Deep groove ball bearing ring grinding error compensation method and system

Abstract

The invention belongs to the technical field of numerical control machine tool control, and discloses a deep groove ball bearing ring grinding error compensation method and system, and the method comprises the steps: monitoring the temperature of a numerical control grinding machine body and a main shaft in real time through a plurality of temperature sensors, and obtaining the temperature field distribution data of a numerical control grinding machine; according to the numerical control grinding machine temperature field distribution data, the temperature gradient of each temperature measurement position is calculated; comparing each temperature gradient obtained by calculation with a preset temperature gradient threshold value, and judging whether a temperature measurement position with the temperature gradient exceeding the temperature gradient threshold value exists or not; when the temperature measurement positions with the temperature gradients exceeding the temperature gradient threshold exist, error compensation parameters of the numerical control system are determined according to the temperature gradients of all the temperature measurement positions, and the grinding machining track and grinding parameters are corrected; therefore, the grinding precision of the deep groove ball bearing ring can be improved.

Description

Technical Field

[0001] The present application relates to the field of CNC machine tool control technology, and in particular to a deep groove ball bearing ring grinding error compensation and system. Background Art

[0002] On automated production lines for automotive parts, the grinding accuracy of thin-walled rings in deep groove ball bearings is a key factor in ensuring bearing performance. However, the thermal deformation generated by CNC grinders during long-term continuous operation has become a major bottleneck restricting improvements in grinding accuracy. Traditional thermal error compensation techniques, such as those that rely on ambient temperature compensation or offline error modeling, often struggle to cope with the complex and dynamic thermal deformation of the grinder itself. Ambient temperature compensation strategies struggle to accurately capture the influence of internal heat sources in the grinder, while offline modeling methods struggle to effectively adapt to the dynamic characteristics of thermal deformation under long-term continuous operation.

[0003] While some existing thermal error compensation methods based on multi-point temperature monitoring exist, these methods typically rely on building complex mathematical models to characterize the complex relationship between the temperature field and thermal deformation. These methods are essentially indirect compensation methods, and their effectiveness is limited by the accuracy of thermal deformation prediction. Prediction errors directly impact the ultimate compensation results.

[0004] In the specific application scenario of large-scale automated bearing production lines for automotive parts, production line cycle requirements are extremely demanding, operators' professional skills are relatively limited, and prolonged continuous operation of CNC grinders can easily lead to significant heat accumulation and thermal deformation of key components. Therefore, designing a method that can quickly, directly, and effectively suppress the adverse effects of thermal deformation on grinding accuracy while avoiding complex modeling and big data analysis has become a key technical challenge in ensuring the grinding accuracy of thin-walled rings and the efficient operation of the production line. Effectively solving this problem will not only significantly improve product quality and significantly reduce scrap rates, but also comprehensively enhance the overall efficiency and economic benefits of the automated production line.

[0005] In view of the above problems, the existing technology is in urgent need of improvement. Summary of the Invention

[0006] The purpose of this application is to provide a deep groove ball bearing ring grinding error compensation and system, which can improve the grinding accuracy of deep groove ball bearing rings.

[0007] In a first aspect, the present application provides a deep groove ball bearing ring grinding error compensation method for an automated bearing production line CNC grinder, the method comprising the following steps:

[0008] S1. Use multiple temperature sensors to monitor the temperature of the CNC grinder bed and spindle in real time and obtain temperature field distribution data of the CNC grinder.

[0009] S2. Calculate the temperature gradient at each temperature measurement location based on the temperature field distribution data of the CNC grinder; the temperature gradient represents the local temperature rise rate of the grinder;

[0010] S3. The calculated temperature gradients are compared with the preset temperature gradient threshold to determine whether there is a temperature gradient exceeding the temperature gradient threshold at the temperature measurement location;

[0011] S4. When there is a temperature measurement position where the temperature gradient exceeds the temperature gradient threshold, determine the error compensation parameters of the numerical control system according to the temperature gradient of each temperature measurement position, and correct the grinding trajectory and grinding parameters.

[0012] Furthermore, the present application also proposes that step S1 includes:

[0013] S101. By means of multiple temperature sensors provided at key points of thermal deformation on the CNC grinding machine bed and the spindle, the temperature of the CNC grinding machine bed and the spindle is monitored in real time to obtain preliminary temperature data; the key points of thermal deformation are predetermined positions that can effectively reflect the thermal deformation situation;

[0014] S102. Use the Kalman filter algorithm to filter the preliminary temperature data to obtain the temperature field distribution data of the CNC grinding machine.

[0015] Furthermore, the present application also proposes that step S2 includes:

[0016] S201. Set multiple time windows of different lengths, calculate the temperature gradient of each temperature measurement position within each time window, and obtain multiple temperature gradient values;

[0017] S202. For each temperature measurement location, calculate the variance of the temperature gradient values ​​under different time windows, and select the time window with the smallest variance as the optimal time window for the temperature measurement location;

[0018] S203. Calculate the temperature gradient at each temperature measurement location based on the optimal time window at each temperature measurement location and using the corresponding temperature field distribution data.

[0019] Furthermore, the present application also proposes that step S3 includes:

[0020] S301. Obtain historical temperature gradient data for each temperature measurement location and calculate the temperature gradient change rate for each temperature measurement location;

[0021] S302. Based on the temperature gradient change rate of each temperature measurement position, a clustering algorithm is used to divide each temperature measurement position into multiple thermal deformation regions, so that the temperature measurement positions within each thermal deformation region have similar temperature gradient change trends;

[0022] S303. For each thermal deformation region, calculate the average temperature gradient of all temperature measurement positions within the thermal deformation region, and determine whether there is a thermal deformation region with an average temperature gradient exceeding a preset temperature gradient threshold;

[0023] S304. When there is a thermal deformation region where the average temperature gradient exceeds the preset temperature gradient threshold, it is determined that there is a temperature measurement position where the temperature gradient exceeds the temperature gradient threshold.

[0024] Furthermore, the present application also proposes that step S301 includes:

[0025] Acquire temperature gradient data of each temperature measurement position periodically acquired since the start of operation of the CNC grinding machine to form a historical temperature gradient data set of each temperature measurement position;

[0026] For each temperature measurement location, the temperature gradient data of the corresponding historical temperature gradient data set is fitted using the least squares method to obtain a fitting curve of the temperature gradient changing with time;

[0027] According to the fitting curve, the temperature gradient change rate of each temperature measurement position at the current moment is calculated.

