Production management method and device for flat surface products

Abstract

The present application relates to the field of image processing technology, and discloses a production management method and device for flat curved surface products. The method comprises: performing feature extraction processing on the surface morphology of the flat curved surface product material to obtain a surface feature topology map; performing non-uniform curvature partitioning processing on the surface feature topology map to obtain a flat curved surface product processing area division scheme and performing variable curvature compensation path planning processing to obtain an adaptive processing trajectory sequence; performing real-time surface error detection processing on the adaptive processing trajectory sequence to obtain local morphology deviation data; performing surface progressive forming compensation processing on the local morphology deviation data to obtain dynamic correction parameters; performing surface quality feature correlation tracing processing on the dynamic correction parameters to obtain a full-cycle quality assessment report for the flat curved surface product. The present disclosure realizes real-time error detection, dynamic compensation and quality tracing in the processing of flat curved surface products, thereby improving the processing accuracy and quality stability of flat curved surface products.

Description

Technical Field

[0001] The present application relates to the field of image processing, and in particular to a production management method and device for flat and curved surface products. Background Art

[0002] Due to its wide application in aerospace, mold manufacturing, precision machinery and other fields, the processing accuracy and quality control of flat curved surface products have always been the research focus of the manufacturing industry. The existing production management method of flat curved surface products mainly uses fixed process parameters for processing control, and adjusts the processing parameters through manual experience or simple online detection during the processing. Although this method has certain operability, it is difficult to meet the processing requirements of high-precision flat curved surface products due to the lack of systematic analysis of surface features and dynamic compensation mechanism. At the same time, the quality control in the existing technology is mainly based on offline detection, which cannot realize real-time quality monitoring and dynamic adjustment during the processing process, resulting in large fluctuations in product quality and high rework rate.

[0003] The main problems with related technologies include: first, traditional flat surface processing methods lack the ability to accurately extract and analyze the surface morphology of products, making it difficult to implement differentiated processing strategies for different curvature areas; second, error compensation in the processing process mostly relies on empirical parameters and lacks a scientific compensation mechanism, making it difficult to ensure the stability of the compensation effect; third, the quality traceability system is imperfect, making it difficult to establish the correlation between process parameters, processing errors and product quality, which restricts the continuous improvement of product quality. Summary of the invention

[0004] The present application provides a production management method and device for flat curved surface products, which are used to realize real-time error detection, dynamic compensation and quality traceability during the processing of flat curved surface products, thereby improving the processing accuracy and quality stability of flat curved surface products.

[0005] In the first aspect, the present application provides a production management method for flat curved surface products, which includes: performing multi-dimensional geometric feature extraction processing on the surface morphology of flat curved surface product materials to obtain a surface feature topological map; performing non-uniform curvature partitioning processing on the surface feature topological map to obtain a flat curved surface product processing area division scheme; performing variable curvature compensation path planning processing on the flat curved surface product processing area division scheme to obtain an adaptive processing trajectory sequence; performing real-time surface error detection processing on the adaptive processing trajectory sequence to obtain local morphology deviation data; performing surface incremental forming compensation processing on the local morphology deviation data to obtain dynamic correction parameters; performing surface quality feature correlation tracing processing on the dynamic correction parameters to obtain a full-cycle quality assessment report for flat curved surface products.

[0006] In a second aspect, the present application provides a production management device for flat curved surface products, the production management device for flat curved surface products comprising:

[0007] The extraction module is used to extract multi-dimensional geometric features of the surface morphology of flat and curved product materials to obtain a surface feature topology map;

[0008] A partitioning module is used to perform non-uniform curvature partitioning on the surface feature topology map to obtain a plan for partitioning the processing area of ​​the flat surface product;

[0009] A planning module, used for performing variable curvature compensation path planning processing on the plan for dividing the processing area of ​​the flat surface product to obtain an adaptive processing trajectory sequence;

[0010] A detection module, used for performing real-time surface error detection processing on the adaptive machining trajectory sequence to obtain local shape deviation data;

[0011] A compensation module, used for performing surface incremental forming compensation processing on the local shape deviation data to obtain dynamic correction parameters;

[0012] The traceability module is used to perform surface quality feature-associated traceability processing on the dynamic correction parameters to obtain a full-cycle quality assessment report for flat and curved surface products.

[0013] In the technical solution provided by this application, by extracting and processing the surface morphology of the flat surface product material in multi-dimensional geometric features, the geometric feature information of the product surface is accurately obtained, providing an accurate data basis for subsequent processing, and by processing the surface feature topology map with non-uniform curvature partitioning, the reasonable partitioning of complex surfaces is realized, ensuring the pertinence of the processing strategy. By performing variable curvature compensation path planning processing on the flat surface product processing area division scheme, an adaptive processing path based on curvature features is established, which improves the processing accuracy and efficiency. Real-time surface error detection processing is performed on the adaptive processing trajectory sequence, which realizes real-time monitoring and error identification during the processing process, and provides timely feedback information for quality control. By performing surface progressive forming compensation processing on the local morphology deviation data, a dynamic compensation mechanism is established, which effectively reduces the cumulative effect of processing errors. Finally, the dynamic correction parameters are traced by the association of surface quality features to form a full-cycle quality assessment report, and a complete quality traceability and improvement system is constructed. Through the organic combination of multiple links, this solution realizes the precise control and quality assurance of the flat surface product processing process, and significantly improves the processing accuracy and quality stability of the product. BRIEF DESCRIPTION OF THE DRAWINGS

[0014] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other accompanying drawings can be obtained based on these accompanying drawings without paying creative work.

[0015] Figure 1 This is a schematic diagram of an embodiment of a production management method for flat curved surface products in an embodiment of the present application;

[0016] Figure 2 This is a schematic diagram of an embodiment of a production management device for flat curved surface products in an embodiment of the present application. DETAILED DESCRIPTION

[0017] The embodiment of the present application provides a production management method and device for flat curved surface products. The terms "first", "second", "third", "fourth", etc. (if any) in the specification and claims of the present application and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchangeable where appropriate, so that the embodiments described here can be implemented in an order other than that illustrated or described here. In addition, the terms "including" or "having" and any variations thereof are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device that includes a series of steps or units is not necessarily limited to those steps or units that are clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0018] For ease of understanding, the specific process of the embodiment of the present application is described below. Figure 1 , an embodiment of the production management method of flat curved surface products in the embodiment of the present application includes:

[0019] Step S101, performing multi-dimensional geometric feature extraction processing on the surface morphology of the flat curved surface product material to obtain a surface feature topology map;

[0020] Step S102, performing non-uniform curvature partitioning processing on the surface feature topology map to obtain a plan for dividing the processing area of ​​the flat surface product;

[0021] Step S103, performing variable curvature compensation path planning processing on the plan for dividing the processing area of ​​the flat surface product to obtain an adaptive processing trajectory sequence;

[0022] Step S104, performing real-time surface error detection processing on the adaptive machining trajectory sequence to obtain local morphology deviation data;

[0023] Step S105, performing surface incremental forming compensation processing on the local shape deviation data to obtain dynamic correction parameters;

[0024] Step S106: Perform surface quality feature correlation tracing processing on the dynamic correction parameters to obtain a full-cycle quality assessment report for flat and curved surface products.

