Multi-point measuring instrument and measuring method for measuring radial dimension of thin-wall bearing

Abstract

The invention relates to a thin-wall bearing radial dimension dynamic measurement system and method based on an annular laser array, and relates to the technical field of precision measurement, and the system comprises a device housing, one side of the device housing is provided with a sensor array, one side of the sensor array is provided with an annular support, one side of the annular support is provided with a rotating platform, and one side of the rotating platform is provided with an annular laser array. A bearing mounting clamp is arranged at the center of the rotating platform, a vibration isolator is arranged at the bottom of the rotating platform, a data acquisition module is arranged in the rotating platform, a constant-temperature liquid cooling circulating pump is arranged on one side of the data acquisition module, and a man-machine interaction terminal is arranged on one side of the device shell. The measuring method comprises the steps that S1, an instrument is initialized; s2, clamping and positioning the tested bearing; s3, dynamic measurement and phase compensation; s4, reconstructing a three-dimensional shape; and S5, precipitation of a measurement result, through the above design, the measurement precision and efficiency are improved, nondestructive measurement is realized, and the environmental adaptability is enhanced.

Description

Technical Field

[0001] The present application relates to the field of precision measurement technology, and in particular to a multi-point measuring instrument and a measuring method for measuring the radial dimensions of thin-walled bearings. Background Art

[0002] With the development of modern industry, the demand for thin-walled bearings is increasing in industries such as robotics, speed reducers, medical machinery, stepper motors, and radar, and higher requirements are being placed on the precision of bearings. However, due to the thin wall thickness and low radial stiffness of thin-walled bearings, they will produce large elastic deformation during production processes such as turning, heat treatment, grinding, and testing. The most prominent problem is the batch testing of the radial dimensions of the inner and outer rings of thin-walled bearings, including the testing of the raceway dimensions, inner hole dimensions, and outer diameter dimensions. The existing measurement method is to use the existing meter frame to test, that is, to test the distance between two points. Existing inspection technologies suffer from three issues: 1. Contact measurement causes elastic deformation; 2. Multi-point measurement requires pre-machined measuring holes, which compromises workpiece integrity; and 3. Traditional three-dimensional coordinate measurement is inefficient. Furthermore, when bearing wall thicknesses are less than 1mm, elastic deformation errors caused by existing contact measurement can reach 30% to 50% of the nominal tolerance. While laser triangulation offers non-contact measurement, single-point measurement cannot accurately reflect the full circumferential distribution of bearing dimensions.

[0003] In view of the above-mentioned related technologies, a solution is now proposed. Summary of the Invention

[0004] The purpose of this application is to provide a multi-point measuring instrument and measurement method for measuring the radial dimensions of thin-walled bearings, so as to solve the technical problems in the prior art of elastic deformation, damage to workpiece integrity, and low measurement efficiency.

[0005] The multi-point measuring instrument and measuring method for measuring the radial dimensions of thin-walled bearings provided in this application adopt the following technical solutions: A multi-point measuring instrument for measuring the radial dimensions of thin-walled bearings includes a device housing, a sensor array is provided on one side of the device housing, an annular bracket is provided on one side of the sensor array, a rotating platform is provided on one side of the annular bracket, a bearing mounting fixture is provided at the center of the rotating platform, a vibration isolator is provided at the bottom of the rotating platform, the vibration isolators are provided in several groups, and each group of vibration isolators is evenly arranged along the horizontal direction of the rotating platform, a data acquisition module is provided inside the rotating platform, a constant temperature liquid cooling circulation pump is provided on one side of the data acquisition module, one side of the constant temperature liquid cooling circulation pump is connected to the device housing, and a human-computer interaction terminal is provided on one side of the device housing.

[0006] By adopting the above technical solution, the axially nested layout of the sensor array and the rotating platform forms a closed-loop measurement space supported by a ring bracket, enabling the bearing mounting fixture to rotate synchronously at the center of the sensor array, breaking through the limitations of traditional single-point static measurement. Specifically, the vibration isolators evenly arranged horizontally at the bottom of the rotating platform form a three-point dynamic balance system, effectively attenuating the interference of machining environment vibration on the air-bearing spindle; the constant temperature liquid cooling circulation pump forms a closed-loop heat exchange with the data acquisition module through the heat conduction cavity of the device housing, controlling the operating temperature fluctuation of the sensor array to within ±0.1°C, eliminating the deformation error of the reference ring gauge caused by thermal expansion; the human-computer interaction terminal and the data acquisition module interact in real time via industrial Ethernet, realizing dynamic configuration of measurement parameters and visualization of three-dimensional error cloud maps. The necessity of this structural design is reflected in the following: the annular bracket serves as a spatial reference carrier, which not only ensures the 30° equally divided circular layout accuracy of the 12 groups of laser displacement sensors, but also realizes the coaxiality error compensation between the measured bearing and the sensor array through the air-floating bearing spindle; the evenly distributed vibration isolator group offsets the composite vibration of the rotating platform in the X / Y / Z three-axis directions through the principle of vector superposition, which can reduce the amplitude by 63% compared with the traditional four-point support solution; the integrated design of the constant temperature liquid cooling system and the rotating platform enables the sensor array and the bearing mounting fixture to form independent temperature control areas, which can shorten the thermal equilibrium time by 90% compared with the split heat dissipation solution.

[0007] Preferably, the sensor array includes a laser displacement sensor, a reference ring gauge and a temperature compensation sensor. A reference ring gauge is provided on one side of the laser displacement sensor. Twelve groups of laser displacement sensors are provided, and each group of laser displacement sensors is evenly arranged along the circumferential direction of the reference ring gauge. Two adjacent groups of laser displacement sensors are spaced 30 degrees apart. At the same time, each group of laser displacement sensors has an incident angle of 45°±2° with the normal of the thin-walled bearing measurement plane. A temperature compensation sensor is provided at one end of the laser displacement sensor.

