Linkage assembling method for automatically heating and positioning assembled bearing

Abstract

The invention provides a linkage assembly method for automatically heating and positioning and assembling a bearing, and belongs to the technical field of industrial products.The method comprises the steps that firstly, assembly parameters are set based on bearing model parameters, and the optimal assembly position is determined through pre-scanning of a three-dimensional vision sensor; an induction heater is used for conducting all-dimensional heating on the bearing, the inner diameter expansion value is calculated in real time through a thermal expansion equation, and meanwhile an infrared temperature sensor monitors temperature distribution of the bearing; when the temperature reaches a target value and the inner diameter expansion meets the requirement, the mechanical arm transfers the bearing to an assembly area; after the laser positioning system assists in accurate centering, the system calculates optimal press-fitting parameters to execute press-in operation, and meanwhile, an assembly force curve is monitored; after completion, uniform cooling treatment is performed to form interference fit; and finally, the assembly quality is detected through rotation testing, comprehensive performance evaluation is carried out, and the technical problem that assembly is not accurate enough due to inaccurate temperature control in the assembly process of the bearing and the long-shaft part is solved.

Description

TECHNICAL FIELD

[0001] The application belongs to the technical field of industrial production, and particularly relates to a linkage assembly method for automatically heating and positioning bearing assembly. BACKGROUND

[0002] Bearing assembly technology is widely used in the fields of mechanical manufacturing, precision instruments, aerospace, etc., and is a key process for ensuring the precision and service life of mechanical equipment. Traditional bearing assembly technology mainly relies on manual operation or semi-automatic assembly system, and adopts simple heating devices to preheat bearings before pressing assembly. This kind of method usually uses oil furnace, hot plate or electric heating ring to heat the bearing, and then implements assembly after positioning by manual operation or simple clamp.

[0003] However, the traditional assembly method has obvious defects in temperature control: the conventional heating methods such as oil bath heating and electric heating ring heating are difficult to accurately control the overall temperature and temperature distribution of the bearing, resulting in uneven thermal expansion of each part of the bearing. If the heating temperature is too high, the performance of the bearing material will change or even be damaged, and if the temperature is too low, it will not achieve enough thermal expansion, causing assembly difficulty or the need for excessive pressing force. In addition, the traditional heating method lacks real-time temperature monitoring and feedback adjustment mechanism, and the operator often relies on experience to judge the heating time, which is difficult to accurately grasp the actual expansion amount of the bearing inner diameter.

[0004] These temperature control problems directly affect the fitting state of the bearing inner diameter and the shaft during the assembly process, and further cause problems such as insufficient assembly precision, unstable interference amount, and shortened service life of the bearing. Especially for the assembly of bearings of high-precision and high-speed rotating equipment, the existing technology cannot solve the problem of accurate control of bearing heating temperature, which ultimately leads to the technical problem of inaccurate assembly. That is, there is a technical problem of inaccurate assembly caused by inaccurate temperature control during the assembly process of the bearing and the long shaft part in the existing technology. SUMMARY

[0005] Therefore, the application provides a linkage assembly method for automatically heating and positioning bearing assembly, which can solve the technical problem of inaccurate assembly caused by inaccurate temperature control during the assembly process of the bearing and the long shaft part in the existing technology.

[0006] The application is implemented as follows:

[0007] The application provides a linkage assembly method for automatically warming and positioning bearings, which comprises the following steps: setting assembly parameters according to bearing model parameters; performing pre-scanning through a three-dimensional visual sensor to determine an optimal assembly position; using an induction heater to perform omnidirectional heating on the bearing, calculating a bearing inner diameter thermal expansion value in real time through a thermal expansion equation, and monitoring the bearing temperature distribution through an infrared temperature sensor to keep the bearing temperature within a thermal expansion threshold range; when the bearing temperature reaches a target temperature and the bearing inner diameter thermal expansion value meets the assembly requirements, triggering a mechanical arm to transfer the bearing to an assembly area; assembling the bearing in the assembly area; after the assembly is completed, uniformly cooling the bearing assembly part to form a stable interference fit between the bearing and a long shaft part; and detecting the assembly quality after the assembly is completed.

[0008] On the basis of the above technical solution, the linkage assembly method for automatically warming and positioning bearings can be further improved as follows:

[0009] The assembly parameter setting comprises a heating temperature range, a heating time, an assembly pressure control value, a positioning accuracy requirement and a linkage assembly speed.

[0010] Further, the thermal expansion equation input comprises a linear expansion coefficient of the bearing material, a bearing initial temperature, a heating target temperature, a preset temperature, a bearing initial inner diameter size and a thermal conductivity coefficient of the bearing material, and the output is an actual inner diameter size of the bearing at the preset temperature and a bearing inner diameter thermal expansion value.

[0011] Further, the step of assembling the bearing in the assembly area specifically comprises the following steps: a laser positioning system collects multi-point coordinate data of the bearing and the long shaft part surface, applies a spatial geometric least square method to construct an assembly evaluation matrix, and controls the mechanical arm to perform optimal trajectory positioning based on the assembly evaluation matrix.

[0012] Further, the assembly evaluation matrix is used to evaluate the coaxiality and perpendicularity of the bearing and the long shaft part, and provides accurate positioning basis for the assembly process.

[0013] Further, the step of assembling the bearing in the assembly area further comprises the following steps: when the relative position deviation value is less than a preset threshold value, the system applies an assembly force optimization function to calculate the best pressing assembly parameters, and a servo pressing assembly mechanism performs bearing pressing operation according to a force displacement curve.

[0014] Further, the relative position deviation value refers to the distance and angle difference between the bearing center axis and the long shaft part center axis, which is calculated by the assembly evaluation matrix and is used to judge whether the bearing and the long shaft part meet the assembly conditions.

[0015] Further, the assembly force optimization function input includes the interference amount of the bearing and the long shaft part, the elastic modulus of the bearing material, the elastic modulus of the long shaft part material, the target assembly depth, and the friction coefficient of the contact surface, and the output is the optimal pressure value and the pressing speed corresponding to each displacement point in the pressing process.

[0016] Further, the step of assembling the bearing in the assembly area further includes: during the pressing process, the pressure sensor continuously monitors the assembly force curve, and if an abnormal force change is detected, the system immediately pauses and issues an alarm.

[0017] Further, the assembly force curve refers to the actual pressure-displacement relationship graph recorded during the bearing pressing process, which is calculated by the assembly force optimization function and used to guide the servo pressing mechanism to perform accurate pressing operation.

[0018] Compared with the prior art, the automatic heating and positioning bearing assembly linkage assembly method has the following advantages: the inductive heater is used to heat the bearing in all directions, the thermal expansion equation is used to calculate the thermal expansion value of the bearing inner diameter in real time, the infrared temperature sensor is used to monitor the temperature distribution of the bearing, the temperature is kept within the thermal expansion threshold range, and accurate temperature control is realized. The method solves the key defects of temperature control in the traditional technology: the high-frequency electromagnetic field generated by the inductive heater realizes rapid and uniform heating of the bearing, avoiding the problem of uneven temperature distribution caused by traditional heating methods; combined with the application of the thermal expansion equation, the system can calculate the actual expansion amount of the bearing inner diameter in real time without relying on experience to judge the heating degree; the whole process monitoring of the infrared temperature sensor ensures that the temperature of the bearing is always kept within the optimal range, preventing overheating damage and avoiding insufficient heating.

[0019] Since the present application realizes accurate temperature control and real-time monitoring of the inner diameter expansion amount during the bearing heating process, the bearing reaches the most suitable thermal expansion state for assembly, and an ideal transition fit is formed between the bearing inner diameter and the long shaft part, successfully solving the technical problem of inaccurate assembly caused by inaccurate temperature control during the bearing and long shaft part assembly process, and significantly improving the precision and reliability of bearing assembly BRIEF DESCRIPTION OF DRAWINGS

[0020] Figure 1 The flowchart of the automatic heating and positioning bearing assembly linkage assembly method;

[0021] Figure 2 The schematic diagram of the heating area of the automatic heating and positioning bearing assembly linkage assembly;

[0022] Figure 3 The schematic diagram of the laser positioning system of the automatic heating and positioning bearing assembly linkage assembly;

[0023] Figure 4A schematic diagram of an assembly area of an automatic heating and positioning assembly bearing linkage assembly;

[0024] Figure 5 A schematic diagram of a cooling system of an automatic heating and positioning assembly bearing linkage assembly;

[0025] In the drawings, the components represented by each reference numeral are listed as follows:

[0026] 10, three-dimensional visual sensor; 11, induction heater; 12, infrared temperature sensor; 13, mechanical arm; 14, laser positioning system; 141, laser; 15, servo press fitting mechanism; 16, pressure sensor; 18, rotation test unit; 181, motor; 182, torque sensor; 183, vibration sensor; 20, bearing; 30, long shaft part. DETAILED DESCRIPTION

[0027] In order to make the purposes, technical solutions and advantages of the embodiments of the present application clearer, the technical solutions in the embodiments of the present application will be described clearly and completely below with reference to the drawings in the embodiments of the present application.