[0028] Furthermore, the present application also proposes that step S302 includes:

[0029] Constructing a multidimensional feature vector for the temperature gradient change rate at each temperature measurement position, wherein the multidimensional feature vector includes the magnitude, direction, and frequency of the temperature gradient change rate;

[0030] The K-means clustering algorithm is used to divide each temperature measurement location into multiple initial thermal deformation regions based on the multidimensional feature vector of each temperature measurement location.

[0031] The Euclidean distance between the multidimensional feature vectors of each temperature measurement position in each initial thermal deformation area is calculated. If the maximum Euclidean distance exceeds the preset distance threshold, the temperature measurement position farthest from the cluster center in the initial thermal deformation area is divided into a new thermal deformation area, and the cluster centers of the remaining initial thermal deformation areas are recalculated. This step is repeated until the maximum Euclidean distance of all initial thermal deformation areas is less than the preset distance threshold, thereby obtaining the final thermal deformation area division result.

[0032] Furthermore, the present application also proposes that step S303 includes:

[0033] For each thermal deformation region, obtain the temperature gradient data of all temperature measurement positions within the thermal deformation region, and calculate the weighted average of each temperature gradient data according to the thermal deformation sensitivity of each temperature measurement position. The higher the thermal deformation sensitivity, the larger the corresponding weight coefficient.

[0034] The weighted average temperature gradient of each thermal deformation region is compared with a preset regional temperature gradient threshold. If the weighted average temperature gradient exceeds the corresponding regional temperature gradient threshold, the corresponding thermal deformation region is determined to be a thermal deformation region whose average temperature gradient exceeds the preset temperature gradient threshold.

[0035] Furthermore, the present application also proposes that step S4 includes:

[0036] S401. For each thermal deformation region, calculate the comprehensive temperature gradient of the thermal deformation region based on the temperature gradient and thermal deformation influence weight at each temperature measurement position within the thermal deformation region; the thermal deformation influence weight is pre-determined based on the degree of influence of each temperature measurement position on the grinding accuracy of the deep groove ball bearing ring, with the greater the influence, the higher the weight;

[0037] S402. According to the integrated temperature gradient of each thermal deformation area, based on the mapping relationship between the integrated temperature gradient obtained in advance and the CNC system error compensation parameters, the CNC system error compensation parameters corresponding to each thermal deformation area are determined; the CNC system error compensation parameters include the tool path offset and the grinding parameter adjustment amount;

[0038] S403. Correct the grinding trajectory and grinding parameters according to the determined CNC system error compensation parameters.

[0039] Furthermore, the present application also proposes that in step S403, the step of correcting the grinding trajectory includes:

[0040] The tool trajectory offset is used as the tool trajectory correction value, and the correction value is low-pass filtered using a Butterworth filter to obtain a smooth tool trajectory correction value;

[0041] The smooth tool trajectory correction amount is superimposed on the original tool trajectory of the CNC system to obtain a corrected grinding trajectory, and the CNC grinding machine is controlled to perform grinding according to the corrected grinding trajectory.

[0042] In a second aspect, the present application also proposes a deep groove ball bearing ring grinding error compensation system for use on a CNC grinder in an automated bearing production line. The system includes multiple temperature sensors, a data collector, and an industrial computer. Each temperature sensor is electrically connected to the data collector, which is in turn electrically connected to the industrial computer.

[0043] Multiple temperature sensors are distributed on the CNC grinding machine bed and spindle to monitor the temperature of the CNC grinding machine bed and spindle in real time and obtain the temperature field distribution data of the CNC grinding machine;

[0044] The data collector is used to send the temperature field distribution data of the CNC grinder to the industrial computer;

[0045] An industrial computer is installed with a compensation control program, wherein the compensation control program is configured with a gradient calculation module, a threshold comparison module and a correction module;

[0046] The gradient calculation module is used to calculate the temperature gradient of each temperature measurement position based on the temperature field distribution data of the CNC grinder; the temperature gradient represents the local temperature rise rate of the grinder;

[0047] The threshold comparison module is used to compare each calculated temperature gradient with a preset temperature gradient threshold to determine whether there is a temperature measurement position where the temperature gradient exceeds the temperature gradient threshold;

[0048] The correction module is used to determine the error compensation parameters of the CNC system and correct the grinding trajectory and grinding parameters according to the temperature gradient of each temperature measurement position when there is a temperature measurement position where the temperature gradient exceeds the temperature gradient threshold.

[0049] Beneficial effect: The deep groove ball bearing ring grinding error compensation method and system provided in this application solves the problem of reduced grinding accuracy caused by thermal deformation of CNC grinders in the prior art by monitoring temperature gradients and performing error compensation, and has the advantage of improving the grinding accuracy of deep groove ball bearing rings. BRIEF DESCRIPTION OF THE DRAWINGS

[0050] Figure 1 This is a flow chart of the deep groove ball bearing ring grinding error compensation method provided in an embodiment of the present application.

[0051] Figure 2 This is a schematic structural diagram of the deep groove ball bearing ring grinding error compensation system provided in an embodiment of the present application.

[0052] Explanation of reference numerals: 1. Temperature sensor; 2. Data collector; 3. Industrial computer; 301. Gradient calculation module; 302. Threshold comparison module; 303. Correction module. DETAILED DESCRIPTION

[0053] The technical solutions in this application will be clearly and completely described below in conjunction with the drawings in this application. Obviously, the described embodiments are only a part of the embodiments of this application, rather than all of the embodiments. The components of the present application generally described and shown in the drawings here can be arranged and designed in various different configurations. Therefore, the following detailed description of the embodiments of the present application provided in the drawings is not intended to limit the scope of the application for which protection is claimed, but merely represents selected embodiments of the present application. Based on the embodiments of the present application, all other embodiments obtained by those skilled in the art without making creative work are within the scope of protection of this application.

[0054] It should be noted that similar reference numerals and letters represent similar items in the following drawings. Therefore, once an item is defined in one drawing, it does not need to be further defined or explained in subsequent drawings. At the same time, in the description of this application, the terms "first", "second", etc. are only used to distinguish the description and should not be understood as indicating or implying relative importance.

