A method and system for axial positioning control of CNC lathe

Abstract

This application relates to the field of CNC machine tool control technology, specifically to a method and system for axial positioning control of a CNC lathe. The method includes: driving a servo motor to perform backlash compensation motion; synchronously sampling to acquire position signals and load current signals; performing a composite judgment based on the position and load current signals: determining whether the change in the position signal along the command direction reaches a preset position threshold, and determining whether the increase in the load current signal relative to a reference current value reaches a preset current threshold; when all composite judgments are satisfied simultaneously, it is determined that the mechanical backlash has been eliminated, the backlash compensation motion is terminated, and the system switches to a dual-closed-loop axial positioning control mode; when any composite judgment is not satisfied, the backlash compensation motion is maintained. This solves the problem of inaccurate compensation and continuous "creeping" of the worktable near the target position, affecting machining accuracy, when the axial movement direction of the CNC lathe reverses for backlash compensation.

Description

Technical Field

[0001] This application relates to the field of CNC machine tool control technology, and more specifically, to a method and system for axial positioning control of a CNC lathe. Background Technology

[0002] In modern precision manufacturing, CNC lathes are responsible for machining high-precision parts, and the performance of their axial positioning control system directly determines the quality of the final product. Typically, to meet stringent precision requirements, these systems employ an advanced control strategy: a dual-closed-loop axial positioning control method based on the fusion of data from a linear encoder and a grating ruler. This method utilizes the high-precision absolute position information provided by the linear encoder and the high-speed relative displacement information provided by the encoder, integrating the advantages of both through a data fusion algorithm. Combined with an inner and outer dual-closed-loop control structure, it aims to achieve precise and stable control of the worktable position, particularly striking a balance between high-speed motion and precise positioning.

[0003] However, in actual industrial production environments, when a CNC lathe continuously performs a precision part machining task for an extended period, the machining process continuously generates a large number of fine metal chips. These chips, carried by the cutting fluid, may not be completely removed and gradually accumulate in the tiny gaps between the grating ruler and the reading head. They mix with the oily components and trace impurities in the cutting fluid, forming a thin, semi-transparent mixture. This mixture causes slight scattering and attenuation of the optical signal path inside the grating ruler's reading head, making the original position signal output by the grating ruler unstable, exhibiting intermittent, small-amplitude fluctuations. These fluctuations are not continuous but rather intermittent, and their amplitude usually hovers around the edge of the system's set normal noise threshold. Therefore, its overall function is not completely lost, and it is difficult for conventional signal anomaly detection mechanisms to immediately identify it as a fault.

[0004] In the axial positioning control of CNC lathes, to ensure high precision, a compensation program for the backlash of the ball screw drive chain is usually preset. When the direction of movement reverses, the servo motor performs a compensating motion at a low speed until it detects that the worktable has overcome the mechanical backlash and begins to move. This judgment process is highly dependent on the position signal of the grating ruler. However, in the critical stage of fine positioning, the grating ruler may produce intermittent fluctuations of small amplitude due to surface contamination or other interference. These fluctuations are similar to actual minute displacement signals and can easily be misinterpreted by the control system as the backlash being eliminated.

[0005] If a misjudgment occurs, the compensation procedure may terminate prematurely, leaving residual gaps unresolved; or it may terminate late due to signal uncertainty, continuing compensation after the gaps have been eliminated, resulting in excessive displacement. Whether premature or delayed, both introduce inaccurate compensation amounts, creating new, minute positional deviations. This deviation prevents the dual-loop control system from stabilizing near the target position because the grating ruler feedback itself is uncertain, and the position loop cannot obtain reliable error signals. The result is a persistent, minute overshoot and oscillation of the worktable near the target point, a phenomenon known as "creep."

[0006] This "creep" causes the worktable to oscillate continuously within a tiny range instead of remaining within the expected static accuracy range. For precision machining tasks requiring high surface quality and dimensional consistency, this instability can significantly reduce machining accuracy and may even lead to scrapped parts or substandard quality.

[0007] To address the aforementioned issues, existing technologies urgently need improvement. Summary of the Invention

[0008] This application discloses an axial positioning control method and system for CNC lathes, aiming to solve the problem in the prior art where, when CNC lathes reverse the axial movement direction to perform clearance compensation, misjudgment is caused by intermittent fluctuations in the grating ruler signal, leading to inaccurate compensation, continuous "creeping" of the worktable near the target position, and affecting machining accuracy.

[0009] The technical solution of this application is as follows:

[0010] In a first aspect, this application discloses an axial positioning control method for a CNC lathe, comprising:

[0011] When the axial movement direction is reversed, the servo motor is driven to perform backlash compensation movement at a preset low speed;

[0012] During the backlash compensation motion, the position signal of the worktable and the load current signal of the servo motor are acquired with a synchronous sampling period.

