Integrated box bottom on-machine measurement point location and path automatic planning method

Through the automatic planning method, the problem of low programming efficiency and poor consistency of the bottom measurement path of the rocket storage tank is solved, and efficient measurement path generation and quality improvement is achieved, which is suitable for complex space box bottom structural parts.

CN119002401APending Publication Date: 2024-11-22SHANGHAI SPACE PRECISION MACHINERY RES INST
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202411032602.5
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-07-30
Publication Date
2024-11-22

AI Technical Summary

Technical Problem

The prior art has problems of inefficiency and poor consistency in the measurement points and path planning of the entire rocket storage box. Especially because a large number of auxiliary lines are required to be manually constructed and the interaction is complicated, resulting in low efficiency in measuring path programming and effect. Rely on the technical level of craftsmen.

Method used

The integrated box bottom automatic planning method is adopted to establish a demand model with dual definition of model geometry and process data, and the latitude sampling line is generated using the busbar and other arc length hierarchical method to judge the interference relationship with the virtual boundary. The interference sampling line is split to generate the final meridian sampling point, and the measurement path is generated in combination with the CAM software.

Benefits of technology

It improves the efficiency and quality of measurement program preparation, simplifies the boundary measurement constraints of the boss boundary, improves the efficiency of the algorithm, is suitable for two measurement modes, provides an intelligent measurement path planning method, and is suitable for complex space box bottom structural parts.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN119002401A_ABST
    Figure CN119002401A_ABST
Patent Text Reader

Abstract

The invention relates to an integrated box bottom on-machine measurement point location and path automatic planning method, which comprises the following steps of: firstly, determining on-machine measurement requirements of parts according to user input, and forming a requirement model; then, non-interference or incomplete-interference weft sampling lines are generated in the sampling space; secondly, radial sampling points are generated on the latitudinal sampling lines; and finally, supporting measurement path generation of a spot measurement mode and a line scanning mode. The method effectively solves the problems that the automation degree of measurement path planning containing the box bottom is low, and measurement adjustment of a boss area is tedious, the programming burden of a measurement program of a technician is remarkably relieved, the utilization rate of a three-dimensional model at the manufacturing end is improved, and the quality problem that the machining precision of a product is not consistent is solved.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The invention belongs to the technical field of integrated numerical control machining and measurement, and relates to an automatic planning method for measuring points and paths of an integral box bottom on a machine. Background Art

[0002] The manufacturing of high-precision complex structural parts for aerospace is the core part of aerospace industry manufacturing. Aerospace products have the typical production characteristics of multiple varieties and variable batches. With the development trend of large-scale, complex and thin-walled complex structural parts in aerospace, the process design and CNC machining and manufacturing requirements of high-precision complex structural parts have been raised to a new level.

[0003] Rocket tanks are the main load-bearing components of launch vehicles. They are large-sized thin-walled structural parts formed from high-specific-strength aluminum alloy sheets, accounting for 80% to 90% of the total weight of the rocket. The tank bottom is located at both ends of the rocket fuel tank and is a key component of the tank. The integral tank bottom is a typical large-scale thin-walled ellipsoidal structure with large geometric dimensions (diameter Φ3.35m), complex shape (local profile error after forming is up to 3mm), thin wall (blank wall thickness is about 12mm, remaining wall thickness is 1.5 to 4mm), high material removal ratio (>70%), long processing cycle, etc., poor processing technology, and difficulty in ensuring the consistency of wall thickness.

[0004] At present, the measurement-processing integration method is widely used for thin-walled and easily deformed parts, and has achieved good results. One of the most important links in the measurement-processing integration is the on-machine measurement of points and path planning. However, the integral tank bottom structure is large in size and contains irregular boss features. During the measurement programming process, a large number of auxiliary lines and auxiliary surfaces need to be manually constructed, and the software interaction is cumbersome, resulting in low efficiency in measurement path programming. In addition, the effect and feasibility of measurement planning largely depend on the technical level and experience level of the process personnel. The technical and experience levels of different personnel are uneven, resulting in poor rationality and consistency of measurement programming parameters and measurement sequence. Summary of the invention

[0005] The technical problem solved by the present invention is to overcome the deficiencies of the prior art and propose an on-machine measurement point position and path automatic planning method for an integral box bottom.

[0006] The solution of the present invention is:

[0007] A method for automatically planning on-machine measurement points and paths of an integral box bottom, comprising:

[0008] Determine the requirements for on-machine measurement of parts based on user input, and establish a demand model with dual definitions of model geometry and process data;

[0009] According to the target sampling density requirements input by the user, the initial latitudinal sampling lines are generated by the generatrix equal arc length layering method, the processing feature virtual boundary is generated by the measurement constraint and the feature boundary, the interference relationship between the sampling line and the virtual boundary is determined, and the completely interfering sampling lines are split to form non-interfering or incompletely interfering latitudinal sampling lines;

[0010] The initial radial sampling points are generated according to the latitudinal sampling lines, and the sampling point positions are updated by using the branching and resampling method for the latitudinal sampling lines with incomplete interference to generate the final meridian sampling points.