[0055] refer to Figure 1 This application proposes a deep groove ball bearing ring grinding error compensation method for an automated bearing production line CNC grinder. The method comprises the following steps:

[0056] S1. Use multiple temperature sensors to monitor the temperature of the CNC grinder bed and spindle in real time and obtain temperature field distribution data of the CNC grinder.

[0057] S2. Calculate the temperature gradient at each temperature measurement location based on the temperature field distribution data of the CNC grinder; the temperature gradient represents the local temperature rise rate of the grinder;

[0058] S3. The calculated temperature gradients are compared with the preset temperature gradient threshold to determine whether there is a temperature gradient exceeding the temperature gradient threshold at the temperature measurement location;

[0059] S4. When there is a temperature measurement position where the temperature gradient exceeds the temperature gradient threshold, determine the error compensation parameters of the numerical control system according to the temperature gradient of each temperature measurement position, and correct the grinding trajectory and grinding parameters.

[0060] Among them, in step S1, multiple temperature sensors are arranged at different positions of the CNC grinding machine bed and the spindle to collect temperature information during the operation of the grinder. These temperature sensors can be thermocouples, thermistors or infrared sensors (for example, for a spindle that needs to rotate, an infrared sensor can be used for temperature measurement), etc.

[0061] In step S2, the temperature gradient is calculated by analyzing the temperature changes at the temperature measurement positions at adjacent or similar time points. For example, the temperature gradient can be calculated using the differential method, that is, the temperature at the previous moment is subtracted from the current moment's temperature, and then divided by the time interval to obtain the rate of change of temperature over time.

[0062] Among them, in step S3, the temperature gradient threshold is pre-set, which represents the upper limit of the temperature change rate allowed by the grinder. The temperature gradient threshold can be determined through experiments or simulation analysis based on factors such as the material properties, structural characteristics and processing accuracy requirements of the grinder.

[0063] Among them, in step S4, the error compensation parameters may include the tool trajectory offset, the grinding feed speed adjustment, the grinding wheel speed adjustment, etc. The determination of the error compensation parameters can be based on the mapping relationship between the temperature gradient and the grinding error. The mapping relationship can be obtained in advance through experimental calibration or finite element analysis. The correction of the grinding process trajectory is achieved by controlling the movement of the tool through the CNC system to offset the grinding error caused by thermal deformation.

[0064] Specifically, this method first uses a temperature sensor network to perceive the temperature distribution status of the CNC grinder bed and spindle in real time, and obtains the dynamic change information of the grinder temperature field. The temperature gradient calculation link converts the absolute value of the temperature field into a gradient value that characterizes the local temperature rise rate. The gradient value can more directly reflect the severity and potential risk of thermal deformation. By setting the temperature gradient threshold, a mechanism is established to quickly determine whether thermal error compensation is needed. When it is detected that the temperature gradient exceeds the threshold, error compensation is immediately started. According to the size and distribution of the temperature gradient, the error compensation parameters of the CNC system are adaptively adjusted, and then the grinding process trajectory and grinding parameters are corrected to achieve real-time suppression of thermal deformation errors. The entire process does not require the construction of a complex and precise mathematical model between the temperature field and thermal deformation, nor does it rely on a large amount of offline experimental data. It simplifies the implementation difficulty of thermal error compensation, improves the response speed and compensation efficiency, and is more suitable for the dual requirements of beat and precision of automated production lines.

[0065] In some embodiments, step S1 includes:

[0066] S101. By means of multiple temperature sensors provided at key points of thermal deformation on the CNC grinding machine bed and the spindle, the temperature of the CNC grinding machine bed and the spindle is monitored in real time to obtain preliminary temperature data; the key points of thermal deformation are predetermined positions that can effectively reflect the thermal deformation situation;

[0067] S102. Use the Kalman filter algorithm to filter the preliminary temperature data to obtain the temperature field distribution data of the CNC grinding machine.

[0068] Among them, in step S101, the key points of thermal deformation can be determined in advance based on the heat source distribution density and thermal deformation sensitivity. Specifically, the heat source distribution density can be determined by analyzing the grinder structure, moving parts and heat generation during the processing. The thermal deformation sensitivity can be determined by finite element analysis or experimental measurement to evaluate the degree of influence of temperature changes at different positions on the key dimensions and processing accuracy of the grinder. The above analysis process can be completed manually or using AI. By arranging sensors at these key points, temperature changes that significantly affect grinding accuracy can be more effectively captured, and preliminary temperature data that can better reflect the actual thermal deformation situation can be obtained.

[0069] In step S102, the Kalman filter algorithm is a recursive state estimation algorithm that can estimate the true state of the system from measurement data containing noise. Specifically, the Kalman filter algorithm establishes a state-space model of the system, combines the system's dynamic model and measurement model, and uses the optimal estimate at the previous moment and the measured value at the current moment to calculate the optimal estimate at the current moment. In this application, the system state can be defined as the actual temperature at each temperature measurement location, and the measured value is the preliminary temperature data collected by the sensor. Kalman filtering can effectively filter out random noise and interference that may be introduced during the sensor measurement process, thereby improving the accuracy and reliability of temperature data.

[0070] Thus, step S101 obtains preliminary temperature data at key positions, and the data at these positions are more representative of the actual thermal deformation situation. Step S102 filters these preliminary data to eliminate noise and improve the accuracy of the data. The combination of these two steps makes the obtained temperature field distribution data of the CNC grinder more effective and accurate. These optimized and accurate temperature field distribution data serve as the basis for subsequent temperature gradient calculation and error compensation, directly improving the reliability and accuracy of subsequent compensation links. Compared with relying solely on raw data or collecting data at non-critical positions, this method that combines optimized sensor layout and filtering processing provides high-quality input for the entire error compensation process, thereby improving the final grinding accuracy.

