Low-coherence interference measuring machine and bearing ring sealing groove size detection method

Abstract

The invention discloses a low-coherence interference measuring machine and a method for detecting the size of a bearing ring sealing groove, and the machine comprises a supporting table, a linear motor disposed at the rear part of the supporting table, a lifting plate connected with a telescoping rod of the linear motor, a transverse guide rail disposed on the front side surface of the lifting plate, a sliding block disposed on the transverse guide rail, and a vertical rod disposed on the sliding block. The vertical rod is arranged on the supporting table, the measuring probe is arranged at the lower end of the vertical rod, the rotary table is arranged on the supporting table, the rotary motor is arranged on the lower surface of the supporting table, and the three-jaw chuck is arranged in the rotary table and used for fixing a measured bearing ring. The measuring probe is aligned to the center position of the three-jaw chuck, and the measuring probe can horizontally emit light to the inner side face of the bearing ring and receive light reflected by the bearing ring; the device has the characteristics of high detection precision, high detection efficiency and long service life of the measuring probe.

Description

Technical Field

[0001] The present invention relates to the technical field of bearing detection, and in particular to a low-coherence interferometric measuring machine and a method for detecting the dimensions of the sealing groove of a bearing ring. Background Art

[0002] Due to the changing operating conditions of bearings, the requirements for sealing performance are getting higher and higher. Therefore, domestic bearing manufacturers have continuously improved the bearing structure and sealing performance. In particular, sealing grooves are provided on the inner and outer rings of bearings, changing the bearing from a structure without a sealing groove to a structure with a sealing groove, which can better solve the bearing sealing performance. Therefore, after the bearing is processed, the geometric measurement of the bearing sealing groove is particularly important.

[0003] Bearing manufacturers usually use templates and gauges to perform light gap detection on the bearing sealing groove to determine whether the dimensions and positions of the sealing groove are qualified, but the actual dimensions cannot be determined.

[0004] Or a contact profilometer is used to detect the bearing sealing groove. It can only detect sealing grooves with larger dimensions. For micro-precision bearings with a sealing groove width and depth both less than 1 mm, the measurement error is large, and it is difficult to meet the measurement requirements.

[0005] In addition, contact measurement is likely to cause scratches on the surface of the workpiece to be measured, resulting in secondary damage, and it is difficult to meet the detection requirements of precision bearings. Summary of the Invention

[0006] The object of the present invention is to overcome the deficiencies of the prior art and provide a low-coherence interferometric measuring machine and a method for detecting the dimensions of the sealing groove of a bearing ring.

[0007] To achieve the above object, the present invention adopts the following technical solutions:

[0008] A low-coherence interferometric measuring machine includes a support table, a linear motor provided at the rear of the support table, a lifting plate connected to the telescopic rod of the linear motor, a transverse guide rail provided on the front side surface of the lifting plate, a slider provided on the transverse guide rail, a vertical rod provided on the slider, a measuring probe provided at the lower end of the vertical rod, a turntable provided on the support table, a rotary motor provided on the lower surface of the support table, and a three-jaw chuck provided in the turntable for fixing the bearing ring to be measured; the measuring probe is aligned with the center position of the three-jaw chuck, and the measuring probe can horizontally emit light rays towards the inner side surface of the bearing ring and receive the light rays reflected by the bearing ring; the rotating shaft of the rotary motor is connected to the lower part of the turntable. The machine further includes a host computer, which is electrically connected to the measuring probe and the rotary motor respectively, and the host computer is electrically connected to the linear motor through a controller. There are several control buttons provided on the controller.

[0009] The present invention adopts optical principles for measurement, and does not require physical contact with the bearing ring being measured, thereby avoiding surface damage or contamination problems caused by contact, which is particularly important for high-precision or fragile bearing surfaces.

[0010] It can provide nanometer or even sub-nanometer measurement accuracy, which is much higher than the measurement accuracy of contact profilometers, and has more advantages in detecting tiny surface defects or morphological changes. It can complete the scanning of large areas or complex surfaces in a short time.