[0011] According to the set measurement mode, define the measurement programming element template, perform measurement process processing, and combine the path generation function of CAM software to generate the final measurement path.

[0012] Preferably, the demand model includes:

[0013] The target sampling density input by the user includes latitudinal sampling density and longitudinal sampling density, and the values ​​are defined by the user interface;

[0014] The sampling space is determined by the three-dimensional model of the box bottom, where the inner surface of the box bottom is used as the sampling reference surface and the outer surface of the box bottom is used as the sampling target surface;

[0015] The size of the measuring system probe and the boundary of the machined feature of the outer surface of the box bottom are used as measurement constraints. During measurement, it is ensured that the probe does not collide with the boundary of the machined feature or that the probe is suspended on the boundary of the feature, so as to improve the integrity and credibility of the measurement data.

[0016] Preferably, the virtual boundary is generated by offsetting the boundary of the characteristic contour inwardly and outwardly on the curved surface, respectively, and the offset value is the sum of the radius of the cylindrical envelope of the probe and the collision accuracy value.

[0017] Preferably, the busbar equal arc length layering method is as follows:

[0018] Get the inner surface generatrix of the part, take the end point of the generatrix as the starting point of segmentation, and the arc length of the generatrix is ​​l p , the arc length interval is d p = l p / ceil(l p / p), where ceil() is a rounding function, with arc length spacing d p The length basis is used to divide the arc length of the generatrix and generate the axis division points, where p is the latitudinal density;

[0019] The height of the split point along the axis is taken as the layer spacing, and a layer plane is generated on the axis. The layer plane is perpendicular to the axis and passes through the split point.

[0020] The layered plane intersects with the inner surface of the part to generate an intersection line, which is the initial latitudinal sampling line.

[0021] Preferably, the interference relationship between the sampling lines and the virtual boundaries is determined, and the completely interfering sampling lines are split to form non-interfering or incompletely interfering latitudinal sampling lines. The specific implementation method is as follows:

[0022] Determine the relationship between the sampling line and the virtual boundary, wherein the relationship includes separation, intersection, and tangency;

[0023] Determine whether the virtual boundary intercepts the sampling line;

[0024] According to the relationship between the sampling line and the virtual boundary and whether the virtual boundary intercepts the sampling line, the geometric relationship between the sampling line and the virtual boundary is marked as complete interference, incomplete interference, and non-interference. The non-interference means that the virtual boundary and the sampling line are separated or the sampling line is only tangent to the outside of the virtual boundary, and the virtual boundary does not intercept the sampling line; incomplete interference means that the virtual boundary partially intercepts the sampling line; complete interference means that the sampling line is completely between the virtual boundaries;

[0025] For completely interfering sampling lines, the sampling lines are split into two parts, and the split sampling lines are in a non-interfering or incompletely interfering state.

[0026] Preferably, the method for generating the final meridional sampling points is as follows:

[0027] c1) Divide the latitudinal sampling line into equally spaced segments according to the meridional sampling density input by the user. The segment spacing should be smaller than the meridional sampling density requirement, and generate initial meridional sampling points;

[0028] c2) The non-interfering latitudinal sampling lines directly use the initial meridional sampling points;

[0029] c3) branching the latitudinal sampling lines of incomplete interference;

[0030] c4) Pair and connect sampling line branches in series, and resample the branched sampling lines. The resampling is done by taking two end points and dividing them into segments with equal spacing.

[0031] Preferably, in step c3), the branching method is the adjacent connection method, with the goal of ensuring the radial point density, and the jump segments of the partially interfering latitudinal sampling lines passing through the virtual boundary gap are replaced by adjacent segments and connecting segments; the latitudinal sampling lines tangent to the inner virtual boundary are formed into two adjacent connected sampling paths.

[0032] Preferably, in step c4), the method of pairing and connecting the sampling line branches in series is as follows:

[0033] For multiple sampling lines generated by interference, determine whether the distance between adjacent sampling lines is less than the pairing distance. If so, use the sampling line series connection method to merge excessive sampling branches; otherwise, do not process it.

[0034] Preferably, the method for generating the final measurement path is as follows:

[0035] d1) according to the set measurement mode, the latitudinal sampling line is used as the sampling route and the longitudinal sampling point is used as the sampling point; the measurement mode includes point measurement or line scanning;

[0036] d2) Use BNF to define measurement programming element templates;

[0037] d3) For the point measurement mode, the height of the measuring point is adaptively adjusted, and the measuring points are sorted according to the C-axis travel constraints of the machine tool to form an "S"-shaped measurement sequence;

[0038] For line scanning measurement mode, the sampling line height is adaptively adjusted, and the measurement line is sorted and set in direction according to the C-axis travel constraints of the machine tool to form an "S"-shaped measurement sequence;

[0039] d4) Based on the tool path generation function of general CAM software, the point measurement mode uses the point processing function to generate the final measurement path, and the line scanning measurement mode uses the curve processing function to generate the final measurement path to complete the path planning.