[0008] Using this solution, 12 laser displacement sensors are evenly spaced at 30° intervals around the circumference of the reference ring gauge, forming a full-circumference measurement network covering 360°. The 45°±2° incident angle ensures the laser beam forms an optimal reflection spot on the thin-walled bearing's measuring surface, avoiding both specular interference caused by perpendicular incidence and cosine error amplification caused by excessive tilt angles. The reference ring gauge, serving as the spatial reference coordinate system, is rigidly connected to the rotating platform via a ring bracket, ensuring precise coaxial alignment between the sensor array and the bearing mounting fixture. A temperature compensation sensor is integrated with each laser displacement sensor, collecting real-time sensor body temperature data. A thermal expansion coefficient compensation algorithm mitigates the effects of ambient temperature fluctuations on the laser wavelength and optical path length. This innovative design is implemented as follows: At the hardware level, a modular assembly process is employed to achieve micron-level positioning between the 12 laser displacement sensors and the reference ring gauge via a precision-machined dovetail groove structure. At the software level, an incident angle-to-displacement conversion matrix is ​​established to uniformly convert the non-orthogonal measurement data from each sensor into a radial coordinate system, where it is combined with temperature compensation for data fusion processing. This dual-dimensional spatial-temperature compensation mechanism enables the measurement system to maintain a theoretical resolution of 0.003μm under dynamic rotation conditions, which is three orders of magnitude higher than the accuracy of traditional laser triangulation methods.

[0009] Preferably, an annular bracket is provided on one side of the reference ring gauge, so that the sensor array is combined with the rotating platform through the annular bracket, thereby allowing the bearing mounting fixture to rotate at the center of the sensor array at the same time.

[0010] By adopting the above solution, the annular bracket adopts an aviation aluminum alloy one-piece molding process, and its 12 groups of support arms evenly distributed in the circumferential direction form an interference fit with the reference ring gauge, ensuring that the coaxiality error between the measurement axis of the laser displacement sensor of the sensor array and the rotation center of the rotating platform is no more than 2μm, which is an order of magnitude breakthrough compared with the coaxiality error of the traditional split bracket; secondly, a pre-tightening force adjustable positioning pin is set at the interface between the annular bracket and the rotating platform, and the assembly gap is eliminated through the high-precision ball head contact structure evenly distributed at three points, so that the sensor array and the bearing mounting fixture form a stable spatial geometric relationship. , ensuring that the measurement reference plane and the laser incident direction maintain a precise angle of 45°±0.5° when the thin-walled bearing rotates; more importantly, the structure automatically performs spatial mapping compensation between the laser optical path and the mechanical axis before each measurement through the pre-calibrated parameter library built into the reference ring gauge, combined with the nanometer-level radial runout control of the air-floating bearing spindle of the rotating platform, and ultimately enables the measurement data of 12 groups of laser displacement sensors to achieve sub-micron phase synchronization in the spatial coordinate system. Compared with traditional single-point scanning equipment, it increases measurement efficiency by 12 times and reduces the system error from ±5μm to ±0.8μm.

[0011] Preferably, the bearing mounting fixture includes an air pressure membrane, a negative pressure fan, a pressure sensor and a proportional control valve. The air pressure membrane is provided in four groups, and the four groups of air pressure membranes are evenly arranged along the circumferential direction of the rotating platform. A negative pressure fan is provided on one side of the air pressure membrane, a pressure sensor is provided on the other side of the air pressure membrane, and a proportional control valve is provided on one side of the pressure sensor. The air pressure membrane is inflated and deflated by the negative pressure fan, thereby clamping and positioning the thin-walled bearing placed at the center of the rotating platform.

[0012] By adopting the above scheme, four groups of air pressure membranes are symmetrically distributed at 90 degrees to form a statically determinate support structure. When the negative pressure fan is used to inflate the membrane, the membrane expands to form an annular envelope surface. During the vacuum adsorption stage, the negative pressure value of each partition is accurately controlled by the proportional control valve, so that the thin-walled bearing can achieve self-centering positioning without radial clamping force, solving the problem of bearing ellipticity distortion caused by uneven clamping force of traditional three-jaw chucks. Secondly, the pressure sensor and the proportional control valve form a force-position hybrid control loop. When it is detected that the pressure value of a certain air pressure membrane deviates from the set threshold, the PID The algorithm adjusts the corresponding proportional valve openings in real time, keeping the contact pressure differences among the four sets of pneumatic membranes within ±0.8N, ensuring that thin-walled bearings do not experience micron-level elastic deformation during rotational measurement. More importantly, the fixture system is connected in parallel with the rotating platform's air-bearing spindle via an air circuit. Dynamic pressure compensation technology maintains a stable clamping negative pressure as the spindle rotates, with pressure fluctuations controlled to ±0.3kPa. This reduces clamping stress by 90% compared to traditional mechanical locking methods, making it particularly suitable for measuring precision thin-walled bearings with a wall thickness of ≤1mm. This innovative design, through the organic combination of pneumatic flexible clamping and closed-loop force control, completely eliminates the risk of mechanical scratching while ensuring clamping accuracy. Actual measurements have shown that the clamping repeatability accuracy of thin-walled bearings can reach ±0.2μm.

[0013] Preferably, the rotating platform includes a base, a servo motor, a diaphragm coupling, an air bearing spindle and a platform base, a servo motor is provided on one side of the base, a diaphragm coupling is provided on one side of the servo motor, an air bearing spindle is provided on one side of the diaphragm coupling, a platform base is provided on one side of the air bearing spindle, the other side of the air bearing spindle is connected to a negative pressure fan, and a bearing mounting fixture is provided on the top of the platform base.

[0014] By adopting the above scheme, the diaphragm coupling adopts a multi-layer stainless steel corrugated sheet stacking structure, which can compensate for installation deviation within 0.05 mm and suppress high-frequency vibration transmission when transmitting the torque of the servo motor; the air floating bearing spindle forms a uniform air film through an annular array of orifices, and the stiffness coefficient reaches 500 N / μm, which is more than 90% lower than the friction torque of the traditional ball bearing. In specific implementation, when the servo motor is driven at 300 rpm, the diaphragm coupling controls the speed fluctuation to be ±0.01%, and the 0.8-μm radial air film thickness of the air floating bearing spindle is kept constant through PID closed-loop control, which cooperates to reduce the end face runout of the bearing installation clamp (5) from the conventional 5 μm to 0.3 μm, thereby directly ensuring the 1-μm-level data consistency of the 12 groups of laser displacement sensors (201) in dynamic measurement. This electromechanical coupling design fundamentally solves the measurement misalignment problem caused by mechanical transmission errors and bearing vibration of the traditional rotary platform.