[0025] It is understandable that the execution subject of the present application may be a production management device for flat and curved surface products, or a terminal or a server, which is not specifically limited here. The present application embodiment is described by taking a server as the execution subject as an example.

[0026] Specifically, a high-precision three-dimensional scanning device is used to collect grid point clouds on the surface of flat curved products to obtain original surface point cloud data. For the original point cloud data obtained, noise filtering technology is used to eliminate measurement noise and extract effective surface feature information. Taking the automobile body cover as an example, the number of original data points obtained through point cloud collection is about 1 million. After noise filtering, 800,000 valid data points are retained. These data points accurately reflect the surface morphology characteristics of the body cover. The local curvature of the filtered point cloud data is calculated to determine the curvature value and distribution characteristics of each point. The data points are evenly distributed on the product surface through resampling. The boundary feature extraction of the uniformly sampled point set is performed to identify the boundary contour of the product, and preliminary partitioning is performed based on the boundary information. Then, the curvature gradient analysis of the partition result is performed, the curvature change characteristics are calculated, and the watershed algorithm is used for regional segmentation to form a candidate partitioning scheme. According to the candidate scheme, key feature points are extracted, and the surface feature topology map is generated through spatial topological reconstruction.

[0027] Based on the generated surface feature topology map, non-uniform curvature partitioning is performed. First, the curvature density distribution is calculated, and the initial region marker is obtained by using the gradient threshold segmentation method. By calculating the curvature change rate, analyzing the regional connectivity, extracting the regional boundary and identifying the curvature mutation point, a feature point sequence is formed. The feature point sequence is clustered, the processing difficulty of each region is evaluated, and the processing path constraints are analyzed. Finally, similar regions are merged, the processing area is planned, and the final processing area division scheme is generated.

[0028] For the divided processing area, variable curvature compensation path planning is performed. The curvature gradient in the area is calculated, the isoline sequence is extracted, and the initial processing path is planned. The path is adjusted and smoothed through compensation calculation and curvature error analysis. The tool posture parameters are calculated, interference check is performed, and avoidance path planning is completed. Finally, the adaptive processing trajectory sequence is generated through speed planning and trajectory fitting. During the processing, the adaptive processing trajectory sequence is subjected to real-time surface error detection. Surface data of the detection point position is collected, measurement noise is filtered, coordinate transformation and reference surface fitting are performed. The morphology error vector is calculated, the spatial distribution characteristics are analyzed, and the main curvature calculation and error accumulation analysis are performed. Local morphology deviation data is obtained through morphology feature extraction and regional statistics.

[0029] According to the acquired local morphology deviation data, the surface progressive forming compensation processing is performed. The deviation data is vectorized and the main error components are extracted. The local curvature sensitivity is analyzed, and the curvature influence coefficient and the initial compensation parameters are calculated. Through the decomposition of the progressive compensation amount and the generation of the compensation path, interference inspection and process parameter mapping are performed, and finally the dynamic correction parameters are obtained. Finally, the dynamic correction parameters are processed by quality feature association and traceability. The timing characteristics are extracted, the parameter sensitivity is analyzed, and the key quality influencing factors are identified. The quality fluctuation law is obtained through the historical data association analysis, and the quality control characteristic quantity is extracted. The process parameters are associated and analyzed, and a defect feature library is established. The feasibility of the improvement measures is verified through quality traceability analysis and improvement plan generation, and finally a full-cycle quality assessment report for flat and curved surface products is formed.

[0030] Taking the precision machining of an aircraft engine blade as an example, the surface of the blade needs to meet the surface roughness requirement of Ra0.4μm. Through the implementation of this solution, the geometric features of the blade surface are first extracted from the original point cloud data, and key feature areas such as the leading edge and trailing edge are identified. According to the curvature distribution characteristics, the blade surface is divided into high curvature areas and low curvature areas, and differentiated processing strategies are formulated for each area. During the processing, through real-time error detection and compensation, the surface roughness of the blade is controlled within Ra0.35μm, and the form and position tolerance is controlled within the range of ±0.02mm.

[0031] In the embodiment of the present application, by extracting and processing the surface morphology of the flat surface product material in multi-dimensional geometric features, the geometric feature information of the product surface is accurately obtained, providing an accurate data basis for subsequent processing, and by processing the surface feature topology map with non-uniform curvature partitioning, the reasonable partitioning of complex surfaces is realized, ensuring the pertinence of the processing strategy. By performing variable curvature compensation path planning processing on the flat surface product processing area division scheme, an adaptive processing path based on curvature features is established, which improves the processing accuracy and efficiency. Real-time surface error detection processing is performed on the adaptive processing trajectory sequence, which realizes real-time monitoring and error identification during the processing process, and provides timely feedback information for quality control. By performing surface progressive forming compensation processing on the local morphology deviation data, a dynamic compensation mechanism is established, which effectively reduces the cumulative effect of processing errors. Finally, the dynamic correction parameters are traced by associating surface quality features to form a full-cycle quality assessment report, and a complete quality traceability and improvement system is constructed. Through the organic combination of multiple links, this scheme realizes the precise control and quality assurance of the flat surface product processing process, and significantly improves the processing accuracy and quality stability of the product.

[0032] In a specific embodiment, the process of executing step S101 may specifically include the following steps:

[0033] (1) Perform grid point cloud collection and processing on the surface of flat and curved products to obtain original surface point cloud data, and perform noise filtering on the original surface point cloud data to obtain filtered point cloud data;

[0034] (2) Perform local curvature calculation on the filtered point cloud data to obtain curvature distribution data, and resample the curvature distribution data to obtain a uniform sampling point set;

[0035] (3) Perform boundary feature extraction on the uniform sampling point set to obtain boundary contour data, and partition the surface area according to the boundary contour data to obtain the initial partition result;

[0036] (4) Perform curvature gradient analysis on the initial partition results to obtain curvature change feature data, and then perform regional segmentation on the curvature change feature data using the watershed algorithm to obtain candidate partition schemes;

[0037] (5) Extract feature points from the surface according to the candidate partitioning scheme to obtain a set of key feature points, and then perform spatial topological reconstruction on the key feature point set to obtain a surface feature topology map.