[0040] An integrated box bottom on-machine measurement point position and path automatic planning system, including a user demand model definition module, a latitudinal sampling line generation module, a longitudinal sampling point generation module, and a dual-mode measurement path planning module;

[0041] User demand model definition module: Determine the requirements for on-machine measurement of parts based on user input and establish a demand model with dual definitions of model geometry and process data;

[0042] Latitude sampling line generation module: according to the target sampling density requirements input by the user, the initial latitude sampling lines are generated by the generatrix equal arc length layering method, the processing feature virtual boundary is generated by the measurement constraint and the feature boundary, the interference relationship between the sampling line and the virtual boundary is determined, and the completely interfering sampling lines are split to form non-interfering or incompletely interfering latitude sampling lines;

[0043] Meridian sampling point generation module: Generates initial radial sampling points based on latitudinal sampling lines, updates sampling point positions by using branching and resampling methods for incompletely interfering latitudinal sampling lines, and generates final meridian sampling points;

[0044] Dual-mode measurement path planning module: According to the set measurement mode, define the measurement programming element template, perform measurement process processing, and combine the CAM path generation function to generate the final measurement path.

[0045] The beneficial effects of the present invention compared with the prior art are:

[0046] (1) The present invention proposes a method for automatically planning the measurement points and paths of an integral box bottom on a machine, which replaces the traditional mode of programming the measurement path by manual interaction, adopts a geometric algorithm to solve the problem of generating a path to avoid boss features, and improves the efficiency and quality of measuring program compilation.

[0047] (2) The present invention proposes the concept of virtual boundary, which is generated by the boss boundary measurement constraint, and simplifies the problem of suspension and collision between the probe and the boss on the outer surface of the box bottom into a "point-line" distance judgment, thereby improving the efficiency of the algorithm.

[0048] (3) The present invention is applicable to the compilation of measurement programs in two modes, adopts a unified measurement path planning engine, and uses programming templates to solidify and apply process knowledge that affects measurement path planning design, automatic decision-making, etc. in the process using digital methods, providing a feasible method and practical path for the intelligent measurement of bottom parts of the box.

[0049] (4) In the process of three-dimensional model-driven measurement process design, the present invention proposes a method for automatically generating digital measurement paths driven by model geometry and algorithms. For complex aerospace box bottom structural parts with irregular shape features and variable wall thickness surfaces, a model-driven measurement path automatic planning method is proposed. This method is easy to develop and apply in general CAM software, solving the application problem of process design with integrated measurement and processing. BRIEF DESCRIPTION OF THE DRAWINGS

[0050] Figure 1 Schematic diagram of the process of measuring points and automatic path planning on the machine for the integral box bottom;

[0051] Figure 2 Schematic diagram of the box bottom parts;

[0052] Figure 3 Schematic diagram of the user input interface;

[0053] Figure 4 Schematic diagram of measurement constraints, where (a) is a schematic diagram of measurement constraints, (b) is a schematic diagram of case 1, and (c) is a schematic diagram of case 2;

[0054] Figure 5 Schematic diagram of busbar equal arc length layering;

[0055] Figure 6 Schematic diagram of virtual boundary generation;

[0056] Figure 7 Schematic diagram of geometric relationships, where (a) and (h) are separation relationships, (b), (d), and (f) are tangent relationships, and (c), (e), and (g) are intersection relationships;

[0057] Figure 8Schematic diagram of the splitting method, where (a) and (c) are schematic diagrams of the sampling line being tangent to the outer and inner virtual boundaries respectively, (b) is the sampling line to be split, and (d) is the superposition of the two tangent sampling lines (a) and (c) formed by splitting;

[0058] Fig. 9 Schematic diagram of the proximity jump method, where (a) is a schematic diagram of case 1, and (b) is a schematic diagram of case 2;

[0059] Fig.10 Schematic diagram of paired and serial sampling line branches;

[0060] Fig.11 Schematic diagram of sequential optimization, where (a) is point-testing sorting method 1, (b) is point-testing sorting method 2, (c) is line-scanning sorting method 1, and (d) is line-scanning sorting method 2;

[0061] Fig.12 Schematic diagram of sampling lines and sampling points, where (a) is the sampling line and (b) is the sampling point;

[0062] Fig.13 Schematic diagram of the measurement path, where (a) is the point measurement path, (b) is the line scanning measurement path, (c) is the measurement height adjustment, and (d) is the boss feature avoidance path. DETAILED DESCRIPTION

[0063] The present invention will be further described below in conjunction with the embodiments.

