triangulated network-based modeling.

7. The method for improving the wall thickness measurement accuracy of high-temperature alloy hollow blade castings according to claim 1, characterized in that: The specific process of fitting the three-dimensional modeling result with the solid model is as follows: According to the rotation matrix and translation matrix obtained in the previous iteration, all the triangular facets in the solid model of the previous iteration are rigidly transformed to obtain the solid model of this iteration; Matching the triangular facets that meet the fitting conditions with each triangular facet in the solid model one by one from the 3D modeling result, and forming a facet pair from the two successfully matched triangular facets; The rotation matrix and translation matrix of this iteration are obtained based on the least square sum of the fitting parameters between the reference patch and the corresponding target patch in all the patch pairs of this iteration; According to the judgment that the absolute value of the difference between the least squares sum of this iteration and the least squares sum of the previous iteration is less than the threshold, it is known to stop the iteration process, and the fitting of the three-dimensional modeling result and the solid model is completed.

8. The method for improving the wall thickness measurement accuracy of high-temperature alloy hollow blade castings according to claim 1, characterized in that: The patch pairs specifically include: The triangular facet in the solid model of this iteration is used as a reference facet, and a triangular facet with the smallest fitting parameter with the reference facet is selected from the three-dimensional modeling result as a pre-selected facet; According to whether the fitting parameters of the pre-selected patch and the reference patch meet a threshold, it is known that the pre-selected patch is a target patch of the reference patch, and the reference patch and the target patch form a patch pair; The fitting parameter is one of the facet spacing or the facet angle.

9. A system for improving the wall thickness measurement accuracy of high-temperature alloy hollow blade castings, characterized in that: include: Prepare the module, design the CT inspection tooling, and use a 6-point coordinate system to clamp the blade to fix the casting so that the stacking axis of the part is perpendicular to the CT turntable during the scanning process; The data scanning module performs a three-dimensional scan of the casting without blind spots to obtain scanning data; The data integration module integrates the scanning data to obtain the 3D scanning point cloud data of the forging; A feature data extraction module, which obtains effective feature data of the three-dimensional scanning point cloud data based on the type of the casting; A modeling method determining module, which determines the modeling method of the casting according to the type of the effective feature data; A modeling result acquisition module, which acquires a three-dimensional modeling result of the casting based on the modeling method and the effective feature data; The measurement result output module fits the 3D modeling results with the solid model and obtains the measurement results based on the wall thickness measurement position.

Images