[0071] Specifically, this solution first uses multiple temperature sensors to monitor the temperatures of the CNC grinder's bed and spindle in real time. These sensors are not randomly placed but, after prior analysis, are located at key thermal deformation points. These key points are determined based on the distribution of heat sources within the grinder and the degree to which temperature changes at different locations affect thermal deformation and machining accuracy. For example, locations near the spindle bearings and the bed guide rails are typically sensitive to thermal deformation, and sensors are preferentially placed in these locations. The raw temperature data collected by the sensors constitutes preliminary temperature data. Due to factors such as sensor measurement and environmental interference, preliminary temperature data may contain noise. To improve data reliability, the preliminary temperature data is processed using a Kalman filter algorithm. The Kalman filter algorithm effectively estimates temperature data closer to the true value from noisy measurement data, resulting in smooth and accurate temperature field distribution data for the CNC grinder. This temperature field distribution data accurately reflects the temperature state of key parts of the grinder, providing reliable input for subsequent temperature gradient calculations. By optimizing the sensor position and performing data filtering, this solution solves the problem of inaccurate or invalid raw temperature data, ensuring the accuracy of subsequent temperature gradient calculation and error compensation, thereby improving the accuracy of the grinding process.

[0072] Through the above technical solution, the present application improves the accuracy and effectiveness of the temperature field distribution data of the CNC grinder, provides a more reliable basis for subsequent temperature gradient calculation and error compensation, and thus improves the accuracy of the grinding process.

[0073] Furthermore, the present application also proposes that step S2 includes:

[0074] S201. Set multiple time windows of different lengths, calculate the temperature gradient of each temperature measurement position within each time window, and obtain multiple temperature gradient values;

[0075] S202. For each temperature measurement location, calculate the variance of the temperature gradient values ​​under different time windows, and select the time window with the smallest variance as the optimal time window for the temperature measurement location;

[0076] S203. Calculate the temperature gradient at each temperature measurement location based on the optimal time window at each temperature measurement location and using the corresponding temperature field distribution data.

[0077] In step S201, the time window length can be set to different lengths, such as 1 second, 5 seconds, or 10 seconds, to capture temperature changes at different time scales. For each temperature sensor's measurement location, within each set time window, a temperature gradient value is calculated using the temperature data within that time window. The temperature gradient can be calculated, for example, by subtracting the temperature value at the beginning of the time window from the temperature value at the end of the time window, and then dividing the result by the length of the time window to obtain the average temperature gradient within the time window; or by performing a straight line fit on the temperature values ​​within the time window, and using the slope of the fitted line as the temperature gradient value.

[0078] Among them, in step S202, for each temperature measurement position, multiple temperature gradient values ​​obtained in step S201 in different time windows are analyzed, and windows of the same length are retrieved, and multiple temperature gradient values ​​recently obtained, including the temperature gradient value currently calculated, are retrieved to perform variance statistics. Specifically, the statistical variance of these temperature gradient values ​​is calculated. The variance reflects the degree of fluctuation of the temperature gradient value in different time windows. The smaller the variance, the more stable the temperature gradient value in the time window. The time window with the smallest variance is selected as the optimal time window for the temperature measurement position, which means that in this time window, the calculation result of the temperature gradient is the most reliable and is least affected by noise or random factors.

[0079] Among them, in step S203, after the optimal time window of each temperature measurement position is determined in step S202, the temperature field distribution data corresponding to the optimal time window is reused to recalculate the temperature gradient of each temperature measurement position to obtain the final temperature gradient value used for subsequent error compensation.

[0080] This method can adaptively select the optimal time window for each temperature measurement location, more accurately calculating the temperature gradient at each location and providing a more precise data foundation for subsequent temperature gradient-based error compensation. This approach can better adapt to the differences in thermal response characteristics of different parts of a CNC grinder, improving the accuracy and adaptability of temperature gradient calculations, thereby enhancing the effectiveness of error compensation and grinding precision.

[0081] In some embodiments, step S3 includes:

[0082] S301. Obtain historical temperature gradient data for each temperature measurement location and calculate the temperature gradient change rate for each temperature measurement location;

[0083] S302. Based on the temperature gradient change rate of each temperature measurement position, a clustering algorithm is used to divide each temperature measurement position into multiple thermal deformation regions, so that the temperature measurement positions within each thermal deformation region have similar temperature gradient change trends;

[0084] S303. For each thermal deformation region, calculate the average temperature gradient of all temperature measurement positions within the thermal deformation region, and determine whether there is a thermal deformation region with an average temperature gradient exceeding a preset temperature gradient threshold;

[0085] S304. When there is a thermal deformation region where the average temperature gradient exceeds the preset temperature gradient threshold, it is determined that there is a temperature measurement position where the temperature gradient exceeds the temperature gradient threshold.

[0086] In step S301, historical temperature gradient data can be periodically acquired and stored in a database starting from the start of the CNC grinding machine operation. A variety of methods can be used to calculate the temperature gradient change rate, such as using the finite difference method to calculate the difference between the temperature gradient at the current moment and the previous moment and then dividing it by the time interval, or using the least squares method to fit a curve showing the temperature gradient changing over time and then taking the derivative of the fitted curve at the current moment to obtain the temperature gradient change rate.

[0087] In step S302, a clustering algorithm is used to analyze the temperature gradient change rate at each temperature measurement location and group locations with similar change trends into the same thermal deformation region. The clustering algorithm can be a K-means algorithm or other suitable clustering methods. Before clustering, feature extraction can be performed on the temperature gradient change rate. For example, features such as the magnitude, direction, and frequency of the change rate can be extracted to construct a multidimensional feature vector to improve clustering effectiveness.

[0088] In step S303, the average temperature gradient is obtained by calculating the average or weighted average of the temperature gradients at all temperature measurement locations within the thermal deformation region. The preset temperature gradient threshold is pre-set based on the thermal characteristics and grinding accuracy requirements of the CNC grinder.

[0089] In step S304, when the average temperature gradient of a certain thermal deformation area exceeds a threshold, the system determines that there is a significant thermal deformation risk in the area and error compensation is required.

