Method for measuring geometric parameters of screw tip tap based on line laser scanning

Abstract

The invention discloses a method for measuring geometric parameters of a screw tip tap based on line laser scanning. The method comprises the following steps: step 1, acquiring point cloud data; step 2, segmenting and correcting point cloud data; and 3, calculating geometric parameters of the screw tip tap. The method is mainly used for solving the technical problems of rapid measurement and parameter calculation of geometric parameters of the screw tip screw tap, considering that thread grinding is the last process in the production process of the screw tip screw tap, rapid and accurate non-contact scanning can be carried out on the screw tip screw tap by utilizing a line laser scanning technology before the thread grinding is the last process; the machining precision of key geometric parameters such as the front angle, the core diameter and the cutting edge inclination angle is detected and evaluated, and a theoretical basis is provided for adjustment and optimization of manufacturing process parameters. The method has the advantages of high measurement efficiency, high measurement information integrity and the like.

Description

Technical Field

[0001] The present invention belongs to the technical field of optical measurement, and specifically relates to a method for measuring geometric parameters of spiral-point taps based on line laser scanning. Background Art

[0002] A spiral-point tap is a special tool mainly used for processing deep threads or through-hole threads, and has the advantages of small tapping torque, not easy to clog chips, high machining accuracy, etc. It is currently widely used in fields such as electronics, instrument manufacturing, and precision machinery. When a spiral-point tap processes an internal thread, the chips are discharged forward along the feed direction, making it easier to start the cutting process and penetrate the material, while reducing machining damage caused by excessive cutting force. Before the thread grinding process in the production of spiral-point taps, how to quickly and accurately measure and obtain the geometric parameters of the cutting edge (including rake angle, core diameter, edge inclination angle, etc.) to ensure the machining accuracy of the tap after thread grinding has always been a major technical problem.

[0003] To detect the geometric parameters of the cutting edge of a spiral-point tap, traditional methods include microscope observation method and probe measurement method. Among them, the microscope observation method requires the measurer to have a high level of manual operation skills, and it is impossible to comprehensively measure the information of the entire spiral-point tap, resulting in low measurement efficiency; the probe measurement method can achieve a certain degree of automation, but limited by the size of the probe, it is often difficult to meet the high-precision measurement requirements of the tiny geometric parameters of the spiral-point tap, and the contact detection is likely to cause damage and deformation to the measured surface, resulting in a decrease in the surface accuracy of the spiral-point tap or forming scratches. Generally speaking, the measurement efficiency of the two methods is low and the measurement accuracy is limited. Summary of the Invention

[0004] In view of this, the purpose of the present invention is to provide a method for measuring geometric parameters of spiral-point taps based on line laser scanning, which uses line laser scanning technology to quickly and accurately perform non-contact scanning on the spiral-point tap, detect and evaluate the machining accuracy of the geometric parameters of the cutting edge, and has the advantages of high measurement efficiency and high integrity of measurement information.

[0005] To achieve the above purpose, the present invention provides the following technical solutions:

[0006] A method for measuring geometric parameters of spiral-point taps based on line laser scanning includes the following steps:

[0007] Step 1: Obtain point cloud data

[0008] 11) Collect point cloud data: Drive the line laser scanner to move uniformly along the Y direction of the axis of the measured tap, and measure at equal intervals to obtain an N×M data matrix; where: N represents the number of scan lines, and M represents the number of measurement points on each scan line;

[0009] 12) Preprocessing of point cloud data: Since the Y coordinates of the data points on the same scan line are the same, the three-dimensional data on each scan line is regarded as two-dimensional data, and missing values and outliers are processed to optimize the point cloud quality and improve data availability;

[0010] Step 2: Point cloud data segmentation and calibration

[0011] 21) Point cloud data segmentation: Project the tap contour containing cutting part information onto the axis plane, extract the projected contour line for change point detection, and divide the point cloud data into cutting part, calibration part, and shank part according to the change point position and tap structure characteristics;

[0012] 22) Point cloud data calibration

[0013] 221) For the point cloud data of the cylindrical part of the tap shank, the least squares method is used to fit the fitting center coordinates and radius of each scan line, and the least squares method is used again to fit the center space fitting line l; if the fitting line l is not parallel to the Y axis, rotate the fitting line l along the X direction and Z direction to make the fitting line l parallel to the Y axis;

[0014] 222) After the fitting line l is parallel to the Y axis, if the fitting line l is parallel to the Y axis but not collinear, translate the fitting line l along the X direction and Z direction to make the fitting line l collinear with the Y axis;

[0015] Step 3: Calculation of geometric parameters of spiral point tap

[0016] 31) Segmentation of flank data and flute data:

[0017] 311) Take the middle position data point of each scan line as the demarcation point and divide it into two parts;

[0018] 312) Find the highest points in the two parts and intercept the point cloud data between the highest points to obtain the flute data, and the remaining point cloud data is the flank data;

[0019] 313) Combine the flank data and flute data segmented from each scan line respectively to obtain the flute and flank;

[0020] 32) Calculate the geometric parameters of the spiral point tap: Based on the segmented flank data and flute data, calculate the geometric parameters of the spiral point tap.

