Numerical control planer type milling machine machining precision prediction method and system and computer

Abstract

The invention relates to the technical field of equipment data processing, and provides a numerical control planer type milling machine machining precision prediction method and system and a computer, and the numerical control planer type milling machine machining precision prediction method comprises the steps: setting a plurality of initial temperature measurement points, and constructing a temperature measurement curve; a plurality of real-time displacements of the motorized spindle are collected, and an error curve is constructed; constructing a plurality of temperature measurement point groups to be determined based on the plurality of temperature measurement point groups, and calculating a plurality of thermal key point coefficients; determining a thermosensitive temperature measurement point group; constructing an electric spindle state model based on historical monitoring data; calculating a basic prediction error based on the plurality of updated temperature values; and obtaining a degradation prediction error based on the updated monitoring data, and obtaining a final prediction error based on the basic prediction error and the degradation prediction error. By screening out the thermosensitive temperature measurement point group, more accurate and efficient thermal error prediction is realized, and the final processing precision prediction is more comprehensive and accurate in combination with prediction of the degradation state of the motorized spindle.

Description

Technical Field

[0001] The present invention relates to the technical field of equipment data processing, and in particular to a method, system and computer for predicting machining accuracy of a numerically controlled gantry milling machine. Background Art

[0002] With the development of science and technology, CNC machine tools are more and more widely used, among which gantry milling machine is an important model in the machine tool family.

[0003] The structure of CNC gantry milling machine is complex, and its components involve various fields such as mechanics, hydraulics, electrical and electronics. In the working state, the working conditions are relatively complex. Therefore, the processing accuracy of CNC gantry milling machine is affected by various factors, and the control of processing accuracy is the key to production quality.

[0004] In the prior art, the analysis of the machining accuracy of CNC gantry milling machines is usually based on geometric errors and thermal errors, and the accuracy of thermal error prediction methods still needs to be improved. In addition, for CNC gantry milling machines that have been operating continuously for a long time, the electric spindle has a degradation process. The electric spindle in a degraded state is not stable enough during the milling process, and is prone to radial jump, resulting in increased machining errors. Therefore, the uncertainty of thermal error prediction and electric spindle status leads to insufficient prediction accuracy of the machining accuracy of CNC gantry milling machines. Summary of the invention

[0005] In view of the deficiencies of the prior art, the purpose of the present invention is to provide a method, system and computer for predicting the machining accuracy of a CNC gantry milling machine. The present invention tests a large number of initial temperature measurement points, groups and calculates the coefficients of thermal key points, selects a number of thermal sensitive temperature measurement points, improves the prediction accuracy of thermal errors, and can improve the efficiency of temperature monitoring, builds an electric spindle state model, and predicts the error caused by the degradation of the electric spindle through the electric spindle state, so as to achieve a more comprehensive and accurate prediction of the machining accuracy of the CNC gantry milling machine. The present invention aims to solve the technical problem of insufficient prediction accuracy of the machining accuracy of the CNC gantry milling machine in the prior art.

[0006] In order to achieve the above object, the present invention is implemented by the following technical solutions: The method for predicting the machining accuracy of a CNC gantry milling machine includes the following steps: Setting a plurality of initial temperature measurement points on a milling machine, operating the milling machine to a thermal equilibrium state, collecting a plurality of initial temperature values ​​through the initial temperature measurement points, and constructing a temperature measurement curve based on the initial temperature values ​​and corresponding time; Collecting a plurality of real-time displacements of the electric spindle through a sensor to obtain a test error value, and constructing an error curve based on the time corresponding to the real-time displacement and the test error value; Based on the similarity between the plurality of temperature measurement curves, the plurality of initial temperature measurement points are divided into a plurality of temperature measurement point groups, based on the plurality of temperature measurement point groups, a plurality of temperature measurement point groups to be determined are constructed, and a plurality of thermal key point coefficients corresponding to the plurality of temperature measurement point groups to be determined are calculated; Comparing a plurality of the thermal key point coefficients, establishing the to-be-determined temperature measurement point group corresponding to the largest thermal key point coefficient as a thermosensitive temperature measurement point group, wherein the thermosensitive temperature measurement point group includes a plurality of thermosensitive temperature measurement points; Acquire historical monitoring data of the electric spindle, and construct an electric spindle state model based on the historical monitoring data; Acquire a plurality of updated temperature values ​​from a plurality of the thermal temperature measuring points, and calculate a basic prediction error based on the plurality of updated temperature values; Updated monitoring data of the electric spindle is obtained, a degradation prediction error is obtained based on the updated monitoring data and the electric spindle state model, and a final prediction error is obtained based on the basic prediction error and the degradation prediction error.

[0007] Further, the step of constructing a plurality of temperature measurement point groups to be determined based on the plurality of temperature measurement point groups includes: Calculating the temperature average values ​​of all the initial temperature values ​​in the temperature measurement point group, selecting the maximum average value from a plurality of the temperature average values, establishing the temperature measurement point group corresponding to the maximum average value as a first temperature measurement point group, and establishing the remaining plurality of the temperature measurement point groups as a plurality of second temperature measurement point groups; Reorganizing a number of the initial temperature measurement points in a number of the second temperature measurement point groups into a number of transition temperature measurement point groups; The first temperature measurement point group and the transition temperature measurement point group constitute a temperature measurement point group to be determined.