[0090] Specifically, during the grinding process of deep groove ball bearing rings, multiple temperature sensors are installed on the CNC grinding machine's bed and spindle to monitor the temperature at each location in real time. The compensation control program periodically reads this temperature data and calculates the current temperature gradient at each temperature measurement location. To determine whether error compensation should be initiated, the program first obtains historical temperature gradient data for each temperature measurement location, such as a series of temperature gradients over a period of time. The program then calculates the rate of change of the temperature gradient at each temperature measurement location, characterizing how quickly the temperature gradient changes over time. Based on these rates of change, the program uses a K-means clustering algorithm to divide the temperature measurement locations into several thermal deformation regions, ensuring that temperature measurement locations within the same region have similar temperature gradient trends. For each thermal deformation region, the program calculates the average temperature gradient of all temperature measurement locations within that region and compares this average temperature gradient with a preset temperature gradient threshold. If the average temperature gradient of any thermal deformation region exceeds the threshold, the CNC grinder is deemed to have significant thermal deformation and requires error compensation. This allows the system to comprehensively consider temperature gradient trends and regional correlations, more accurately determining thermal deformation states, avoiding misjudgments and providing a reliable basis for subsequent error compensation. In practice, localized hotspots—specific areas where hot spots can cause abnormally high temperature gradients—may have a minimal impact on overall grinding accuracy. Applying blanket compensation could introduce new errors. Using the thermal deformation region as a whole for threshold determination effectively addresses this issue.

[0091] Preferably, step S301 may include:

[0092] Acquire temperature gradient data of each temperature measurement position periodically acquired since the start of operation of the CNC grinding machine to form a historical temperature gradient data set of each temperature measurement position;

[0093] For each temperature measurement location, the temperature gradient data of the corresponding historical temperature gradient data set is fitted using the least squares method to obtain a fitting curve of the temperature gradient changing with time;

[0094] According to the fitting curve, the temperature gradient change rate of each temperature measurement position at the current moment is calculated.

[0095] Among them, periodic acquisition refers to the cyclic collection of temperature gradient data at preset time intervals. The historical temperature gradient data set is used to store a collection of temperature gradient values ​​periodically recorded at each temperature measurement position since the CNC grinder started running. Least squares fitting is used to process the historical temperature gradient data set for each temperature measurement position. Through the least squares method, a best fit curve can be obtained, which can reflect the overall trend of temperature gradient changes over time and reduce the impact of data noise. The fitting curve represents the functional relationship of temperature gradient changes over time. The temperature gradient change rate is the derivative of the fitting curve at the current moment, which indicates the speed and direction (increase or decrease) of the temperature gradient change at the current moment.

[0096] Specifically, by periodically collecting and recording temperature gradient data, the system accumulates historical temperature information for each temperature measurement location. For each temperature measurement location, the least squares method is used to fit this historical data, aiming to develop a curve model that smoothly reflects the temperature gradient's changing trend. This curve model effectively filters out noise interference and highlights the regularity of temperature gradient changes. After obtaining the fitted curve, the slope of the curve at the current point in time is calculated to determine the temperature gradient change rate. The temperature gradient change rate can sensitively reflect whether the current temperature gradient is accelerating, decelerating, or leveling off. Compared to using only the current temperature gradient value, the temperature gradient change rate provides dynamic information on temperature gradient changes, enabling more accurate and timely identification of abnormal temperature gradient variations. This provides a more reliable basis for subsequent error compensation decisions, thereby improving the timeliness and accuracy of error compensation.

[0097] Furthermore, step S302 may include:

[0098] Constructing a multidimensional feature vector for the temperature gradient change rate at each temperature measurement position, wherein the multidimensional feature vector includes the magnitude, direction, and frequency of the temperature gradient change rate;

[0099] The K-means clustering algorithm is used to divide each temperature measurement location into multiple initial thermal deformation regions based on the multidimensional feature vector of each temperature measurement location.

[0100] The Euclidean distance between the multidimensional feature vectors of each temperature measurement position in each initial thermal deformation area is calculated. If the maximum Euclidean distance exceeds the preset distance threshold, the temperature measurement position farthest from the cluster center in the initial thermal deformation area is divided into a new thermal deformation area, and the cluster centers of the remaining initial thermal deformation areas are recalculated. This step is repeated until the maximum Euclidean distance of all initial thermal deformation areas is less than the preset distance threshold, thereby obtaining the final thermal deformation area division result.

[0101] Among them, constructing a multidimensional feature vector for the temperature gradient change rate at each temperature measurement position can be specifically implemented as follows:

[0102] First, determine the rate of change of the temperature gradient. This can be done by analyzing historical temperature gradient data. For example, using the least squares method to fit a curve showing the temperature gradient changing over time, then calculating the slope of the fitted curve at the current moment and taking its absolute value.

[0103] Secondly, determine the direction of the temperature gradient change rate. The direction can refer to whether the temperature gradient change is positively increasing or negatively decreasing. For example, if the slope of the temperature gradient fitting curve at the current moment is positive, the direction is positively increasing, otherwise it is negatively decreasing;

[0104] Next, determine the frequency of the temperature gradient change rate. Frequency can characterize the speed of the periodic change of the temperature gradient. For example, historical temperature gradient data can be analyzed to calculate the periodicity of the temperature gradient change, or methods such as fast Fourier transform can be used to extract the main frequency components of the temperature gradient change.

[0105] Thus, through the above steps, a multidimensional feature vector including the magnitude, direction and frequency of the temperature gradient change rate can be constructed for subsequent cluster analysis.

[0106] The clustering process of the K-means clustering algorithm is an existing technology and will not be described in detail here.

[0107] Finally, to improve the accuracy of thermal deformation zone division, this solution also introduces an iterative optimization process. By calculating the Euclidean distance between the multidimensional feature vectors of each temperature measurement position within each initial thermal deformation zone, the degree of discreteness of the temperature gradient change characteristics within the zone is evaluated. If the maximum Euclidean distance within the zone exceeds the preset distance threshold, the zone division is considered inaccurate, and the temperature measurement position farthest from the cluster center needs to be divided out to form a new thermal deformation zone, and the cluster center calculation is performed again. This iterative optimization process ensures that the final thermal deformation zone division result is more reasonable and the thermal deformation characteristics of the temperature measurement positions within the zone are more consistent.

[0108] Through the above steps, the influence of accidental errors of individual temperature measurement positions on the thermal deformation area division results can be effectively reduced, the accuracy and reliability of the thermal deformation area division can be improved, and the foundation can be laid for subsequent error compensation based on the thermal deformation area.

[0109] Furthermore, step S303 may include:

[0110] For each thermal deformation region, obtain the temperature gradient data of all temperature measurement positions within the thermal deformation region, and calculate the weighted average of each temperature gradient data according to the thermal deformation sensitivity of each temperature measurement position. The higher the thermal deformation sensitivity, the larger the corresponding weight coefficient.