[0011] Preferably, the lifting plate is rectangular, the transverse guide rail includes a U-shaped plate opening forward, two cross bars arranged on the U-shaped plate, the sliders are slidably connected to the two cross bars respectively, and the U-shaped plate is fixedly connected to the front surface of the lifting plate.

[0012] A method for detecting the size of a bearing ring sealing groove of a low-coherence interferometer measuring machine includes the steps of detecting the width of the sealing groove:

[0013] Step 1: Place the bearing ring to be tested in the three-jaw chuck, use the top surface of the jaws of the three-jaw chuck as the reference surface, and adjust the bearing ring to keep it flush with the reference surface;

[0014] Step 2: The staff manipulates the control button to control the measuring probe to move downward until the light shines 1 mm above the sealing groove of the bearing ring. The upper computer records the position 1 of the upper end of the telescopic rod of the linear motor at this time.

[0015] Step 3: The staff manipulates the control button to control the measuring probe to move downward until the light is irradiated to 1 mm below the sealing groove of the bearing ring. The upper computer records the position 2 of the upper end of the telescopic rod of the linear motor at this time.

[0016] Step 4, the host computer sets the scanning range of the low-coherence interferometer to be from position 1 to position 2, the scanning mode is linear scanning, that is, data is collected vertically, the turntable does not rotate, and the scanning speed is set;

[0017] Step 4-1, the host computer controls the linear motor to drive the measuring probe to move downward, so that the upper end of the telescopic rod moves from position 1 to position 2, and the measuring probe emits light horizontally to the inner side of the bearing ring and receives the light reflected from the inner side; the host computer uses the reflected light to obtain the detected point cloud data, and generates a point cloud based on the point cloud data;

[0018] Step 4-2, the host computer uses the ISS point cloud corner detection algorithm to extract the corner points of the point cloud;

[0019] Step 4-3, selecting the point clouds at the upper side and the lower side of the sealing groove, and fitting the straight line L1 corresponding to the point cloud of the upper side of the sealing groove and the straight line L2 corresponding to the point cloud of the lower side of the sealing groove by using the point cloud straight line fitting algorithm;

[0020] Step 4-4: The host computer calculates the distance D1 between L1 and L2, and D1 is the width value of the sealing groove at the current angle.

[0021] Step 5: Let i1 = 0. Taking the current scanning position as the starting point, select measurement positions at M - 1 angles along the circumferential direction of the bearing race, where M ≥ 3.

[0022] Increase the value of i1 by 1. If i > M - 1, then go to Step 6; otherwise, the host computer controls the turntable to rotate to the position at the i1-th angle, and return to Step 4-1.

[0023] Step 6: The host computer obtains D1 at M measurement angles, removes the maximum and minimum values among all D1, and calculates the average value of the remaining M - 2 D1. The obtained result is the final width measurement value of the detected sealing groove.

[0024] The detection method of the present invention has high detection accuracy, can perform large - batch detections, saving time and effort; it can effectively avoid probe damage and ensure the safety of the detected bearing race.

[0025] Preferably, between Step 4-2 and Step 4-3, the following steps are further included: The sparse point cloud data on the upper and lower sides of the sealing groove are both processed as follows:

[0026] Step 11: For each point P i2 , use the k - d tree nearest neighbor search to find its nearest neighbor point P j2 (x j2 , y j2 , z j2 );

[0027] Step 12: Calculate the weight w j2 according to the distance d j2 between each point and its nearest neighbor point;

[0028]

[0029] where p is a positive number that controls the attenuation rate of the weight, and the weight of the nearest neighbor point can be adjusted by changing the value of p;

[0030] Step 13: Calculate the coordinates of the insertion point using the weights of the nearest neighbor points

[0031]

[0032] where m is the number of nearest neighbor points;

[0033] Step 14: Insert the calculated insertion point into the sparse point cloud data.

[0034] During the measurement process, since the angles formed by the two side surfaces of the sealing groove and the light of the measurement probe are too large, it affects the reception of the reflected light, resulting in less data measured for the positions of the two side surfaces of the sealing groove and a relatively sparse point cloud. Therefore, the present invention supplements and inserts the sparse point cloud, improves the accuracy of fitting the corresponding straight lines of the two side surfaces of the sealing groove by the subsequent point cloud, and further improves the measurement accuracy.