[0038] Specifically, the surface feature extraction process of flat curved products first performs grid point cloud acquisition and processing, and uses a high-precision three-dimensional scanning device to scan the surface of the flat curved product. The scanning device projects a specific coded grating stripe on the product surface through the principle of structured light projection, and uses multiple CCD cameras to collect the deformed image of the grating stripe from different angles. The three-dimensional point cloud data of the product surface is reconstructed through the triangulation principle, and each point contains spatial coordinate (x, y, z) information. The original point cloud data is subjected to noise filtering, and the bilateral filtering algorithm is used to eliminate the measurement noise. The algorithm considers both the spatial distance weight and the grayscale similarity weight, which not only retains the edge features but also smoothes the noise.

[0039] The local curvature is calculated for the filtered point cloud data, a local coordinate system is established for each point, a local quadratic surface is fitted, and the principal curvature and Gaussian curvature are calculated. The curvature calculation formula is:

[0040] ,

[0041] Among them, K is the Gaussian curvature, L, N, M are the quadratic basic form coefficients of the surface, and E, G, F are the first basic form coefficients of the surface. The curvature distribution data is resampled. Based on the degree of change of the local curvature value, the sampling density is increased in the area where the curvature changes dramatically, and the sampling density is appropriately reduced in the flat area to ensure that the sampling points are evenly distributed and accurately reflect the surface characteristics. The boundary feature extraction is performed on the uniform sampling point set. First, the normal vector field of the point cloud data is calculated, and the boundary points are identified by analyzing the mutation of the normal vector. A continuous point sequence is found in the boundary point set to form a closed boundary contour. According to the boundary contour data, the surface is partitioned by the region growing method, and the region with similar curvature is selected as the seed point, which is gradually expanded to form the initial partition.

[0042] Perform curvature gradient analysis on the initial partition results and calculate the curvature change rate between adjacent areas. The curvature gradient calculation formula is:

[0043] ,

[0044] in, is the curvature gradient, and n is the direction of the normal vector. The curvature change feature data is segmented by the watershed algorithm, and the surface is divided into several sub-regions with similar curvature characteristics. The watershed algorithm regards the curvature gradient as the terrain height, and starts the flooding simulation from the local minimum point. When different waters meet, a watershed line is formed to achieve the natural segmentation of the region. Feature points are extracted according to the candidate partition scheme, and points with significant curvature changes at the partition boundary are selected as key feature points. The key feature point set is spatially topologically reconstructed, the connection relationship between the points is established, and a topological network that characterizes the surface features is constructed. By analyzing the spatial distribution and connection relationship of the feature points, a complete surface feature topology map is obtained.

[0045] Taking the compressor blade of an aircraft engine as an example, its working surface is a complex three-dimensional surface. The original data of 800×600 sampling points are obtained through grid point cloud acquisition, and about 420,000 valid data points are retained after bilateral filtering. The local curvature calculation shows that the Gaussian curvature value of the leading edge area of ​​the blade is between 0.5 and 0.8. The curvature value in the middle area is between 0.1 and 0.3 After resampling, 120,000 evenly distributed feature points were obtained. Boundary feature extraction identified key contours such as the leading edge and trailing edge of the blade, and the initial partitioning divided the blade surface into 15 sub-areas. Through curvature gradient analysis, it was found that the curvature gradient value between adjacent areas was between 0.2 and 0.5. The entire feature extraction process fully considers the geometric characteristics of flat and curved surface products, and ensures the accuracy and reliability of feature extraction through multi-level data processing.

[0046] In a specific embodiment, the process of executing step S102 may specifically include the following steps:

[0047] (1) Calculate the curvature density of the surface feature topology map to obtain curvature density distribution data, and perform gradient threshold segmentation on the curvature density distribution data to obtain the initial area mark;

[0048] (2) Calculate the curvature change rate of the initial region mark to obtain the change rate feature data, and perform regional connectivity analysis on the change rate feature data to obtain the connected domain feature;

[0049] (3) Perform boundary extraction processing on the connected domain features to obtain regional boundary line data, and perform curvature mutation point identification processing on the regional boundary line data to obtain a feature point sequence;

[0050] (4) Performing regional clustering processing on the feature point sequence to obtain processing difficulty partition data, and performing processing path constraint analysis on the processing difficulty partition data to obtain process constraint conditions;

[0051] (5) The process constraints are regionally merged to obtain a benchmark partition sequence, and the processing area planning is performed based on the benchmark partition sequence to obtain a processing area division scheme for flat and curved surface products.

[0052] Specifically, the curvature density is calculated, and the distribution of curvature values ​​in the local area of ​​each point is calculated. Specifically, a spherical neighborhood with a fixed radius is set around each point, and the curvature values ​​of all points in the neighborhood are counted to form a curvature density histogram. According to the distribution characteristics of the histogram, a statistical method is used to determine the curvature density threshold, and the surface area is divided into different curvature levels. The curvature density distribution data is segmented by gradient threshold, and an adaptive threshold method is used to dynamically adjust the segmentation threshold according to the curvature change of the local area to obtain the initial area marking result. After obtaining the initial area marking, it is necessary to analyze the curvature change between each area. By calculating the curvature change rate between adjacent areas, the change rate data describing the regional transition characteristics is obtained. The calculation of the curvature change rate takes into account the spatial position relationship and curvature gradient change between regions, and regional connectivity analysis is performed based on the change rate data. The connectivity analysis uses a regional growing algorithm, starting from the initial seed area, and gradually merges adjacent areas with similar curvature characteristics to form a complete connected domain feature.

[0053] The connected domain features are processed by boundary extraction. By analyzing the boundary features between connected domains, clear regional boundaries are extracted. The regional boundary data contains the spatial position information and geometric features of the partition boundary. The curvature mutation point is identified on the boundary line. The curvature mutation point refers to the location where the curvature value changes significantly. The calculation formula is:

[0054] ,

[0055] in, is the curvature mutation value, is the curvature value of the current point, is the curvature value of adjacent points, is the mutation threshold, and m is the detection window size. By identifying these mutation points, a feature point sequence is formed for subsequent region division.