[0021] Furthermore, in the step 12), the method steps for preprocessing the point cloud data are as follows:

[0022] 121) Marking and processing of invalid points;

[0023] 122) Marking and processing of outliers.

[0024] Further, in the step 121), the method steps for marking and processing invalid points are as follows:

[0025] 1211) Based on the "minimum" bounding box, identify and eliminate the data points that are irrelevant to the measured tap;

[0026] 1212) Identify the points with the value of -99.9999 in the data matrix as invalid points and mark them as NaN;

[0027] 1213) Calculate the number of invalid points in each scan line;

[0028] 1214) Determine whether the ratio of the number of invalid points in the scan line to the total number of data points exceeds the set first threshold: if so, delete the scan line and execute step 1215); if not, execute step 1215);

[0029] 1214) Delete the invalid points in each scan line;

[0030] 1215) Use cubic spline interpolation to complete the data points in the scan line to obtain the data after processing the invalid points.

[0031] Further, in the step 122), the method steps for marking and processing outliers are as follows:

[0032] 1221) Traverse the scan line and calculate the data residuals for the data of each scan line using a Gaussian filter;

[0033] 1222) Determine whether the data residual exceeds the set second threshold: if so, the corresponding data point is an outlier and execute step 1223); if not, execute step 1224);

[0034] 1223) Eliminate the outliers;

[0035] 1224) Use cubic spline interpolation to complete the data points in the scan line to obtain the data after processing the outliers for the t-th time;

[0036] 1225) Determine whether the current iteration number t is equal to the set maximum iteration number T max : if so, execute step 1226); if not, execute step 1221);

[0037] 1226) Output the data.

[0038] Further, in the step 21), the method steps for detecting the position and number of change points are as follows:

[0039] 211) Assume that there are m change points in the projected contour data, and define the set of change point position sequences as:

[0040] S(k,m) = {τ0:m+1 = τ 0< τ 1< … < τ m< τ m+1}

[0041] where τ0 = 0, τ m+1 = k, and k represents the number of data points;

[0042] 212) According to the statistical characteristics of the projected contour line, the cost function is determined as the sum of squared residuals of linear fitting. Considering that the number and position of change points are unknown, a penalty term ζ is introduced, and its objective function is obtained as:

[0043]

[0044] where, represents the projected contour data from τ j-1 + 1 to τ j ;

[0045] 213) Let k = 1, 2, …, n, and substitute it into F(k) to iteratively find the optimal set of change points:

[0046] τ 1:m = (τ1, …, τ m )

[0047] where n represents the total amount of projected contour data.

[0048] Furthermore, during the change point detection process, a pruning strategy is introduced to eliminate the data points that cannot be change points in each iteration from the search space, reducing computational redundancy.

[0049] Furthermore, in step 32), the geometric parameters of the unthreaded ground tap include the basic major diameter d, the core diameter D of the calibration part, the rake angle γ p , the length l4 of the inclined edge, the edge inclination angle λ, and the edge back width m.

[0050] Furthermore, the calculation method of the basic major diameter d is as follows:

[0051] Perform least squares fitting of a circle on each scan line of the edge back data to obtain the basic major diameter information on the scan line; perform outlier processing on the basic major diameter information, and take the mean of the basic major diameter information after outlier processing as the basic major diameter d of the tap;

[0052] The calculation method of the core diameter D of the calibration part is as follows:

[0053] For each scan line of the flank data, perform a least squares fitting of a circle to obtain the basic major diameter information on the scan line; use the center of the basic major diameter d as a reference point, calculate the distance from each point in the flute part to the center, and record the minimum distance. Twice this minimum distance is the core diameter; traverse all scan lines to obtain a series of core diameter data; perform outlier processing on the obtained core diameter data, and after processing, take the average of the remaining core diameter data, and use this average as the core diameter D of the standard part of the tap.

[0054] Helix angle γ p The calculation method is as follows:

[0055] Extract the rake face data from the flute data;

[0056] Perform a least squares fitting on each scan line of the flank data to obtain the center coordinate O, the basic major diameter d, and the tap minor diameter design parameter d1, and then obtain the intersection point p1 of the basic major diameter d and the rake face data, and the intersection point p2 of the tap minor diameter design parameter d1 and the rake face data; obtain the straight line 1 from the intersection points p1 and p2, and obtain the straight line 2 from the fitting center O and the intersection point p1; calculate the included angle between the straight line 1 and the straight line 2 to obtain the helix angle γ p ;

[0057] The calculation method of the inclined edge length l4 is as follows:

[0058] Measure the axial length of the cutting part of the tap, which is the inclined edge length l4;

[0059] The calculation method of the edge inclination angle λ is as follows:

[0060] Select the point s at l5 / 2, where l5 is the cutting cone length, and use two adjacent points to solve the tangent at the selected point s:

[0061]

[0062] where s1 = (x1, y1, z1) and s2 = (x2, y2, z2) respectively represent the coordinates of two adjacent points to the selected point s, and t represents the tangent direction vector of the cutting edge curve at the selected point;

[0063] The axis direction vector is represented as J y = [0 1 0], and the unit normal vector of all planes passing through the axis is represented as n = [cosθ 0 sinθ]; where θ represents the angle between n and the positive direction of the X-axis, and the range of θ is [-π π]; obtain the projection unit vector of the tangent on the plane from the tangent and the plane unit normal vector:

[0064]

[0065] Calculate the projection angle α between the tangent and the axis from the axis direction vector and the projection unit vector:

[0066]

[0067] Among them, α is a unary function containing θ; the extreme value within the domain is obtained according to the extreme value theorem, and the maximum value among the endpoint values and extreme values is selected, and this maximum value is the edge inclination angle λ.

[0068] The calculation method of the flank width m is as follows:

[0069] The spiral point tap is rotated at equal angles in sequence, and the rotation angle each time is 2π / N t , N t is the number of cutting teeth of the spiral point tap; the point cloud data is collected once every time the spiral point tap rotates; the first set of measurement data obtained from the first measurement is selected as the fixed point cloud, and the remaining N t -1 sets of measurement data are used as the moving point cloud, and the NDT algorithm is used to achieve accurate point cloud registration; the flank data of the calibration part in each group of measurement data is extracted respectively, and the flank data is subjected to a rotation transformation, and the rotation matrix is:

[0070]

[0071] In the formula, R i represents the rotation matrix of the i-th measurement data rotating counterclockwise around the Y axis; represents the rotation angle, and the range of i is [1, N t ;

[0072] Mark the contour lines of each part of the flank data and merge the flank data of each part;

[0073] Perform outlier processing on the extracted contour lines;

[0074] Perform linear fitting on the marked contour lines and add labels to the fitting lines according to the rotation direction;

[0075] Select a certain Y coordinate at equal intervals, calculate the coordinate values corresponding to each fitting line at this Y coordinate; calculate the distances between the corresponding points of the adjacent two lines at the corresponding Y coordinates respectively, and take the average value of the calculated distances to obtain the flank width m of each flank.

[0076] The beneficial effects of the present invention are as follows:

[0077] The method for measuring the geometric parameters of a spiral point tap based on line laser scanning according to the present invention takes into account that thread grinding is the last process in the production process of a spiral point tap. Before this, the line laser scanning technology can be used to perform rapid and accurate non-contact scanning on it, detect and evaluate the machining accuracy of key geometric parameters such as the rake angle, core diameter, and edge inclination angle, and provide a theoretical basis for the adjustment and optimization of manufacturing process parameters. This method has the advantages of high measurement efficiency and high integrity of measurement information. Description of the Drawings

[0078] To make the objectives, technical solutions, and beneficial effects of the present invention clearer, the present invention provides the following drawings for description:

[0079] Figure 1 It is a flowchart of the method for measuring geometric parameters of a spiral-point tap based on line laser scanning according to the present invention;

[0080] Figure 2 It is a flowchart of preprocessing of point cloud data;

[0081] Figure 3 It is the projection relationship of the tap axis plane;

[0082] Figure 4 It is the plane projection of the contour line;

[0083] Figure 5 It is a schematic diagram of the geometric parameters of the tap;

[0084] Figure 6 It is a flowchart of measuring the core diameter of the calibration part of the spiral-point tap;

[0085] Figure 7 It is a schematic diagram of marking the flank profile line;

[0086] Figure 8 It is a comparison chart of outlier processing based on a Gaussian filter; (a) original measurement data; (b) after outlier processing;

[0087] Figure 9 It is the point cloud segmentation based on the structural characteristics of the tap; (a) tap shank; (b) calibration part; (c) cutting part;

[0088] Figure 10 It is a comparison chart before and after tilt correction of the spiral-point tap; (a) original measurement data; (b) after tilt correction;

[0089] Figure 11 It is a schematic diagram of separating the flank and flute profile of the cutting part of the tap; (a) flank data; (b) flute data. Detailed Embodiments

[0090] The present invention will be further described below in conjunction with the drawings and specific embodiments, so that those skilled in the art can better understand the present invention and implement it, but the embodiments given are not intended to limit the present invention.

[0091] As Figure 1 shown, the method for measuring geometric parameters of a spiral-point tap based on line laser scanning in this embodiment includes the following steps.

[0092] Step 1: Obtain point cloud data

[0093] 11) Acquisition of point cloud data:

[0094] The line laser scanning technology captures point cloud data by continuously moving the scanning line along a specific surface, obtaining three-dimensional information of the object or scene. The height information of each scanning line is recorded in a CSV file, denoted as the Z coordinate. During the measurement, the line laser scanner is driven to move uniformly along the Y direction of the axis of the measured tap, and an N×M data matrix is obtained by equidistant measurement; where: N represents the number of scanning lines; M represents the number of measurement points on each scanning line, that is, the contour resolution, with the unit of points / contour. The spacing L between two adjacent scanning lines is:

[0095]

[0096] In the formula, v is the moving speed of the scanner; f s is the sampling frequency.