[0008] Furthermore, the temperature measurement point group to be determined includes a plurality of temperature measurement points to be determined, and the step of calculating a plurality of thermal key point coefficients corresponding to the plurality of temperature measurement point groups to be determined includes: Acquire a plurality of the initial temperature values ​​at the same time as the test error value through a plurality of the temperature measurement points to be determined, and establish them as a plurality of temperature values ​​to be regressed, perform linear regression on the plurality of temperature values ​​to be regressed, so as to obtain a predicted error value, and the group of temperature measurement points to be determined corresponds to a plurality of the predicted error values; Based on the plurality of test error values, the temperature measurement point group to be determined and the plurality of prediction error values ​​corresponding to the temperature measurement point group to be determined, a thermal key point coefficient corresponding to the temperature measurement point group to be determined is calculated.

[0009] Furthermore, the formula of the thermal critical point coefficient is:

[0010] in, represents the thermal key point coefficient, represents the number of test error values, Indicates the number of temperature measurement points to be determined in the temperature measurement point group to be determined. , Indicates The test error value, Indicates The prediction error value corresponding to the test error value is Represents the average of several test error values.

[0011] Furthermore, the historical monitoring data includes the historical radial runout of the electric spindle and the historical front and rear bearing vibration signals.

[0012] Furthermore, the formula of the electric spindle state model is:

[0013] in, Indicates that the electric spindle is The state of the moment, Indicates that the electric spindle is The state of the moment, represents the diffusion coefficient, represents the standard Brownian motion of the healthy phase of the electric spindle, Indicates the initial moment of the electric spindle working. Indicates the moment when the electric spindle switches from the healthy stage to the slow degradation stage. Indicates that the electric spindle is The state of the moment, represents the drift coefficient, represents the standard Brownian motion of the slow degradation stage of the electric spindle, Indicates the moment when the electric spindle switches from the slow degradation stage to the rapid degradation stage. Indicates that the electric spindle is The state of the moment, Represents the standard Brownian motion of the motorized spindle during the rapid degradation phase.

[0014] Furthermore, the step of calculating the basic prediction error based on the plurality of updated temperature values ​​comprises: Based on the temperature measurement curves and the error curves corresponding to the thermosensitive temperature measurement points, a multivariate linear regression is performed with the initial temperature values ​​as independent variables and the test error value as the dependent variable to obtain an error linear regression model; A plurality of the updated temperature values ​​are input into the error linear regression model to obtain a basic prediction error.

[0015] Furthermore, the step of obtaining the degradation prediction error based on the updated monitoring data and the electric spindle state model includes: Performing Kalman filtering on the update monitoring data to obtain the update state of the electric spindle; A degradation prediction error is obtained based on the electric spindle update state and the electric spindle state model.

[0016] A CNC gantry milling machine machining accuracy prediction system is applied to the CNC gantry milling machine machining accuracy prediction method as described in the above technical solution, and the system comprises: A first test module is used to set a plurality of initial temperature measurement points on a milling machine, operate the milling machine to a thermal equilibrium state, collect a plurality of initial temperature values ​​through the initial temperature measurement points, and construct a temperature measurement curve based on the initial temperature values ​​and corresponding time; A second test module is used to collect a number of real-time displacements of the electric spindle through a sensor to obtain a test error value, and to construct an error curve based on the time corresponding to the real-time displacement and the test error value; A determination module, for dividing the initial temperature measurement points into a plurality of temperature measurement point groups based on the similarity between the temperature measurement curves, constructing a plurality of temperature measurement point groups to be determined based on the temperature measurement point groups, and calculating a plurality of thermal key point coefficients corresponding to the plurality of temperature measurement point groups to be determined; An establishing module, used for comparing a plurality of the thermal key point coefficients, establishing the to-be-determined temperature measurement point group corresponding to the largest thermal key point coefficient as a thermal sensitive temperature measurement point group, wherein the thermal sensitive temperature measurement point group includes a plurality of thermal sensitive temperature measurement points; A construction module is used to obtain historical monitoring data of the electric spindle and construct an electric spindle state model based on the historical monitoring data; A first prediction module, used for obtaining a plurality of updated temperature values ​​from a plurality of the thermal temperature measurement points, and calculating a basic prediction error based on the plurality of updated temperature values; The second prediction module is used to obtain updated monitoring data of the electric spindle, obtain a degradation prediction error based on the updated monitoring data and the electric spindle state model, and obtain a final prediction error based on the basic prediction error and the degradation prediction error.

[0017] A computer comprises a memory, a processor and a computer program stored in the memory and executable on the processor, wherein when the processor executes the computer program, the method for predicting machining accuracy of a CNC gantry milling machine as described in the above technical solution is implemented.