[0035] Preferably, after detecting the width of the sealing groove, the following depth detection of the sealing groove is carried out:

[0036] Step 7-1, the upper computer controls the linear motor to drive the measurement probe to move downward, so that the upper end of the telescopic rod moves from position 1 to position 2, the measurement probe horizontally emits light to the inner side surface of the bearing race, and receives the light reflected by the inner side surface; the upper computer obtains the detected point cloud data by using the reflected light and generates a point cloud according to the point cloud data.

[0037] Step 7-2, the upper computer extracts the corner points of the point cloud by using the ISS point cloud corner point detection algorithm.

[0038] Step 7-3, select the point clouds corresponding to the top surface and the bottom surface of the sealing groove, and use the point cloud fitting straight line algorithm based on the least square method to fit the straight line L3 corresponding to the top surface point cloud of the sealing groove and the straight line L4 corresponding to the bottom surface point cloud of the sealing groove.

[0039] Step 7-4, the upper computer calculates the distance D2 between L3 and L4, and D2 is the width value of the sealing groove at the current angle.

[0040] Step 7-5, set i3 = 0, take the current scanning position as the starting point, and select the measurement positions at M-1 angles along the circumferential direction of the bearing race, where M≥3.

[0041] Increase the value of i3 by 1. If i3 > M-1, then go to step 7-6; otherwise, the upper computer controls the turntable to rotate to the position at the i3th angle and returns to step 7-1.

[0042] Step 7-6, the upper computer obtains D2 at M measurement angles, removes the maximum value and the minimum value among all D2, and calculates the average value of the remaining M-2 D2. The obtained result is the final depth measurement value of the detected sealing groove.

[0043] Preferably, step 4-2 includes the following steps:

[0044] Step 61, construct a covariance matrix:

[0045] For each point p in the point cloud i4 and its neighborhood points both construct the covariance matrix C of the two.

[0046] Among them, p j4 ∈R 3 ,p i4 is the point coordinates inside, is the centroid of the center point, is the total number of internal points;

[0047] Step 62, eigenvalue calculation:

[0048] The host computer performs eigenvalue decomposition on C using the method of constructing a characteristic polynomial to obtain eigenvalues λ 1 , λ 2 , λ 3 , and sets λ 1 ≥λ 2 ≥λ 3 ≥0 and the corresponding eigenvectors;

[0049] Step 63, significance determination:

[0050] The host computer calculates the eigenvalue ratios T 21 and T 32 :

[0051]

[0052] Among them, λ 1 is the largest eigenvalue, corresponding to the variation of the data in the main direction; λ 2 is the second largest eigenvalue, corresponding to the variation of the data in the second main direction; λ 3 is the smallest eigenvalue, corresponding to the variation of the data in the smallest main direction;

[0053] If T 21 or T 32 exceeds the set threshold, then p i4 is considered a point cloud corner point, otherwise, p i4 is not considered a point cloud corner point.

[0054] Extract the corner points of the point cloud corresponding to the seal groove part, which is convenient for subsequent point cloud segmentation. By selecting the point cloud of each part of the seal groove, the obtained redundant point cloud can be removed, and only the point cloud corresponding to the seal groove part is retained, making the measurement result more accurate.

[0055] Preferably, before step 1, the following calibration steps are also included:

[0056] Place the standard ring gauge in the three-jaw chuck, and take the top surface of the chuck jaw of the three-jaw chuck as the reference plane, and adjust the standard ring gauge to make it horizontal with the reference plane;

[0057] The operator controls the downward movement of the measurement probe through the control button, so that the light emitted by the measurement probe irradiates the inner ring surface of the standard ring gauge, and receives the light reflected by the inner ring surface of the standard ring gauge;

[0058] The host computer sets the scanning mode to circular scanning and sets the scanning speed; the host computer controls the rotation motor to work and starts scanning;

[0059] After the scanning is completed, the host computer generates the calibrated point cloud from the measured point cloud data, and the host computer calculates the circle fitted by the calibrated point cloud and calculates the diameter D3 of the circle.