[0111] The weighted average temperature gradient of each thermal deformation region is compared with a preset regional temperature gradient threshold. If the weighted average temperature gradient exceeds the corresponding regional temperature gradient threshold, the corresponding thermal deformation region is determined to be a thermal deformation region whose average temperature gradient exceeds the preset temperature gradient threshold.

[0112] To address the issue of insufficient precision in the average temperature gradient calculation method, the embodiment introduces a thermal deformation sensitivity parameter. Specifically, in step S303, the thermal deformation sensitivity of each temperature measurement location must first be determined. Thermal deformation sensitivity can be determined in advance through various methods, such as experimental analysis, finite element simulation, or expert experience. For example, the impact of a unit temperature change at each temperature measurement location on grinding accuracy can be analyzed. The greater the impact, the higher the thermal deformation sensitivity of that location. After determining the thermal deformation sensitivity, when calculating the average temperature gradient of each thermal deformation region, the temperature gradient data of each temperature measurement location within that region are weighted averaged. The weighted average can be calculated by multiplying the temperature gradient value of each temperature measurement location by its corresponding thermal deformation sensitivity weight coefficient, summing all the products, and finally dividing by the sum of the weight coefficients to obtain a weighted average temperature gradient value. The weight coefficient is proportional to the thermal deformation sensitivity; that is, the higher the thermal deformation sensitivity of a location, the larger the corresponding weight coefficient. Therefore, during the weighted average process, changes in the temperature gradient at locations with high thermal deformation sensitivity have a greater impact on the final weighted average temperature gradient value. Through this weighted averaging method, the impact of temperature gradient changes at thermal deformation-sensitive locations on the overall temperature gradient level of the region can be more accurately reflected.

[0113] Among them, multiple divided areas can be set in advance, the regional temperature gradient threshold of each divided area can be determined, and the center point position of the corresponding divided area can be recorded. The regional temperature gradient threshold of each divided area can be pre-set according to factors such as the thermal characteristics of the grinder, processing accuracy requirements, and historical operation data; in the actual working process, according to the real-time thermal deformation area clustering results, the distance from the center point of each thermal deformation area to the center point of each divided area is calculated, and the regional temperature gradient threshold of the divided area corresponding to the minimum distance is used as the regional temperature gradient threshold corresponding to the thermal deformation area. If the weighted average temperature gradient of a certain thermal deformation area exceeds the corresponding regional temperature gradient threshold, the thermal deformation area is determined to be an area where the average temperature gradient exceeds the threshold. This judgment method takes into account the differences in thermal deformation sensitivity at different positions, and can more accurately identify thermal deformation areas that have a greater impact on grinding accuracy, providing a more reliable basis for subsequent error compensation.

[0114] In some embodiments, step S4 includes:

[0115] S401. For each thermal deformation region, calculate the comprehensive temperature gradient of the thermal deformation region based on the temperature gradient and thermal deformation influence weight at each temperature measurement position within the thermal deformation region; the thermal deformation influence weight is pre-determined based on the degree of influence of each temperature measurement position on the grinding accuracy of the deep groove ball bearing ring, with the greater the influence, the higher the weight;

[0116] S402. According to the integrated temperature gradient of each thermal deformation area, based on the mapping relationship between the integrated temperature gradient obtained in advance and the CNC system error compensation parameters, the CNC system error compensation parameters corresponding to each thermal deformation area are determined; the CNC system error compensation parameters include the tool path offset and the grinding parameter adjustment amount;

[0117] S403. Correct the grinding trajectory and grinding parameters according to the determined CNC system error compensation parameters.

[0118] Among them, in step S401, the calculation of the comprehensive temperature gradient can be specifically implemented as follows: first, determine the thermal deformation influence weight of each temperature measurement position. These weights are pre-set and determined based on the degree of influence of each temperature measurement position on the grinding accuracy of the deep groove ball bearing ring. For example, the grinding machine structure and thermal deformation characteristics can be analyzed to identify the key positions that have a greater impact on the grinding accuracy and assign higher weights. The weights of positions with a greater degree of influence are also set higher. Then, for each thermal deformation area, the temperature gradient data of all temperature measurement positions in the area are collected. Finally, each temperature gradient is multiplied by its corresponding thermal deformation influence weight by the weighted average method, and the products are summed to obtain the comprehensive temperature gradient of the thermal deformation area. This weighted average comprehensively considers the temperature change rate of each position and its degree of influence on the grinding accuracy.

[0119] In step S402, the error compensation parameter determination process can be performed as follows: a calibration experiment is performed in advance to establish a mapping relationship between the comprehensive temperature gradients of multiple pre-set divided areas and the error compensation parameters of the CNC system. In the calibration experiment, the temperature field of the grinding machine is manually controlled to generate different comprehensive temperature gradients in the divided areas, and the grinding errors under these gradients are measured. By analyzing the experimental data, a corresponding relationship between the comprehensive temperature gradient of each divided area and the tool path offset and grinding parameter adjustment can be established. For example, this mapping relationship can be represented by a lookup table or a fitting function. In the actual grinding process, after the comprehensive temperature gradient of each thermal deformation area is calculated, for each thermal deformation area, the divided area closest to the center of the thermal deformation area is used as the corresponding divided area of ​​the thermal deformation area. Then, a weighted calculation is performed on the comprehensive temperature gradients of all thermal deformation areas corresponding to the same divided area (the smaller the distance from the center of the divided area, the greater the weight). The comprehensive temperature gradient of the divided area is obtained. Based on the comprehensive temperature gradient of each divided area, the corresponding tool path offset and grinding parameter adjustment can be quickly obtained through table lookup or function calculation.

[0120] Among them, the tool trajectory offset can be the offset of the tool in the X, Y, and Z directions, which is used to correct the motion trajectory of the tool. The grinding parameter adjustment amount can include the adjustment amount of parameters such as the grinding wheel speed, feed speed, and grinding depth, which are used to optimize the grinding process and reduce the error caused by thermal deformation. When correcting the grinding processing trajectory, the following steps can be adopted: First, the determined tool trajectory offset and the smoothed grinding parameter adjustment amount are applied to the CNC system. The tool trajectory offset can be directly superimposed on the original tool trajectory program to achieve real-time correction of the tool trajectory. The grinding parameter adjustment amount can be sent to the parameter control module of the CNC system to adjust the grinding parameters in real time. As a result, the CNC grinder will perform processing according to the corrected tool trajectory and grinding parameters in subsequent grinding processes, thereby compensating for the grinding error caused by thermal deformation and improving processing accuracy.