[0018] Compared with the prior art, the beneficial effects of the present invention are as follows: by setting a plurality of the initial temperature measurement points and running the milling machine for testing, collecting test data, constructing a plurality of the temperature measurement curves, and grouping the temperature measurement curves with similarity into the same group, that is, dividing a plurality of the initial temperature measurement points into a plurality of the temperature measurement point groups, selecting a plurality of the initial temperature measurement points from the plurality of the temperature measurement point groups for reorganization in a plurality of combinations, that is, forming a plurality of the temperature measurement point groups to be determined, and selecting the optimal temperature measurement point combination with thermal sensitivity according to the thermal key point coefficients, the temperature changes of a plurality of the thermistor temperature measurement points in the thermistor temperature measurement point group have the highest correlation with the deformation of the electric spindle caused by temperature, and predicting the thermal error through the data of the thermistor temperature measurement points, which greatly reduces the error. The accuracy of thermal error prediction is improved, and the total number of the thermistor temperature measuring points is greatly reduced compared with the total number of the initial temperature measuring points. In the working state of the milling machine, only the data of the thermistor temperature measuring points are monitored, which saves costs and has high efficiency while ensuring the accuracy of thermal error prediction. By constructing the electric spindle state model, the state analysis of the electric spindle on the milling machine that has been running for a long time in production is carried out. The electric spindle ages and degrades over time, and its state change will bring about certain degradation errors, which is easy to produce radial jumps during the machining process. Based on the updated monitoring data, the current state of the electric spindle is identified, and the future state and existing degradation errors can be predicted, which are finally combined with the predicted thermal errors to form a more accurate and comprehensive machining accuracy prediction. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] Figure 1 It is a flowchart of a method for predicting machining accuracy of a CNC gantry milling machine in a first embodiment of the present invention; Figure 2 It is a structural block diagram of a CNC gantry milling machine machining accuracy prediction system in a second embodiment of the present invention; The following specific implementation manner will further illustrate the present invention in conjunction with the above-mentioned drawings. DETAILED DESCRIPTION

[0020] In order to facilitate the understanding of the present invention, the present invention will be described more fully below with reference to the relevant drawings. Several embodiments of the present invention are given in the drawings. However, the present invention can be implemented in many different forms and is not limited to the embodiments described herein. On the contrary, the purpose of providing these embodiments is to make the disclosure of the present invention more thorough and comprehensive.

[0021] It should be noted that when an element is referred to as being "fixed to" another element, it may be directly on the other element or there may be a central element. When an element is considered to be "connected to" another element, it may be directly connected to the other element or there may be a central element at the same time. The terms "vertical", "horizontal", "left", "right" and similar expressions used herein are for illustrative purposes only.

[0022] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as those commonly understood by those skilled in the art to which the present invention belongs. The terms used herein in the specification of the present invention are only for the purpose of describing specific embodiments and are not intended to limit the present invention. The term "and / or" used herein includes any and all combinations of one or more of the related listed items.

[0023] See also Figure 1 The method for predicting machining accuracy of a CNC gantry milling machine in the first embodiment of the present invention comprises the following steps: Step S10: setting a plurality of initial temperature measurement points on the milling machine, operating the milling machine to a thermal equilibrium state, collecting a plurality of initial temperature values ​​through the initial temperature measurement points, and constructing a temperature measurement curve based on the initial temperature values ​​and corresponding time; The geometric error of a CNC gantry milling machine can be measured by a laser interferometer, and the geometric error of the milling machine does not change with temperature. It can be obtained based on the factory data of each model of milling machine. During operation, the parts of the milling machine expand and contract due to temperature changes. Although it is difficult to identify with the naked eye, it will have a great impact on the machining accuracy of the CNC gantry milling machine. The spindle of the CNC gantry milling machine is more affected by the heat source than the XYZ axis worktable and other machine tool parts of the CNC gantry milling machine under working conditions, and the thermal deformation is greater. The machining error caused by the thermal deformation of the electric spindle accounts for a very large proportion of the overall thermal error of the machine tool. In this embodiment, a PT100 temperature sensor is used, and the initial temperature measuring points are arranged on the spindle box, the spindle base, the spindle coolant outlet, the spindle coolant inlet, the left end of the spindle lower bearing, the right end of the spindle lower bearing, the spindle motor, the spindle upper bearing, the front side of the milling machine bed, the back side of the milling machine bed, the contact position between the guide rail and the bed, and the external ambient temperature. The milling machine simulates the real working conditions on the way to the thermal equilibrium state. When the temperature of each of the initial temperature measuring points does not increase significantly, it indicates that the milling machine has reached the thermal equilibrium state, and the temperature measurement curve is constructed for the temperature changes of all the initial temperature measuring points.