[0064] The present invention proposes a method for automatically planning the measurement points and paths of an integral box bottom on a machine, and the implementation steps are as follows:

[0065] a) User requirements model definition;

[0066] The requirements for on-machine measurement of parts are determined based on user input, forming a requirement model with dual definitions of model geometry and process data. The requirement model includes three parts: target sampling density input by the user, sampling space formed by the inner and outer surfaces of the box bottom, and measurement constraints formed by the probe size and processing feature boundaries.

[0067] The user demand model of box bottom measurement-processing integration is expressed as

[0068] Dem Etm =(Density(m,p),Geo,Check(r,b))

[0069] Among them, Dem EtmIt represents the user demand information for the bottom processing of the box. Density (m, p) represents the sampling density of the bottom measurement. m is the density in the longitudinal direction, and p is the density in the latitudinal direction. The value is greater than 0 and the unit is mm. Geo is the sampling space. The sampling space mainly refers to the surface geometry, axis, surface adjacency relationship and other information of the inner and outer surfaces of the bottom of the box, among which the inner surface is the reference surface. Check represents the measurement constraints, including the probe size and the processing feature boundary. The probe size is the radius r of the probe cylindrical envelope. The processing feature boundary refers to the boundary of the processing feature on the outer surface of the bottom of the box, which is denoted as b. Since the outer surface of the bottom of the box includes thinning area and thickening area, and the thickening area is higher than the thinning area, it is regarded as a boss feature. The distribution and shape of the boss feature are different. The processing feature here mainly refers to the boss feature in the thickening area, hereinafter referred to as the boss.

[0070] According to the user demand model of box bottom measurement-processing integration, Figure 3 In the user interface shown, the warp density m and the weft density p are set, and the values ​​are defined by user interaction as input. At the same time, the inner and outer surfaces of the three-dimensional model of the bottom of the box are selected as the geometric input of the system.

[0071] The sampling space is determined by the three-dimensional model of the box bottom, in which the inner surface of the box bottom is used as the sampling reference surface, and the outer surface of the box bottom is used as the sampling target surface. The inner surface of the box bottom is a standard ellipsoidal surface. The surface can be verified according to the mathematical expression of the ellipsoidal surface to check the geometric quality of the reference surface. The box bottom parts processing technology targeted by the present invention is "first inner shape and then outer shape", with the inner surface as the reference, which is convenient for part processing alignment and subsequent measurement path generation. With the inner surface as the reference, the outer boss contour is processed according to the box bottom model requirements and the wall thickness requirements are guaranteed. The outer surface of the box bottom is the sampling target surface, and the measurement space is open, which is convenient for subsequent measurement.

[0072] The size of the measuring system probe and the boundary of the machined feature of the outer surface of the box bottom are used as measurement constraints. The relationship between the radius r of the probe cylindrical envelope and the boundary b of the machined feature is as follows: Figure 4 shown. Figure 4 (a) is a schematic diagram of the measurement constraint. Due to the characteristics of the wall thickness measurement probe, the probe is suspended on the feature boundary, and the thickness measurement data is inaccurate. Figure 4 (b) shows situation 1: When measuring, the probe collides with the boundary of the machining feature, which will damage the probe and affect the measurement operation. Figure 4 (c) shows situation 2. Therefore, avoiding the two situations of hanging in the air and collision can improve the integrity and credibility of the measurement data.

[0073] b) Generation of latitudinal sampling lines.

[0074] According to the user demand model, the initial latitudinal sampling line is generated by the busbar equal arc length stratification method; the busbar equal arc length stratification method first obtains the busbar of the inner surface of the part, takes the endpoint of the busbar as the starting segmentation point, and the busbar arc length is l p , the arc length interval is d p = l p / ceil(l p / p), where ceil() is a rounding function, with arc length spacing d p The length basis is used to divide the arc length of the busbar and generate the axis division point. Then, the height of the division point along the axis is used as the layer spacing, and the layer plane is generated on the axis. The layer plane is perpendicular to the axis and passes through the division point, such as Figure 5 As shown; finally, the layered plane intersects with the inner surface of the part to generate an intersection line, which is the initial latitudinal sampling line.

[0075] The virtual boundary is generated by the measurement constraint and the feature boundary. The virtual boundary is generated by offsetting the boundary of the feature contour inward and outward on the surface to be measured. The virtual boundary line is two curves, and the offset value is the sum of the radius r of the probe cylindrical envelope and the collision accuracy value δ, where the collision accuracy value δ>0, such as Figure 6 shown.

[0076] Determine the geometric relationship between the virtual boundary and the latitudinal sampling line. The geometric relationship is divided into three types: separation, intersection, and tangency. For example, Figure 7 (a) and (h) are in a disjoint relationship. Figure 7 (b), (d), and (f) are tangent relationships. Figure 7 Among them, (c), (e), and (g) are intersection relationships. Figure 7 , Figure 8 The solid line in the middle is the latitudinal sampling line, and the dotted line is the virtual boundary of the feature contour, which is simplified and replaced by a circle. The virtual boundaries appear in pairs, called the inner and outer virtual boundaries.