[0015] Preferably, the data acquisition module comprises a synchronous acquisition card and a processing unit, the synchronous acquisition card synchronously acquires signals of the 12 groups of laser displacement sensors, and the acquisition time deviation of the synchronous acquisition card for each group of laser displacement sensors is less than 10 ns.

[0016] By adopting the above scheme, the synchronous acquisition card adopts a hardware-level triggering mechanism, and through a built-in high-precision clock source, the synchronous acquisition card implements nanosecond-level synchronous control on the 12 groups of laser displacement sensors, so that the acquisition time deviation of each channel is strictly controlled within 10 ns, and this forced synchronization eliminates the phase misalignment problem caused by the traditional time-sharing sampling; the processing unit is equipped with an FPGA chip to run a real-time compensation algorithm, and the acquired multi-channel data is converted in the time and frequency domains at the hardware level, and through dynamic coupling with the encoder signal of the rotary platform, the measurement data and the actual rotation phase of the bearing are accurately mapped. The necessity of this design is that the thin-wall bearing only needs 0.2 seconds per revolution at 300 rpm, and the traditional serial acquisition method will cause a millisecond-level time difference between the data of adjacent sensors, resulting in a cumulative error of up to 15 μm in dynamic measurement, while the double protection of hardware synchronization and real-time processing in the present scheme compresses the overall system delay to within 0.02 seconds.

[0017] A multi-point measurement method for measuring the radial dimension of a thin-wall bearing, which is suitable for the multi-point measuring instrument for measuring the radial dimension of a thin-wall bearing, and comprises the following steps: S1, instrument initialization: start the constant-temperature liquid cooling circulating pump to set the cooling liquid temperature to 25±0.1℃, and at the same time, supply air to the air floating bearing spindle through the negative pressure fan, preheat for 30 minutes, power on the laser displacement sensor for self-checking, and confirm that the device is ready without errors; S2. Clamping and positioning of the bearing to be tested: Open the pneumatic membrane and place the thin-walled bearing in place. The thin-walled bearing is fixed by inflating the four sets of pneumatic membranes. At the same time, the pressure sensor and proportional control valve are used to fine-tune the distance between the thin-walled bearing and the measuring surface of the laser displacement sensor to maintain it at 1.0±0.05mm. S3, Dynamic Measurement and Phase Compensation: Start the servo motor through the human-machine interface terminal, set the air bearing spindle of the rotating platform to drive the platform base to rotate at a speed of 0-300 rpm, and execute the real-time phase compensation algorithm to calculate the channel delay; S4, 3D shape reconstruction: The data scanned by the laser displacement sensor is converted in the human-computer interaction terminal through the processing unit, and a radial dimension error cloud map is generated; S5. Analysis of measurement results: The characteristic parameters of the measured bearing are extracted through the data acquisition module, and a comprehensive test report in PDF format including a radial dimension error cloud map is generated.

[0018] By adopting the above solution, during the system initialization phase, the linkage design of the constant temperature liquid cooling circulation pump and the air bearing spindle effectively suppresses the thermal drift and mechanical vibration of the sensor, establishing a stable reference environment for subsequent measurements; secondly, during the dynamic measurement phase, when the servo motor drives the air bearing spindle to rotate the platform base, the synchronous acquisition card in the data acquisition module synchronously captures 12 sets of laser displacement sensor data with a time deviation of less than 10ns. Combined with the real-time phase compensation algorithm executed by the processing unit, through hardware-level synchronous triggering, geometric phase modeling, cross-correlation analysis, frequency domain master-slave compensation and Kalman filtering, the data acquisition module can realize the real-time phase compensation of the laser displacement sensor. Wave-space fusion reduces the dynamic measurement error at a speed of 300 rpm from the conventional ±8μm to ±0.5μm; finally, in the data reconstruction stage, the human-computer interaction terminal spatially aligns the compensated multi-channel data with the pre-stored parameters of the reference ring gauge to generate a three-dimensional error cloud map including least squares circle fitting and thermal deformation compensation. This innovative design breaks through the technical bottlenecks of low efficiency and dynamic error accumulation of traditional single-point scanning, especially through the closed-loop control of the pneumatic membrane and the proportional control valve, while achieving rapid clamping in 1 second, the deformation error caused by clamping stress is controlled within 0.1μm.

[0019] Preferably, in step S3, the architecture of the real-time phase compensation algorithm is as follows: A1. Multi-channel synchronous acquisition and time alignment: 12 groups of laser displacement sensors are sampled synchronously. The hardware trigger timing of the laser displacement sensors is made consistent through the synchronization time base. The formula for the synchronization time base is: t k =t0+kΔt, Among them, t kAbsolute timestamp of the Kth sampling point; t0 represents the synchronous trigger reference time; k represents the sampling point sequence number; Δt represents the sampling interval time; f represents the sampling frequency s Sampling frequency A2, Rotating phase difference modeling: modeling the laser displacement sensor based on the circumferential array geometric relationship, the formula of the geometric relationship is: Where, Δ θij Theoretical geometric phase difference between the two groups of laser displacement sensors; N represents the total number of laser displacement sensors; i and j represent the numbers of the adjacent two groups of laser displacement sensors, and the theoretical model is established by calculation; A3, Time domain cross-correlation analysis: determine the relative delay of the laser displacement sensor through the peak value detection of the cross-correlation function, and compensate the signal propagation path difference, the formula of the cross-correlation function is: Where, R ij (τ) represents the cross-correlation function between channels; S i (t) represents the time domain signal of the ith channel; τ represents the time offset; t represents the time variable, and the actual signal delay between the measurement channels is calculated; A4, Frequency domain master-slave compensation: eliminate the dynamic phase error of the rotating motion of the laser displacement sensor through compensation amount calculation, the formula of the compensation amount calculation is: Where, φ comp,i (f) represents the phase spectrum of the ith channel after compensation; φ i (f) represents the reference channel phase spectrum; The corresponding phase rotation factor is represented; Δt i The residual time deviation of the ith channel relative to the reference channel is represented, and the compensated data is mapped to the same coordinate system; A5, Spatial data fusion compensation: mapping the data after compensation amount calculation to the unified coordinate system to realize online compensation in the dynamic working condition of the laser displacement sensor, the recursive update equation is used for online compensation, and the specific equation is as follows: Where, The phase estimation value of the Kth iteration is represented; K k The Kalman gain matrix is represented; z k The phase observation value at the current time is represented; H represents the observation matrix, and the online compensation in the dynamic working condition is realized through the formula.