[0056] The feature point sequence is subjected to regional clustering, and a density-based clustering method is used to aggregate points with similar spatial distribution characteristics into regions. In the clustering process, the spatial position and curvature characteristics of the points are considered at the same time to ensure that the clustering results reflect the processing difficulty characteristics of the surface. The processing path constraint analysis is performed on the processing difficulty partition data obtained by clustering, and detailed process constraints are generated by considering process factors such as tool accessibility and processing allowance. Based on the process constraints, adjacent partitions are reasonably merged. During the merging process, it is necessary to balance the region size and processing difficulty to avoid excessive segmentation or merging resulting in reduced processing efficiency. The benchmark partition sequence is obtained by regional merging as the basis for the final processing area planning. According to the benchmark partition sequence, a detailed processing area division plan is formulated by comprehensively considering the processing efficiency and quality requirements.

[0057] Taking the processing of a large mold surface as an example, the total area of ​​the surface is 2500×1500mm. Through the curvature density calculation, it is found that the curvature density value at the corner of the surface reaches 0.8~1.2 , while the curvature density of the flat area is only 0.1~0.3 The curvature density distribution is segmented by gradient threshold, using 0.4 As the segmentation threshold, the profile is initially divided into 25 regions. The curvature change rate between adjacent regions is calculated, and the average change rate is 0.15 , the maximum value reaches 0.45 at the corner Through connectivity analysis, similar regions were merged to obtain 12 main connected domains. A total of 156 curvature mutation points were identified on the region boundaries, and the curvature change value at the mutation point exceeded the set threshold of 0.3. Based on these feature points, regional clustering was performed, and considering the machining constraint of tool diameter 25 mm, the profile was finally divided into 8 machining areas, among which the high curvature area (curvature>0.6 ) The minimum processing area shall not be less than 100×100mm to ensure the continuity and efficiency of processing.

[0058] In a specific embodiment, the process of executing step S103 may specifically include the following steps:

[0059] (1) The curvature gradient is calculated for the plan of processing area division of flat surface products to obtain the regional curvature distribution map, and the equipotential line is extracted from the regional curvature distribution map to obtain the curvature contour line sequence;

[0060] (2) Perform path density planning on the curvature contour line sequence to obtain an initial processing path set, and perform compensation calculation on the initial processing path set to obtain path compensation parameters;

[0061] (3) Perform curvature error distribution analysis on the path compensation parameters to obtain the compensation adjustment amount, and perform path smoothing on the compensation adjustment amount to obtain smooth trajectory data;

[0062] (4) Perform tool posture calculation on the smooth trajectory data to obtain trajectory posture parameters, and perform interference check on the trajectory posture parameters to obtain an avoidance path point set;

[0063] (5) Perform velocity planning on the avoidance path point set to obtain motion control parameters, and then perform trajectory fitting on the motion control parameters to obtain an adaptive machining trajectory sequence.

[0064] Specifically, the curvature gradient is calculated for the processing area division scheme, and the curvature change trend inside each area is calculated by the finite difference method. The curvature gradient data reflects the change characteristics of the surface shape, and the regional curvature distribution map is generated based on these data. The regional curvature distribution map is processed by equipotential line extraction, and the contour line tracking algorithm is used to extract contour lines at different curvature value levels. By setting a suitable curvature interval, a series of curvature contour line sequences reflecting the surface characteristics are formed. After obtaining the curvature contour line sequence, path density planning is required. Path density planning determines the path spacing based on the degree of curvature change, increases the path density in areas with drastic curvature changes, and appropriately reduces the path density in areas with gentle curvature. The initial processing path set is obtained through path density planning, which includes the tool center point position information. The compensation amount is calculated for the initial processing path set, considering factors such as tool radius and contact point position, and the compensation value of each path point is calculated to generate path compensation parameters.

[0065] The curvature error distribution of the path compensation parameters is analyzed, and the distribution of the machining error after compensation is evaluated according to the local geometric characteristics of the surface. By analyzing the error distribution characteristics, the compensation adjustment amount is calculated to ensure that the machining accuracy meets the requirements. The compensation adjustment amount is smoothed by using the spline curve fitting method to eliminate the sudden changes in the path and obtain smooth trajectory data.

[0066] The tool posture calculation is performed on the smooth trajectory data to determine the spatial posture of the tool at each processing point. The tool posture calculation is based on the normal vector control method, and its calculation formula is:

[0067] × + ( ),

[0068] in, is the tool axial vector, is the surface normal vector, is the feed direction vector, is the velocity vector, is the tilt angle adjustment coefficient. After obtaining the trajectory posture parameters, interference check processing is performed. Through spatial geometric analysis, the interference between the tool and the workpiece and the fixture is detected to generate an avoidance path point set. Finally, the avoidance path point set is processed by speed planning. Based on the curvature characteristics and processing requirements, the feed speed at each path point is calculated. The feed speed is reduced in areas with large curvature changes to ensure processing quality. The motion control parameters are obtained through speed planning, and trajectory fitting processing is performed to generate an adaptive processing trajectory sequence that meets the dynamic characteristics requirements.

[0069] Taking the processing of a certain aircraft engine turbine blade as an example, the blade working surface area is 200×150mm. Through the curvature gradient calculation, it is found that the curvature gradient value of the leading edge area of ​​the blade reaches 0.6 , while the curvature gradient in the middle area is only 0.15 The equipotential line extraction was performed with a value of 0.1 The interval is 0.5 mm, and a total of 15 curvature contours are extracted. The path density planning sets the tool path spacing in the leading edge area to 0.3 mm and in the middle area to 0.8 mm. The initial planning obtains 2800 basic processing paths. The compensation calculation considers the radius of the ball-end cutter R=3 mm, and the maximum compensation value reaches 0.12 mm. Error distribution analysis shows that the theoretical processing error after compensation is controlled within the range of ±0.01 mm. The tool axial angle data is obtained by posture calculation. The tool inclination angle ranges from 15° to 25° in the leading edge area, and remains at 5° to 10° in the middle area. The interference check identifies 12 potential interference positions, which are successfully avoided by adjusting the avoidance distance of 0.5 mm. The speed planning sets the feed speed in the leading edge area to 800 mm / min and in the middle area to 2000 mm / min. The final generated processing trajectory sequence contains about 35,000 control points, which effectively ensures the processing accuracy and efficiency of the blade.