[0077] According to the interception of the sampling line between two virtual boundaries, the geometric relationship is marked as complete interference, incomplete interference, and no interference. Figure 7 (a), (b), and (h) are non-interference, the virtual boundary and the sampling line are separated or the sampling line is only tangent to the outside of the virtual boundary, and the virtual boundary does not intercept the sampling line; Figure 7 (c), (d), and (e) are partial interferences, where the virtual boundary partially intercepts the sampling line. Figure 7 (f) and (g) are complete interference, and the sampling lines are completely between the virtual boundaries.

[0078] For completely interfering sampling lines, the sampling lines are split into two parts. The split sampling lines are in a non-interfering or incompletely interfering state and are used as the final generated latitudinal sampling lines. Figure 8(a) and (c) are two special tangent relationships, where the sampling line is tangent to the outer and inner virtual boundaries respectively. Figure 8 (b) is the sampling line to be split. The splitting method is to split the sampling line enclosed by the virtual boundary into two sampling lines tangent to the inner and outer virtual boundaries, that is, split to form Figure 8 The superposition of the two tangent relationship sampling lines (a) and (c) is shown in Figure 1. Figure 8 As shown in (d).

[0079] c) Generation of meridian sampling points.

[0080] The initial radial sampling points are generated according to the latitudinal sampling lines. The sampling points of the non-interfering latitudinal sampling lines are directly inherited. The branching and resampling methods are used to update the sampling point positions of the incompletely interfering latitudinal sampling lines, so that the final sampling points are relatively evenly distributed and meet the sampling density requirements.

[0081] The specific method is as follows:

[0082] The latitudinal sampling line is divided into equally spaced segments according to the longitudinal sampling density. Since the latitudinal sampling line is closed, the geometric coordinates of the starting point and the end point of the sampling line are the same. The arc length of the latitudinal sampling line is taken as l m , the spacing between equally spaced segments is d m = l m / ceil(l m / m), where ceil() is the upward rounding function, ensuring that d m <m, meeting the sampling density requirements.

[0083] For non-interfering latitudinal sampling lines, the initial sampling points are directly used;

[0084] For the partially interfering latitudinal sampling lines, the branching method is the adjacent connection method, which replaces the jump segments of the partially interfering latitudinal sampling lines passing through the virtual boundary gap with adjacent segments and connection segments, such as Fig. 9 (a) is the enlarged view of the branch of case 1. Different from Fig. 9 Case 1 in (a) is a partially interfering latitudinal sampling line that passes through the inner virtual boundary and forms two adjacent connected sampling paths. Fig. 9 In case 2 (b), sampling line 1 is tangent to the inner virtual boundary, which is a partial interference latitudinal sampling line, forming two adjacent connected sampling paths, while sampling line 2 is tangent to the outer virtual boundary, which is a partial interference and does not need to be processed. Fig. 9 In the method, the maximum spacing of the measurement path formed by the proximity connection method is the spacing between the outer and inner virtual boundaries, that is, 2(r+δ). The longitudinal density m set during path planning requires m>2(r+δ).

[0085] Pairing and series connection of sampling line branches: for multiple sampling lines generated by interference, determine whether the distance between adjacent sampling lines is less than the pairing distance. If so, the sampling line series connection method is used to merge excessive sampling branches; otherwise, no processing is performed. Fig.10 Then, the branch sampling line is resampled by first taking the two end points and then equally spaced segmentation, where the maximum spacing of the equally spaced segments is d m , and the sampling points generated thereby are used as the final sampling points.

[0086] d) Dual-mode measurement path planning.

[0087] According to the set measurement mode point measurement or line scanning, the sampling line and sampling point in the above steps are used as the geometric reference for the measurement path planning of point measurement and line scanning respectively. Point measurement uses the sampling point as the measurement point, and measures at a constant height between the part to be measured; line scanning uses the sampling line as the measurement path, and measures at a constant speed and constant height between the part to be measured.

[0088] Measurement path planning is the process of converting measurement mode, measurement requirements, and measurement process into CAM measurement tool path. According to the characteristics, process, and manufacturing activity measurement programming elements of the bottom surface of the box, a digital description template describing the measurement process is defined, which helps to decompose the implicit knowledge of process planning into explicit knowledge and assist in completing measurement programming. The point measurement and line scanning programming templates defined by BNF are as follows:

[0089] <Box bottom measurement path planning>::=(<Measurement requirements>, <Measurement mode>, <Measurement process>)

[0090] <Measurement requirements>::=(<Material>, <Surface roughness>, <Probe assembly>, <Auxiliary requirements>);

[0091] <Material>::=(<Aluminum Alloy>, <Stainless Steel>, <Composite Material>, ...);

[0092] <Surface roughness>::=(<rough surface>, <semi-glossy surface>, <glossy surface>);