[0028] As Figure 1 shown is a flowchart of an automatic heating and positioning assembly bearing linkage assembly method provided by the present application, wherein the method comprises the following steps:

[0029] S01, setting assembly parameters according to bearing model parameters, including heating temperature range, heating time, assembly pressure control value, positioning accuracy requirement and linkage assembly speed;

[0030] S02, pre-scanning the bearing and the long shaft part by a high-precision three-dimensional visual sensor to obtain bearing inner diameter, bearing outer diameter, bearing thickness and long shaft part outer diameter data, and determine the optimal assembly position;

[0031] S03, heating the bearing in all directions by an induction heater, calculating the bearing inner diameter thermal expansion value in real time through the thermal expansion equation, and monitoring the bearing temperature distribution by an infrared temperature sensor to keep it within the thermal expansion threshold value range;

[0032] S04, when the bearing temperature reaches the target temperature and the bearing inner diameter thermal expansion value meets the assembly requirement, triggering a high-precision mechanical arm to transfer the bearing from the heating area to the assembly area;

[0033] S05, the laser positioning system of the assembly area is started, multi-point coordinate data of the bearing and the long shaft part surface are collected, a spatial geometric least squares method is applied to construct an assembly evaluation matrix, and the mechanical arm is controlled based on the assembly evaluation matrix to position the optimal trajectory;

[0034] S06, when the relative position deviation value is less than a preset threshold value, the system applies an assembly force optimization function to calculate optimal press-fitting parameters, and the servo press-fitting mechanism performs bearing press-in operation according to the optimized force-displacement curve;

[0035] S07, during the press-fitting process, the pressure sensor continuously monitors the assembly force curve, and if an abnormal force change is detected, the system immediately suspends and issues an alarm;

[0036] S08, after the assembly is completed, the automatic cooling system is started to uniformly cool the bearing assembly part, so that the bearing and the long shaft part form a stable interference fit;

[0037] S09, after the assembly is completed, the rotation test unit is started to detect the bearing rotation resistance and vibration spectrum to determine whether the assembly quality is qualified;

[0038] S10, the system constructs an assembly quality evaluation model based on the assembly base matrix, the assembly variation matrix, the rotation harmonic variation matrix and the harmonic factor to comprehensively evaluate the performance of the assembled bearing system.

[0039] Among them, the three-dimensional visual sensor refers to a sensing device that obtains three-dimensional spatial information of an object through multiple high-resolution cameras combined with structured light or laser scanning technology, with an accuracy of 10μm.

[0040] Among them, the induction heater refers to a heating device that generates a high-frequency electromagnetic field using electromagnetic induction principles to generate eddy currents inside the metal bearing to generate heat by itself, with high heating efficiency and uniform temperature.

[0041] Among them, the thermal expansion equation is used to calculate the expansion amount of the inner diameter of the bearing during heating. The input includes the linear expansion coefficient of the bearing material, the initial temperature of the bearing, the target heating temperature, the preset temperature, the initial inner diameter size of the bearing, and the thermal conductivity coefficient of the bearing material. The output is the actual inner diameter size of the bearing at the preset temperature and the thermal expansion value of the bearing inner diameter. Among them, the preset temperature is a variable, and the output is the data corresponding to this preset temperature.

[0042] Among them, the thermal expansion threshold range refers to the upper and lower limits of the temperature during the heating process of the bearing, which is determined by the heating temperature range set in step S01, and is used to ensure that the bearing will not be damaged due to overheating or difficult to assemble due to insufficient temperature.

[0043] Among them, the assembly evaluation matrix refers to a spatial relative position relationship matrix constructed by analyzing the multi-point coordinate data of the bearing and the long shaft part surface, which is used to evaluate the coaxiality and perpendicularity of the bearing and the long shaft part, and provides accurate positioning basis for the assembly process.

[0044] Wherein, the relative position deviation value refers to the distance and angle difference between the bearing center axis and the long shaft part center axis, which is calculated by the assembly evaluation matrix in step S05, and is used to judge whether the bearing and the long shaft part meet the assembly condition.

[0045] Wherein, the assembly force optimization function is used to calculate the optimal force displacement relationship curve in the bearing press-fitting process, the input includes the interference amount of the bearing and the long shaft part, the elastic modulus of the bearing material, the elastic modulus of the long shaft part material, the target assembly depth and the contact surface friction coefficient, and the output is the optimal pressure value and the press-fitting speed corresponding to each displacement point in the press-fitting process, which ensures the smoothness of the assembly process and does not damage the parts.

[0046] Wherein, the force displacement curve refers to the relationship curve between displacement and corresponding applied force in the bearing press-fitting process, which is calculated by the assembly force optimization function in step S06, and is used to guide the servo press-fitting mechanism to perform accurate press-fitting operation.

[0047] Wherein, the assembly force curve refers to the actual pressure and displacement relationship diagram recorded in the bearing press-fitting process, which is collected by the pressure sensor in step S07 in real time, and is used to evaluate whether the assembly process is smooth, whether there are abnormal conditions such as jamming or deformation, etc.

[0048] Wherein, the rotation test unit refers to a test device that drives the assembled bearing to rotate through a motor, and collects the bearing rotation resistance and vibration spectrum through a torque sensor and a vibration sensor.

[0049] Wherein, the vibration spectrum refers to the amplitude distribution of each frequency vibration signal generated by the bearing in the rotation process, which is collected by the rotation test unit in step S09, and is used to evaluate the assembly quality and running state of the bearing.

[0050] Wherein, the interference fit refers to the fitting method that the outer diameter of the shaft is larger than the inner diameter of the bearing, which needs to apply a certain pressure to assemble at room temperature, and can be assembled without pressure or with small pressure by heating the bearing to make its inner diameter expand.

[0051] Wherein, the assembly area refers to the working space provided with a precision positioning system, a press-fitting mechanism and various sensors, which is used to complete the accurate centering and press-fitting of the bearing and the long shaft part.

[0052] Wherein, the assembly basic matrix refers to the standard value matrix of each key parameter in the ideal assembly state of the bearing, which contains the bearing rotation resistance reference value, the vibration spectrum reference value, the coaxiality reference value and the axial clearance reference value, etc., which are used as the reference standard for evaluating the assembly quality.

[0053] Wherein, the assembly variation matrix refers to the matrix composed of the deviation values of each parameter measured after the actual assembly is completed and the parameters of the assembly basic matrix, which is used to quantify the parameter changes caused by the assembly process and reflect the stability of the assembly quality.

[0054] Among them, the rotational harmonic matrix refers to the matrix of the change law of the vibration characteristics of the bearing at different speeds, which contains the amplitude and phase information of each vibration frequency component changing with the speed, and is used to predict the dynamic performance of the bearing under actual working conditions.

[0055] Among them, the harmonic change factor refers to the degree of deviation between the vibration frequency component generated by the bearing system during rotation and the theoretical frequency. It is an important indicator for evaluating the quality of bearing assembly. The smaller the harmonic change factor, the higher the assembly accuracy and the better the stability of the bearing system.

[0056] The specific implementation of the above steps is described in detail below. Figures 2-5 As shown, the specific implementation of step S01 is that during the initialization phase of the linked assembly system, the bearing material, size, and precision grade information is obtained from the bearing model parameter matching database. The system automatically sets the heating temperature range to 60% to 80% of the lower limit of the bearing material's phase transition temperature based on the bearing model. The heating temperature range for typical bearing steel materials is 80°C to 120°C. The heating time is also set to the bearing wall thickness divided by the material's thermal conductivity, multiplied by a heating factor of 1.5 to 2.0 to ensure uniform heat distribution. The assembly pressure is controlled to 10% to 15% of the bearing's rated static load to avoid damage to the bearing due to excessive pressure. The positioning accuracy requirement is automatically set to 1 / 3 of the bearing's precision grade, with a positioning accuracy of ±20μm for ordinary precision bearings and ±5μm for high-precision bearings. The linked assembly speed is adaptively adjusted based on the bearing size and precision grade, generally set to 0.5mm / s to 2mm / s. This step, by accurately setting assembly parameters, provides a basic parameter basis for subsequent assembly processes.