[0024] Step S20: collecting a plurality of real-time displacements of the electric spindle through a sensor to obtain a test error value, and constructing an error curve based on the time corresponding to the real-time displacement and the test error value; Preferably, the real-time displacement of the electric spindle is measured by a displacement sensor using a five-point method. Specifically, two of the displacement sensors are arranged in the X-axis direction of the electric spindle, and two of the displacement sensors are arranged in the Y-axis direction of the electric spindle. A steel test rod is provided on the electric spindle, and one displacement sensor is arranged in the Z-axis direction of the steel test rod to measure the thermal elongation error of the electric spindle. The initial temperature value and the test error value are collected at the same time interval. The obtained real-time displacements are converted to obtain the rotation error around the X-axis, the rotation error around the Y-axis and the thermal elongation error of the electric spindle. It can be understood that at the same moment, one test error value corresponds to the temperature of all the initial temperature measurement points at that moment.

[0025] Step S30: dividing the initial temperature measurement points into a plurality of temperature measurement point groups based on the similarities between the temperature measurement curves, constructing a plurality of temperature measurement point groups to be determined based on the temperature measurement point groups, and calculating a plurality of thermal key point coefficients corresponding to the plurality of temperature measurement point groups to be determined; Preferably, the initial temperature measurement points with similar temperature changes can be judged according to the shape of the temperature measurement curve, or the K-means algorithm can be used to perform cluster analysis on the sample data of all the initial temperature measurement points to form groups. Specifically, the present embodiment adopts a cluster analysis method. To predict the thermal error, data at the thermosensitive points need to be collected. The thermosensitive points are the key temperature measurement points with the highest correlation with the thermal deformation of the electric spindle. There are usually multiple thermosensitive points. To select the optimal thermosensitive point, the temperature measurement point with the greatest impact on the thermal error needs to be selected from the multiple similar initial temperature measurement points in the temperature measurement point group.

[0026] The step S30 comprises: S310: Calculate the temperature average values ​​of all the initial temperature values ​​in the temperature measurement point group, select the maximum average value from a plurality of the temperature average values, establish the temperature measurement point group corresponding to the maximum average value as a first temperature measurement point group, and establish the remaining plurality of the temperature measurement point groups as a plurality of second temperature measurement point groups; Preferably, since the electric spindle follows the principle of thermal expansion and contraction, all temperature measuring points in the temperature measuring point group with the largest overall temperature increase are established as thermosensitive points, that is, all the initial temperature measuring points in the first temperature measuring point group are thermosensitive points.

[0027] S320: reorganizing a number of the initial temperature measurement points in a number of the second temperature measurement point groups into a number of transition temperature measurement point groups; S330: The first temperature measurement point group and the transition temperature measurement point group form a temperature measurement point group to be determined; Preferably, the method for specifically forming the temperature measurement point group to be determined is exemplified as follows: several of the temperature measurement point groups are specifically three groups, namely A, B, and C. Group A includes two initial temperature measurement points, T1 and T2; Group B includes two initial temperature measurement points, T3 and T4; Group C includes three initial temperature measurement points, T5, T6, and T7. Among the temperature measurement results, the average temperature rise of Group C is the highest among the three groups. Then, Group C is directly established as the first temperature measurement point group, and the three initial temperature measurement points, T5, T6, and T7, are directly established as thermal sensitive points. Both Group A and Group B are established as the second temperature measurement point groups. One initial temperature measurement point is selected from each of Group A and Group B, and through permutation and combination, multiple transition temperature measurement point groups are re - formed. In this example, specifically four transition temperature measurement point groups are re - formed, namely: T1, T3 group, T1, T4 group, T2, T3 group, and T2, T4 group; The four transition temperature measurement point groups and the first temperature measurement point group form four temperature measurement point groups to be determined, namely: T1, T3, T5, T6, T7 group, T1, T4, T5, T6, T7 group, T2, T3, T5, T6, T7 group, T2, T4, T5, T6, T7 group.

[0028] The temperature measurement point group to be determined includes several temperature measurement points to be determined, and step S30 further includes: S340: Obtain several initial temperature values at the same moment as the test error value through several temperature measurement points to be determined, and establish them as several temperature values to be regressed. Perform linear regression on several temperature values to be regressed to obtain a prediction error value. The temperature measurement point group to be determined corresponds to several prediction error values; Preferably, for example, at the moment when the milling machine runs for one hour, the five initial temperature values corresponding to the five temperature measurement points to be determined in the group of T1, T3, T5, T6, T7 at this moment are all established as the temperature values to be regressed. Based on the curves of these five temperature measurement points to be determined and the error curve for fitting, several regression coefficients and correction coefficients of the linear regression equation are obtained. According to the linear regression equation, the prediction error value when the milling machine runs for one hour is calculated. The total number of prediction error values corresponding to each temperature measurement point group to be determined is the same as the total number of test error values, and they are all arranged according to time points.

[0029] S350: Based on several test error values, the temperature measurement point group to be determined, and several prediction error values corresponding to the temperature measurement point group to be determined, calculate the thermal key point coefficient corresponding to the temperature measurement point group to be determined.

[0030] The formula for the thermal key point coefficient is:

[0031] Wherein, represents the thermal key point coefficient, represents the number of test error values, Indicates the number of temperature measurement points to be determined in the temperature measurement point group to be determined. , Indicates The test error value, Indicates The prediction error value corresponding to the test error value is Represents the average of several test error values.