[0093] <probe equipment>::=(<probe size>, <probe accuracy>, <probe type>, <lift-off distance>, <machine tool handle>);

[0094] <auxiliary requirements>::=(<spacing requirements>, <coupling agent>, <stabilization time>, ...);

[0095] <Measurement mode>::=(<Point measurement>, <Line scanning>);

[0096] <Point measurement type>::=(<Measurement point height>, <Point sorting>, <Point entry and exit>, ...);

[0097] <Line scan mode>::=(<measurement line height>, <line order>, <line scan speed>, ...);

[0098] <Measurement process>::=(<Processing feature>, <Measurement parameter>, <Path connection>, <Tool advance and retract>);

[0099] <Processing feature>::=(<Thinning area surface>, <Boss>, <Chamfer>);

[0100] <Measurement parameters>::=(<feed speed>, <measurement interval>, <holding time>, ...);

[0101] <path connection>::=(<straight line>, <polyline>, <arc>, <curve>);

[0102] <feed and retract>::=(<feed and retract length>, <feed and retract speed>, <feed and retract path>, <safety plane>);

[0103] The measurement height is adjusted in the point measurement mode. For contact measurement, the measurement point can be used directly; for non-contact measurement, the point is offset outward along the normal vector of the outer surface by a certain measurement spacing to obtain the measurement point, where the measurement spacing must be less than the maximum spacing requirement allowed by the probe.

[0104] The process of processing adaptive sorting in the point measurement mode is as follows: first, the generated sampling points are sorted in rows and columns, and the starting point and end point of each row of sampling points are on the same busbar; then, the measurement points are sorted according to the C-axis travel constraint of the machine tool. Since the machine tool operating system has different processing methods for 360° rotating bodies and the C-axis exceeds the maximum travel of 360°, considering compatibility and error prevention, the measurement points are measured in a reciprocating manner of [0°, 360°) to form an "S"-shaped measurement sequence, such as Fig.11 As shown in (a) and (b), P s is the starting point of measurement, P e D is the measurement end point. p is the latitudinal measurement direction, D m is the radial measurement direction; finally, the final point trajectory is generated.

[0105] The measurement height is adjusted in the line scanning measurement mode. For non-contact measurement, the sampling line is offset outward along the normal vector of the outer surface by a certain measurement spacing to obtain the measurement line, where the measurement spacing must be less than the maximum spacing requirement allowed by the probe.

[0106] The process of line scanning measurement mode processing adaptive sorting is as follows: first, connect the generated sampling lines, and unify the starting point of each line to the same busbar; then, set the measurement direction, and the tool direction includes the latitudinal measurement direction D p , Meridional measurement direction Dm , according to the C-axis travel constraint of the machine tool, the measurement lines are sorted to prevent the C-axis of the machine tool from winding around the axis. The measurement lines are measured line by line in a reciprocating manner to form an "S" shaped measurement sequence, such as Fig.11 As shown in (c) and (d), P s is the starting point of the measuring line, P e is the end point of the measuring line, D p is the latitudinal measurement direction, D m The direction is measured for each row of measured lines; finally, the final trajectory curve is generated.

[0107] Based on the path generation function of general CAM software, the point measurement mode uses point processing to generate the measurement path, and the line scanning measurement mode uses curve processing to generate the path, completing the planning and implementation of the route measurement path.

[0108] The present invention also provides an integrated box bottom on-machine measurement point and path automatic planning system based on the above method, including a user demand model definition module, a latitudinal sampling line generation module, a longitudinal sampling point generation module, and a dual-mode measurement path planning module.

[0109] User requirement model definition module: Determines the requirements for on-machine measurement of parts based on user input and establishes a requirement model with dual definitions of model geometry and process data.

[0110] Latitude sampling line generation module: According to the target sampling density requirements input by the user, the initial latitude sampling lines are generated by the generatrix equal arc length layering method, the processing feature virtual boundary is generated by the measurement constraint and the feature boundary, the interference relationship between the sampling line and the virtual boundary is judged, and the completely interfering sampling lines are split to form non-interfering or incompletely interfering latitude sampling lines.

[0111] Meridian sampling point generation module: Generate initial radial sampling points based on the latitudinal sampling lines, use branching and resampling methods to update the sampling point positions for the incompletely interfering latitudinal sampling lines, and generate the final meridian sampling points.

[0112] Dual-mode measurement path planning module: According to the set measurement mode, define the measurement programming element template, perform measurement process processing, and combine the CAM path generation function to generate the final measurement path.

[0113] Example:

[0114] like Figure 2 The test parts of the embodiment are shown. On the box bottom part, the maximum diameter is 3350mm, there are multiple boss structures on the outer surface, the allowable deviation of the wall thickness is 0-0.3mm, the warp density m is 100mm, the weft density p is 100mm, the inner surface is selected as the reference surface, the radius r of the probe cylindrical envelope is 15mm, and the collision accuracy value δ is 1mm.