[0057] The specific implementation of step S02 is to use multiple groups of high-precision industrial cameras equipped with structured light projectors to perform a 360° full-scale scan of the bearings and long axis parts. First, the bearings are collected from multiple angles to obtain at least 200 point cloud data, and a digital model of the bearing is constructed through a three-dimensional point cloud reconstruction algorithm. The Gaussian surface fitting algorithm is used to extract the bearing inner diameter data, and the least squares circle fitting algorithm is used to extract the bearing outer diameter data. The flatness analysis algorithm is used to measure the bearing thickness; then the long axis parts are also collected and reconstructed with point clouds, and the cylindricity analysis algorithm is used to extract the shaft diameter data; finally, based on the obtained dimensional parameters and combined with the assembly requirements, the optimal assembly position coordinates of the bearing on the long axis are calculated through the spatial geometric constraint optimization method. This position should optimize the mechanical properties of the bearing after assembly. The system acquisition accuracy is 10μm, and the point cloud density is 10μm per cm 2 No less than 500 points are required to ensure the accuracy of dimensional measurement. This step uses high-precision visual inspection to obtain key dimensional data of bearings and long shaft parts, providing the spatial position basis for precise assembly.

[0058] The specific implementation of step S03 is to heat the bearing using an electromagnetic induction heater, which has a working frequency in the range of 10 kHz to 30 kHz, generates an alternating electromagnetic field to generate eddy currents inside the bearing material and self-heats. During the heating process, the system uses the linear expansion coefficient equation of the bearing material to calculate the thermal expansion value of the bearing inner diameter in real time, and the equation considers the linear expansion coefficient a of the bearing material (typical value is 11.8×10 -6 / ℃ to 13.5×10 -6 / ℃), the initial temperature T0, the current temperature T, the initial inner diameter D0, and the material thermal conductivity coefficient λ, and calculates the current bearing inner diameter D=D0×(1+α×(T-T0)). At the same time, the infrared temperature sensor scans the bearing surface temperature distribution at a frequency of 5 Hz to form a temperature field mapping, ensuring that the temperature difference of each part of the bearing does not exceed 15℃, keeping it within the thermal expansion threshold range, and preventing the bearing from deforming due to uneven expansion. This step realizes the controllable thermal expansion of the bearing inner diameter by accurately controlling the bearing heating process, creating conditions for non-destructive assembly.

[0059] The specific implementation of step S04 is that the system compares the current temperature of the bearing with the target temperature, and when the temperature reaches the set value and the deviation does not exceed ±2℃, and the thermal expansion value of the bearing inner diameter reaches or exceeds the interference amount with the shaft diameter and does not exceed 120% of the interference amount, the mechanical arm is triggered to perform the material taking action. The mechanical arm adopts six-axis servo control, with positioning accuracy better than ±0.03mm, repeat positioning accuracy better than ±0.01mm, and motion trajectory generated by five times spline interpolation algorithm to ensure smooth motion; the mechanical arm is equipped with a temperature isolation clamp with a heat resistance temperature not less than 200℃, and has a pressure sensing function, and the clamping force is adaptively adjusted in the range of 10N to 50N to avoid deformed bearings. During the material taking process, the mechanical arm performs real-time bearing posture stability analysis, and automatically adjusts the clamping force if the posture changes exceed the preset threshold. This step realizes the safe transfer of the bearing from the heating area to the assembly area, maintains the thermal expansion state of the bearing and ensures the posture stability.

[0060] The specific implementation of step S05 is that after the laser positioning system of the assembly area is started, a high-precision laser beam with a wavelength of 635 nm is used to scan the inner and outer surfaces of the bearing and the long shaft surface, and at least 36 uniformly distributed spatial coordinate points are collected. The system applies spatial geometric least squares method to construct the center axis equation of the bearing and the long shaft, calculates the distance and angular deviation between the two axes, and forms a 4x4 homogeneous transformation assembly evaluation matrix, which contains the relative position and attitude information of the bearing and the long shaft parts. Based on the matrix, the system uses gradient descent algorithm combined with cross-entropy optimization method to calculate the optimal motion trajectory of the mechanical arm, so that the bearing center axis and the long shaft center axis have the highest coincidence degree. The spatial resolution of the laser positioning system is 5 μm, and the sampling frequency is not less than 200 Hz, which ensures the real-time and accuracy of the positioning process. This step realizes the accurate centering positioning of the bearing and the long shaft through high-precision spatial coordinate measurement and geometric analysis, and creates conditions for subsequent press fitting.

[0061] The specific implementation of step S06 is that when the relative position deviation value of the bearing center axis and the long shaft part center axis is less than 15 μm and the angular deviation is less than 0.01°, the system determines that the positioning condition is met and enters the press fitting stage. At this time, the system applies an assembly force optimization function to calculate the best press fitting parameters, which considers the interference amount δ (typical value is 0.01% to 0.03% of the shaft diameter) of the bearing and the long shaft part, the elastic modulus E1 (about 2.1x10 5 MPa) of the bearing material, the elastic modulus E2 of the long shaft material, the target assembly depth L and the friction coefficient μ (about 0.15 to 0.3) of the contact surface, and calculates the pressure value F i and the corresponding press fitting speed v i of each displacement point in the press fitting process through finite element analysis method combined with Lame equation to generate the optimal force displacement curve. The servo press fitting mechanism executes accurate press fitting according to the curve, and the press fitting force control accuracy is better than ±2%, and the displacement control accuracy is better than ±5 μm. This step realizes the lossless press fitting of the bearing by scientifically calculating the best press fitting parameters and accurately executing, and ensures the assembly quality.

[0062] The specific implementation of step S07 is that during the press-fitting process, the system continuously collects real-time pressure data through a high-precision pressure sensor (range 0-10 kN, accuracy level 0.1 level) arranged on the press-fitting head, with a sampling frequency of not less than 1000 Hz, and simultaneously collects displacement data to construct a real-time assembly force curve. The system compares the real-time force curve with the theoretical force-displacement curve calculated in step S06, and if it detects a sudden change in assembly force (change rate exceeding 20% / mm) or absolute value deviation exceeding 15% of the theoretical value, it is determined to be abnormal. At this time, the system immediately suspends the press-fitting operation, with a servo motor deceleration time of not more than 0.1 s, and simultaneously issues an audible and visual alarm, displaying abnormal position and force value information on the operation interface. The system uses wavelet analysis method to extract features of the force curve to assist in determining the cause of the abnormality, such as bearing jamming, deformation, or foreign matter interference, etc. This step detects assembly abnormalities in real time by monitoring the change in press-fitting force, prevents damage to parts caused by forced assembly, and ensures assembly safety.

[0063] The specific implementation of step S08 is that after assembly is completed, the automatic cooling system is started to uniformly cool the bearing assembly part. The cooling system uses a temperature-adjustable cold air circulation method, with a wind speed controlled within the range of 3 m / s to 5 m / s, forming a ring-shaped air flow field surrounding the bearing periphery, and a temperature gradient controlled within the range of 3°C / min to 5°C / min to avoid stress concentration caused by rapid cooling. At the same time, the system monitors the temperature distribution of the bearing surface through an infrared thermal imager to ensure uniformity of cooling, with a temperature difference of not more than 10°C between different points. When the bearing temperature drops to near room temperature (25°C±5°C), the system monitors the interface stress wave signals between the bearing and the shaft through acoustic emission sensors to determine the stability of the interference fit. The cooling process generally lasts for 8 to 15 minutes, allowing the bearing inner diameter to shrink and form a stable interference fit with the long shaft part, with the interface pressure meeting the design requirements. This step controls the cooling process scientifically to form a stable interference connection between the bearing and the shaft, ensuring the connection strength and stiffness.

[0064] The specific implementation of step S09 is that after the assembly is completed and cooled and stabilized, the rotating test unit starts to detect the assembly quality. The unit includes a variable frequency servo motor, a torque sensor and a vibration sensor. First, a step acceleration method is used to rotate the bearing system at 10%, 30%, 50%, 70% and 100% of the rated speed, and each speed is maintained for 30 seconds. During rotation, the torque sensor (accuracy 0.1% F.S) collects real-time bearing rotation resistance data, while the three-axis acceleration sensor (frequency response range 1 Hz to 10 kHz) collects vibration signals, with a sampling frequency set to 2.56 times the upper limit frequency of vibration. The system applies a fast Fourier transform algorithm to analyze the frequency spectrum of the vibration signal, extracts characteristic frequency components, including bearing cage frequency, rolling element passing frequency, inner and outer ring characteristic frequency, etc. The system determines whether the assembly quality is qualified according to the preset standard (generally, the bearing rotation resistance is not more than 120% of the rated value, and the characteristic frequency amplitude is not more than 150% of the reference value). This step comprehensively evaluates the running performance of the assembled bearing through dynamic testing method, and verifies the assembly quality.

[0065] The specific implementation of step S10 is that the system constructs an assembly quality comprehensive evaluation model based on a multi-dimensional matrix analysis method. First, the system establishes an assembly base matrix, including standard parameters such as bearing rotation resistance reference value (usually 90% to 110% of the nominal value of the bearing model), vibration spectrum reference value, coaxiality reference value (usually not more than 10 μm) and axial clearance reference value (usually 0.01 mm to 0.03 mm); then calculate the assembly variation matrix, that is, the difference matrix between the measured value and the reference value; then analyze the rotation harmonic variation matrix, that is, the variation law of the vibration characteristics of the bearing at each speed, including the amplitude and phase information of the main frequency and harmonic frequency; finally, calculate the harmonic variation factor, that is, the deviation degree of the actual vibration frequency from the theoretical frequency, and the harmonic variation factor threshold is usually set between 0.05 and 0.15. The system reduces the dimension of the above matrix data by principal component analysis method, and combines with fuzzy comprehensive judgment method to obtain the final score of the assembly quality. The score standard is divided into four levels: excellent (more than 90 points), good (80 to 90 points), qualified (60 to 80 points) and unqualified (less than 60 points). This step comprehensively evaluates the assembly quality through mathematical model construction, and provides scientific and quantitative quality criteria.