[0032] Step S40: comparing a plurality of thermal key point coefficients, establishing the to-be-determined temperature measurement point group corresponding to the largest thermal key point coefficient as a thermosensitive temperature measurement point group, wherein the thermosensitive temperature measurement point group includes a plurality of thermosensitive temperature measurement points; Preferably, the thermistor temperature measuring point is a thermosensitive point. It is understandable that the total number of thermistor points is less than the total number of the initial temperature measuring points. In the monitoring of the actual working conditions, only the thermistor temperature measuring points are monitored to effectively predict the thermal error of the milling machine. The thermistor temperature measuring points have been quantitatively analyzed to select the optimal solution, which is beneficial to greatly improve the accuracy of thermal error prediction, while saving the cost of using sensors, reducing the amount of data collected and processed during the prediction process, and making the prediction more efficient.

[0033] Step S50: acquiring historical monitoring data of the electric spindle, and constructing an electric spindle state model based on the historical monitoring data; It is understandable that the performance of the electric spindle will degrade over time due to aging, wear and other degradation phenomena. Specifically, wear of the spindle system bearings, looseness of the motor rotor and other degradation will cause increased rotational error and intensified vibration, usually resulting in micron-level errors. Such errors have a negative impact on CNC gantry milling machines with high precision requirements. Therefore, establishing the electric spindle state model is beneficial to evaluating the state of the electric spindle and predicting the errors caused by degradation, making the prediction of the machining accuracy of the CNC gantry milling machine more comprehensive and accurate.

[0034] The historical monitoring data includes the historical radial runout of the electric spindle and the historical front and rear bearing vibration signals.

[0035] Preferably, the radial runout is obtained by collecting the axis trajectory of the electric spindle at 25 mm from the shaft end through two mutually perpendicular displacement sensors, and the vibration signals of the front and rear bearings are collected by an acceleration sensor installed on the outer shell at the bearing position. The unit of the radial runout is micron, and the unit of the vibration amount in the vibration signal is g.

[0036] The formula of the electric spindle state model is:

[0037] in, Indicates that the electric spindle is The state of the moment, Indicates that the electric spindle is The state of the moment, represents the diffusion coefficient, represents the standard Brownian motion of the healthy phase of the electric spindle, Indicates the initial moment of the electric spindle working. Indicates the moment when the electric spindle switches from the healthy stage to the slow degradation stage. Indicates that the electric spindle is The state of the moment, represents the drift coefficient, represents the standard Brownian motion of the slow degradation stage of the electric spindle, Indicates the moment when the electric spindle switches from the slow degradation stage to the rapid degradation stage. Indicates that the electric spindle is The state of the moment, Represents the standard Brownian motion of the motorized spindle during the rapid degradation phase.

[0038] Preferably, the electric spindle state model is established based on a state space model, and the change of the electric spindle state includes three stages, namely, a healthy stage, a slow degradation stage and a rapid degradation stage. In the healthy stage, the electric spindle runs smoothly, and the vibration index of the electric spindle reflected by the degradation state of the electric spindle is near a constant value, and the vibration index is further reflected as a radial jump amount, etc. When the electric spindle is in the slow degradation stage, the electric spindle can still be used normally, and the vibration index of the electric spindle changes linearly and uniformly. When the electric spindle is in the rapid degradation stage, the vibration index will no longer maintain a uniform change. The parameters and change points of each stage in the electric spindle state model are obtained through multiple simulation experiments and training of the historical monitoring data. Specifically, in this embodiment, the change point from the healthy stage to the slow degradation stage is 41.5% of the total service life of the electric spindle, and the change point from the slow degradation stage to the rapid degradation stage is 79% of the total service life of the electric spindle.

[0039] Step S60: obtaining a plurality of updated temperature values ​​from a plurality of the thermal temperature measuring points, and calculating a basic prediction error based on the plurality of updated temperature values; Preferably, the unit of the basic prediction error is micrometer. It can be understood that the number of temperature measurement points to be monitored is reduced, the thermal error is predicted more efficiently, and the basic prediction error is beneficial to provide data analysis for error compensation.

[0040] The step S60 comprises: S610: Based on the temperature measurement curves and the error curves corresponding to the thermosensitive temperature measurement points, a plurality of initial temperature values ​​are used as independent variables, and the test error value is used as a dependent variable, and a multivariate linear regression is performed to obtain an error linear regression model; Preferably, the temperature data of the thermistor is used to reflect the processing error caused by the temperature change, and fitting is performed based on the multiple temperature measurement curves and the error curve that have been acquired. Based on the multivariate linear regression model and a large amount of discrete data sampled from the curve, multiple regression coefficients and correction coefficients of the multivariate linear regression model are obtained to obtain the error linear regression model. The total number of regression coefficients is greater than the total number of thermistor temperature measurement points, and different thermistor temperature measurement points correspond to different regression coefficients.

[0041] S620: Inputting a plurality of the updated temperature values ​​into the error linear regression model to obtain a basic prediction error.