[0115] There are 19 sampling lines generated by the busbar equal arc length layering method. The generated sampling lines are as follows: Fig.12 The measurement path planning engine is run, and a geometric algorithm is used to solve the automatic generation of feature avoidance points containing boss features. 1280 sampling points are generated, as shown in Fig.12 As shown in (b).

[0116] The bottom parts of the box are programmed using general CAM software. The point measurement mode uses the point processing function to generate the measurement path, and the line scanning measurement mode uses the curve processing function to generate the path to complete the planning and implementation of the route measurement path. Fig.13 shown.

[0117] The present invention, firstly, determines the demand for on-machine measurement of parts according to user input, forms a demand model with dual definition of model geometry and process data, including three parts: target sampling density input by the user, sampling space formed by inner and outer surfaces of the bottom of the box, and measurement constraints formed by probe size and processing feature boundary; then, generates latitudinal sampling lines in the sampling space, generates initial latitudinal sampling lines according to sampling density requirements by adopting the generatrix equal arc length layering method, generates processing feature virtual boundary by measurement constraint, judges the interference relationship between sampling lines and virtual boundary, splits completely interfering sampling lines, and forms non-interfering or incompletely interfering latitudinal sampling lines; secondly, generates radial sampling points on the latitudinal sampling lines, and updates the sampling positions of latitudinal sampling lines with interference points by adopting branching and resampling methods, so that all sampling points avoid interference positions; finally, measurement path planning and generation, supports measurement path generation of two modes, point measurement mode and line scanning mode, and in the generation process of the two modes, defines measurement programming element templates respectively, performs special measurement process processing such as measurement height adjustment, boss feature avoidance, adaptive sorting, and generates the final measurement path.

[0118] Aiming at the problems of easy deformation of profile processing and excessive wall thickness in the machining process of large-scale integral box bottom parts of aerospace, the method of the present invention aims at the current situation that the number of iterations of "machining-measurement-fitting-remachining" is large and the measurement programming efficiency does not meet the production cycle requirements in the implementation of measurement-machining integration technology. A method and system for automatic planning of measurement points and paths for the integral box bottom on the machine are proposed, which effectively solves the problems of low automation level of measurement path planning for the box bottom and cumbersome measurement and adjustment of the boss area, significantly reduces the programming burden of measurement procedures for process personnel, improves the utilization rate of three-dimensional models at the manufacturing end, and avoids quality problems of inconsistent product machining accuracy.

[0119] The contents not described in detail in the specification of the present invention belong to the common knowledge of those skilled in the art.

Claims

1. A method for automatically planning the measurement points and paths of an integral box bottom on a machine, characterized in that: include: Determine the requirements for on-machine measurement of parts based on user input, and establish a demand model with dual definitions of model geometry and process data; According to the target sampling density requirements input by the user, the initial latitudinal sampling lines are generated by the generatrix equal arc length layering method, the processing feature virtual boundary is generated by the measurement constraint and the feature boundary, the interference relationship between the sampling line and the virtual boundary is determined, and the completely interfering sampling lines are split to form non-interfering or incompletely interfering latitudinal sampling lines; The initial radial sampling points are generated according to the latitudinal sampling lines, and the sampling point positions are updated by using the branching and resampling method for the latitudinal sampling lines with incomplete interference to generate the final meridian sampling points. According to the set measurement mode, define the measurement programming element template, perform measurement process processing, and combine the path generation function of CAM software to generate the final measurement path.

2. According to claim 1, a method for automatically planning the measurement points and paths of an integral box bottom on a machine, characterized in that: The demand model includes: The target sampling density input by the user includes latitudinal sampling density and longitudinal sampling density, and the values ​​are defined by the user interface; The sampling space is determined by the three-dimensional model of the box bottom, where the inner surface of the box bottom is used as the sampling reference surface and the outer surface of the box bottom is used as the sampling target surface; The size of the measuring system probe and the boundary of the machined feature of the outer surface of the box bottom are used as measurement constraints. During measurement, it is ensured that the probe does not collide with the boundary of the machined feature or that the probe is suspended on the boundary of the feature, so as to improve the integrity and credibility of the measurement data.

3. The method for automatically planning the measurement points and paths of an integral box bottom on a machine according to claim 1, characterized in that: The virtual boundary is generated by offsetting the boundary of the feature contour inwardly and outwardly on the curved surface, respectively, and the offset value is the sum of the radius of the cylindrical envelope of the probe and the collision accuracy value.

4. The method for automatically planning the measurement points and paths of an integral box bottom on a machine according to claim 1, characterized in that: The busbar equal arc length layering method is as follows: Get the inner surface generatrix of the part, take the end point of the generatrix as the starting point of segmentation, and the arc length of the generatrix is ​​l p , the arc length interval is d p = l p / ceil(l p / p), where ceil() is a rounding function, with arc length spacing d p The length basis is used to divide the arc length of the generatrix and generate the axis division points, where p is the latitudinal density; The height of the split point along the axis is taken as the layer spacing, and a layer plane is generated on the axis. The layer plane is perpendicular to the axis and passes through the split point. The layered plane intersects with the inner surface of the part to generate an intersection line, which is the initial latitudinal sampling line.