[0070] In a specific embodiment, the process of executing step S104 may specifically include the following steps:

[0071] (1) Perform surface data acquisition and processing on the detection point positions in the adaptive machining trajectory sequence to obtain measured point cloud data, and perform measurement noise filtering on the measured point cloud data to obtain effective measurement data;

[0072] (2) Perform local coordinate transformation on the effective measurement data to obtain standardized measurement data, and perform reference surface fitting on the standardized measurement data to obtain the theoretical reference surface;

[0073] (3) Calculate the deviation vector of the theoretical reference surface to obtain the morphology error vector, and perform spatial distribution analysis on the morphology error vector to obtain the error distribution characteristics;

[0074] (4) Perform principal curvature calculation on the error distribution characteristics to obtain local curvature feature quantities, and perform error accumulation analysis on the local curvature feature quantities to obtain the error transmission law;

[0075] (5) The error transmission law is processed by morphological feature extraction to obtain key error features, and the key error features are processed by regional statistics to obtain local morphological deviation data.

[0076] Specifically, the surface data of the detection points in the adaptive processing trajectory sequence is collected, and a high-precision laser scanner is used to obtain real-time measurement data during the processing process. The laser scanner uses the triangulation principle to emit lasers to each detection point and receive reflected signals to calculate the three-dimensional coordinate information of the point. The measured point cloud data obtained contains a large amount of random measurement noise, which needs to be filtered. The median filter algorithm is used to reduce the noise of the measurement data, effectively remove outliers while retaining edge features, and obtain accurate and effective measurement data. The effective measurement data is processed by local coordinate transformation, and the measurement data is uniformly converted to the standard coordinate system. The coordinate transformation process includes two steps: translation and rotation. First, the reference point is determined as the coordinate origin, and then the measurement data is aligned to the standard position through rigid body transformation. The reference surface is fitted for the standardized measurement data, and the least squares method is used to fit the ideal reference surface. The fitting process takes into account the continuity and smoothness of the surface to ensure that the fitting result can accurately reflect the design intent of the product.

[0077] The deviation vector of the theoretical reference plane is calculated, and the normal distance from each measurement point to the reference plane is calculated to obtain the vector field describing the topographic error. The deviation vector contains not only the size information of the error, but also the direction information of the error. The spatial distribution analysis of the topographic error vector is performed, and the spatial distribution law of the error is analyzed by statistical methods, including the mean, standard deviation and spatial correlation of the error, so as to obtain the complete error distribution characteristics. The principal curvature of the error distribution characteristics is calculated. The principal curvature reflects the curvature of the surface in each direction. The calculation formula is:

[0078] ,

[0079] in, are the principal curvature values, , , is the second-order partial derivative of the surface, , is the first-order partial derivative of the surface. The local curvature feature is obtained by calculating the principal curvature, and the error accumulation analysis is performed on the local curvature feature to study the transmission and accumulation law of errors in the processing process. The morphological features are extracted according to the error transmission law to identify the key error feature points and feature areas. The feature extraction process is based on the error amplitude and spatial distribution characteristics, focusing on the area where the error changes dramatically. The key error features are subjected to regional statistical processing, and the error statistics in each feature area are calculated to obtain the local morphological deviation data comprehensively.

[0080] Taking the processing of a precision optical reflector as an example, the reflector has an aperture of 500mm and the surface accuracy is required to be better than λ / 4 (λ=632.8nm). During the processing, a detection point is set every 0.5mm, and a total of 400×400 measurement points are collected for measured data. Through noise filtering, 95% of the measurement noise is removed, and the repeatability of the measurement data is better than 0.1μm. The coordinate transformation uses the three-point positioning method to convert the measured data to the standard coordinate system, and the transformation error is less than 0.05μm. The reference surface fitting uses the fifth-order surface equation, and the root mean square value of the fitting residual is 0.08μm. The deviation vector calculation shows that the maximum morphological error is 0.35μm, and the spatial distribution of the error shows obvious ring characteristics. Through the calculation of the principal curvature, it is found that the principal curvature value in the edge area of ​​the reflector reaches 0.002 , which is three times that of the central area. Error accumulation analysis shows that the processing errors are mainly concentrated in the area where the curvature changes sharply, and show an obvious cumulative effect. The local morphology deviation data finally obtained accurately reflects the processing quality status of the reflector.

[0081] In a specific embodiment, the process of executing step S105 may specifically include the following steps:

[0082] (1) Perform error distribution vectorization processing on the local morphology deviation data to obtain the morphology error field data, and perform feature decomposition processing on the morphology error field data to obtain the main error components;

[0083] (2) Perform local curvature sensitivity analysis on the main error components to obtain the curvature influence coefficient, and calculate the compensation amount of the curvature influence coefficient to obtain the initial compensation parameter;

[0084] (3) The initial compensation parameters are decomposed into progressive compensation amounts to obtain a step-by-step compensation sequence, and the step-by-step compensation sequence is processed for compensation path generation to obtain a compensation machining path;

[0085] (4) Perform interference check on the compensation processing path to obtain a feasible compensation path, and perform process parameter mapping on the feasible compensation path to obtain compensation process parameters;

[0086] (5) The compensation process parameters are analyzed and processed according to the forming process to obtain the compensation update amount, and the compensation update amount is comprehensively processed to obtain the dynamic correction parameters.

[0087] Specifically, the incremental forming compensation process of flat surface products first performs error distribution vectorization on the local morphology deviation data, and converts the deviation value of each measurement point into an error vector in three-dimensional space. A continuous morphology error field is constructed through a spatial interpolation algorithm, and the morphology error field data is characteristically decomposed using an orthogonal decomposition method. The characteristic decomposition process decomposes the error field into error components of different spatial frequencies, identifies the dominant error mode, and extracts the main error components. A local curvature sensitivity analysis is performed on the main error components to study the degree of influence of curvature changes on processing errors. By establishing an error-curvature response relationship, the curvature influence coefficient that describes the degree of curvature influence is obtained. The compensation amount is calculated for the curvature influence coefficient, and the compensation amount calculation formula is:

[0088] ,

[0089] in, is the compensation amount, is the weight coefficient of the i-th error component, is the corresponding curvature influence coefficient, is the attenuation factor, is the spatial distance, is the number of main error components. According to the compensation calculation results, the initial compensation parameters are obtained. The initial compensation parameters are decomposed by progressive compensation, and the total compensation is decomposed into multiple compensation steps according to the processing procedures. The decomposition process takes into account the uniformity of material removal and processing stability to avoid processing instability caused by excessive single compensation. The compensation path generation process is performed on the step-by-step compensation sequence, and the compensation processing path is designed based on the equivalent removal principle to ensure the uniformity and continuity of the compensation processing.