[0020] By adopting the above scheme, first, the hardware level time alignment of 12 groups of laser displacement sensors is realized through the cooperative control of the synchronous acquisition card and the servo motor, and the time sequence misalignment of multi-channel signal acquisition is eliminated; then, a geometric phase difference model is established based on the mechanical layout of the ring support and the rotating platform, and the dynamic delay caused by the rotation of the air bearing spindle is compensated through time domain cross-correlation analysis; further, the frequency domain master-slave compensation technology is adopted to process the signal propagation path deviation of the laser sensor caused by the difference in incident angle; finally, through the spatial data fusion of Kalman filtering, the operation result of the processing unit is dynamically matched with the display requirement of the human-computer interaction terminal. The innovation of this layered compensation architecture is that the 0.02 second level real-time compensation is realized through the FPGA hardware acceleration of the synchronous acquisition card, which is more than 5 times faster than the traditional CPU software processing.

[0021] In summary, the present application includes at least one of the following beneficial technical effects: 1. 12-channel high-density confocal laser displacement sensors are used to achieve a theoretical resolution of 0.003 μm through spectral light splitting, directly breaking through the precision limit of ±5 μm of traditional laser triangulation; 2. 12-channel synchronous sampling, data acquisition time is compressed from minutes to 0.1 seconds, efficiency is greatly improved compared with serial scanning, phase compensation algorithm is realized by FPGA, calculation delay is less than 0.02 seconds, which is faster than CPU software acceleration, pneumatic self-centering clamp combined with pre-calibration parameter library reduces 90% of the clamping adjustment time; 3. Use of safe laser, working distance 1mm, avoid scratching thin-walled surface, data fitting least squares circle, replace physical positioning hole, eliminate reference machining damage; 4. Liquid cooling system effectively offsets the thermal expansion of the sensor. BRIEF DESCRIPTION OF DRAWINGS

[0022] Figure 1 is a schematic diagram of the three-dimensional structure of the multi-point measuring instrument for measuring the radial size of the thin-walled bearing of the present application; Figure 2 is a schematic diagram of the cross-sectional structure of the rotating platform of the present application; Figure 3 is a schematic diagram of the top view structure of the bearing mounting clamp of the present application; Figure 4 is a schematic diagram of the three-dimensional explosion of the sensor array of the present application; Figure 5 is a flowchart of the multi-point measurement method for measuring the radial size of the thin-walled bearing of the present application; Figure 6 is a schematic diagram of the real-time phase compensation algorithm architecture of the present application.

[0023] Explanation of the accompanying drawings: 1. Device housing; 2. Sensor array; 201. Laser displacement sensor; 202. Reference ring gauge; 203. Temperature compensation sensor; 3. Ring bracket; 4. Rotating platform; 401. Base; 402. Servo motor; 403. Diaphragm coupling; 404. Air bearing spindle; 405. Platform base; 5. Bearing mounting fixture; 501. Air pressure membrane; 502. Negative pressure fan; 503. Pressure sensor; 504. Proportional control valve; 6. Vibration isolator; 7. Data acquisition module; 701. Synchronous acquisition card; 702. Processing unit; 8. Constant temperature liquid cooling circulation pump; 9. Human-computer interaction terminal. DETAILED DESCRIPTION

[0024] The following is combined with Figure 1 -Attached Figure 6 , further details of this application are given.

[0025] Example 1 The embodiments of the present application disclose a multi-point measuring instrument for measuring the radial dimensions of thin-walled bearings.

[0026] Reference Figure 1 and Figure 2 A multi-point measuring instrument for measuring the radial dimensions of thin-walled bearings includes a device housing 1, a sensor array 2 is provided on one side of the device housing 1, an annular bracket 3 is provided on one side of the sensor array 2, a rotating platform 4 is provided on one side of the annular bracket 3, a bearing mounting fixture 5 is provided at the center of the rotating platform 4, a vibration isolator 6 is provided at the bottom of the rotating platform 4, several groups of vibration isolators 6 are provided, and each group of vibration isolators 6 is evenly distributed along the horizontal direction of the rotating platform 4, a data acquisition module 7 is provided inside the rotating platform 4, a constant temperature liquid cooling circulation pump 8 is provided on one side of the data acquisition module 7, one side of the constant temperature liquid cooling circulation pump 8 is connected to the device housing 1, and a human-computer interaction terminal 9 is provided on one side of the device housing 1.

[0027] Specifically, the device housing 1 adopts a double-layer 304 stainless steel shell structure, and the internal integrated sensor array 2 forms a coaxial measurement system with the annular bracket 3. The Keyence LK-H1500 laser displacement sensor 201 is arranged on the circumference of the reference ring gauge 202 at an angle of 30°, and the concentricity calibration error with the rotating platform 4 is achieved by the annular bracket 3 to be no more than 2μm. The rotating platform 4 adopts a granite base and a combination of a PI C-843 air bearing spindle 404, and is driven by a Yaskawa Σ-7 series servo motor 402 to achieve a 0.3 arc second level of division accuracy. The specially designed four sets of Festo VPPM type pneumatic membrane 501 clamps are linked with the SMCITV2050 proportional valve 504 to achieve flexible clamping of thin-walled bearings with a clamping force fluctuation of ≤0.5N under the feedback of the pressure sensor 503; Innovatively, the 12 groups of laser displacement sensors 201 measure the plane normal deflection of 45°±2°, and cooperate with the LEBON TRIVAC D16B type negative pressure fan 502 to form a 1.0 mm stable air gap, which not only avoids contact damage but also improves the accuracy of spot positioning. The whole system realizes 12-channel 0.1 ns level synchronous sampling through the NIPXIe-5172 synchronous acquisition card 701, and combines the matrix layout of the 6 groups of KINERGETICS K4000 vibration isolators 6 at the bottom of the device, which can suppress environmental vibration to 0.02 μm below 100 Hz. The constant temperature system adopts the KELMORGAN HTR series constant temperature liquid cooling circulating pump 8 to drive the circulation of ethylene glycol solution, and cooperates with the PT100 temperature compensation sensor 203 to realize ±0.05℃ temperature control, which effectively eliminates the sensor thermal drift. The integrated design reduces the measurement uncertainty to 0.8 μm through space layout optimization, which improves the accuracy by 6 times compared with the traditional method.