[0042] Preferably, a plurality of the updated temperature values ​​are used as independent variables, and the updated temperature values ​​collected at different thermistor temperature measuring points are multiplied by corresponding regression coefficients, and the basic prediction error is calculated according to the error linear regression model.

[0043] Step S70: acquiring updated monitoring data of the electric spindle, obtaining a degradation prediction error based on the updated monitoring data and the electric spindle state model, and obtaining a final prediction error based on the basic prediction error and the degradation prediction error.

[0044] Preferably, the potential degradation process of the state of the electric spindle system can be described by the electric spindle state model, and the observed data can be associated with the state of the electric spindle by establishing a linear observation equation, that is, the degradation-related indicators are associated with the state of the electric spindle, that is, the vibration indicators reflected by the monitoring data are associated with the state of the electric spindle, and the observation equation obtains the observation state based on the observation matrix and the observation noise, and the observation matrix is ​​obtained based on the updated monitoring data. It can be understood that the final prediction error includes the consideration of thermal errors and the degradation state of the electric spindle, and the machining accuracy prediction of the CNC gantry milling machine is more comprehensive and accurate.

[0045] The step S70 includes: S710: Perform Kalman filtering on the updated monitoring data to obtain an updated state of the electric spindle; Preferably, for the electric spindle in the healthy stage, a constant-based Kalman filter is used to filter out observation noise, for the electric spindle in the slowly degraded stage, a linear model-based Kalman filter is used, and for the electric spindle in the rapidly degraded stage, a nonlinear-based Kalman filter is used.

[0046] S720: Obtain a degradation prediction error based on the updated state of the electric spindle and the electric spindle state model.

[0047] Preferably, the degradation prediction error reflects the predicted radial jump amount on the electric spindle.

[0048] Please refer to Figure 2 , the machining accuracy prediction system of the numerically controlled gantry milling machine described in the second embodiment of the present invention is applied to the machining accuracy prediction method of the numerically controlled gantry milling machine described in the first embodiment above. The system includes: The first test module 10 is used to set a plurality of initial temperature measurement points on the milling machine, run the milling machine to the thermal equilibrium state, collect a plurality of initial temperature values through the initial temperature measurement points, and construct a temperature measurement curve based on the initial temperature values and the corresponding time; The second test module 20 is used to collect a plurality of real-time displacements of the electric spindle through sensors to obtain test error values, and construct an error curve based on the time corresponding to the real-time displacements and the test error values; The determination module 30 is used to divide a plurality of the initial temperature measurement points into a plurality of temperature measurement point groups based on the similarity between a plurality of the temperature measurement curves, construct a plurality of to-be-determined temperature measurement point groups based on the plurality of temperature measurement point groups, and calculate a plurality of thermal key point coefficients corresponding to the plurality of to-be-determined temperature measurement point groups; The determination module 30 includes: The first unit is used to calculate the temperature average value of all the initial temperature values in the temperature measurement point group, select the maximum average value from a plurality of the temperature average values, establish the temperature measurement point group corresponding to the maximum average value as the first temperature measurement point group, and establish the remaining plurality of temperature measurement point groups as a plurality of second temperature measurement point groups; The second unit is used to reorganize a plurality of the initial temperature measurement points in the plurality of second temperature measurement point groups into a plurality of transition temperature measurement point groups; The third unit is used to form a to-be-determined temperature measurement point group from the first temperature measurement point group and the transition temperature measurement point group; In the determination module 30, the to-be-determined temperature measurement point group includes a plurality of to-be-determined temperature measurement points; The determination module 30 further includes: The fourth unit is used to obtain a plurality of the initial temperature values at the same moment as the test error values through the plurality of to-be-determined temperature measurement points, and establish them as a plurality of to-be-regressed temperature values, perform linear regression on the plurality of to-be-regressed temperature values to obtain a prediction error value, and the to-be-determined temperature measurement point group corresponds to a plurality of the prediction error values; The fifth unit is configured to calculate a thermal key point coefficient corresponding to the temperature measurement point group to be determined based on a plurality of the test error values, the temperature measurement point group to be determined, and a plurality of the prediction error values corresponding to the temperature measurement point group to be determined.

[0049] In the determination module 30, the formula for the thermal key point coefficient is:

[0050] Wherein, represents the thermal key point coefficient, represents the number of test error values, represents the number of temperature measurement points to be determined in the temperature measurement point group to be determined, , represents the th test error value, represents the prediction error value corresponding to the th test error value, represents the average value of a plurality of test error values.

[0051] The establishment module 40 is configured to compare a plurality of the thermal key point coefficients, and establish the temperature measurement point group corresponding to the largest thermal key point coefficient as the thermal sensitive temperature measurement point group, and the thermal sensitive temperature measurement point group includes a plurality of thermal sensitive temperature measurement points; The construction module 50 is configured to obtain historical monitoring data of the electric spindle, and construct an electric spindle state model based on the historical monitoring data; In the construction module 50, the historical monitoring data includes the historical radial runout amount of the electric spindle and the historical vibration signals of the front and rear bearings.