[0066] The above technical solution constitutes a linked assembly device for automatically heating and positioning bearings. The device includes a high-precision 3D vision sensor 10, an induction heater 11, an infrared temperature sensor 12, a high-precision robotic arm 13, a laser positioning system 14, a servo press-fit mechanism 15, a pressure sensor 16, an automatic cooling system, a rotational testing unit 18, and a control system for performing assembly basic matrix, assembly variation matrix, rotational harmonic matrix, and harmonic factor analysis. The high-precision 3D vision sensor is used to obtain dimensional data of the bearing 20 and the longitudinal shaft component 30. The induction heater, in conjunction with the infrared temperature sensor, achieves uniform heating and temperature monitoring of the bearing. The high-precision robotic arm is responsible for transferring the bearing from the heating area to the assembly area. The laser positioning system constructs an assembly assessment matrix and guides the robotic arm for precise positioning. The servo press-fit mechanism, combined with the pressure sensor, performs the bearing press-fit operation according to an optimized force-displacement curve and monitors assembly force changes in real time. The automatic cooling system uniformly cools the assembly area. The rotational testing unit detects the bearing's rotational resistance and vibration spectrum. The control system is responsible for executing a comprehensive assembly quality assessment algorithm and providing a final quality grade.

[0067] The mathematical model or calculation process involved in the present invention is described in detail below.

[0068] In step S02, a 3D point cloud reconstruction algorithm is used to construct a digital model of the bearing. The specific calculation process is as follows:

[0069] First, the point cloud data of the bearing is obtained by a high-precision 3D vision sensor. The data can be represented as a point set P = {p1, p2, ..., p n}, where p i is the coordinate of a point in three-dimensional space, denoted as p i =(x i ,y i , z i Then, the least squares circle fitting algorithm is used to extract the outer diameter of the bearing, which is specifically expressed as follows:

[0070]

[0071] In the formula, (x c ,y c ) is the coordinate of the center of the fitted circle; r is the radius of the fitted circle; n is the number of points in the point cloud; f(p i ) is point p i Distance function to the fitted circle.

[0072] For the extraction of the bearing inner diameter, the Gaussian surface fitting algorithm is used, which is expressed as:

[0073]

[0074] In the formula, a, b, c, d, e, f, g are undetermined coefficients; m is the number of points in the inner diameter surface point cloud; F(p i ) is the distance function of point p i to the fitted surface.

[0075] The parameter acquisition method of the two algorithms is as follows: the point cloud data P is obtained by collecting through a three-dimensional vision sensor. The collection steps include: first, setting the imaging parameters of the three-dimensional vision sensor, including the exposure time of 5ms to 10ms and the light intensity of 70% to 90% of the bearing material reflection characteristic; second, the sensor collects at least 12 different angles around the bearing, and not less than 20 point clouds are obtained at each angle; finally, the point cloud data collected at all angles is registered and fused to form a complete bearing three-dimensional point cloud model. The coefficients of the fitting algorithm are solved by an iterative optimization method, the initial value is estimated according to the spatial distribution characteristics of the point cloud, and the iteration termination condition is that the residual change is less than 10 -6 or the number of iterations exceeds 1000 times.

[0076] In step S03, the bearing thermal expansion equation is used to calculate the bearing inner diameter thermal expansion value, which is specifically represented as follows:

[0077] D=D0·(1+α·(T-T0))·(1-β·(T-T0 2 +γ·(T-T0) 3 );

[0078] In the formula, D is the current bearing inner diameter; D0 is the initial inner diameter; a is the linear expansion coefficient of the material, and the typical value is 11.8×10 -6 / ℃ to 13.5×10 -6 / ℃; T is the current temperature; T0 is the initial temperature; β is the second-order temperature correction coefficient, and the typical value is 0.5×10 -8 / ℃ 2 ; γ is the third-order temperature correction coefficient, and the typical value is 0.3×10 -10 / ℃ 3 .

[0079] The temperature field distribution monitoring is calculated by using the heat diffusion equation, which is represented as:

[0080]

[0081] In the formula, T is the temperature field function; t is the time; κ is the thermal diffusion coefficient, λ is the thermal conductivity coefficient; p is the material density; c p is the specific heat capacity; q is the internal heat source density generated by induction heating. The equation is solved by the finite difference method, and the bearing is divided into at least 100 grid units, and the time step is not greater than 0.1s.

[0082] The acquisition method of these parameters is: the linear expansion coefficient α, the thermal conductivity coefficient λ, the density ρ and the specific heat capacity c of the bearing material p Obtained by querying the material database; the initial inner diameter D0 is obtained by visual measurement in step S02; the temperature T is measured in real time by an infrared temperature sensor, with a sampling frequency of 5Hz; the second-order temperature correction coefficient β and the third-order temperature correction coefficient γ are obtained by experimental calibration, and the calibration steps include: heating the standard bearing sample under standard environment at different temperature gradients, recording the actual inner diameter size at different temperatures, and fitting the two parameters by the least square method.

[0083] In step S05, the assembly evaluation matrix is a 4x4 homogeneous transformation matrix, which is used to describe the relative position relationship between the bearing and the long shaft part, and is specifically represented as follows:

[0084]

[0085] In the formula, r ij (i, j = 1, 2, 3) is the rotation matrix element, which describes the rotation relationship of the bearing coordinate system relative to the long shaft coordinate system; t x , t y , t z is the translation vector element, which describes the translation amount of the bearing center relative to the long shaft center.

[0086] The rotation matrix element can be expressed by Euler angles:

[0087]

[0088] In the formula, α, β, γ are the rotation angles around the x, y, z axes respectively.

[0089] The relative position deviation value calculation method is:

[0090]

[0091] In the formula, δ d is the distance deviation between the bearing center axis and the long shaft part center axis; δ θ is the angle deviation between the two axes.

[0092] The acquisition method of these parameters is: the bearing and long shaft surface coordinate points are obtained by laser positioning system acquisition, and the acquisition steps include: first, the laser 141 emits a laser beam with a wavelength of 635nm to scan the bearing inner and outer surfaces and the long shaft surface; second, the high-speed CCD camera captures the reflected light signal, and the surface coordinate points are calculated by the triangulation principle; finally, the collected coordinate points are filtered to remove noise points. The center axes of the bearing and the long shaft are fitted by the spatial geometric least square method, and then the distance and angle relationship between the two axes are calculated to construct the assembly evaluation matrix.

[0093] In step S06, the assembly force optimization function is used to calculate the optimal press fitting parameters, which is expressed as follows:

[0094] F(x) = π μ E δ L (1 - e -k·x / L ) (1 + η v(x));

[0095] In the formula, F(x) is the press fitting force required when the press fitting depth is x; μ is the contact surface friction coefficient, and the typical value is 0.15 to 0.3; E is the equivalent elastic modulus, E1 is the elastic modulus of the bearing material, E2 is the elastic modulus of the long axis material; δ is the interference amount, and the typical value is 0.01% to 0.03% of the shaft diameter; L is the bearing width; k is the press fitting coefficient, which is related to the material and surface state, and the typical value is 2 to 5; η is the speed correction coefficient, and the typical value is 0.05 to 0.2; v(x) is the press fitting speed function.

[0096] The press fitting speed function is defined as:

[0097]

[0098] In the formula, v max is the maximum allowed press fitting speed, which is set in step S01; the function makes the speed low at the beginning and the end of the press fitting and high in the middle, and also considers smooth transition.

[0099] The acquisition method of these parameters is as follows: the contact surface friction coefficient μ is obtained by surface roughness measurement and table lookup; the elastic moduli E1 and E2 are obtained by querying the material database; the interference amount δ is calculated from the visual measurement result in step S02, δ = R shaft -R bearing , where R shaft is the long axis radius, and R bearing is the bearing inner diameter radius; the press fitting coefficient k is determined by small batch test, and the test steps include: press fitting test of sample bearings under different pressures, recording the pressure and displacement relationship, and fitting to obtain the k value; the speed correction coefficient η is determined by dynamic press fitting test, and the test steps include: press fitting test at different press fitting speeds, analyzing the influence of speed on the press fitting force, and fitting to obtain the η value.