[0090] Interference check is performed on the compensation processing path to analyze the interference between the tool and the workpiece during the compensation processing. Interference check includes collision detection between the tool body, tool holder and fixture and the workpiece surface. By adjusting the tool posture and path to avoid the interference area, a feasible compensation path is obtained. Process parameter mapping is performed on the feasible compensation path to convert the path information into specific processing parameters, including process parameters such as spindle speed, feed speed and cutting depth. The forming process of the compensation process parameters is analyzed to study the material removal law and shape evolution process during the compensation processing. By analyzing the forming effect under different compensation parameters, the compensation update amount is obtained for dynamic adjustment of the compensation strategy. The compensation update amount is subjected to parameter comprehensive processing, and the final dynamic correction parameters are generated by comprehensively considering the processing accuracy requirements and process stability.

[0091] Taking the precision machining of a large reflector antenna as an example, the antenna diameter is 3 meters, and the surface accuracy is required to be better than 0.05mm. The initial measurement found that the maximum morphological error reached 0.15mm. Through error distribution vectorization processing, the measurement data was converted into an error vector field of 100×100 points. The characteristic decomposition extracted three main error components, accounting for 65%, 25% and 10% of the total error respectively. The curvature sensitivity analysis showed that when the curvature value was greater than 0.01 The error sensitivity coefficient to curvature change is 1.8 in the area. According to the compensation calculation formula, the processing area is divided into 5 compensation areas, and the maximum compensation is 0.12mm. The progressive compensation process is implemented in 3 steps, and the compensation amount of each step is controlled within 0.04mm. The compensation path planning adopts an equidistant spiral path with a path spacing of 1.5mm. The interference check was carried out under the condition of ball-end tool R=5mm, and 4 positions where the tool posture needed to be adjusted were found. The process parameter mapping sets the spindle speed to 4000rpm and the feed speed range is 800-2000mm / min. After 3 iterative compensation processes, the final morphology error was controlled within 0.04mm, meeting the processing accuracy requirements. The entire compensation process reflects the characteristics of progressive forming and effectively ensures the stability and accuracy of processing.

[0092] In a specific embodiment, the process of executing step S106 may specifically include the following steps:

[0093] (1) Perform time series feature extraction on the dynamic correction parameters to obtain the correction parameter change sequence, and perform parameter sensitivity analysis on the correction parameter change sequence to obtain the key quality influencing factors;

[0094] (2) Perform historical data correlation processing on key quality influencing factors to obtain the quality fluctuation law, and perform feature value extraction processing on the quality fluctuation law to obtain the quality control feature quantity;

[0095] (3) Performing processing on the quality control feature quantity during the processing process to obtain a process parameter association map, and classifying the process parameter association map into quality defects to obtain a defect feature library;

[0096] (4) Perform quality traceability analysis on the defect feature library to obtain the defect cause chain, and generate quality improvement plans for the defect cause chain to obtain quality control suggestions;

[0097] (5) Conduct feasibility verification on quality control suggestions to obtain a set of quality improvement measures, and evaluate and summarize the set of quality improvement measures to obtain a full-cycle quality evaluation report for flat and curved surface products.

[0098] Specifically, the quality feature correlation traceability of flat and curved products first extracts the time series features of the dynamic correction parameters and analyzes the trend of parameter changes over time. The wavelet decomposition method is used to perform multi-scale analysis on the correction parameter sequence to extract short-term fluctuations, medium-term trends and long-term change characteristics. Parameter sensitivity analysis is performed on the extracted correction parameter change sequence, and the key factors that have a significant impact on product quality are identified through correlation analysis and variance analysis. The sensitivity analysis focuses on the degree of influence of parameter changes on quality characteristics. By establishing an impact measurement model, the quality impact weight of each parameter is quantified, and finally the key quality influencing factors are determined. Historical data correlation analysis is performed on the determined key quality influencing factors, and the quality data of historical processing batches are collected and organized. The periodic changes and fluctuation characteristics of quality characteristics are studied through time series analysis methods. Mathematical modeling is performed on the quality fluctuation law, and quality fluctuation measurement indicators are established:

[0099] ,

[0100] Among them, Q is the quality fluctuation index, is the weight coefficient of the jth quality feature, is the actual mass value, is the target quality value, is the time decay factor, t is the time interval, and m is the number of quality features. Through feature value extraction processing, a series of control feature quantities reflecting the quality status are obtained. The quality control feature quantities are mapped during the processing process to establish the corresponding relationship between the feature quantities and the processing parameters. Through process capability analysis, the processing stability and quality consistency under different process parameter combinations are evaluated to form a complete process parameter association map. The process parameter association map is used to classify quality defects, classify them according to the nature, severity and frequency of defects, and establish a standardized defect feature library.

[0101] Conduct quality traceability analysis based on the defect feature library, and use the cause-effect chain analysis method to trace the formation mechanism and impact path of each type of defect. By establishing a defect cause map, clarify the key process links and influencing factors of defects. Conduct a systematic analysis of the defect cause chain, formulate targeted improvement plans, and form detailed quality control suggestions. Conduct feasibility verification of quality control suggestions, and evaluate the technical feasibility and economic rationality of the improvement plan. Verify the improvement effect through small batch tests, conduct statistical analysis of the verification results, and screen out effective improvement measures. Conduct a systematic evaluation of the quality improvement measure set to form a comprehensive quality evaluation report.

[0102] Taking the processing of a certain aircraft engine compressor blade as an example, the blade requires the profile error to be controlled within ±0.02mm. Through time series feature extraction, it is found that the dynamic correction parameters show obvious periodic fluctuations in the process of continuous processing of 50 samples, and the fluctuation period is about 10 pieces. Parameter sensitivity analysis shows that tool wear and thermal deformation are the main factors affecting quality, with sensitivities of 0.65 and 0.45 respectively. Historical data analysis shows that during the 8-hour continuous processing process, the blade profile error gradually increases with the processing time, and the quality fluctuation index Q increases from the initial 0.015mm to 0.035mm. Process parameter mapping found that when the spindle speed exceeds 6000rpm, the error caused by thermal deformation increases significantly. Quality defect analysis identified a total of 5 major defect modes, among which the progressive error caused by tool wear accounts for 45%, and the periodic error caused by thermal deformation accounts for 30%. Through traceability analysis, the optimization scheme of spindle cooling and tool replacement strategy was determined. Improvement measures include: replacing the tool every 8 pieces, controlling the spindle speed within 5500rpm, and increasing the cooling time between processing. After implementing the improvement measures, the blade qualification rate increased from 92% to 98%, and the profile error was stably controlled within the range of ±0.018mm. The final quality assessment report fully recorded the improvement process and effect verification data.