[0060] The size (50mm inner diameter) of the standard ring gauge is known and of high precision. After calibration, the measured value is aligned with the true value of the standard ring gauge to correct the systematic errors caused by mechanical assembly, probe offset or optical system distortion, etc.

[0061] During the calibration process, the optical measurement ratio (for example, the correspondence between pixels and actual dimensions) is determined, which is used to correct the measured point cloud data to ensure that the scanned point cloud data is consistent with the true geometric dimensions.

[0062] Therefore, the present invention has the following beneficial effects: convenient operation, high detection accuracy, high detection efficiency, low detection cost; long service life of the measurement probe, and good safety of the detected bearing raceway. Description of the Drawings

[0063] Figure 1 is a schematic structural diagram of the present invention;

[0064] Figure 2 is a schematic structural diagram of the bearing raceway of the present invention;

[0065] Figure 3 is a cross-sectional view of the bearing raceway of the present invention;

[0066] Figure 4 is a schematic structural diagram of the standard ring gauge used in the present invention. Detailed Embodiments

[0067] The following further describes the present invention in conjunction with the drawings and specific embodiments.

[0068] As Figures 1 - 3The illustrated embodiment is a low-coherence interferometric measuring machine, comprising a support table 1, a linear motor 11 disposed at the rear of the support table, a lifting plate 12 connected to the telescopic rod of the linear motor, a transverse guide rail 13 disposed on the front side surface of the lifting plate, a slider 14 disposed on the transverse guide rail, a vertical rod 15 disposed on the slider, a measuring probe 16 disposed at the lower end of the vertical rod, a turntable 2 disposed on the support table, a rotary motor disposed on the lower surface of the support table, and a three-jaw chuck 21 disposed in the turntable for fixing the bearing race to be measured; the measuring probe is aligned with the center position of the three-jaw chuck, and the measuring probe can horizontally emit light rays towards the inner side surface of the bearing race 1000 and receive the light rays reflected by the bearing race; the rotating shaft of the rotary motor is connected to the lower part of the turntable, and further comprising a host computer 3, the host computer is electrically connected to the measuring probe and the rotary motor respectively, the host computer is electrically connected to the linear motor through a controller 31, and a plurality of control buttons are provided on the controller.

[0069] The lifting plate is rectangular. The transverse guide rail includes a U-shaped plate 131 with a forward opening, two cross bars 132 disposed on the U-shaped plate, and the sliders are respectively slidably connected to the two cross bars, and the U-shaped plate is fixedly connected to the front surface of the lifting plate.

[0070] A method for detecting the dimensions of the sealing groove of the bearing race of a low-coherence interferometric measuring machine, comprising a step of detecting the width of the sealing groove:

[0071] Detect the three-jaw chuck to check whether its center is aligned with the axis of the measuring probe;

[0072] Take the 50mm standard ring gauge 1100 as shown in Figure 4 as the measurement object to calibrate the low-coherence interferometric measuring machine:

[0073] Place the standard ring gauge in the three-jaw chuck, and take the top surface of the jaws of the three-jaw chuck as the reference surface, and adjust the standard ring gauge to keep it horizontal with the reference surface;

[0074] Manipulate the control buttons to control the downward movement of the measuring probe until the light rays emitted by the measuring probe irradiate the inner ring surface of the standard ring gauge;

[0075] The host computer sets the scanning mode as circular scanning, that is, only the turntable rotates, and sets the scanning speed; the host computer controls the rotary motor to work and starts scanning;

[0076] After the scanning is completed, the host computer generates a calibrated point cloud from the measured point cloud data; the host computer calculates the circle fitted by the calibrated point cloud to obtain the diameter D3 of the circle;

[0077] The host computer calculates:

[0078] ΔD = 50mm - D3

[0079] k = D3 / 50mm.

[0080] After calibration is completed, operate the control button to move the measurement probe to a safe position.

[0081] Remove impurities and stains on the surface of the bearing race to be measured.