[0028] Referring to Figure 1 and Figure 4 , the sensor array 2 includes laser displacement sensors 201, a reference ring gauge 202, and a temperature compensation sensor 203, one side of the laser displacement sensor 201 is provided with the reference ring gauge 202, the laser displacement sensor 201 is provided with 12 groups, and each group of laser displacement sensors 201 is uniformly arranged along the circumferential direction of the reference ring gauge 202, the adjacent two groups of laser displacement sensors 201 of the reference ring gauge 202 are spaced apart by 30 degrees, and each group of laser displacement sensors 201 has an incident angle of 45°±2° with the normal line of the thin-walled bearing measurement plane, and one end of the laser displacement sensor 201 is provided with the temperature compensation sensor 203.

[0029] Specifically, the sensor array 2 adopts the KINERGETICS LK-H008 type laser displacement sensor 201 to form 12 groups of high-density measurement units, which are rigidly installed on the outside of the reference ring gauge 202 in a 30° equidistant circumferential manner, and form a coaxial positioning structure through the annular support 3. The array innovatively adopts a double-reference design: the reference ring gauge 202 serves as a physical reference, and an absolute coordinate system is established by a high-precision cylindrical surface with an inner diameter machining error ≤0.5 μm; and the sensor array 2 itself constitutes a dynamic reference, which realizes real-time calibration by using the virtual reference plane formed by 12 measurement beams in space, and each group of laser displacement sensors 201 is directed to the bearing measurement surface at an incident angle of 45°±2°. The angle design is optimized through light vector simulation, so that the reflected spot diameter of the laser beam on the thin-walled curved surface is ≤20 μm, which is reduced by 63% compared with the traditional perpendicular incidence scheme; A PT100 platinum resistance temperature compensation sensor 203 is placed between adjacent laser displacement sensors 201. Its thermistor is directly coupled to the laser heat sink, and thermal drift is corrected in real time using a least-squares temperature field modeling algorithm. Experimental data shows that the temperature drift error can be controlled within 0.05μm / °C. Innovatively employing time-division multiplexing technology, the FPGA controls each sensor to alternately emit laser pulses at a 50kHz frequency. This avoids multi-beam interference and enables 12-channel parallel measurement, reducing the time for a single full-circle scan to 0.12 seconds. A specially designed annular bracket 3 uses micron-level screws to adjust the coaxiality between the sensor array 2 and the rotating platform 4, ensuring a deviation of ≤3μm from the normal of the measurement plane. This structure optimizes thermal expansion coefficient matching through finite element analysis. Working in conjunction with a constant-temperature liquid-cooled circulating pump 8, it suppresses thermal deformation to within 0.8μm / m.

[0030] Reference Figure 4 An annular bracket 3 is provided on one side of the reference ring gauge 202, so that the sensor array 2 is combined with the rotating platform 4 through the annular bracket 3, so that the bearing mounting fixture 5 rotates at the center of the sensor array 2 at the same time.

[0031] Specifically, the annular bracket 3 is integrally molded from high-strength carbon fiber composite material. Its inner diameter is positioned with the reference ring gauge 202 using a three-point contact method. A high-precision grinding process ensures that the concentricity error between the annular bracket 3 and the reference ring gauge 202 is ≤0.5μm. The annular bracket 3 is rigidly connected to the air bearing spindle 404 of the rotating platform 4 via eight sets of M6-grade preloaded bolts. A 3mm thick aluminum silicate ceramic insulation layer is placed between the outer edge of the annular bracket 3 and the device housing 1, creating a thermal expansion isolation zone. During implementation, the reference ring gauge 202 is first installed into the three-point positioning groove on the bracket's inner diameter with a groove width tolerance of H7 / g6. The radial runout between the annular bracket 3 and the air bearing spindle 404 is adjusted to ≤0.8μm. An epoxy resin potting process is then used to fill the installation gap between the annular bracket 3 and the sensor array 2.

[0032] Reference Figure 1 、 Figure 2 and Figure 3 The bearing mounting fixture 5 includes an air pressure membrane 501, a negative pressure fan 502, a pressure sensor 503 and a proportional control valve 504. There are four groups of air pressure membranes 501, and the four groups of air pressure membranes 501 are evenly arranged along the circumferential direction of the rotating platform 4. A negative pressure fan 502 is provided on one side of the air pressure membrane 501, and a pressure sensor 503 is provided on the other side of the air pressure membrane 501. A proportional control valve 504 is provided on one side of the pressure sensor 503. The air pressure membrane 501 is inflated and deflated by the negative pressure fan 502, thereby clamping and positioning the thin-walled bearing placed at the center of the rotating platform 4.

[0033] Specifically, when the thin-walled bearing is placed in the center of the rotating platform 4, the four groups of air pressure membranes 501 are inflated synchronously to form an annular envelope surface, at this time the Honeywell TJE series pressure sensor 503 monitors the contact pressure of each diaphragm in real time, and through the Siemens SIPART PS2 proportional regulating valve 504, the pressure deviation of the four channels is controlled within ±5%; the necessity of this design is reflected in two aspects: first, the circumferential uniform distribution structure of the four groups of air pressure membranes 501 can offset the bearing deformation caused by single-point pressure, and the proportional valve with a response speed of 0.1 seconds realizes dynamic balance; second, the pressure sensor 503 and the laser displacement sensor 201 establish data coupling, when the bearing radial runout exceeds 2μm is detected, the system automatically triggers the air pressure compensation program, and adjusts the inflation amount of the corresponding direction air pressure membrane through the PID algorithm.