[0097] 12) Preprocessing of point cloud data:

[0098] On the same scanning line, the Y coordinates are the same, and the X coordinates of each data point can be calculated according to the contour resolution M and the X vector range of the scanner. By integrating all the measurement point information, the complete three-dimensional point cloud of the tap can be reconstructed.

[0099] Since the Y coordinates of the data points on the same scanning line are the same, the three-dimensional data on each scanning line is regarded as two-dimensional data, and missing values and outliers are processed to optimize the point cloud quality and improve data availability.

[0100] Specifically, as Figure 2 shown, in this embodiment, the method steps for preprocessing the point cloud data are:

[0101] 121) Marking and processing of invalid points. The method steps for marking and processing invalid points are:

[0102] 1211) Based on the "minimum" bounding box, identify and eliminate the data points irrelevant to the measured tap, reducing the redundancy of coordinate information;

[0103] 1212) Identify the points with a value of -99.9999 in the data matrix as invalid points and mark them as NaN;

[0104] 1213) Calculate the number of invalid points on each scanning line;

[0105] 1214) Determine whether the ratio of the number of invalid points in the scanning line to the total number of data points exceeds the set first threshold: if so, delete the scanning line and execute step 1215); if not, execute step 1215);

[0106] 1214) Delete the invalid points in each scan line; use cubic spline interpolation to complement the data points in the scan line to obtain the data after processing the invalid points.

[0107] In this embodiment, the first threshold is set to 50%.

[0108] 122) Marking and processing of outliers. The method steps for marking and processing outliers in this embodiment are as follows:

[0109] 1221) Traverse the scan line and calculate the data residuals for the data of each scan line using a Gaussian filter;

[0110] 1222) Determine whether the data residual exceeds the set second threshold: if so, the corresponding data point is an outlier, and step 1223) is executed; if not, step 1224) is executed;

[0111] 1223) Eliminate the outliers;

[0112] 1224) Use cubic spline interpolation to complement the data points in the scan line to obtain the data after processing the outliers for the t-th time;

[0113] 1225) Determine whether the current iteration number t is equal to the set maximum iteration number T max : if so, step 1226) is executed; if not, step 1221) is executed;

[0114] 1227) Output the data.

[0115] In this embodiment, the maximum iteration number T max = 2.

[0116] Step two: Point cloud data segmentation and calibration

[0117] 21) Point cloud data segmentation:

[0118] The spiral point tap consists of a cutting part, a calibration part, and a shank part. To accurately calculate the geometric parameters of the tap, it needs to be accurately segmented. Specifically, project the tap contour containing the cutting part information onto the axis plane, as Figure 3 shown. Extract the projected contour line for change point detection. The projected contour line consists of multiple straight lines connected end to end, as Figure 4 shown. According to the change point position and the tap structure characteristics, segment the point cloud data into a cutting part, a calibration part, and a shank part. Figure 4 In, the AB, BC, and CD segments represent the shank part of the tap, the DE segment represents the calibration part, and the EF and FG segments represent the cutting part.

[0119] The change point information in the projection contour reflects the structural composition of the spiral tap, and its quantity and position need to be detected. Specifically, in this embodiment, the method steps for detecting the position and quantity of the change points are:

[0120] 211) Assuming that there are m change points in the projection contour data, the set of change point position sequence is defined as:

[0121] S(k,m)={τ 0:m+1 =τ 0< τ 1< …<τ m< τ m+1}

[0122] Where τ0=0, τ m+1 =k, k represents the number of data points;

[0123] 212) According to the statistical characteristics of the projection contour, the cost function Determined as the residual sum of squares of linear fitting, considering the unknown number and location of change points, the penalty term ζ is introduced, and the objective function is obtained as follows:

[0124]

[0125] in, Indicates that from τ j-1 +1 to τ j Projection profile data;

[0126] 213) Let k = 1, 2, ..., n, and substitute it into F(k) to iteratively find the optimal set of change points:

[0127] τ 1:m =(τ1,…,τ m )

[0128] Where n represents the total amount of projection profile data.

[0129] In a preferred implementation of this embodiment, a pruning strategy is introduced during the change point detection process to remove data points that are unlikely to become change points from the search space during each iteration, thereby reducing computational redundancy.

[0130] 22) Point cloud data correction

[0131] In view of the fact that there are certain inaccuracies in the installation of the scanner and the tap being measured, the scanning data may be tilted, thus affecting the accuracy of the geometric parameter measurement. Therefore, it is necessary to rotate and translate the tool axis to align it with the sensor coordinate axis to ensure accurate measurement of the tap geometric parameters.