[0082] Step 1: Place the bearing race in the three-jaw chuck. Taking the top surface of the jaws of the three-jaw chuck as the reference plane, adjust the bearing race to make it flush with the reference plane.

[0083] Step 2: The operator operates the control button to control the measurement probe to move downward until the light shines 1 mm above the sealing groove of the bearing race. The upper computer records the position 1 of the upper end of the telescopic rod of the linear motor at this time.

[0084] Step 3: The operator operates the control button to control the measurement probe to move downward until the light shines 1 mm below the sealing groove of the bearing race. The upper computer records the position 2 of the upper end of the telescopic rod of the linear motor at this time.

[0085] Step 4: The upper computer sets the scanning range of the low-coherence interferometry measuring machine from position 1 to position 2, and the scanning method is linear scanning, that is, collecting data vertically, and the turntable does not rotate, and sets the scanning speed.

[0086] Step 4-1: The upper computer controls the linear motor to drive the measurement probe to move downward, so that the upper end of the telescopic rod moves from position 1 to position 2. The measurement probe horizontally emits light to the inner side of the bearing race and receives the light reflected from the inner side. The upper computer obtains the detected point cloud data by using the reflected light and generates a point cloud according to the point cloud data.

[0087] Step 4-2: The upper computer uses the ISS point cloud corner detection algorithm to extract the corners of the point cloud.

[0088] Step 4-2-1: Construct a covariance matrix:

[0089] For each point p in the point cloud i4 and its neighboring points both construct the covariance matrix C of the two.

[0090]

[0091] where p j4 ∈R 3 and p i4 is the point coordinates inside, is the centroid of the center point, is the total number of points inside;

[0092] Step 4-2-2: Calculate eigenvalues:

[0093] The host computer performs eigenvalue decomposition on C using the method of constructing a characteristic polynomial to obtain eigenvalues λ 1 , λ 2 , λ 3 , and sets λ 1 ≥λ 2 ≥λ 3 ≥0 and the corresponding eigenvectors;

[0094] Step 63, significance determination:

[0095] The host computer calculates the eigenvalue ratio T 21 and T 32 :

[0096]

[0097] where λ 1 is the largest eigenvalue, corresponding to the variation of the data in the main direction; λ 2 is the second largest eigenvalue, corresponding to the variation of the data in the second main direction; λ 3 is the smallest eigenvalue, corresponding to the variation of the data in the smallest main direction;

[0098] If T 21 or T 32 exceeds the set threshold, then p i4 is considered a point cloud corner point, otherwise, p i4 is not a point cloud corner point.

[0099] The following processing is performed on the sparse point cloud data on the upper and lower sides of the sealing groove 1001:

[0100] Step 11, for each point P i2 , use k-d tree nearest neighbor search to find its nearest neighbor point P j2 (x j2 , y j2 , z j2 );

[0101] Step 12, calculate the weight w j2 according to the distance d j2 from each point to its nearest neighbor point;

[0102]

[0103] where p is a positive number that controls the attenuation rate of the weight, and the weight of the nearest neighbor point can be adjusted by changing the size of p;

[0104] Step 13, calculate the coordinates of the insertion point using the weight of the nearest neighbor point

[0105]

[0106] Among them, m is the number of nearest neighbors;

[0107] Step 14, insert the calculated insertion points into the sparse point cloud data.

[0108] Step 4-3, select the point clouds on the upper and lower sides of the sealing groove 1001, and use the point cloud fitting line algorithm to fit the straight line L1 corresponding to the point cloud on the upper side of the sealing groove and the straight line L2 corresponding to the point cloud on the lower side of the sealing groove.