[0100] In step S10, the assembly quality evaluation involves the calculation of multiple matrices and factors. First, the assembly base matrix is expressed as:

[0101]

[0102] In the formula, B F is the bearing rotation resistance reference value; B V is the vibration spectrum reference value; B Cis the coaxiality reference value; B A is the axial clearance reference value; B Fi , B Vi , B Ci , B Ai is the reference parameter value corresponding to different rotational speeds.

[0103] The assembly variation matrix calculation formula is:

[0104]

[0105] In the formula, M F , M V , M C , M A , M Fi , M Vi , M Ci , M Ai is the actual measurement value.

[0106] The rotational harmonic variation matrix is represented as:

[0107]

[0108] In the formula, A ij is the amplitude of the jth vibration frequency component at the ith rotational speed; φ ij is the corresponding phase; j is the imaginary unit, m is the number of test rotational speed levels; n is the number of vibration frequency components considered.

[0109] The harmonic variation factor calculation formula is:

[0110]

[0111] In the formula, f ij is the actual vibration frequency; is the theoretical vibration frequency; is the reference amplitude. The harmonic variation factor threshold is usually set between 0.05 and 0.15, and the smaller the value, the higher the assembly quality.

[0112] The final assembly quality score is calculated by principal component analysis and fuzzy comprehensive evaluation method, represented as:

[0113]

[0114] In the formula, Q is the final quality score; w i is the weight of the ith principal component; λ i is the eigenvalue of the ith principal component; S i is the score of the ith principal component; k is the number of principal components selected, usually the number of principal components corresponding to the cumulative contribution rate of eigenvalues reaching more than 85%.

[0115] The parameter acquisition method of these matrices and factors is as follows: the parameters in the assembled basic matrix B are obtained through standard bearing tests or are acquired from data provided by the bearing manufacturer; the actual measured values are collected by the rotating test unit in step S09, which includes a variable frequency servo motor 181, a torque sensor 182, and a vibration sensor 183. The measurement steps include: first, rotating tests are sequentially performed on the bearing system at 10%, 30%, 50%, 70%, and 100% of the rated speed; second, the torque sensor collects the rotating resistance data; and finally, the three-axis acceleration sensor collects the vibration signal. Theoretical vibration frequency The theoretical vibration frequency is calculated according to the bearing structure parameters and kinematic relationships; and the weights and eigenvalues of the principal component analysis are determined through statistical analysis of historical data.

[0116] Optionally, in step S05, a spatial geometric least squares method is used to construct the central axis equation of the bearing and the long axis, which is specifically represented as follows:

[0117] For the central axis of the bearing, assume its equation as:

[0118]

[0119] In the formula, (x0, y0, z0) is the coordinate of a point on the axis; and (l, m, n) is the direction vector of the axis.

[0120] The parameters are solved by the least squares method, and the objective function is:

[0121]

[0122] In the formula, p is the number of collected coordinate points; d i is the distance from the ith point to the axis, and the calculation formula is:

[0123]

[0124] For the central axis of the long axis, the same method is used to solve its parameters. These parameters are obtained by solving a nonlinear least squares problem, and the Levenberg-Marquardt algorithm is used for iterative optimization. The initial value is set according to the roughly estimated axis direction, and the iteration termination condition is that the residual change is less than 10 -5 or the number of iterations exceeds 500.

[0125] In step S07, the anomaly detection of the assembled force curve uses the following equation:

[0126] ΔF(x) = |F measuerd (x) - F theoretical (x) |;

[0127]

[0128] F(x) = F(x-1) + ΔF(x) measured (x) is the actual press-fit force measured at displacement x; F theoretical (x) is the theoretical press-fit force calculated in step S06; ΔF(x) is the absolute deviation; η F (x) is the relative deviation percentage; is the rate of change of the assembly force.

[0129] The abnormality judgment condition is: if η F (x) > 15% or the system judges as abnormal, where F max is the maximum press-fit force, Δx mm represents the displacement unit in millimeters. These parameters are obtained by real-time acquisition through a pressure sensor with a sampling frequency not less than 1000 Hz, and the sensor is installed on the press-fit head to directly measure the force data during the press-fit process.

[0130] Specifically, the principle of the present application is: the technical principle of the present application is mainly based on the accurate control and real-time monitoring of the thermal expansion characteristics of the bearing. The basic principle of bearing assembly is to temporarily increase the inner diameter of the bearing by using the thermal expansion effect to facilitate installation, and then realize the interference fit by cooling. The key is to accurately control the bearing temperature to make the inner diameter expansion amount exactly meet the assembly requirements without excessive heating.

[0131] The induction heater plays a core role in the present application, and its working principle is to generate a high-frequency alternating magnetic field by electromagnetic induction to form an eddy current inside the bearing, and convert electrical energy into heat energy through eddy current loss. Compared with traditional heating methods, induction heating can quickly heat the bearing and generate heat from the inside, making the temperature distribution of each part of the bearing more uniform. This heating method avoids the problem of uneven heat conduction of traditional heating devices, laying a foundation for realizing accurate thermal expansion control.

[0132] The application of the thermal expansion equation is another key technical point of the present application. The system establishes a mathematical model of bearing thermal expansion according to the linear expansion coefficient of the bearing material, the initial temperature of the bearing, the target heating temperature, the initial inner diameter size of the bearing, and the thermal conductivity coefficient of the bearing material. Through this model, the system can calculate the actual inner diameter size of the bearing at different temperatures and its expansion value in real time, providing accurate quantitative basis for temperature control and avoiding subjective judgment of the heating degree in traditional methods.

[0133] The real-time monitoring mechanism of the infrared temperature sensor provides closed-loop feedback for temperature control. The system monitors the temperature distribution of each part of the bearing in real time through multi-point temperature collection, ensuring that the bearing temperature always remains within the preset thermal expansion threshold range. When the temperature is detected to be close to the upper limit, the system automatically adjusts the heating power; when the temperature reaches the target value and the inner diameter expansion amount meets the assembly requirements, the system immediately triggers the subsequent assembly process to avoid excessive heating.

[0134] The organic combination of these technical principles enables the present application to accurately control the bearing heating temperature and the inner diameter expansion amount, ensuring that an ideal fit state is formed between the bearing and the long shaft part, thereby solving the problem of inaccurate assembly caused by inaccurate temperature control and improving the precision and stability of bearing assembly.

[0135] A specific embodiment 1 of the present application is provided below, and the specific implementation of each step in embodiment 1 is described in detail as follows.

[0136] In this embodiment, the specific implementation of step S01 is the same as described above and will not be described in detail here.

[0137] In this embodiment, the specific implementation of step S02 is to use multiple groups of high-precision industrial cameras equipped with structured light projectors to perform 360° omnidirectional scanning of the bearing and the long shaft part. First, the bearing is collected at multiple angles to obtain at least 200 point cloud data, a bearing digital model is constructed through a three-dimensional point cloud reconstruction algorithm, a Gaussian surface fitting algorithm is applied to extract the bearing inner diameter data, a least squares circle fitting algorithm is used to extract the bearing outer diameter data, and a flatness analysis algorithm is used to measure the bearing thickness; then the long shaft part is also subjected to point cloud collection and reconstruction, and a cylindrical degree analysis algorithm is used to extract the shaft diameter data; finally, based on the obtained size parameters, the optimal assembly position coordinates of the bearing on the long shaft are calculated through a spatial geometric constraint optimization method according to the assembly requirements, which should make the mechanical properties of the bearing optimal after assembly. The specific calculation process of the least squares circle fitting algorithm is as follows:

[0138]

[0139] In the formula, (x c , y c ) is the center coordinates of the fitted circle; r is the radius of the fitted circle; n is the number of points in the point cloud; f(p i ) is the distance function from point p i to the fitted circle. This algorithm minimizes the sum of the squares of the distances of all sampling points to the fitted circle to solve the optimal circle parameters, which is suitable for accurate measurement of the bearing outer diameter. For the extraction of the bearing inner diameter, a Gaussian surface fitting algorithm is used:

[0140]

[0141] where a, b, c, d, e, f, g are undetermined coefficients; m is the number of points in the inner diameter surface point cloud; F(p i ) is the distance function of point p i to the fitted surface. The algorithm solves the surface parameters by establishing the quadratic surface equation of the inner diameter surface and minimizing the sum of the squared distances of the point cloud to the surface, and is suitable for accurate measurement of complex inner diameter shapes. The system has a collection accuracy of 10 μm, and the point cloud density is not less than 500 points per cm 2 , ensuring the accuracy of size measurement. This step obtains key size data of the bearing and long shaft parts through high-precision visual detection, providing a spatial position basis for precise assembly.