[0132] 221) For the point cloud data of the cylindrical part of the tap shank, the least squares method is used to fit the center coordinates and radius of each scan line, and the least squares method is used again to fit the spatial fitting line l of the centers; if the fitting line l is not parallel to the Y-axis, then rotate the fitting line l along the X direction and the Z direction to make the fitting line l parallel to the Y-axis.

[0133] Specifically, the method of X-axis rotation correction is as follows: First, project l onto the XOY plane and calculate the angles α and β1 between it and the positive directions of the X-axis and Y-axis; then, judge the rotation direction according to the size of the angle α: rotate counterclockwise when α < 90° and rotate clockwise when α > 90°; finally, take β1 as the rotation angle and rotate l around the Z-axis to obtain the spatial line l1, making it perpendicular to the X-axis, and obtain the rotation matrix R1.

[0134] The method of Z-axis rotation correction is as follows: Calculate the angle between l1 and the positive direction of the Z-axis. If l1 is not perpendicular to the Z-axis, it needs to be further rotated and transformed; first, project l1 onto the ZOY plane and calculate the angles γ and β2 between it and the positive directions of the Z-axis and Y-axis; then, judge the rotation direction according to γ and rotate l1 around the X-axis by the angle β2 to obtain the spatial line l2, making it parallel to the Y-axis, and obtain the rotation matrix R2.

[0135] 222) Translation correction: After the fitting line l is parallel to the Y-axis, if the fitting line l is parallel to the Y-axis but not collinear, then translate the fitting line l along the X direction and the Y direction to make the fitting line l collinear with the Y-axis. Specifically, if the tap axis is still not completely aligned with the sensor coordinate axis, it needs to be translated; calculate that the displacements of l2 in the X direction and Z direction with respect to the Y-axis are x d and z d , and translate l2 in the corresponding directions by x d and z d respectively, to make it completely aligned with the sensor coordinate axis, and obtain the translation matrix T1.

[0136] Combining the above steps, the tilt correction matrix is M = T1·R2·R1. Applying it to the complete point cloud data can complete the tilt correction of the tap.

[0137] Step 3: Calculation of the geometric parameters of the spiral-point tap

[0138] In the previous steps, the point cloud data of the tap has been segmented into three main parts, and the key focus is on the spiral-point groove. For more accurate subsequent calculations, it is also necessary to further segment and isolate the point clouds of the flank and groove parts in the spiral-point groove point cloud. Considering the characteristics of the tap structure, continue to adopt the two-dimensional data processing strategy, and the method steps are as follows:

[0139] 31) Segmentation of flank data and groove data:

[0140] (311) Take the data point at the middle position of each scan line as the demarcation point and divide it into two parts;

[0141] (312) Find the highest points in the two parts and intercept the point cloud data between the highest points to obtain the groove data, and the remaining point cloud data is the flank data;

[0142] (313) Combine the flank data and groove data segmented from each scan line respectively to obtain the groove and the flank;

[0143] (32) Calculate the geometric parameters of the spiral-point tap:

[0144] Based on the segmented flank data and groove data, calculate the geometric parameters of the spiral-point tap to provide a basis for more precise machining process adjustment. Specifically, as Figure 5 shown, the geometric parameters of the spiral-point tap without thread grinding include the basic major diameter d, the core diameter D of the calibration part, the rake angle γ p , the length l4 of the inclined edge, the edge inclination angle λ and the flank width m.

[0145] (1) The calculation method of the basic major diameter d is:

[0146] The basic major diameter refers to the basic size of the major diameter of the tap thread. Perform least squares fitting of a circle on each scan line of the flank data to obtain the basic major diameter information on the scan line. To reduce the influence of external factors on the measurement results, perform outlier processing on the basic major diameter information, and take the mean value of the basic major diameter information after outlier processing as the basic major diameter d of the tap.

[0147] (2) The calculation method of the core diameter D of the calibration part is:

[0148] The calibration part of the spiral-point tap, that is, the straight groove part, and its core diameter refers to the diameter of an arc tangent to the bottom of the groove at a given point on the axis of this part. In an ideal situation, the core diameter of any cross-section of the standard part of the tap should be equal. However, in actual production, affected by various factors such as machining errors, machine tool accuracy, and measurement deviations, the core diameters of each cross-section may be different, resulting in a certain degree of runout. The measurement process of the core diameter of the calibration part of the spiral-point tap is as Figure 6 shown:

[0149] Perform least squares fitting of a circle on each scan line of the flank data to obtain the basic major diameter information on the scan line; use the center of the circle of the basic major diameter d as the reference point, calculate the distance from each point in the groove part to the center of the circle, and record the minimum value of the distance. Twice this minimum distance value is the core diameter; traverse all scan lines to obtain a series of core diameter data; perform outlier processing on the obtained core diameter data, and take the average value of the remaining core diameter data after processing as the core diameter D of the standard part of the tap.