[0109] The calculation method of L1 is as follows:

[0110] Step 101, select the point cloud corresponding to the upper side of the sealing groove, and set the point-direction equation of L1 corresponding to the point cloud as

[0111]

[0112] Among them, m, n, p are the direction vectors of L1, and (x 0 , y 0 , z 0 ) is any point on the straight line;

[0113] Step 102, convert the point-direction equation into a parametric equation:

[0114]

[0115] Let Then the above formula can be converted into

[0116]

[0117] So far, the six geometric parameters of L1 are converted into four geometric parameters;

[0118] Step 103, solve the target straight line model by the matrix method:

[0119] Suppose represents the approximate value obtained, then the difference between it and the measured values x i5 , y i5 and Q x , Q y is

[0120]

[0121] Based on the least square criterion, to obtain the minimum values of Q x and Q y , take the derivatives of the above formula with respect to a, b, c, d and set them to 0, and we can get:

[0122]

[0123] Let Then the above formula is transformed into:

[0124] MM T A = MX, MM T B = MY

[0125] where A = [a, b] T ; X = [x 1 , …, x t T ; B = [c, d] T ; Y = [y 1 , …, y t T ;

[0126] The host computer solves the above formula to obtain the values of a, b, c, and d, and then obtains L1;

[0127] The calculation method of L2 is as follows:

[0128] Select the point cloud corresponding to the lower side of the seal groove, and repeat steps 101 to 103 to obtain L2.

[0129] Step 4-4, the host computer calculates the distance D1 between L1 and L2, and D1 is the width value of the seal groove at the current angle;

[0130] Step 5, let i1 = 0, take the current scanning position as the starting point, and select the measurement positions at M - 1 angles along the circumferential direction of the bearing ring, where M ≥ 3;

[0131] Increase the value of i1 by 1. If i > M - 1, then go to step 6; otherwise, the host computer controls the turntable to rotate to the position at the i1th angle, and return to step 4-1;

[0132] Step 6, the host computer obtains D1 at M measurement angles, removes the maximum and minimum values among all D1, and calculates the average value of the remaining M - 2 D1. The obtained result is the final width measurement value of the detected seal groove.

[0133] After the width detection of the seal groove is completed, the following depth detection of the seal groove is carried out:

[0134] Step 7-1, the host computer controls the linear motor to drive the measurement probe to move downward, so that the upper end of the telescopic rod moves from position 1 to position 2. The measurement probe horizontally emits light to the inner side of the bearing ring and receives the light reflected from the inner side; the host computer obtains the detected point cloud data using the reflected light and generates a point cloud according to the point cloud data;

[0135] ​​Step 7-2: The host computer extracts the corner points of the point cloud using the ISS point cloud corner detection algorithm;

[0136] Step 7-3: Select the point clouds corresponding to the top surface and the bottom surface of the sealing groove, and use the point cloud fitting line algorithm based on the least squares method to fit the line L3 corresponding to the top surface point cloud of the sealing groove and the line L4 corresponding to the bottom surface point cloud of the sealing groove;

[0137] The calculation methods of L3 and L4 are the same as that of L1.

[0138] Step 7-4: The host computer calculates the distance D2 between L3 and L4, and D2 is the width value of the sealing groove at the current angle;

[0139] Step 7-5: Let i3 = 0, take the current scanning position as the starting point, and select the measurement positions at M - 1 angles along the circumferential direction of the bearing race, where M ≥ 3;

[0140] Increase the value of i3 by 1. If i3 > M - 1, then go to Step 7-6; otherwise, the host computer controls the turntable to rotate to the position at the i3-th angle, and return to Step 7-1;

[0141] Step 7-6: The host computer obtains D2 at M measurement angles, removes the maximum value and the minimum value among all D2, and calculates the average value of the remaining M - 2 D2. The obtained result is the final depth measurement value at the raceway 1002 of the ring shown in Figure 2 、 Figure 3 The width measurement value is corrected using the following formula to obtain the final width measurement value Dfinal:

[0142] Dfinal = Draw1 × k + ΔD

[0143] Draw1 represents the width measurement value.

[0144] The depth measurement value is corrected using the following formula to obtain the final depth measurement value Dfina2:

[0145] Dfina2 = Draw2 × k + ΔD

[0146] Draw2 represents the depth measurement value.

[0147] Draw2 represents the depth measurement value.

[0148] The above are only the preferred embodiments of the present invention, and are not intended to limit the present invention. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principles of the present invention shall be included within the protection scope of the present invention.