5. The method for automatically planning the measurement points and paths of an integral box bottom on a machine according to claim 1, characterized in that: Determine the interference relationship between the sampling line and the virtual boundary, split the completely interfering sampling lines, and form non-interfering or incompletely interfering latitudinal sampling lines. The specific implementation method is as follows: Determine the relationship between the sampling line and the virtual boundary, wherein the relationship includes separation, intersection, and tangency; Determine whether the virtual boundary intercepts the sampling line; According to the relationship between the sampling line and the virtual boundary and whether the virtual boundary intercepts the sampling line, the geometric relationship between the sampling line and the virtual boundary is marked as complete interference, incomplete interference, and non-interference. The non-interference means that the virtual boundary and the sampling line are separated or the sampling line is only tangent to the outside of the virtual boundary, and the virtual boundary does not intercept the sampling line; incomplete interference means that the virtual boundary partially intercepts the sampling line; complete interference means that the sampling line is completely between the virtual boundaries; For completely interfering sampling lines, the sampling lines are split into two parts, and the split sampling lines are in a non-interfering or incompletely interfering state.

6. The method for automatically planning the position and path of the integral box bottom on-machine according to claim 1, characterized in that: The method to generate the final meridional sampling points is as follows: c1) Divide the latitudinal sampling line into equally spaced segments according to the meridional sampling density input by the user. The segment spacing should be smaller than the meridional sampling density requirement, and generate initial meridional sampling points; c2) The non-interfering latitudinal sampling lines directly use the initial meridional sampling points; c3) branching the latitudinal sampling lines of incomplete interference; c4) Pair and connect sampling line branches in series, and resample the branched sampling lines. The resampling is done by taking two end points and dividing them into segments with equal spacing.

7. The method for automatically planning the measurement points and paths of an integral box bottom on a machine according to claim 6, characterized in that: In step c3), the branching method is the adjacent connection method, with the goal of ensuring the radial point density. The jump segments of the partially interfering latitudinal sampling lines passing through the virtual boundary gap are replaced by adjacent segments and connecting segments; the latitudinal sampling lines tangent to the inner virtual boundary are formed into two adjacent connected sampling paths.

8. The method for automatically planning the position and path of the integral box bottom on-machine according to claim 6, characterized in that: In step c4), the method of pairing and connecting the sampling line branches in series is as follows: For multiple sampling lines generated by interference, determine whether the distance between adjacent sampling lines is less than the pairing distance. If so, use the sampling line series connection method to merge excessive sampling branches; otherwise, do not process it.

9. The method for automatically planning the measurement points and paths of an integral box bottom on a machine according to claim 1, characterized in that: The method to generate the final measurement path is as follows: d1) according to the set measurement mode, the latitudinal sampling line is used as the sampling route and the longitudinal sampling point is used as the sampling point; the measurement mode includes point measurement or line scanning; d2) Use BNF to define measurement programming element templates; d3) For the point measurement mode, the height of the measuring point is adaptively adjusted, and the measuring points are sorted according to the C-axis travel constraints of the machine tool to form an "S"-shaped measurement sequence; For line scanning measurement mode, the sampling line height is adaptively adjusted, and the measurement line is sorted and set in direction according to the C-axis travel constraints of the machine tool to form an "S"-shaped measurement sequence; d4) Based on the tool path generation function of general CAM software, the point measurement mode uses the point processing function to generate the final measurement path, and the line scanning measurement mode uses the curve processing function to generate the final measurement path to complete the path planning.

10. An integrated box bottom on-machine measurement point position and path automatic planning system, characterized in that: It includes user demand model definition module, latitudinal sampling line generation module, longitudinal sampling point generation module, and dual-mode measurement path planning module; User demand model definition module: Determine the requirements for on-machine measurement of parts based on user input and establish a demand model with dual definitions of model geometry and process data; Latitude sampling line generation module: according to the target sampling density requirements input by the user, the initial latitude sampling lines are generated by the generatrix equal arc length layering method, the processing feature virtual boundary is generated by the measurement constraint and the feature boundary, the interference relationship between the sampling line and the virtual boundary is determined, and the completely interfering sampling lines are split to form non-interfering or incompletely interfering latitude sampling lines; Meridian sampling point generation module: Generates initial radial sampling points based on latitudinal sampling lines, updates sampling point positions by using branching and resampling methods for incompletely interfering latitudinal sampling lines, and generates final meridian sampling points; Dual-mode measurement path planning module: According to the set measurement mode, define the measurement programming element template, perform measurement process processing, and combine the CAM path generation function to generate the final measurement path.