[0013] Based on the position signal and load current signal, a combined judgment is made: within the preset duration window, it is determined whether the change in the position signal along the command direction reaches the preset position threshold, and whether the increase in the load current signal relative to the reference current value before the backlash compensation movement reaches the preset current threshold.

[0014] When all conditions of the composite judgment are met simultaneously, it is determined that the mechanical backlash has been eliminated, the backlash compensation motion is terminated, and the system switches to the normal dual closed-loop axial positioning control mode.

[0015] If any one of the composite judgments is not satisfied, maintain the reverse backlash compensation motion.

[0016] Furthermore, composite judgments also include:

[0017] A micro-amplitude high-frequency oscillation command is superimposed on the servo motor to synchronously acquire the load current signal;

[0018] The load current signal is high-pass filtered to obtain the filtered load current signal.

[0019] Perform spectrum analysis on the filtered load current signal to extract the current oscillation amplitude corresponding to the frequency of the micro-amplitude high-frequency oscillation command.

[0020] Determine whether the current oscillation amplitude continuously exceeds the preset rigid contact oscillation amplitude threshold.

[0021] Further, determining whether the current oscillation amplitude continuously exceeds a preset rigid contact oscillation amplitude threshold includes:

[0022] Before performing the backlash compensation motion, a micro-amplitude high-frequency oscillation command is superimposed on the servo motor, and the load current signal of the servo motor is acquired simultaneously.

[0023] The load current signal is subjected to high-pass filtering to obtain the filtered load current signal. The spectrum analysis of the filtered load current signal is then performed to obtain the current oscillation amplitude corresponding to the frequency of the micro-amplitude high-frequency oscillation command.

[0024] The current oscillation amplitude is calibrated based on the machine tool operating parameters, and the rigid contact oscillation amplitude threshold is adjusted accordingly to the threshold under the current operating conditions for composite judgment.

[0025] Furthermore, the current oscillation amplitude is calibrated based on the machine tool operating parameters, and the rigid contact oscillation amplitude threshold is adjusted accordingly to the threshold under the current operating conditions, including:

[0026] Monitor machine tool operating parameters, which include at least one of the following: ambient temperature, cumulative running time, cumulative axial travel, spindle speed, and feed rate;

[0027] Based on the machine tool operating parameters and the preset parameter influence weight set, calculate the comprehensive rigidity correction factor;

[0028] The threshold value of rigid contact oscillation amplitude is dynamically adjusted based on the current oscillation amplitude and the comprehensive rigidity correction factor.

[0029] Furthermore, the method also includes:

[0030] Before performing the backlash compensation motion, the actual backlash value and friction torque parameters of the ball screw are measured by micro-amplitude probe motion. Based on the actual backlash value and friction torque parameters, a backlash crossing optimization trajectory is generated. The backlash compensation motion is completed along the backlash crossing optimization trajectory. After the backlash compensation motion is completed, the system smoothly switches to the normal dual closed-loop axial positioning control mode.

[0031] Furthermore, based on the machine tool operating parameters and the preset parameter influence weight set, a comprehensive rigidity correction factor is calculated, including:

[0032] Before performing the backlash compensation motion, multiple independent rigidity correction components are calculated based on the machine tool operating parameters and parameter influence weight set;

[0033] Dynamically adjust the nonlinear combination coefficients based on machine tool operating parameters;

[0034] The adjusted nonlinear combination coefficients are used to nonlinearly combine multiple independent rigid correction components to obtain a comprehensive rigid correction factor. The nonlinear combination is an adaptive nonlinear combination.

[0035] Furthermore, the nonlinear combination coefficients are dynamically adjusted based on the machine tool operating parameters, including:

[0036] Based on the machine tool operating parameters, calculate the dynamic influence of each operating parameter on the nonlinear combination coefficients;

[0037] Based on the degree of dynamic influence, the nonlinear combination coefficients between multiple rigid correction components are adjusted in real time to reflect the true coupling relationship between the rigid correction components under the current working conditions.

[0038] Furthermore, based on the machine tool operating parameters, the dynamic influence of each operating parameter on the nonlinear combination coefficients is calculated, including:

[0039] To address the measurement noise of machine tool operating parameters, the monitored operating parameters are sampled at multiple points and processed using a moving average to obtain the processed operating parameters.

[0040] To address the measurement delay of each operating parameter, the processed operating parameters are timestamped and synchronized to obtain calibrated operating parameters.

[0041] Based on the calibrated operating parameters, the dynamic influence of each operating parameter on the nonlinear combination coefficients is calculated.

[0042] Furthermore, the dynamic influence of each operating parameter on the nonlinear combination coefficients is calculated, including:

[0043] A trend model is built based on historical operating data to predict the changing trends of machine tool operating parameters and obtain the operating parameter trends.

[0044] The trend of operating parameters is combined with the calibrated operating parameters and current oscillation amplitude to generate a prediction correction factor;

[0045] Based on the prediction correction factor, feedforward compensation is performed to correct the dynamic influence of each operating parameter on the nonlinear combination coefficient, so as to compensate for the dynamic deviation caused by the lag of each operating parameter.