Claims

1. A low coherence interferometer measuring machine, characterized in that: The invention comprises a support platform (1), a linear motor (11) arranged at the rear of the support platform, a lifting plate (12) connected to the telescopic rod of the linear motor, a transverse guide rail (13) arranged on the front side of the lifting plate, a slider (14) arranged on the transverse guide rail, a vertical rod (15) arranged on the slider, a measuring probe (16) arranged at the lower end of the vertical rod, a turntable (2) arranged on the support platform, a rotary motor arranged on the lower surface of the support platform, and a three-claw chuck (21) arranged in the turntable for fixing the bearing ring to be measured; the measuring probe is aligned with the center position of the three-claw chuck, and the measuring probe can emit light horizontally to the inner side of the bearing ring and receive light reflected by the bearing ring; the rotating shaft of the rotary motor is connected to the lower part of the turntable, and also comprises an upper computer (3), the upper computer is electrically connected to the measuring probe and the rotary motor respectively, and the upper computer is electrically connected to the linear motor through a controller, and the controller is provided with a plurality of control buttons.

2. The low-coherence interferometer measuring machine according to claim 1, characterized in that: The lifting plate is rectangular, and the transverse guide rail comprises a U-shaped plate (131) opening forward, two cross bars (132) arranged on the U-shaped plate, the sliders are respectively slidably connected to the two cross bars, and the U-shaped plate is fixedly connected to the front surface of the lifting plate.

3. A method for detecting the size of a bearing ring sealing groove based on the low coherence interferometer measuring machine according to claim 1, characterized in that: The width detection steps of the sealing groove include: Step 1: Place the bearing ring to be tested in the three-jaw chuck, use the top surface of the jaws of the three-jaw chuck as the reference surface, and adjust the bearing ring to keep it flush with the reference surface; Step 2: The staff manipulates the control button to control the measuring probe to move downward until the light shines 1 mm above the sealing groove of the bearing ring. The upper computer records the position 1 of the upper end of the telescopic rod of the linear motor at this time. Step 3: The staff manipulates the control button to control the measuring probe to move downward until the light is irradiated to 1 mm below the sealing groove of the bearing ring. The upper computer records the position 2 of the upper end of the telescopic rod of the linear motor at this time. Step 4, the host computer sets the scanning range of the low-coherence interferometer to be from position 1 to position 2, the scanning mode is linear scanning, that is, data is collected vertically, the turntable does not rotate, and the scanning speed is set; Step 4-1, the host computer controls the linear motor to drive the measuring probe to move downward, so that the upper end of the telescopic rod moves from position 1 to position 2, and the measuring probe emits light horizontally to the inner side of the bearing ring and receives the light reflected from the inner side; the host computer uses the reflected light to obtain the detected point cloud data, and generates a point cloud based on the point cloud data; Step 4-2, the host computer uses the ISS point cloud corner detection algorithm to extract the corner points of the point cloud; Step 4-3, selecting the point clouds at the upper side and the lower side of the sealing groove, and fitting the straight line L1 corresponding to the point cloud of the upper side of the sealing groove and the straight line L2 corresponding to the point cloud of the lower side of the sealing groove by using the point cloud straight line fitting algorithm; Step 4-4, the host computer calculates the distance D1 between L1 and L2, where D1 is the width of the sealing groove at the current angle; Step 5, set i1=0, take the current scanning position as the starting point, select M-1 measuring positions along the circumferential direction of the bearing ring, M≥3; Increase the value of i1 by 1. If i>M-1, go to step 6; otherwise, the host computer controls the turntable to rotate to the position of the i1th angle and returns to step 4-1; Step 6: The host computer obtains D1 of M measurement angles, removes the maximum and minimum values ​​of all D1, and calculates the average value of the remaining M-2 D1. The result is the final width measurement value of the sealing groove obtained by detection.

4. The method for detecting the size of the bearing ring sealing groove of the low coherence interferometer measuring machine according to claim 3 is characterized in that: The following steps are also included between step 4-2 and step 4-3: the sparse point cloud data of the upper side and the lower side of the sealing groove are processed as follows: Step 11, for each point P i2 , use the kd tree nearest neighbor to search for its nearest neighbor point P j2 (x j2 ,y j2 ,z j2 ); Step 12: According to the distance d from each point to its nearest neighbor j2 Calculate the weight w j2 ; Among them, p is a positive number that controls the decay rate of the weight. The weight of the neighboring points can be adjusted by changing the size of p; Step 13: Calculate the coordinates of the insertion point using the weights of the nearest neighbor points Where m is the number of nearest neighbor points; Step 14: insert the calculated insertion points into the sparse point cloud data.