[0103] The above describes the production management method of the flat curved surface product in the embodiment of the present application. The following describes the production management device of the flat curved surface product in the embodiment of the present application. Figure 2 In the embodiment of the present application, an embodiment of the production management device for flat curved surface products includes:

[0104] Extraction module 201, used to perform multi-dimensional geometric feature extraction processing on the surface morphology of flat surface product materials to obtain a surface feature topology map;

[0105] A partitioning module 202 is used to perform non-uniform curvature partitioning processing on the surface feature topology map to obtain a plan for partitioning the processing area of ​​the flat surface product;

[0106] Planning module 203, used for performing variable curvature compensation path planning processing on the plan for dividing the processing area of ​​the flat surface product to obtain an adaptive processing trajectory sequence;

[0107] A detection module 204 is used to perform real-time surface error detection processing on the adaptive machining trajectory sequence to obtain local shape deviation data;

[0108] A compensation module 205, for performing surface incremental forming compensation processing on the local shape deviation data to obtain dynamic correction parameters;

[0109] The tracing module 206 is used to perform surface quality feature-related tracing processing on the dynamic correction parameters to obtain a full-cycle quality assessment report for flat and curved surface products.

[0110] Through the cooperation of the above components, the geometric feature information of the product surface is accurately obtained by extracting multi-dimensional geometric features from the surface morphology of the flat surface product material, providing an accurate data basis for subsequent processing, and the surface feature topology map is processed by non-uniform curvature partitioning to achieve reasonable partitioning of complex surfaces and ensure the pertinence of the processing strategy. By planning the variable curvature compensation path for the flat surface product processing area division scheme, an adaptive processing path based on curvature features is established to improve the processing accuracy and efficiency. Real-time surface error detection and processing of the adaptive processing trajectory sequence are carried out to achieve real-time monitoring and error identification during the processing process, providing timely feedback information for quality control. By performing surface incremental forming compensation processing on the local morphology deviation data, a dynamic compensation mechanism is established to effectively reduce the cumulative effect of processing errors. Finally, the dynamic correction parameters are traced by the surface quality feature association to form a full-cycle quality assessment report, and a complete quality traceability and improvement system is constructed. Through the organic combination of multiple links, this scheme realizes the precise control and quality assurance of the flat surface product processing process, and significantly improves the processing accuracy and quality stability of the product.

[0111] As described above, the above embodiments are only used to illustrate the technical solutions of the present application, rather than to limit them. Although the present application has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present application.

Claims

1. A production management method for flat curved surface products, characterized in that: The production management method of the flat curved surface product comprises: Perform multi-dimensional geometric feature extraction on the surface morphology of flat and curved product materials to obtain the surface feature topology map; The surface feature topology map is subjected to non-uniform curvature partitioning processing to obtain a plan for dividing the processing area of ​​the flat surface product, including: performing curvature density calculation processing on the surface feature topology map to obtain curvature density distribution data, and performing gradient threshold segmentation processing on the curvature density distribution data to obtain an initial area mark; performing curvature change rate calculation processing on the initial area mark to obtain change rate feature data, and performing regional connectivity analysis processing on the change rate feature data to obtain a connected domain feature; performing boundary extraction processing on the connected domain feature to obtain regional boundary line data, and performing curvature mutation point recognition processing on the regional boundary line data to obtain a feature point sequence; performing regional clustering processing on the feature point sequence to obtain processing difficulty partition data, and performing processing path constraint analysis processing on the processing difficulty partition data to obtain process constraint conditions; performing regional merging processing on the process constraint conditions to obtain a benchmark partition sequence, and performing processing area planning processing based on the benchmark partition sequence to obtain a plan for dividing the processing area of ​​the flat surface product; The planar surface product processing area division scheme is subjected to variable curvature compensation path planning processing to obtain an adaptive processing trajectory sequence, including: performing curvature gradient calculation processing on the planar surface product processing area division scheme to obtain a regional curvature distribution map, and performing equipotential line extraction processing on the regional curvature distribution map to obtain a curvature contour line sequence; performing path density planning processing on the curvature contour line sequence to obtain an initial processing path set, and performing compensation amount calculation processing on the initial processing path set to obtain path compensation parameters; performing curvature error distribution analysis processing on the path compensation parameters to obtain compensation adjustment amounts, and performing path smoothing processing on the compensation adjustment amounts to obtain smooth trajectory data; performing tool posture calculation processing on the smooth trajectory data to obtain trajectory posture parameters, and performing interference checking processing on the trajectory posture parameters to obtain an avoidance path point set; performing speed planning processing on the avoidance path point set to obtain motion control parameters, and performing trajectory fitting processing on the motion control parameters to obtain an adaptive processing trajectory sequence; The adaptive machining trajectory sequence is subjected to real-time surface error detection processing to obtain local morphology deviation data, including: performing surface data acquisition processing on the detection point positions in the adaptive machining trajectory sequence to obtain measured point cloud data, and performing measurement noise filtering processing on the measured point cloud data to obtain effective measurement data; performing local coordinate transformation processing on the effective measurement data to obtain standardized measurement data, and performing reference plane fitting processing on the standardized measurement data to obtain a theoretical reference plane; performing deviation vector calculation processing on the theoretical reference plane to obtain a morphology error vector, and performing spatial distribution analysis processing on the morphology error vector to obtain error distribution characteristics; performing principal curvature calculation processing on the error distribution characteristics to obtain local curvature feature quantities, and performing error accumulation analysis processing on the local curvature feature quantities to obtain error transmission laws; performing morphology feature extraction processing on the error transmission laws to obtain key error features, and performing regional statistical processing on the key error features to obtain local morphology deviation data; Performing surface incremental forming compensation processing on the local shape deviation data to obtain dynamic correction parameters; The dynamic correction parameters are traced back to the surface quality characteristics to obtain a full-cycle quality assessment report for flat and curved surface products.