[0142] The specific implementation of step S03 is to heat the bearing using an electromagnetic induction heater. The heater has a working frequency in the range of 10 kHz to 30 kHz, generating an alternating electromagnetic field to cause eddy current in the bearing material and self-heating. During the heating process, the system uses the linear expansion coefficient equation of the bearing material to calculate the thermal expansion value of the bearing inner diameter in real time. The equation takes into account the linear expansion coefficient a of the bearing material (typical value: 11.8×10 -6 / ℃ to 13.5×10 -6 / ℃), the initial temperature T0, the current temperature T, the initial inner diameter D0, and the material thermal conductivity coefficient λ. The thermal expansion equation is specifically expressed as follows:

[0143] D=D0·(1+α·(T-T0))·(1-β·(T-T0) 2 +γ·(T-T0) 3 );

[0144] where D is the current bearing inner diameter; D0 is the initial inner diameter; a is the material linear expansion coefficient, with a typical value of 11.8×10 -6 / ℃ to 13.5×10 -6 / ℃; T is the current temperature; T0 is the initial temperature; β is the second-order temperature correction coefficient, with a typical value of 0.5×10 -8 / ℃ 2 ; and γ is the third-order temperature correction coefficient, with a typical value of 0.3×10 -10 / ℃ 3 . This equation not only considers the linear thermal expansion characteristics of the material, but also modifies the nonlinear thermal expansion effect at high temperatures by introducing high-order terms, improving the size prediction accuracy. At the same time, the infrared temperature sensor scans the bearing surface temperature distribution at a frequency of 5 Hz to form a temperature field mapping, ensuring that the temperature difference of each part of the bearing does not exceed 15℃, keeping it within the thermal expansion threshold range, preventing the bearing from deforming due to uneven expansion. The temperature field distribution monitoring uses the heat diffusion equation to calculate:

[0145]

[0146] where T is the temperature field function; t is the time; x is the thermal diffusivity, λ is the thermal conductivity; ρ is the material density; c p is the specific heat capacity; q is the internal heat source density, which is generated by induction heating. This equation realizes the accurate prediction of the temperature field distribution by describing the heat conduction process inside the bearing, helping to control the uniformity of heating. This step realizes the controllable thermal expansion of the inner diameter of the bearing by accurately controlling the bearing heating process, creating conditions for non-destructive assembly.

[0147] The specific implementation of step S04 is that the system triggers the mechanical arm to perform the picking action when the temperature reaches the set value and the deviation does not exceed ±2℃, and the thermal expansion value of the bearing inner diameter reaches or exceeds the interference amount with the shaft diameter and does not exceed 120% of the interference amount. The mechanical arm adopts six-axis servo control, with positioning accuracy better than ±0.03mm and repeat positioning accuracy better than ±0.01mm. The motion trajectory is generated using a quintic spline interpolation algorithm to ensure smooth motion. The mechanical arm is equipped with a temperature isolation clamp with a heat resistance temperature of not less than 200℃, and has a pressure sensing function, with the clamping force adaptively adjusted within the range of 10N to 50N to avoid deforming the bearing. During the picking process, the mechanical arm performs real-time bearing posture stability analysis, and automatically adjusts the clamping force if the posture changes exceed the preset threshold. This step realizes the safe transfer of the bearing from the heating area to the assembly area, maintains the thermal expansion state of the bearing and ensures the posture stability.

[0148] The specific implementation of step S05 is that after the laser positioning system of the assembly area is started, a high-precision laser beam with a wavelength of 635nm is used to scan the inner and outer surfaces of the bearing and the long shaft surface, and at least 36 uniformly distributed spatial coordinate points are collected. The system applies spatial geometric least squares method to construct the center axis equation of the bearing and the long shaft, calculates the distance and angle deviation between the two axis lines, and forms a 4x4 homogeneous transformation assembly evaluation matrix, which contains the relative position and posture information of the bearing and the long shaft parts. For the bearing center axis, assume its equation as:

[0149]

[0150] where (x0, y0, z0) is the coordinate of a point on the axis; (l, m, n) is the direction vector of the axis. The parameters are solved by least squares method, and the objective function is:

[0151]

[0152] where p is the number of collected coordinate points; d i is the distance from the ith point to the axis, and the calculation formula is:

[0153]

[0154] The method is suitable for high-precision axis extraction by solving the axis parameters by minimizing the sum of squares of distances of all sampling points to the fitted axis. The assembly evaluation matrix is expressed as:

[0155]

[0156] In the formula, r ij (i, j = 1, 2, 3) is the rotation matrix element, which describes the rotation relationship of the bearing coordinate system relative to the long axis coordinate system; t x , t y , t z are translation vector elements, which describe the translation amount of the bearing center relative to the long axis center. The rotation matrix element can be expressed by Euler angles:

[0157]

[0158] In the formula, α, β, γ are the rotation angles around the x, y, z axes, respectively. The relative position deviation value calculation method is:

[0159]

[0160] In the formula, δ d is the distance deviation between the bearing center axis and the long axis part center axis; δ θ is the angle deviation between the two axes. This matrix describes the relative position relationship between the bearing and the long axis, providing a mathematical basis for accurate positioning. Based on this matrix, the system uses the gradient descent algorithm combined with the cross-entropy optimization method to calculate the optimal motion trajectory of the mechanical arm, so that the bearing center axis and the long axis center axis have the highest coincidence degree. The spatial resolution of the laser positioning system is 5 μm, and the sampling frequency is not less than 200 Hz, ensuring the real-time and accuracy of the positioning process. This step realizes the accurate centering positioning of the bearing and the long axis through high-precision spatial coordinate measurement and geometric analysis, creating conditions for subsequent press fitting.

[0161] The specific implementation of step S06 is that when the relative position deviation value between the bearing center axis and the long axis part center axis is less than 15 μm and the angle deviation is less than 0.01°, the system determines that the positioning condition is met and enters the press fitting stage. At this time, the system applies the assembly force optimization function to calculate the optimal press fitting parameters, which takes into account the interference amount δ (typical value is 0.01% to 0.03% of the shaft diameter) of the bearing and the long axis part, the elastic modulus E1 (about 2.1 × 10 5 MPa) of the bearing material, the elastic modulus E2 of the long axis material, the target assembly depth L, and the friction coefficient μ (about 0.15 to 0.3) of the contact surface. The assembly force optimization function is specifically expressed as follows:

[0162] F(x) = π·μ·E·δ·L·(1-e-kx / L )·(1+η·v(x));

[0163] where F(x) is the required press-in force at the press-in depth x; μ is the friction coefficient of the contact surface, typically 0.15 to 0.3; E is the equivalent elastic modulus, E1 is the elastic modulus of the bearing material, E2 is the elastic modulus of the long axis material; δ is the interference, typically 0.01% to 0.03% of the shaft diameter; L is the bearing width; k is the press-in coefficient, related to the material and surface state, typically 2 to 5; η is the speed correction coefficient, typically 0.05 to 0.2; v(x) is the press-in speed function. The press-in speed function is defined as:

[0164]

[0165] where v max is the maximum allowed press-in speed, set by step S01. This function ensures smooth changes in force during the press-in process, avoiding impact loads, while taking into account the variation of friction force during the press-in process and the speed influence. Through finite element analysis method combined with Lame equation, the pressure value F i and the corresponding press-in speed v i to be applied at each displacement point during the press-in process are calculated, generating the optimal force-displacement curve. The servo press-in mechanism executes precise press-in according to this curve, with press-in force control accuracy better than ±2% and displacement control accuracy better than ±5μm. This step realizes non-destructive press-in of bearings by scientifically calculating the optimal press-in parameters and accurately executing them, ensuring assembly quality.

[0166] The specific implementation of step S07 is that during the press-in process, the system continuously collects real-time pressure data through the high-precision pressure sensor (range 0 to 10kN, accuracy level 0.1 level) configured on the press-in head, with a sampling frequency not less than 1000Hz, while collecting displacement data, to construct a real-time assembly force curve. The system compares the real-time force curve with the theoretical force-displacement curve calculated in step S06, and uses the following equation for anomaly detection:

[0167] ΔF(x)=|F measured (x)-F theoretical (x)|

[0168]

[0169] where F measured (x) is the actual press-in force measured at displacement x; F theoretical (x) is the theoretical press-in force calculated in step S06; ΔF(x) is the absolute deviation; η F (x) is the relative deviation percentage; The assembly force rate of change is assembled. If abnormal force change (relative deviation η F (x) > 15% or rate of change The system immediately suspends the press-fitting operation, the servo motor deceleration time is not more than 0.1s, and at the same time, an audible and visual alarm is issued, and the abnormal position and force value information is displayed on the operation interface. The system uses wavelet analysis method to extract the characteristics of the force curve, identifies the abnormal mode by decomposing the force curve into different frequency components, and assists in judging the abnormal reasons such as bearing jamming, deformation or foreign matter interference. This step detects assembly abnormalities in time by monitoring the press-fitting force change in real time, prevents damage to parts caused by forced assembly, and ensures assembly safety.