[0150] (3) The rake angle γ pThe calculation method is as follows:

[0151] The rake angle of the tap refers to the angle between the front face (the line connecting the intersections of the major diameter and the minor diameter with the front cutting face) and a plane passing through the center line of the tap and the crest of the thread, and is usually measured at the l4 / 2 position. Since the actual major diameter of the tap with unground threads cannot be obtained, when calculating the rake angle of the spiral tip, in this embodiment, the angle between the line connecting the design parameters of the basic major diameter and the minor diameter of the tap with the intersection point of the curve front and the line connecting the intersection point of the basic major diameter and the curved surface with the center of the tap is taken as the rake angle. The method is as follows:

[0152] Separate the flank and groove profile data of the spiral tip groove data, and extract the front cutting face data from the groove profile data;

[0153] Perform least squares fitting on each scan line of the flank data to obtain the center coordinate O, the basic major diameter d, and the design parameter d1 of the minor diameter of the tap, and then obtain the intersection point p1 of the basic major diameter d and the front cutting face data, as well as the intersection point p2 of the design parameter d1 of the minor diameter of the tap and the front cutting face data; obtain the straight line 1 from the intersection points p1 and p2, and obtain the straight line 2 from the fitting center O and the intersection point p1; calculate the included angle between the straight line 1 and the straight line 2 to obtain the rake angle γ p .

[0154] (4) The calculation method of the inclined edge length l4 is as follows:

[0155] The inclined edge length refers to the axial length between the radial cross-sectional planes at the beginning and end of the cutting edge. In the above steps, the cutting part of the tap has been separated and independent, and by measuring the axial distance between its first and last end faces, the inclined edge length can be obtained. That is, in this embodiment, measuring the axial length of the cutting part of the tap is the inclined edge length l4.

[0156] (5) The calculation method of the cutting edge inclination angle λ is as follows:

[0157] The cutting edge inclination angle refers to the maximum projection angle between the tangent line of the selected point on the cutting edge and the axis. Generally, λ = 10°. to 15°. Referring to the GB / T 28254—2012 standard, the selected point s is at l5 / 2, and l5 is the cutting cone length. Since the parametric expression of the cutting edge curve cannot be directly obtained, it is necessary to use two adjacent points to solve the tangent line at the selected point s:

[0158]

[0159] Among them, s1 = (x1, y1, z1) and s2 = (x2, y2, z2) respectively represent the coordinates of two points adjacent to the selected point s, and t represents the tangent direction vector of the cutting edge curve at the selected point;

[0160] The axis direction vector is represented as J y= [0 1 0], and the unit normal vectors of all planes passing through the axis are expressed as n = [cosθ 0 sinθ]; where θ represents the angle between n and the positive direction of the X-axis, and the range of θ is [-π, π]; the projection unit vector of the tangent on the plane is obtained from the tangent and the unit normal vector of the plane:

[0161]

[0162] The projection angle α between the tangent and the axis is calculated from the axis direction vector and the projection unit vector:

[0163]

[0164] where α is a unary function containing θ;

[0165] By performing the inverse cosine solution on the above formula, the projection angle α can be obtained. α is a unary function containing θ. According to the extreme value theorem, the extreme values within the domain can be obtained, and the maximum value among the endpoint values and the extreme values is selected, and the maximum value is the rake angle λ.

[0166] (6) The calculation method of the flank width m is as follows:

[0167] The flank width of the tap refers to the chord width between the cutting edge and the opposite edge. Due to the existence of the spiral point groove, the flank width of the cutting part is variable. Therefore, the focus of the present invention is the flank width of the calibration part of the spiral point tap. In actual measurement, considering the axisymmetric characteristics and the number of cutting teeth of the tap, the method of measuring by equally angular rotation in batches is adopted. The measurement of the flank width is based on the complete tap data. Therefore, it is necessary to splice the point cloud data. The specific process is as follows:

[0168] Rotate the spiral point tap at equal angles in turn, and the rotation angle each time is 2π / N t , N t is the number of cutting teeth of the spiral point tap; each time the spiral point tap rotates, the point cloud data is collected once; the first set of measurement data obtained from the first measurement is selected as the fixed point cloud, and the remaining N t -1 sets of measurement data are used as the moving point cloud, and the NDT algorithm is used to achieve accurate point cloud registration; the flank data of the calibration part in each set of measurement data is extracted respectively, and the flank data is subjected to rotation transformation, and the rotation matrix is:

[0169]

[0170] In the formula, R i represents the rotation matrix of the i-th measurement data rotating counterclockwise around the Y-axis; represents the rotation angle, and the range of i is [1, N t ;

[0171] After the rotation transformation, outline the flank data of each part and merge the flank data of each part;

[0172] To avoid the influence of outliers, process the extracted outline for outliers;

[0173] Perform linear fitting on the marked outline and add labels to the fitting lines according to the rotation direction, such as Figure 7 shown;

[0174] Select a certain Y coordinate at equal intervals and calculate the corresponding coordinate values of each fitting line at this Y coordinate; calculate the distances between the corresponding points of the Y coordinates of two adjacent lines (such as 1 and 2, 3 and 4, etc.) respectively, and take the average of the calculated distances to obtain the flank width m of each flank.