2. The production management method of flat curved surface products according to claim 1, characterized in that: The multi-dimensional geometric feature extraction process is performed on the surface morphology of the flat surface product material to obtain the surface feature topology map, including: Performing grid point cloud acquisition processing on the surface of the flat curved product to obtain original surface point cloud data, and performing noise filtering processing on the original surface point cloud data to obtain filtered point cloud data; Performing local curvature calculation processing on the filtered point cloud data to obtain curvature distribution data, and resampling the curvature distribution data to obtain a uniform sampling point set; Performing boundary feature extraction processing on the uniform sampling point set to obtain boundary contour data, and performing partition processing on the surface area according to the boundary contour data to obtain an initial partition result; Performing curvature gradient analysis on the initial partition results to obtain curvature change feature data, and performing regional segmentation on the curvature change feature data using a watershed algorithm to obtain candidate partition schemes; According to the candidate partitioning scheme, feature point extraction is performed on the surface to obtain a key feature point set, and spatial topology reconstruction is performed on the key feature point set to obtain a surface feature topology map.

3. The production management method of flat curved surface products according to claim 1, characterized in that: The step of performing surface incremental forming compensation processing on the local shape deviation data to obtain dynamic correction parameters includes: Performing error distribution vectorization processing on the local shape deviation data to obtain shape error field data, and performing feature decomposition processing on the shape error field data to obtain main error components; Performing local curvature sensitivity analysis processing on the main error component to obtain a curvature influence coefficient, and performing compensation amount calculation processing on the curvature influence coefficient to obtain an initial compensation parameter; Performing progressive compensation amount decomposition processing on the initial compensation parameters to obtain a step-by-step compensation sequence, and performing compensation path generation processing on the step-by-step compensation sequence to obtain a compensation processing path; Performing interference checking processing on the compensation processing path to obtain a feasible compensation path, and performing process parameter mapping processing on the feasible compensation path to obtain compensation process parameters; The compensation process parameters are subjected to forming process analysis to obtain compensation update amounts, and the compensation update amounts are subjected to parameter comprehensive processing to obtain dynamic correction parameters.

4. The production management method of flat curved surface products according to claim 1, characterized in that: The dynamic correction parameters are subjected to surface quality feature correlation tracing processing to obtain a full-cycle quality assessment report for flat surface products, including: Performing time series feature extraction processing on the dynamic correction parameters to obtain a correction parameter change sequence, and performing parameter sensitivity analysis processing on the correction parameter change sequence to obtain key quality influencing factors; Performing historical data correlation processing on the key quality influencing factors to obtain quality fluctuation rules, and performing characteristic value extraction processing on the quality fluctuation rules to obtain quality control characteristic quantities; Performing a processing process mapping process on the quality control feature quantity to obtain a process parameter association map, and performing a quality defect classification process on the process parameter association map to obtain a defect feature library; Performing quality traceability analysis on the defect feature library to obtain a defect cause chain, and generating a quality improvement plan for the defect cause chain to obtain quality control suggestions; The quality control suggestions are subjected to feasibility verification to obtain a set of quality improvement measures, and the set of quality improvement measures is evaluated and summarized to obtain a full-cycle quality evaluation report for flat and curved surface products.

5. A production management device for flat curved surface products, used to implement the production management method for flat curved surface products as claimed in any one of claims 1 to 4, characterized in that: The production management device of the flat curved surface product comprises: The extraction module is used to extract multi-dimensional geometric features of the surface morphology of flat and curved product materials to obtain a surface feature topology map; A partitioning module is used to perform non-uniform curvature partitioning processing on the surface feature topology map to obtain a plan for dividing the processing area of ​​a flat surface product, including: performing curvature density calculation processing on the surface feature topology map to obtain curvature density distribution data, and performing gradient threshold segmentation processing on the curvature density distribution data to obtain an initial area mark; performing curvature change rate calculation processing on the initial area mark to obtain change rate feature data, and performing regional connectivity analysis processing on the change rate feature data to obtain a connected domain feature; performing boundary extraction processing on the connected domain feature to obtain regional boundary line data, and performing curvature mutation point identification processing on the regional boundary line data to obtain a feature point sequence; performing regional clustering processing on the feature point sequence to obtain processing difficulty partition data, and performing processing path constraint analysis processing on the processing difficulty partition data to obtain process constraint conditions; performing regional merging processing on the process constraint conditions to obtain a benchmark partition sequence, and performing processing area planning processing based on the benchmark partition sequence to obtain a plan for dividing the processing area of ​​a flat surface product; A planning module is used to perform variable curvature compensation path planning processing on the plan for dividing the processing area of ​​the flat surface product to obtain an adaptive processing trajectory sequence, including: performing curvature gradient calculation processing on the plan for dividing the processing area of ​​the flat surface product to obtain a regional curvature distribution map, and performing equipotential line extraction processing on the regional curvature distribution map to obtain a curvature contour line sequence; performing path density planning processing on the curvature contour line sequence to obtain an initial processing path set, and performing compensation amount calculation processing on the initial processing path set to obtain path compensation parameters; performing curvature error distribution analysis processing on the path compensation parameters to obtain compensation adjustment amounts, and performing path smoothing processing on the compensation adjustment amounts to obtain smooth trajectory data; performing tool posture calculation processing on the smooth trajectory data to obtain trajectory posture parameters, and performing interference checking processing on the trajectory posture parameters to obtain an avoidance path point set; performing speed planning processing on the avoidance path point set to obtain motion control parameters, and performing trajectory fitting processing on the motion control parameters to obtain an adaptive processing trajectory sequence; The detection module is used to perform real-time surface error detection processing on the adaptive processing trajectory sequence to obtain local morphology deviation data, including: performing surface data acquisition processing on the detection point positions in the adaptive processing trajectory sequence to obtain measured point cloud data, and performing measurement noise filtering processing on the measured point cloud data to obtain effective measurement data; performing local coordinate transformation processing on the effective measurement data to obtain standardized measurement data, and performing reference plane fitting processing on the standardized measurement data to obtain a theoretical reference plane; performing deviation vector calculation processing on the theoretical reference plane to obtain a morphology error vector, and performing spatial distribution analysis processing on the morphology error vector to obtain error distribution characteristics; performing principal curvature calculation processing on the error distribution characteristics to obtain local curvature feature quantities, and performing error accumulation analysis processing on the local curvature feature quantities to obtain error transmission laws; performing morphology feature extraction processing on the error transmission laws to obtain key error features, and performing regional statistical processing on the key error features to obtain local morphology deviation data; A compensation module, used for performing surface incremental forming compensation processing on the local shape deviation data to obtain dynamic correction parameters; The traceability module is used to perform surface quality feature-associated traceability processing on the dynamic correction parameters to obtain a full-cycle quality assessment report for flat and curved surface products.

Images