[0052] The formula for the electric spindle state model is:

[0053] Wherein, represents the state of the electric spindle at moment, represents the state of the electric spindle at moment, represents the diffusion coefficient, represents the standard Brownian motion in the healthy stage of the electric spindle, represents the initial moment when the electric spindle starts to work, represents the moment when the electric spindle transitions from the healthy stage to the slow degradation stage, represents the state of the electric spindle at moment, represents the drift coefficient, represents the standard Brownian motion in the slow degradation stage of the electric spindle, Indicates the moment when the electric spindle switches from the slow degradation stage to the rapid degradation stage. Indicates that the electric spindle is The state of the moment, Represents the standard Brownian motion of the motorized spindle during the rapid degradation phase.

[0054] A first prediction module 60 is used to obtain a plurality of updated temperature values ​​from a plurality of the thermal temperature measurement points, and calculate a basic prediction error based on the plurality of updated temperature values; The first prediction module 60 includes: A sixth unit is used to perform multiple linear regression based on the temperature measurement curves and the error curves corresponding to the thermal temperature measurement points, with the initial temperature values ​​as independent variables and the test error value as a dependent variable, to obtain an error linear regression model; The seventh unit is used to input a plurality of the updated temperature values ​​into the error linear regression model to obtain a basic prediction error.

[0055] The second prediction module 70 is used to obtain updated monitoring data of the electric spindle, obtain a degradation prediction error based on the updated monitoring data and the electric spindle state model, and obtain a final prediction error based on the basic prediction error and the degradation prediction error.

[0056] The second prediction module 70 includes: An eighth unit is used for performing Kalman filtering on the update monitoring data to obtain an update state of the electric spindle; A ninth unit is used to obtain a degradation prediction error based on the electric spindle update state and the electric spindle state model.

[0057] The third embodiment of the present invention provides a computer, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, the method for predicting machining accuracy of a CNC gantry milling machine as described in the first embodiment above is implemented.

[0058] In the description of this specification, the description with reference to the terms "one embodiment", "some embodiments", "examples", "specific examples", or "some examples" means that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present invention. In this specification, the schematic representation of the above terms does not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described may be combined in any one or more embodiments or examples in a suitable manner.

[0059] The above-mentioned embodiments only express several implementation methods of the present invention, and the description thereof is relatively specific and detailed, but it cannot be understood as limiting the scope of the patent of the present invention. It should be pointed out that, for ordinary technicians in this field, several variations and improvements can be made without departing from the concept of the present invention, which all belong to the protection scope of the present invention. Therefore, the protection scope of the patent of the present invention shall be subject to the attached claims.

Claims

1. A method for predicting machining accuracy of a CNC gantry milling machine, characterized in that: The steps include: Setting a plurality of initial temperature measurement points on a milling machine, operating the milling machine to a thermal equilibrium state, collecting a plurality of initial temperature values ​​through the initial temperature measurement points, and constructing a temperature measurement curve based on the initial temperature values ​​and corresponding time; Collecting a plurality of real-time displacements of the electric spindle through a sensor to obtain a test error value, and constructing an error curve based on the time corresponding to the real-time displacement and the test error value; Based on the similarity between the plurality of temperature measurement curves, the plurality of initial temperature measurement points are divided into a plurality of temperature measurement point groups, based on the plurality of temperature measurement point groups, a plurality of temperature measurement point groups to be determined are constructed, and a plurality of thermal key point coefficients corresponding to the plurality of temperature measurement point groups to be determined are calculated; Comparing a plurality of the thermal key point coefficients, establishing the to-be-determined temperature measurement point group corresponding to the largest thermal key point coefficient as a thermosensitive temperature measurement point group, wherein the thermosensitive temperature measurement point group includes a plurality of thermosensitive temperature measurement points; Acquire historical monitoring data of the electric spindle, and construct an electric spindle state model based on the historical monitoring data; Acquire a plurality of updated temperature values ​​from a plurality of the thermal temperature measuring points, and calculate a basic prediction error based on the plurality of updated temperature values; Updated monitoring data of the electric spindle is obtained, a degradation prediction error is obtained based on the updated monitoring data and the electric spindle state model, and a final prediction error is obtained based on the basic prediction error and the degradation prediction error.

2. The method for predicting machining accuracy of a CNC gantry milling machine according to claim 1, characterized in that: The step of constructing a plurality of temperature measurement point groups to be determined based on the plurality of temperature measurement point groups comprises: Calculating the temperature average values ​​of all the initial temperature values ​​in the temperature measurement point group, selecting the maximum average value from a plurality of the temperature average values, establishing the temperature measurement point group corresponding to the maximum average value as a first temperature measurement point group, and establishing the remaining plurality of the temperature measurement point groups as a plurality of second temperature measurement point groups; Reorganizing a number of the initial temperature measurement points in a number of the second temperature measurement point groups into a number of transition temperature measurement point groups; The first temperature measurement point group and the transition temperature measurement point group constitute a temperature measurement point group to be determined.