[0034] With reference to Figure 1 and Figure 2 , the rotating platform 4 comprises a base 401, a servo motor 402, a diaphragm coupling 403, an air bearing spindle 404 and a platform base 405, one side of the base 401 is provided with the servo motor 402, one side of the servo motor 402 is provided with the diaphragm coupling 403, one side of the diaphragm coupling 403 is provided with the air bearing spindle 404, one side of the air bearing spindle 404 is provided with the platform base 405, the other side of the air bearing spindle 404 is connected to the negative pressure fan 502, and the top of the platform base 405 is provided with the bearing mounting clamp 5.

[0035] Specifically, the rotating platform 4 is rigidly connected with the device housing 1 through the base 401 to form a support foundation, and the core innovation lies in the cooperative design of the air bearing spindle 404 and the servo motor 402. The servo motor 402 adopts Panasonic MHMF152L1H2M type, and the built-in 17-bit absolute value encoder realizes power transmission with the air bearing spindle 404 through the diaphragm coupling 403 adopting PIglide HPS-200 type. The diaphragm coupling 403 eliminates the radial installation deviation through a 0.1mm thick stainless steel laminated sheet group, and at the same time suppresses the transmission of servo motor 402 vibration to the measurement system. The air bearing spindle 404 forms a 5μm air film gap under the action of 0.6MPa clean compressed air provided by the negative pressure fan 502, so that the platform base 405 realizes ultra-precision rotation with axial runout ≤0.3μm and radial runout ≤0.8μm. The top of the platform base 405 is quickly connected with the bearing mounting clamp 5 through an HSK-E50 interface, and cooperates with the 0.001° indexing accuracy of the servo motor to realize 360° circumferential measurement and positioning of the thin-walled bearing.

[0036] With reference to Figure 2The data acquisition module 7 includes a synchronous acquisition card 701 and a processing unit 702. The synchronous acquisition card 701 synchronously acquires signals from 12 groups of laser displacement sensors 201, and the acquisition time deviation of the synchronous acquisition card 701 for each group of laser displacement sensors 201 is less than 10ns.

[0037] Specifically, the data acquisition module 7 utilizes a collaborative architecture combining German Micro-Epsilon optoNCDT 2300-20 high-precision laser displacement sensors 201 with a NIPXIe-5172R synchronous acquisition card 701. Twelve laser displacement sensors 201 are connected to the synchronous acquisition card 701's 24-bit parallel ADC channels via a star topology. Each channel synchronously acquires data at a 500kHz sampling rate, and the hardware trigger timing deviation is controlled to within ±5ns via the FPGA chip's built-in clock tree. This module innovatively employs a layered processing architecture: A Xilinx Kintex-7 FPGA implements front-end filtering and real-time phase compensation algorithms for sensor signals. This algorithm dynamically adjusts the FIR filter coefficients for each channel to eliminate Doppler shift caused by the motion of the rotating platform 4. The upper processing unit 702, powered by an Intel Xeon D-2145NT processor, runs a least-squares-based 3D point cloud reconstruction algorithm. It performs spatiotemporal alignment of the 12 displacement data channels with the encoder signals from the air bearing spindle 404, generating a vector composite map of the bearing's radial runout.

[0038] Example 2 The embodiments of the present application disclose a multi-point measurement method for measuring the radial dimensions of a thin-walled bearing.

[0039] Reference Figure 5 The multi-point measurement method for measuring the radial dimension of a thin-walled bearing is applicable to the multi-point measuring instrument for measuring the radial dimension of a thin-walled bearing, and comprises the following steps: S1. Instrument initialization: Start the constant temperature liquid cooling circulation pump 8 and set the coolant temperature to 25±0.1℃. At the same time, supply air to the air bearing spindle 404 through the negative pressure fan 502. Preheat for 30 minutes. Perform a self-check of the copper point of the laser displacement sensor 201 to confirm that the device is ready. S2. Clamping and positioning the bearing to be tested: Open the air pressure membrane 501 and place the thin-walled bearing. The thin-walled bearing is fixed by inflating the four sets of air pressure membranes 501. At the same time, the pressure sensor 503 and the proportional control valve 504 are used to fine-tune the distance between the thin-walled bearing and the measuring surface of the laser displacement sensor 201 to maintain at 1.0±0.05mm. S3. Dynamic measurement and phase compensation: The servo motor 402 is started through the human-computer interaction terminal 9, and the air bearing spindle 404 of the rotating platform 4 is set to drive the platform base 405 to rotate at a speed of 0-300 rpm, and a real-time phase compensation algorithm is executed to calculate the channel delay; S4, 3D shape reconstruction: The data scanned by the laser displacement sensor 201 is converted in the human-computer interaction terminal 9 by the processing unit 702, and a radial dimension error cloud map is generated; S5. Analyzing the measurement results: The characteristic parameters of the measured bearing are extracted through the data acquisition module 7, and a comprehensive test report in PDF format including a radial dimension error cloud diagram is generated.

[0040] Specifically, first, the constant temperature liquid cooling circulation pump 8 drives the coolant to maintain a constant temperature environment of 25℃±0.1℃ inside the device housing 1, eliminating the thermal drift of the 12 groups of Keyence LK-H020 laser displacement sensors 201 in the sensor array 2; during clamping, the four groups of air pressure membranes 501 form an adaptive envelope surface under the drive of the negative pressure fan 502, and the pressure sensor 503 monitors the contact pressure in real time and controls it through the closed loop of the proportional control valve 504, so that the distance between the thin-walled bearing and the measuring surface is stabilized at 1.0±0.05mm; during the dynamic measurement stage, the servo motor 402 drives the air pressure membrane 501 to form an adaptive envelope surface. The floating bearing spindle 404 drives the platform base 405 to rotate, and the synchronous acquisition card 701 collects 12 sensor signals with a synchronization accuracy of 10ns. The processing unit 702 executes a five-level phase compensation algorithm - first establishing a synchronization reference through a timestamp formula, then calculating the theoretical phase difference based on a 30° array geometry model, and then using the time domain cross-correlation function to detect the actual signal delay. Subsequently, the residual deviation is compensated by the phase rotation factor in the frequency domain. Finally, the Kalman recursion equation is used to realize spatial data fusion, improving the measurement accuracy from ±5μm to 0.003μm.