[0121] As a result, the grinding error caused by thermal deformation of the CNC grinder can be quickly and directly compensated, thereby improving the grinding accuracy and production efficiency of deep groove ball bearing rings.

[0122] Preferably, in step S403, the step of correcting the grinding trajectory may include:

[0123] The tool trajectory offset is used as the correction value for calculating the tool trajectory, and the correction value is low-pass filtered using a Butterworth filter to obtain a smooth tool trajectory correction value.

[0124] The smooth tool trajectory correction amount is superimposed on the original tool trajectory of the CNC system to obtain a corrected grinding trajectory, and the CNC grinding machine is controlled to perform grinding according to the corrected grinding trajectory.

[0125] Among them, the tool trajectory offset is used as the correction amount of the tool trajectory, which represents the amplitude and direction of the tool trajectory that needs to be adjusted. In order to avoid the problem of trajectory mutation that may be caused by the direct application of the correction amount, the Butterworth filter is introduced to process the correction amount. Specifically, the Butterworth filter is configured as a low-pass filter, whose function is to filter out the high-frequency components in the correction amount and retain the low-frequency components, thereby obtaining a smooth correction amount output. The filtered smooth correction amount is then superimposed on the original tool trajectory of the CNC system to generate a corrected grinding process trajectory. The CNC grinding machine finally performs the actual grinding process operation according to this smoothed corrected trajectory. The specific parameters of the filter, such as the filter order and cutoff frequency, can be adjusted according to the actual application requirements and the dynamic characteristics of the system to optimize the filtering effect and smoothness.

[0126] Specifically, in order to solve the problem of sudden changes or oscillations in the tool trajectory correction, this solution adopts a low-pass filtering method. First, based on the error compensation parameters determined in the previous step, the correction amount of the tool trajectory at each control point is calculated. These correction amounts may contain high-frequency noise caused by changes in temperature gradients or other factors. In order to eliminate these high-frequency noises and ensure the smoothness of the tool trajectory, the calculated correction amount is input into the Butterworth low-pass filter. The Butterworth filter can effectively attenuate high-frequency signals and make the output signal smoother and more continuous. The filtered smoothed tool trajectory correction amount is superimposed on the original tool trajectory coordinates of the CNC system to obtain the final grinding trajectory. The CNC system controls the movement of the grinder according to the corrected grinding trajectory and performs the grinding process. Through low-pass filtering, the sudden changes and oscillations in the correction amount are eliminated, the smoothness of the tool trajectory is guaranteed, and the degradation of the machining surface quality and the loss of precision due to the unstable trajectory are avoided.

[0127] Among them, when correcting the grinding parameters, the calculated grinding parameter adjustment amount can be directly superimposed on the original grinding parameters.

[0128] refer to Figure 2 , the application also proposes a deep groove ball bearing ring grinding error compensation system for an automated bearing production line CNC grinder, the system comprising a plurality of temperature sensors 1, a data collector 2 and an industrial computer 3, each temperature sensor 1 is electrically connected to the data collector 2, and the data collector 2 is electrically connected to the industrial computer 3;

[0129] Multiple temperature sensors 1 are distributed on the CNC grinding machine bed and spindle, and are used to monitor the temperature of the CNC grinding machine bed and spindle in real time to obtain the temperature field distribution data of the CNC grinding machine (for details, please refer to step S1 above);

[0130] The data collector 2 is used to send the temperature field distribution data of the CNC grinding machine to the industrial computer 3;

[0131] The industrial computer 3 is installed with a compensation control program, which is configured with a gradient calculation module 301, a threshold comparison module 302 and a correction module 303;

[0132] The gradient calculation module 301 is used to calculate the temperature gradient at each temperature measurement position based on the temperature field distribution data of the CNC grinding machine; the temperature gradient represents the local temperature rise rate of the grinding machine (for details, please refer to step S2 above);

[0133] The threshold comparison module 302 is used to compare the calculated temperature gradients with a preset temperature gradient threshold to determine whether there is a temperature measurement location where the temperature gradient exceeds the temperature gradient threshold (for details, please refer to step S3 above);

[0134] The correction module 303 is used to determine the error compensation parameters of the CNC system according to the temperature gradient of each temperature measurement position when there is a temperature measurement position where the temperature gradient exceeds the temperature gradient threshold, and to correct the grinding process trajectory and grinding parameters (for details, please refer to step S4 above).

[0135] The above description is merely an embodiment of the present application and is not intended to limit the scope of protection of the present application. For those skilled in the art, various modifications and variations of the present application are possible. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present application shall be included in the scope of protection of the present application.

Claims

1. A deep groove ball bearing ring grinding error compensation method for an automated bearing production line CNC grinder, characterized in that: The steps of the method include: S1. Use multiple temperature sensors to monitor the temperature of the CNC grinder bed and spindle in real time and obtain temperature field distribution data of the CNC grinder. S2. Calculate the temperature gradient at each temperature measurement location based on the temperature field distribution data of the CNC grinder; the temperature gradient represents the local temperature rise rate of the grinder; S3. The calculated temperature gradients are compared with the preset temperature gradient threshold to determine whether there is a temperature gradient exceeding the temperature gradient threshold at the temperature measurement location; S4. When there is a temperature measurement position where the temperature gradient exceeds the temperature gradient threshold, determine the error compensation parameters of the numerical control system according to the temperature gradient of each temperature measurement position, and correct the grinding trajectory and grinding parameters.

2. A deep groove ball bearing ring grinding error compensation method according to claim 1, characterized in that: Step S1 includes: S101. By means of multiple temperature sensors provided at key points of thermal deformation on the CNC grinding machine bed and the spindle, the temperature of the CNC grinding machine bed and the spindle is monitored in real time to obtain preliminary temperature data; the key points of thermal deformation are predetermined positions that can effectively reflect the thermal deformation situation; S102. Use the Kalman filter algorithm to filter the preliminary temperature data to obtain the temperature field distribution data of the CNC grinding machine.