[0170] The specific implementation of step S08 is that after the assembly is completed, the automatic cooling system is started to uniformly cool the bearing assembly part. The cooling system adopts a cold air circulation mode with adjustable temperature, the wind speed is controlled within the range of 3m / s to 5m / s, a ring-shaped air flow field is formed to surround the periphery of the bearing, and the temperature gradient is controlled within the range of 3℃ / min to 5℃ / min to avoid stress concentration caused by rapid cooling. At the same time, the system monitors the temperature distribution of the bearing surface through an infrared thermal imager to ensure the uniformity of cooling, and the temperature difference of each point is not more than 10℃. When the temperature of the bearing decreases to the room temperature (25℃±5℃), the system monitors the interface stress wave signals between the bearing and the shaft through the acoustic emission sensor to judge the stability of the interference fit. The cooling process generally lasts for 8 minutes to 15 minutes, so that the inner diameter of the bearing shrinks and forms a stable interference fit with the long shaft part, and the interface pressure reaches the design requirement. This step controls the cooling process scientifically to form a stable interference connection between the bearing and the shaft, and ensures the connection strength and stiffness.

[0171] The specific implementation of step S09 is that after the assembly is completed and the cooling is stable, the rotation test unit starts to detect the assembly quality. The unit includes a variable frequency servo motor, a torque sensor and a vibration sensor. First, a step acceleration method is used to rotate the bearing system at 10%, 30%, 50%, 70% and 100% of the rated speed, and each speed is maintained for 30 seconds. During the rotation process, the torque sensor (accuracy 0.1% F.S) collects bearing rotation resistance data in real time, and at the same time, a three-axis acceleration sensor (frequency response range 1Hz to 10kHz) collects vibration signals, and the sampling frequency is set to 2.56 times the upper limit frequency of vibration. The system applies fast Fourier transform algorithm to frequency spectrum analysis of the vibration signal, extracts characteristic frequency components, including bearing cage frequency, rolling element passing frequency, inner and outer ring characteristic frequency, etc. The system judges whether the assembly quality is qualified according to the preset qualified standard (generally, the bearing rotation resistance is not more than 120% of the rated value, and the characteristic frequency amplitude is not more than 150% of the reference value). This step comprehensively evaluates the running performance of the bearing after assembly through dynamic test method, and verifies the assembly quality.

[0172] The specific implementation of step S10 is that the system constructs an assembly quality comprehensive evaluation model based on a multi-dimensional matrix analysis method. First, the system establishes an assembly basic matrix, which is expressed as:

[0173]

[0174] In the formula, B F is a bearing rotation resistance reference value; B V is a vibration frequency spectrum reference value; B C is a coaxial degree reference value; B A is an axial clearance reference value; B Fi , B Vi , B Ci , B Ai (i=1, 2,...) are reference parameter values corresponding to different rotating speeds. Then, the system calculates an assembly variation matrix:

[0175]

[0176] In the formula, M F , M V , M C , M A , M Fi , M Vi , M Ci , M Ai are actual measurement values. Next, the system analyzes a rotating harmonic variation matrix:

[0177]

[0178] In the formula, A ij is an amplitude of a jth vibration frequency component at an ith rotating speed; φ ij is a corresponding phase; j is an imaginary unit, m is a test rotating speed level number; and n is a vibration frequency component number considered. Finally, the system calculates a harmonic variation factor:

[0179]

[0180] In the formula, f ij is an actual vibration frequency; is a theoretical vibration frequency; is a reference amplitude. The harmonic variation factor threshold is usually set to be between 0.05 and 0.15, and the smaller the value is, the higher the assembly quality is. Finally, the system calculates an assembly quality score through principal component analysis and a fuzzy comprehensive evaluation method:

[0181]

[0182] In the formula, Q is the final quality score; w i is a weight of an ith principal component; and λ iis the eigenvalue of the ith principal component; S i is the score of the ith principal component; k is the number of principal components selected, usually the number of principal components corresponding to the cumulative contribution rate of eigenvalues reaching more than 85%. The scoring criteria are divided into four levels: excellent (more than 90 points), good (80 to 90 points), qualified (60 to 80 points), and unqualified (less than 60 points). This matrix analysis method comprehensively considers multiple key performance indicators in the bearing assembly process, quantitatively processes and comprehensively evaluates these indicators through a mathematical model, and provides an objective and accurate assembly quality evaluation result. This step provides a scientific and quantitative quality criterion by building a comprehensive assembly quality evaluation model through a mathematical model.

[0183] The automatic heating and positioning assembly bearing linkage assembly device involved in this embodiment includes a high-precision three-dimensional visual sensor, an induction heater, an infrared temperature sensor, a high-precision mechanical arm, a laser positioning system, a servo press-fitting mechanism, a pressure sensor, an automatic cooling system, a rotation test unit, and a control system for executing assembly base matrix, assembly variation matrix, rotation harmonic variation matrix, and harmonic factor analysis. The high-precision three-dimensional visual sensor is used to obtain size data of the bearing and the long shaft part; the induction heater cooperates with the infrared temperature sensor to realize uniform heating and temperature monitoring of the bearing; the high-precision mechanical arm is responsible for transferring the bearing from the heating area to the assembly area; the laser positioning system is used to build an assembly evaluation matrix and guide the mechanical arm for precise positioning; the servo press-fitting mechanism combines the pressure sensor to perform bearing press-in operation according to the optimized force-displacement curve and monitor the assembly force change in real time; the automatic cooling system uniformly cools the assembly part; the rotation test unit is used to detect the bearing rotation resistance and vibration spectrum; and the control system is responsible for executing the comprehensive evaluation algorithm of assembly quality and giving the final quality grade. Through the cooperative work of each functional unit, the entire linkage assembly device realizes automation and intelligentization of the entire bearing assembly process, improves assembly precision and efficiency, and guarantees the stability and reliability of assembly quality. The device is suitable for high-precision assembly of various bearings, especially for high-precision bearings with high assembly precision requirements and difficult assembly, and has wide application prospects in the fields of aerospace, precision machine tools, and automobile manufacturing.

[0184] In order to better understand and implement the present application, the following provides an embodiment 2 of a specific application scenario of the present application: researchers designed an automatic heating and positioning assembly bearing linkage assembly system to solve the problem of high assembly precision requirement and difficult operation of high-speed bearings of an aero-engine, and applied it in the assembly of a main bearing of an aero-engine. This application involves a type 7320B angular contact ball bearing that needs to be assembled onto the engine main shaft, with extremely high assembly precision requirements.

[0185] During the assembly parameter setting phase, the system first reads the parameter information of the 7320B angular contact ball bearing in the bearing database, including the inner diameter of 100 mm, the outer diameter of 215 mm, the thickness of 47 mm, the GCr15 material (linear expansion coefficient α = 12.5 × 10 -6 / °C). The system automatically sets the heating temperature range to 90°C to 110°C, the heating time to 240 seconds, the assembly pressure control value to 12% of the bearing's rated static load (156kN), or 18.72kN, the positioning accuracy requirement to ±4μm, and the linkage assembly speed to 1.2mm / s. These parameters are shown in Table 1:

[0186] Table 17320B angular contact ball bearing assembly parameter setting table

[0187] Parameter name Parameter value Bearing inner diameter 100 mm Bearing outer diameter 215 mm Bearing thickness 47 mm Bearing material GCr15 Linear expansion coefficient <![CDATA[12.5×10 -6 / ℃]]> Heating temperature range 90℃～110℃ Heating time 240s Assembly pressure control value 18.72 kN Positioning accuracy requirement ± 4 μm Linkage assembly speed 1.2 mm / s

[0188] During the pre-scan phase, a 3D vision sensor scanned the bearing and engine main shaft, collecting 250 point cloud data points. Using a least-squares circular fitting algorithm, the bearing's actual inner diameter was calculated to be 99.982mm, its outer diameter to be 214.996mm, and its thickness to be 46.994mm. The engine main shaft's outer diameter is 100.024mm. Based on this, the interference fit was calculated to be 0.042mm (0.042% of the shaft diameter), confirming the optimal bearing assembly position on the shaft to be 652.00mm from the shaft shoulder.

[0189] During the heating phase, the induction heater heats the bearing at a frequency of 20kHz. The system monitors the bearing temperature in real time and calculates the inner diameter thermal expansion value. When the bearing temperature rises to 105°C, the inner diameter thermal expansion is calculated as follows: D = 99.982 × (1 + 12.5 × 10 -6 ×(105-22))×(1-0.48×10 -8 ×(105-22) 2 +0.25×10 -10 ×(105-22) 3 )=100.078mm

[0190] At this point, the inner diameter thermal expansion value is 0.096mm, more than double the interference fit of 0.042mm, meeting the assembly requirements. Infrared temperature sensor monitoring shows that the maximum temperature difference between various parts of the bearing is 8.6°C, and the temperature distribution is uniform within the 15°C threshold range. The thermal expansion process and temperature monitoring data are shown in Table 2:

[0191] Table 2 Bearing heating process temperature and inner diameter monitoring data

[0192] Time (s) Average temperature (°C) Maximum temperature difference (°C) Inner diameter calculated value (mm) Thermal expansion value (mm) 0 22 0.5 99.982 0 60 46 3.2 100.008 0.026 120 68 5.7 100.032 0.050 180 89 7.4 100.057 0.075 240 105 8.6 100.078 0.096

[0193] When the bearing temperature reaches 105℃ and the inner diameter thermal expansion meets the requirements, the high-precision mechanical arm transfers the bearing from the heating area to the assembly area, and the whole transfer process takes 4.6 seconds. The clamping force of the temperature isolation clamp at the end of the mechanical arm is automatically adjusted to 32N, ensuring the stability of the bearing during the transfer process.