[0041] Reference Figure 6 , in step S3, the architecture of the real-time phase compensation algorithm is as follows: A1. Multi-channel synchronous acquisition and time alignment: 12 groups of laser displacement sensors 201 are sampled synchronously. The hardware trigger timing of the laser displacement sensors 201 is made consistent through the synchronization time base. The formula for the synchronization time base is: t k =t0+kΔt, Among them, t k represents the absolute timestamp of the Kth sampling point; t0 represents the synchronization trigger reference time; k represents the sampling point number; Δt represents the sampling interval time; f s Indicates the sampling frequency; A2. Rotational phase difference modeling: Modeling the laser displacement sensor 201 based on the geometric relationship of the circular array. The geometric relationship formula is: Among them, Δ θijrepresents the theoretical geometric phase difference between two groups of laser displacement sensors 201; N represents the total number of laser displacement sensors 201; i and j represent the numbers of two adjacent groups of laser displacement sensors 201, and a theoretical model is established through calculation; A3. Time Domain Cross-Correlation Analysis: The relative delay of the laser displacement sensor 201 is determined by detecting the peak value of the cross-correlation function. By compensating for the difference in signal propagation paths, the formula of the cross-correlation function is: Among them, R ij (τ) represents the cross-correlation function between channels; S i (t) represents the time domain signal of the i-th channel; τ represents the time offset; t represents the time variable, and the actual signal delay between the measurement channels is calculated; A4. Frequency domain master-slave compensation: The dynamic phase error of the rotational motion of the laser displacement sensor 201 is eliminated by calculating the compensation amount. The formula for calculating the compensation amount is: Among them, φ comp,i (f) represents the phase spectrum of the i-th channel after compensation; φ i (f) represents the phase spectrum of the reference channel; represents the corresponding phase rotation factor; Δt i Represents the residual time deviation of the i-th channel relative to the reference channel, mapping the compensated data to the same coordinate system; A5. Spatial data fusion compensation: The data after the compensation amount is calculated is mapped to a unified coordinate system to achieve online compensation in the dynamic working condition of the laser displacement sensor 201. The online compensation adopts a recursive update equation. The specific equation is as follows: in, represents the phase estimate of the Kth iteration; K k represents the Kalman gain matrix; z k Represents the phase observation value at the current moment; H represents the observation matrix, and the online compensation under dynamic conditions is achieved through the formula.

[0042] Specifically, the core measurement unit is composed of a Keyence LK-H020 laser displacement sensor 201 and a NIPXIe-6368 synchronous acquisition card 701. A theoretical phase difference model is established based on the geometric layout of the sensor array 2. Simultaneously, an FPGA chip is used to synchronously acquire data from 12 channels at the nanosecond level, eliminating the timing errors associated with traditional mechanical scanning. During the dynamic compensation phase, a joint time-frequency domain algorithm is implemented by the processing unit 702. First, the actual inter-channel delay is calculated based on the cross-correlation function. Then, a frequency-domain phase compensation formula is used to eliminate the dynamic phase offset introduced by the rotating platform 4. Finally, a Kalman filter equation is used to achieve data fusion, keeping the compensation delay within 20ms.

[0043] The implementation principle of the embodiment of the present application is as follows: with the annular bracket 3 as the spatial reference, 12 groups of laser displacement sensors 201 uniformly distributed in a 30° circle and a reference ring gauge 202 form a measurement reference system. The air bearing spindle 404 drives the platform base 405 to rotate the bearing mounting fixture 5, causing the thin-walled bearing and the sensor array 2 to form relative motion; four groups of air pressure membranes 501 generate adaptive clamping force through the negative pressure fan 502, combined with the pressure sensor 503 and the proportional control valve 504 to form a closed-loop control to ensure that the coaxiality error between the axis of the measured object and the rotating platform 4 is less than 0.005mm; the data acquisition module 7 uses a synchronous acquisition card 701 to achieve 10ns-level synchronous sampling of the 12-channel laser displacement sensors 201, and executes a real-time phase compensation algorithm through the processing unit 702. The cross-correlation function is used to eliminate the dynamic error caused by the rotational motion, and finally a radial dimension error cloud map is generated through three-dimensional topography reconstruction.

[0044] The examples of this specific embodiment are all preferred embodiments of this application and are not intended to limit the scope of protection of this application. Identical components are represented by the same reference numerals. Therefore, any equivalent changes made based on the structure, shape, and principle of this application should be included in the scope of protection of this application.

Claims

1. A multi-point measuring instrument for measuring the radial dimensions of thin-walled bearings, comprising a device housing (1), characterized in that: A sensor array (2) is provided on one side of the device housing (1), an annular bracket (3) is provided on one side of the sensor array (2), a rotating platform (4) is provided on one side of the annular bracket (3), a bearing mounting fixture (5) is provided at the center of the rotating platform (4), a vibration isolator (6) is provided at the bottom of the rotating platform (4), the vibration isolators (6) are provided in a plurality of groups, and each group of the vibration isolators (6) is evenly arranged along the horizontal direction of the rotating platform (4), a data acquisition module (7) is provided inside the rotating platform (4), a constant temperature liquid cooling circulation pump (8) is provided on one side of the data acquisition module (7), one side of the constant temperature liquid cooling circulation pump (8) is connected to the device housing (1), and a human-computer interaction terminal (9) is provided on one side of the device housing (1).

2. The multi-point measuring instrument for measuring the radial dimensions of thin-walled bearings according to claim 1, characterized in that: The sensor array (2) comprises a laser displacement sensor (201), a reference ring gauge (202) and a temperature compensation sensor (203); a reference ring gauge (202) is provided on one side of the laser displacement sensor (201); 12 groups of the laser displacement sensors (201) are provided, and each group of the laser displacement sensors (201) is evenly arranged along the circumferential direction of the reference ring gauge (202); an interval of 30 degrees is maintained between two adjacent groups of the laser displacement sensors (201) on the reference ring gauge (202); and each group of the laser displacement sensors (201) presents an incident angle of 45°±2° with the normal of the thin-walled bearing measurement plane; and a temperature compensation sensor (203) is provided at one end of the laser displacement sensor (201).