5. The method for detecting the size of the bearing ring sealing groove of the low coherence interferometer measuring machine according to claim 3 is characterized in that: After the seal groove width detection is completed, the seal groove depth detection is performed as follows: Step 7-1, the host computer controls the linear motor to drive the measuring probe to move downward, so that the upper end of the telescopic rod moves from position 1 to position 2, and the measuring probe emits light horizontally to the inner side of the bearing ring and receives the light reflected from the inner side; the host computer uses the reflected light to obtain the detected point cloud data, and generates a point cloud based on the point cloud data; Step 7-2, the host computer uses the ISS point cloud corner detection algorithm to extract the corner points of the point cloud; Step 7-3, select the point clouds corresponding to the top and bottom surfaces of the sealing groove, and use the point cloud straight line fitting algorithm based on the least squares method to fit the straight line L3 corresponding to the top surface point cloud of the sealing groove and the straight line L4 corresponding to the bottom surface point cloud of the sealing groove; Step 7-4, the host computer calculates the distance D2 between L3 and L4, where D2 is the width of the sealing groove at the current angle; Step 7-5, set i3=0, take the current scanning position as the starting point, select M-1 measuring positions along the circumferential direction of the bearing ring, M≥3; Increase the value of i3 by 1. If i3>M-1, go to step 7-6; otherwise, the host computer controls the turntable to rotate to the position of the i3th angle and returns to step 7-1; Step 7-6, the host computer obtains D2 of M measurement angles, removes the maximum and minimum values ​​of all D2, and calculates the average value of the remaining M-2 D2. The result is the final depth measurement value of the sealing groove obtained by detection.

6. The method for detecting the size of the bearing ring sealing groove of the low coherence interferometer measuring machine according to claim 3 or 4, characterized in that the steps 4-2 includes the following steps: Step 61, construct the covariance matrix: For every point p in the point cloud i4 and its neighboring points The covariance matrix C of both is constructed; Among them, p j4 ∈R 3 , p i4 yes The coordinates of the points inside, yes The centroid of the center point, yes The total number of inliers; Step 62, eigenvalue calculation: The host computer uses the method of constructing characteristic polynomials to perform eigenvalue decomposition on C, obtains eigenvalues ​​λ1, λ2, λ3, and sets λ1≥λ2≥λ3≥0 and the corresponding eigenvectors; Step 63, significance determination: The host computer calculates the eigenvalue ratio T 21 and T 32 : Among them, λ1 is the maximum eigenvalue, corresponding to the variation of the data in the main direction; λ2 is the secondary eigenvalue, corresponding to the variation of the data in the second main direction; λ3 is the minimum eigenvalue, corresponding to the variation of the data in the minimum main direction; If T 21 or T 32 If it exceeds the set threshold, it is considered that p i4 is the point cloud corner point, otherwise, it is considered that p i4 Not a point cloud corner.

7. The method for detecting the size of the bearing ring sealing groove of the low coherence interferometer measuring machine according to claim 3 is characterized in that: Before step 1, the following calibration steps are also included: Place the standard ring gauge in the three-jaw chuck, take the top surface of the jaws of the three-jaw chuck as the reference surface, and adjust the standard ring gauge to keep it level with the reference surface; The staff controls the measuring probe to move downward through the control button, so that the light emitted by the measuring probe is irradiated to the inner surface of the standard ring gauge, and the light reflected by the inner surface of the standard ring gauge is received; The host computer sets the scanning mode to circular scanning and sets the scanning speed; the host computer controls the rotating motor to work and starts scanning; After the scanning is completed, the host computer generates a calibration point cloud from the measured point cloud data, calculates the circle fitted by the calibration point cloud, and calculates the diameter D3 of the circle.

Images