[0194] After the laser positioning system in the assembly area is started, the bearing and the spindle surface are scanned, and 48 spatial coordinate points are collected. Through the spatial geometric least squares method, the system establishes an assembly evaluation matrix and calculates the initial relative position deviation value of the bearing and the shaft as follows: distance deviation δ d = 0.235mm, angle deviation δ θ = 0.036°. The mechanical arm adjusts the bearing position according to the optimal trajectory, and after three fine adjustments, the final position deviation is reduced to δ d = 0.003mm, and the angle deviation is reduced to δ θ = 0.002°, which meets the assembly conditions. The positioning process data is shown in Table 3:

[0195] Table 3 Adjustment data table of bearing positioning process

[0196] Adjustment times Distance deviation (mm) Angle deviation (°) Whether the assembly condition is met Initial state 0.235 0.036 No First adjustment 0.064 0.012 No Second adjustment 0.018 0.005 No Third adjustment 0.003 0.002 Yes

[0197] After reaching the positioning requirements, the system calculates the best pressing parameters: the contact surface friction coefficient μ = 0.22, and the equivalent elastic modulus E = 1.05 × 10 5 MPa. According to the assembly force optimization function, a detailed force-displacement curve is generated, the initial pressure of the pressing is set to 3.86kN, the maximum pressure in the middle stage is 16.24kN, and the final pressure is 12.58kN. The pressing speed is set to 0.4mm / s at the beginning and end, and the maximum speed in the middle stage is 1.2mm / s.

[0198] During the pressing process, the pressure sensor monitors the change of the assembly force in real time. When the pressing depth is 29.6mm, a sudden change in pressure is detected, with a change rate of 23.5% / mm, which exceeds the threshold of 20% / mm. The system immediately suspends the operation and alarms, and after inspection, it is found that there is a small foreign matter in the assembly area. After removing the foreign matter, the system is restarted and the remaining pressing process is successfully completed.

[0199] After the assembly is completed, the cooling system is started, the cooling air speed is set to 4.2m / s, and a ring-shaped air flow field is formed to surround the bearing. The temperature gradient is controlled at 4.3℃ / min, and the temperature difference of each point of the bearing is kept within 7.8℃. The cooling process lasts for 11 minutes, and the bearing temperature drops to 26℃, forming a stable interference fit.

[0200] The assembled bearing system was tested by a rotating test unit, and was tested at five rotating speeds of 540 rpm, 1620 rpm, 2700 rpm, 3780 rpm and 5400 rpm in turn, and each rotating speed was maintained for 30 seconds. The torque sensor measured that the rotating resistance of the bearing was 0.32 Nm, 0.36 Nm, 0.42 Nm, 0.48 Nm and 0.56 Nm respectively, all within the rated value (0.6 Nm). The vibration frequency spectrum data collected by the vibration sensor was analyzed by fast Fourier transform, and the amplitude of each frequency component did not exceed 132% of the reference value, meeting the qualified standard of 150%.

[0201] Finally, the system constructed an assembly quality evaluation model based on a multi-dimensional matrix analysis method. According to the assembly basic matrix, the assembly variation matrix and the rotation harmonic variation matrix, the harmonic variation factor σ = 0.068 was calculated, which was within the qualified range of 0.05 to 0.15. Through principal component analysis and fuzzy comprehensive evaluation, the final assembly quality score was 86.4 points, belonging to the "good" level of assembly quality.

[0202] Compared with the traditional manual heating assembly, the system has significant improvement in assembly precision, efficiency and stability. The traditional method relies on the experience of workers to judge the heating temperature and time, and the temperature control error is usually ±10℃, which leads to uneven thermal expansion of the inner diameter of the bearing; positioning mainly relies on visual observation and manual adjustment, and the positioning accuracy is generally about ±0.1mm; the pressing process lacks force control, which easily causes damage to the bearing. The automatic heating and positioning assembly system of the application reduces the temperature control error to within ±2℃, improves the positioning accuracy to the micron level, and accurately and reliably controls the pressing force, significantly reducing the assembly defect rate. In addition, the whole assembly process time is shortened from 45 minutes to 18 minutes, greatly improving the production efficiency, and the consistency of the assembly quality is also guaranteed, which realizes a substantial improvement in the assembly technology level of high-precision bearings.

[0203] It should be noted that the variables involved in the application are explained in detail as shown in Table 4.

[0204] Table 4 Variable Explanation Table

[0205]

[0206]

[0207] The above is only a specific embodiment of the application, but the protection scope of the application is not limited thereto, and any person skilled in the art can easily think of changes or replacements within the technical range disclosed by the application, which should be covered within the protection scope of the application.

Claims

1. A linkage assembly method for automatically heating and positioning bearings, characterized in that: include: Set assembly parameters according to bearing model parameters; perform pre-scanning with a 3D vision sensor to determine the optimal assembly position; The bearing is heated in all directions using an induction heater, and the thermal expansion value of the bearing inner diameter is calculated in real time using the thermal expansion equation. At the same time, an infrared temperature sensor monitors the bearing temperature distribution to keep it within the thermal expansion threshold range. When the bearing temperature reaches the target temperature and the thermal expansion value of the bearing inner diameter meets the assembly requirements, the robotic arm is triggered to transfer the bearing to the assembly area. The bearing is assembled in the assembly area. After assembly, the bearing assembly part is evenly cooled to ensure a stable interference fit between the bearing and the long shaft parts. After assembly, the assembly quality is inspected.

2. A linkage assembly method for automatic heating and positioning of bearings according to claim 1, characterized in that: The assembly parameter settings include heating temperature range, heating time, assembly pressure control value, positioning accuracy requirements and linkage assembly speed.

3. The method for automatically heating and positioning a bearing according to claim 2, characterized in that: The input of the thermal expansion equation includes the linear expansion coefficient of the bearing material, the initial bearing temperature, the heating target temperature, the preset temperature, the initial inner diameter size of the bearing, and the thermal conductivity coefficient of the bearing material. The output is the actual inner diameter size of the bearing at the preset temperature and the thermal expansion value of the bearing inner diameter.

4. A linkage assembly method for automatic heating and positioning of bearings according to claim 3, characterized in that: The steps for assembling bearings in the assembly area are specifically: a laser positioning system collects multi-point coordinate data of the bearing and the long axis part surface, applies spatial geometric least squares method to construct an assembly evaluation matrix, and controls the robotic arm to perform optimal trajectory positioning based on the assembly evaluation matrix.

5. The method for automatically heating and positioning a bearing according to claim 4, characterized in that: The assembly assessment matrix is ​​used to evaluate the coaxiality and perpendicularity of the bearing and the long shaft part, providing a precise positioning basis for the assembly process.

6. The method for automatically heating and positioning a bearing according to claim 5, characterized in that: The step of assembling the bearing in the assembly area also includes: when the relative position deviation value is less than a preset threshold, the system applies the assembly force optimization function to calculate the optimal press-fitting parameters, and the servo press-fitting mechanism performs the bearing pressing operation according to the force displacement curve.

7. A linkage assembly method for automatic heating and positioning of bearings according to claim 6, characterized in that: The relative position deviation value refers to the distance and angle difference between the center axis of the bearing and the center axis of the long shaft part, which is calculated by the assembly evaluation matrix and is used to determine whether the bearing and the long shaft part meet the assembly conditions.

8. The linkage assembly method for automatically heating and positioning bearings according to claim 7, characterized in that: The input of the assembly force optimization function includes the interference fit between the bearing and the long shaft part, the elastic modulus of the bearing material, the elastic modulus of the long shaft part material, the target assembly depth and the contact surface friction coefficient. The output is the optimal pressure value and pressing speed corresponding to each displacement point during the pressing process.

9. The linkage assembly method for automatically heating and positioning bearings according to claim 8, characterized in that: The step of assembling the bearing in the assembly area also includes: during the press-fitting process, the pressure sensor continuously monitors the assembly force curve, and if an abnormal force change is detected, the system immediately pauses and issues an alarm.

10. A linkage assembly method for automatic heating and positioning of bearings according to claim 9, characterized in that: The assembly force curve refers to the actual pressure-displacement relationship diagram recorded during the bearing press-fitting process, which is calculated by the assembly force optimization function and is used to guide the servo press-fitting mechanism to perform precise press-fitting operations.

Images