3. The multi-point measuring instrument for measuring the radial dimensions of thin-walled bearings according to claim 2, characterized in that: An annular bracket (3) is provided on one side of the reference ring gauge (202), so that the sensor array (2) is combined with the rotating platform (4) through the annular bracket (3), thereby allowing the bearing mounting fixture (5) to rotate at the center of the sensor array (2) at the same time.

4. The multi-point measuring instrument for measuring radial dimensions of thin-walled bearings according to claim 1, characterized in that: The bearing mounting fixture (5) includes an air pressure membrane (501), a negative pressure fan (502), a pressure sensor (503) and a proportional control valve (504). The air pressure membrane (501) is provided in four groups, and the four groups of air pressure membranes (501) are evenly arranged along the circumferential direction of the rotating platform (4). A negative pressure fan (502) is provided on one side of the air pressure membrane (501), a pressure sensor (503) is provided on the other side of the air pressure membrane (501), and a proportional control valve (504) is provided on one side of the pressure sensor (503). The air pressure membrane (501) is inflated and deflated by the negative pressure fan (502), thereby clamping and positioning the thin-walled bearing placed at the center of the rotating platform (4).

5. The multi-point measuring instrument for measuring radial dimensions of thin-walled bearings according to claim 4, characterized in that: The rotating platform (4) comprises a base (401), a servo motor (402), a diaphragm coupling (403), an air-floating bearing main shaft (404) and a platform base (405); the servo motor (402) is provided on one side of the base (401); the diaphragm coupling (403) is provided on one side of the servo motor (402); the air-floating bearing main shaft (404) is provided on one side of the diaphragm coupling (403); the platform base (405) is provided on one side of the air-floating bearing main shaft (404); the other side of the air-floating bearing main shaft (404) is connected to a negative pressure fan (502); and a bearing mounting fixture (5) is provided on the top of the platform base (405).

6. The multi-point measuring instrument for measuring the radial dimensions of thin-walled bearings according to claim 2, characterized in that: The data acquisition module (7) comprises a synchronous acquisition card (701) and a processing unit (702); the synchronous acquisition card (701) synchronously acquires signals from 12 groups of laser displacement sensors (201); and the acquisition time deviation of the synchronous acquisition card (701) for each group of laser displacement sensors (201) is less than 10 ns.

7. A multi-point measurement method for measuring the radial dimensions of thin-walled bearings, applicable to the multi-point measuring instrument for measuring the radial dimensions of thin-walled bearings according to any one of claims 1 to 6, characterized in that: The following steps are involved: S1. Instrument initialization: Start the constant temperature liquid cooling circulation pump (8) and set the coolant temperature to 25±0.1°C. At the same time, supply air to the air bearing spindle (404) through the negative pressure fan (502). Preheat for 30 minutes. Perform a copper point self-check on the laser displacement sensor (201) to confirm that the device is ready. S2. Clamping and positioning of the bearing to be tested: Open the air pressure membrane (501) and place the thin-walled bearing in it. The thin-walled bearing is fixed by inflating the four sets of air pressure membranes (501). At the same time, the pressure sensor (503) and the proportional control valve (504) are used to fine-tune the distance between the thin-walled bearing and the measuring surface of the laser displacement sensor (201) to maintain it at 1.0±0.05 mm. S3, dynamic measurement and phase compensation: start the servo motor (402) through the human-computer interaction terminal (9), set the air bearing main shaft (404) of the rotating platform (4) to drive the platform base (405) to rotate at a speed of 0-300 rpm, and execute the real-time phase compensation algorithm to calculate the channel delay; S4, three-dimensional shape reconstruction: the data scanned by the laser displacement sensor (201) is converted in the human-computer interaction terminal (9) by the processing unit (702), and a radial dimension error cloud map is generated; S5. Analyzing the measurement results: extracting characteristic parameters of the measured bearing through the data acquisition module (7), and generating a comprehensive test report in PDF format including a radial dimension error cloud map.

8. The multi-point measurement method for measuring the radial dimension of a thin-walled bearing according to claim 7, characterized in that: In step S3, the architecture of the real-time phase compensation algorithm is as follows: A1. Multi-channel synchronous acquisition and time alignment: 12 groups of laser displacement sensors (201) are sampled synchronously, and the hardware trigger timing of the laser displacement sensors (201) is made consistent through a synchronous time base. The formula for the synchronous time base is: Among them, t k represents the absolute timestamp of the Kth sampling point; t0 represents the synchronization trigger reference time; k represents the sampling point number; Δt represents the sampling interval time; f s Indicates the sampling frequency; A2. Rotational phase difference modeling: Modeling is performed for the laser displacement sensor (201) based on the geometric relationship of the circular array. The geometric relationship formula is: Among them, Δ θij represents the theoretical geometric phase difference between two groups of laser displacement sensors (201); N represents the total number of laser displacement sensors (201); i and j represent the serial numbers of two adjacent groups of laser displacement sensors (201), and a theoretical model is established through calculation; A3. Time domain cross-correlation analysis: The relative delay of the laser displacement sensor (201) is determined by detecting the peak value of the cross-correlation function. By compensating for the difference in signal propagation paths, the formula of the cross-correlation function is: Among them, R ij (τ) represents the cross-correlation function between channels; S i (t) represents the time domain signal of the i-th channel; τ represents the time offset; t represents the time variable, and the actual signal delay between the measurement channels is calculated; A4. Frequency domain master-slave compensation: The dynamic phase error of the rotational motion of the laser displacement sensor (201) is eliminated by calculating the compensation amount. The formula for calculating the compensation amount is: Among them, φ comp,i (f) represents the phase spectrum of the i-th channel after compensation; φ i (f) represents the phase spectrum of the reference channel; represents the corresponding phase rotation factor; Δt i Represents the residual time deviation of the i-th channel relative to the reference channel, mapping the compensated data to the same coordinate system; A5. Spatial data fusion compensation: The data after the compensation amount is calculated is mapped to a unified coordinate system to achieve online compensation in the dynamic working condition of the laser displacement sensor (201). The online compensation adopts a recursive update equation. The specific equation is as follows: in, represents the phase estimate of the Kth iteration; K k represents the Kalman gain matrix; z k Represents the phase observation value at the current moment; H represents the observation matrix, and the online compensation under dynamic conditions is achieved through the formula.

Images