[0175] For the convenience of those skilled in the art to better understand the present invention, the technical solution of the present invention will be further described below in conjunction with measurement examples.

[0176] Referring to Step 1, outliers in the measurement data can be removed, such as Figure 8 shown.

[0177] Then, referring to Step 2, perform point cloud segmentation and tilt correction on the processed data based on the structural characteristics of the tap, such as Figure 9 and Figure 10 shown.

[0178] Finally, referring to Step 3, separate the flank and flute data, such as Figure 11 shown, and accurately measure and calculate the geometric parameters of the spiral point tap. The specific measurement parameters are shown in Table 1.

[0179] Table 1 Measurement Parameters of Spiral Point Tap

[0180]

[0181] The above-described embodiments are only preferred embodiments given to fully illustrate the present invention, and the protection scope of the present invention is not limited thereto. Equivalent substitutions or transformations made by those skilled in the art on the basis of the present invention are all within the protection scope of the present invention. The protection scope of the present invention is subject to the claims.

Claims

1. A method for measuring geometric parameters of a spiral-point tap based on line laser scanning, characterized in that: It includes the following steps: Step 1: Obtain point cloud data 11) Collect point cloud data: Drive the line laser scanner to move uniformly along the Y direction of the axis of the tap to be measured, and measure at equal intervals to obtain an N×M data matrix; where: N represents the number of scan lines, and M represents the number of measurement points on each scan line; 12) Preprocess the point cloud data: Since the Y coordinates of the data points on the same scan line are the same, regard the three-dimensional data on each scan line as two-dimensional data, and perform missing value and outlier processing to optimize the point cloud quality and improve data availability; Step 2: Segment and correct the point cloud data 21) Segment the point cloud data: Project the tap contour containing cutting part information onto the axis plane, extract the projected contour line for change point detection, and according to the change point position and tap structure characteristics, segment the point cloud data into a cutting part, a calibration part, and a shank part; 22) Correct the point cloud data 221) For the point cloud data of the cylindrical part of the tap shank, use the least squares method to fit the center coordinates and radius of each scan line, and then use the least squares method again to fit the spatial fitting line l of the centers; if the fitting line l is not parallel to the Y axis, rotate the fitting line l along the X direction and the Z direction to make the fitting line l parallel to the Y axis; 222) After the fitting line l is parallel to the Y axis, if the fitting line l is parallel to the Y axis but not collinear, translate the fitting line l along the X direction and the Z direction to make the fitting line l collinear with the Y axis; Step 3: Calculate the geometric parameters of the spiral point tap 31) Segment the flank data and the groove data: 311) Use the middle position data point of each scan line as the demarcation point and segment it into two parts; 312) Find the highest points in the two parts and intercept the point cloud data between the highest points to obtain the groove data, and the remaining point cloud data is the flank data; 313) Combine the flank data and the groove data segmented from each scan line respectively to obtain the groove and the flank; 32) Calculate the geometric parameters of the spiral point tap: Based on the segmented flank data and groove data, calculate the geometric parameters of the spiral point tap.

2. The geometric parameter measurement method of the spiral point tap based on line laser scanning according to claim 1, characterized in that: In the said step 12), the method steps for preprocessing the point cloud data are: 121) Mark and process invalid points; 122) Mark and process outliers.

3. The geometric parameter measurement method of a spiral point tap based on line laser scanning according to claim 2, characterized in that: In the said step 121), the method steps for marking and processing invalid points are: 1211) Based on the "minimum" bounding box, identify and remove the data points irrelevant to the tap to be measured; 1212) Identify the points with a value of -99.9999 in the data matrix as invalid points and mark them as NaN; 1213) Calculate the number of invalid points on each scan line; 1214) Judge whether the ratio of the number of invalid points in the scan line to the total number of data points exceeds the set first threshold: if so, delete this scan line and execute step 1215); if not, execute step 1215); 1214) Delete the invalid points in each scan line; 1215) Use cubic spline interpolation to complement the data points in the scan line to obtain the data after invalid point processing.

4. The geometric parameter measurement method of the spiral pointed tap based on line laser scanning according to claim 2, wherein: In the said step 122), the method steps for marking and processing outliers are: 1221) Traverse the scan lines, and use a Gaussian filter to calculate the data residuals for the data of each scan line respectively. 1222) Determine whether the data residual exceeds a set second threshold: If so, the corresponding data point is an abnormal point, and step 1223) is executed; if not, step 1224) is executed. 1223) Remove the abnormal points. 1224) Use cubic spline interpolation to complement the data points in the scan line to obtain the data after the t-th abnormal point processing. 1225) Determine whether the current iteration number t is equal to the set maximum iteration number T max : If so, execute step 1226); if not, execute step 1221); 1226) Output the data.