3. The method for predicting machining accuracy of a CNC gantry milling machine according to claim 1, characterized in that: The temperature measurement point group to be determined includes a plurality of temperature measurement points to be determined, and the step of calculating a plurality of thermal key point coefficients corresponding to the plurality of temperature measurement point groups to be determined includes: Acquire a plurality of the initial temperature values ​​at the same time as the test error value through a plurality of the temperature measurement points to be determined, and establish them as a plurality of temperature values ​​to be regressed, perform linear regression on the plurality of temperature values ​​to be regressed, so as to obtain a predicted error value, and the group of temperature measurement points to be determined corresponds to a plurality of the predicted error values; Based on the plurality of test error values, the temperature measurement point group to be determined and the plurality of prediction error values ​​corresponding to the temperature measurement point group to be determined, a thermal key point coefficient corresponding to the temperature measurement point group to be determined is calculated.

4. The method for predicting machining accuracy of a CNC gantry milling machine according to claim 3, characterized in that: The formula of the thermal critical point coefficient is: in, represents the thermal key point coefficient, represents the number of test error values, Indicates the number of temperature measurement points to be determined in the temperature measurement point group to be determined. , Indicates The test error value, Indicates The prediction error value corresponding to the test error value is Represents the average of several test error values.

5. The method for predicting machining accuracy of a CNC gantry milling machine according to claim 1, characterized in that: The historical monitoring data includes the historical radial runout of the electric spindle and the historical front and rear bearing vibration signals.

6. The method for predicting machining accuracy of a CNC gantry milling machine according to claim 1, characterized in that: The formula of the electric spindle state model is: in, Indicates that the electric spindle is The state of the moment, Indicates that the electric spindle is The state of the moment, represents the diffusion coefficient, represents the standard Brownian motion of the healthy phase of the electric spindle, Indicates the initial moment of the electric spindle working. Indicates the moment when the electric spindle switches from the healthy stage to the slow degradation stage. Indicates that the electric spindle is The state of the moment, represents the drift coefficient, represents the standard Brownian motion of the slow degradation stage of the electric spindle, Indicates the moment when the electric spindle switches from the slow degradation stage to the rapid degradation stage. Indicates that the electric spindle is The state of the moment, Represents the standard Brownian motion of the motorized spindle during the rapid degradation phase.

7. The method for predicting machining accuracy of a CNC gantry milling machine according to claim 1, characterized in that: The step of calculating the basic prediction error based on the plurality of updated temperature values ​​comprises: Based on the temperature measurement curves and the error curves corresponding to the thermosensitive temperature measurement points, a multivariate linear regression is performed with the initial temperature values ​​as independent variables and the test error value as the dependent variable to obtain an error linear regression model; A plurality of the updated temperature values ​​are input into the error linear regression model to obtain a basic prediction error.

8. The method for predicting machining accuracy of a CNC gantry milling machine according to claim 1, characterized in that: The step of obtaining the degradation prediction error based on the updated monitoring data and the electric spindle state model comprises: Performing Kalman filtering on the update monitoring data to obtain the update state of the electric spindle; A degradation prediction error is obtained based on the electric spindle update state and the electric spindle state model.

9. A CNC gantry milling machine machining accuracy prediction system, applied to a CNC gantry milling machine machining accuracy prediction method as claimed in any one of claims 1 to 8, characterized in that: The system comprises: A first test module is used to set a plurality of initial temperature measurement points on a milling machine, operate the milling machine to a thermal equilibrium state, collect a plurality of initial temperature values ​​through the initial temperature measurement points, and construct a temperature measurement curve based on the initial temperature values ​​and corresponding time; A second test module is used to collect a number of real-time displacements of the electric spindle through a sensor to obtain a test error value, and to construct an error curve based on the time corresponding to the real-time displacement and the test error value; A determination module, for dividing the initial temperature measurement points into a plurality of temperature measurement point groups based on the similarity between the temperature measurement curves, constructing a plurality of temperature measurement point groups to be determined based on the temperature measurement point groups, and calculating a plurality of thermal key point coefficients corresponding to the plurality of temperature measurement point groups to be determined; An establishing module, used for comparing a plurality of the thermal key point coefficients, establishing the to-be-determined temperature measurement point group corresponding to the largest thermal key point coefficient as a thermal sensitive temperature measurement point group, wherein the thermal sensitive temperature measurement point group includes a plurality of thermal sensitive temperature measurement points; A construction module is used to obtain historical monitoring data of the electric spindle and construct an electric spindle state model based on the historical monitoring data; A first prediction module, used for obtaining a plurality of updated temperature values ​​from a plurality of the thermal temperature measurement points, and calculating a basic prediction error based on the plurality of updated temperature values; The second prediction module is used to obtain updated monitoring data of the electric spindle, obtain a degradation prediction error based on the updated monitoring data and the electric spindle state model, and obtain a final prediction error based on the basic prediction error and the degradation prediction error.

10. A computer comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein: When the processor executes the computer program, the method for predicting machining accuracy of a CNC gantry milling machine as claimed in any one of claims 1 to 8 is implemented.

Images