3. The method for compensating grinding errors of deep groove ball bearing rings according to claim 1, characterized in that: Step S2 includes: S201. Set multiple time windows of different lengths, calculate the temperature gradient of each temperature measurement position within each time window, and obtain multiple temperature gradient values; S202. For each temperature measurement location, calculate the variance of the temperature gradient values ​​under different time windows, and select the time window with the smallest variance as the optimal time window for the temperature measurement location; S203. Calculate the temperature gradient at each temperature measurement location based on the optimal time window at each temperature measurement location and using the corresponding temperature field distribution data.

4. A deep groove ball bearing ring grinding error compensation method according to claim 1, characterized in that: Step S3 includes: S301. Obtain historical temperature gradient data for each temperature measurement location and calculate the temperature gradient change rate for each temperature measurement location; S302. Based on the temperature gradient change rate of each temperature measurement position, a clustering algorithm is used to divide each temperature measurement position into multiple thermal deformation regions, so that the temperature measurement positions within each thermal deformation region have similar temperature gradient change trends; S303. For each thermal deformation region, calculate the average temperature gradient of all temperature measurement positions within the thermal deformation region, and determine whether there is a thermal deformation region with an average temperature gradient exceeding a preset temperature gradient threshold; S304. When there is a thermal deformation region where the average temperature gradient exceeds the preset temperature gradient threshold, it is determined that there is a temperature measurement position where the temperature gradient exceeds the temperature gradient threshold.

5. A deep groove ball bearing ring grinding error compensation method according to claim 4, characterized in that: Step S301 includes: Acquire temperature gradient data of each temperature measurement position periodically acquired since the start of operation of the CNC grinding machine to form a historical temperature gradient data set of each temperature measurement position; For each temperature measurement location, the temperature gradient data of the corresponding historical temperature gradient data set is fitted using the least squares method to obtain a fitting curve of the temperature gradient changing with time; According to the fitting curve, the temperature gradient change rate of each temperature measurement position at the current moment is calculated.

6. A deep groove ball bearing ring grinding error compensation method according to claim 4, characterized in that: Step S302 includes: Constructing a multidimensional feature vector for the temperature gradient change rate at each temperature measurement position, wherein the multidimensional feature vector includes the magnitude, direction, and frequency of the temperature gradient change rate; The K-means clustering algorithm is used to divide each temperature measurement location into multiple initial thermal deformation regions based on the multidimensional feature vector of each temperature measurement location. The Euclidean distance between the multidimensional feature vectors of each temperature measurement position in each initial thermal deformation area is calculated. If the maximum Euclidean distance exceeds the preset distance threshold, the temperature measurement position farthest from the cluster center in the initial thermal deformation area is divided into a new thermal deformation area, and the cluster centers of the remaining initial thermal deformation areas are recalculated. This step is repeated until the maximum Euclidean distance of all initial thermal deformation areas is less than the preset distance threshold, thereby obtaining the final thermal deformation area division result.

7. The method for compensating grinding errors of deep groove ball bearing rings according to claim 4, characterized in that: Step S303 includes: For each thermal deformation region, obtain the temperature gradient data of all temperature measurement positions within the thermal deformation region, and calculate the weighted average of each temperature gradient data according to the thermal deformation sensitivity of each temperature measurement position. The higher the thermal deformation sensitivity, the larger the corresponding weight coefficient. The weighted average temperature gradient of each thermal deformation region is compared with a preset regional temperature gradient threshold. If the weighted average temperature gradient exceeds the corresponding regional temperature gradient threshold, the corresponding thermal deformation region is determined to be a thermal deformation region whose average temperature gradient exceeds the preset temperature gradient threshold.

8. The method for compensating grinding errors of deep groove ball bearing rings according to claim 4, characterized in that: Step S4 includes: S401. For each thermal deformation region, calculate the comprehensive temperature gradient of the thermal deformation region based on the temperature gradient and thermal deformation influence weight at each temperature measurement position within the thermal deformation region; the thermal deformation influence weight is pre-determined based on the degree of influence of each temperature measurement position on the grinding accuracy of the deep groove ball bearing ring, with the greater the influence, the higher the weight; S402. According to the integrated temperature gradient of each thermal deformation area, based on the mapping relationship between the integrated temperature gradient obtained in advance and the CNC system error compensation parameters, the CNC system error compensation parameters corresponding to each thermal deformation area are determined; the CNC system error compensation parameters include the tool path offset and the grinding parameter adjustment amount; S403. Correct the grinding trajectory and grinding parameters according to the determined CNC system error compensation parameters.

9. A deep groove ball bearing ring grinding error compensation method according to claim 8, characterized in that: In step S403, the step of correcting the grinding trajectory includes: The tool trajectory offset is used as the tool trajectory correction value, and the correction value is low-pass filtered using a Butterworth filter to obtain a smooth tool trajectory correction value; The smooth tool trajectory correction amount is superimposed on the original tool trajectory of the CNC system to obtain a corrected grinding trajectory, and the CNC grinding machine is controlled to perform grinding according to the corrected grinding trajectory.

10. A deep groove ball bearing ring grinding error compensation system for use in a CNC grinding machine for an automated bearing production line, characterized in that: The system includes multiple temperature sensors, a data collector and an industrial computer. Each temperature sensor is electrically connected to the data collector, and the data collector is electrically connected to the industrial computer. Multiple temperature sensors are distributed on the CNC grinding machine bed and spindle to monitor the temperature of the CNC grinding machine bed and spindle in real time and obtain the temperature field distribution data of the CNC grinding machine; The data collector is used to send the temperature field distribution data of the CNC grinding machine to the industrial computer; An industrial computer is installed with a compensation control program, wherein the compensation control program is configured with a gradient calculation module, a threshold comparison module and a correction module; The gradient calculation module is used to calculate the temperature gradient of each temperature measurement position based on the temperature field distribution data of the CNC grinder; the temperature gradient represents the local temperature rise rate of the grinder; The threshold comparison module is used to compare each calculated temperature gradient with a preset temperature gradient threshold to determine whether there is a temperature measurement position where the temperature gradient exceeds the temperature gradient threshold; The correction module is used to determine the error compensation parameters of the CNC system and correct the grinding trajectory and grinding parameters according to the temperature gradient of each temperature measurement position when there is a temperature measurement position where the temperature gradient exceeds the temperature gradient threshold.

Images