5. The geometric parameter measurement method of the spiral point tap based on line laser scanning according to claim 1, characterized in that: In step 21), the method steps for detecting the position and quantity of change points are as follows: 211) Assume that there are m change points in the projection contour data, and define the change point position sequence set as: S(k,m) = {τ 0:m+1 = τ0 < τ1 < … < τ m < τ m+1} where τ0 = 0, τ m+1 = k, where k represents the number of data points; 212) According to the statistical characteristics of the projection contour line, the cost function is determined as the sum of squared residuals of linear fitting. Considering that the number and position of the change points are unknown, a penalty term ζ is introduced, and its objective function is obtained as follows: Among them, represents the projection contour data from τ j-1 +1 to τ j ; 213) Let k = 1, 2,..., n, and substitute it into F(k) to iteratively find the optimal change point set: τ 1:m =(τ1,…,τ m ) where n represents the total amount of projection contour data.

6. The method for measuring geometric parameters of a spiral-point tap based on line laser scanning according to claim 5, wherein: During the change point detection process, a pruning strategy is introduced to remove the data points that cannot be change points in each iteration process from the search space, reducing the calculation redundancy.

7. The geometric parameter measurement method of the spiral point tap based on line laser scanning according to claim 1, characterized in that: In the said step 32), the geometric parameters of the unthreaded ground screw tip tap include the basic major diameter d, the core diameter D of the calibration part, and the rake angle γ p , the length l4 of the inclined edge, the edge inclination angle λ, and the edge back width m.

8. The geometric parameter measurement method of the spiral tip tap based on line laser scanning according to claim 7, characterized in that: The calculation method of the basic major diameter d is as follows: Perform least squares fitting of a circle on each scan line of the flank data to obtain the basic major diameter information on the scan line; perform outlier processing on the basic major diameter information, and take the mean value of the basic major diameter information after outlier processing as the basic major diameter d of the tap. The calculation method of the calibrated partial core diameter D is as follows: Perform least squares fitting of a circle on each scan line of the flank data to obtain the basic major diameter information on the scan line; use the center of the basic major diameter d as a reference point, calculate the distance from each point in the groove part to the center of the circle, and record the minimum value of the distance. Twice the minimum value of the distance is the core diameter; traverse all scan lines to obtain a series of core diameter data; perform outlier processing on the obtained core diameter data, and take the average value of the remaining core diameter data after processing as the core diameter D of the standard part of the tap. Rake angle γ p is calculated as follows: Extract the rake face data from the groove data. Perform least squares fitting on each scan line of the flank data to obtain the center coordinate O, the basic major diameter d, and the tap minor diameter design parameter d1. Furthermore, obtain the intersection point p1 of the basic major diameter d and the rake face data, and the intersection point p2 of the tap minor diameter design parameter d1 and the rake face data; obtain the straight line 1 from the intersection points p1 and p2, and obtain the straight line 2 from the fitted center O and the intersection point p1; calculate the included angle between the straight line 1 and the straight line 2 to obtain the rake angle γ p ; The calculation method of the inclined edge length l4 is as follows: Measure the axial length of the cutting part of the tap, which is the inclined edge length l4. The calculation method of the rake angle λ is as follows: Use two adjacent points to solve the tangent line at the selected point s: where s1 = (x1, y1, z1) and s2 = (x2, y2, z2) respectively represent the coordinates of two points adjacent to the selected point s, and t represents the tangent direction vector of the cutting edge curve at the selected point. The axial direction vector is represented as J y = [0 1 0], and the unit normal vectors of all planes passing through the axis are represented as n = [cosθ 0 sinθ]; where θ represents the angle between n and the positive direction of the X-axis, and the range of θ is [-π, π]; the projection unit vector of the tangent on the plane is obtained from the tangent and the plane unit normal vector: Calculate the projection angle α between the tangent line and the axis from the axis direction vector and the projection unit vector: where α is a unary function containing θ; obtain the extreme value within the domain according to the extreme value theorem, and select the maximum value among the endpoint values and the extreme values. This maximum value is the rake angle λ. The calculation method of the flank width m is as follows: Rotate the spiral point tap at equal angles in sequence, with the rotation angle being 2π / N each time t , N t is the number of cutting teeth of the spiral point tap; collect the point cloud data once every time the spiral point tap rotates; select the first set of measurement data obtained from the first measurement as the fixed point cloud, and the remaining N t -1 sets of measurement data as the moving point cloud, and use the NDT algorithm to achieve accurate point cloud registration; extract the flank data of the calibration part in each set of measurement data respectively, and perform a rotation transformation on the flank data. The rotation matrix is: Wherein, R i represents the rotation matrix for the i-th measurement data to rotate counterclockwise around the Y-axis; represents the rotation angle, and the range of i is [1, N t ; Mark the contour lines of each part of the flank data and merge each part of the flank data. Perform outlier processing on the extracted contour lines. Perform linear fitting on the marked contour lines and add labels to the fitting lines according to the rotation direction. Select a certain Y coordinate at equal intervals and calculate the corresponding coordinate values of each fitted line at this Y coordinate; calculate the distances between the corresponding points of adjacent lines at the corresponding Y coordinates respectively, and take the average of the calculated distances to obtain the flank width m of each flank.

Images