[0046] Secondly, this application also discloses an axial positioning control system for a CNC lathe, comprising:

[0047] The drive module is used to drive the servo motor to perform backlash compensation motion at a preset low speed when the axial movement direction is reversed.

[0048] The acquisition module is used to acquire the position signal of the worktable and the load current signal of the servo motor at a synchronous sampling period during the backlash compensation motion.

[0049] The judgment module is used to make a combined judgment based on the position signal and the load current signal: within a preset duration window, it judges whether the change of the position signal along the command direction reaches a preset position threshold, and judges whether the increase of the load current signal relative to the reference current value before the backlash compensation movement reaches a preset current threshold.

[0050] The termination module is used to determine that the mechanical backlash has been eliminated when all the composite judgment items are met at the same time, terminate the backlash compensation motion, and switch to the normal dual closed-loop axial positioning control mode.

[0051] The switching module is used to maintain the backlash compensation motion when any one of the composite judgments is not satisfied.

[0052] Beneficial Effects: This application proposes an axial positioning control method for CNC lathes. The core of this method is to drive a servo motor at a preset low speed to perform backlash compensation motion when the axial movement direction reverses. During this process, the position signal of the worktable and the load current signal of the servo motor are acquired at a synchronous sampling period. Based on this, a composite judgment mechanism based on the position signal and load current signal is introduced. This solves the problem of misjudgment of backlash compensation caused by intermittent fluctuations in the grating ruler signal in existing technologies, thus avoiding continuous "creeping" of the worktable near the target position. By introducing the load current signal as an auxiliary judgment basis, the shortcomings of a single position signal under interference are compensated, more reliably reflecting the force state of the mechanical system. The composite judgment mechanism can effectively identify and ignore false fluctuations, improving the accuracy and robustness of the judgment. Thirdly, by precisely controlling the compensation termination timing, a smooth switch with the dual closed-loop control mode is achieved, avoiding insufficient or excessive compensation. In summary, this method significantly improves the axial positioning accuracy and machining stability of CNC lathes, and has important application value for precision manufacturing. Attached Figure Description

[0053] Figure 1 This is a flowchart illustrating an axial positioning control method for a CNC lathe provided in this application.

[0054] Figure 2 This application provides a schematic diagram of the axial positioning control system module structure for a CNC lathe.

[0055] In the diagram: 1. Driver module; 2. Acquisition module; 3. Judgment module; 4. Termination module; 5. Switching module. Detailed Implementation

[0056] The technical solutions of this application will now be clearly and completely described with reference to the accompanying drawings. Obviously, the described embodiments are merely some embodiments of this application, and not all embodiments. The components of this application described and shown in the accompanying drawings can generally be arranged and designed in various different configurations. Therefore, the following detailed description of the embodiments of this application provided in the accompanying drawings is not intended to limit the scope of the claimed application, but merely to illustrate selected embodiments of this application. All other embodiments obtained by those skilled in the art based on the embodiments of this application without inventive effort are within the scope of protection of this application.

[0057] It should be noted that similar reference numerals and letters in the following figures indicate similar items; therefore, once an item is defined in one figure, it does not need to be further defined and explained in subsequent figures. Furthermore, in the description of this application, terms such as "first," "second," etc., are used only to distinguish descriptions and should not be construed as indicating or implying relative importance.

[0058] Reference Figure 1 This application proposes an axial positioning control method for a CNC lathe, comprising:

[0059] S1000: When the axial movement direction is reversed, the servo motor is driven to perform backlash compensation movement at a preset low speed;

[0060] S2000: During the backlash compensation motion, the position signal of the worktable and the load current signal of the servo motor are acquired with a synchronous sampling period.

[0061] S3000: Based on position signal and load current signal, perform composite judgment: within a preset duration window, determine whether the change of position signal along the command direction reaches a preset position threshold, and determine whether the increase of load current signal relative to the reference current value before the backlash compensation movement reaches a preset current threshold.

[0062] S4000: When all the conditions of the composite judgment are met at the same time, it is determined that the mechanical backlash has been eliminated, the reverse backlash compensation motion is terminated, and the normal dual closed-loop axial positioning control mode is switched.

[0063] S5000: When any one of the composite judgments is not satisfied, maintain the backlash compensation motion.

[0064] Specifically, axial motion reversal refers to the process by which the movement direction of the CNC lathe table changes from one direction to the opposite direction when performing precise positioning or eliminating mechanical backlash. For example, axial motion reversal occurs when the table switches from positive to negative motion, or vice versa.

[0065] Servo motors are the actuators in CNC lathes that drive the worktable to make precise movements. Their output torque and speed are precisely regulated by the control system.

[0066] Backlash compensation motion refers to a pre-set, small-range reverse motion performed when the axial movement direction is reversed in order to eliminate the inherent mechanical backlash in transmission chains such as ball screws. Its purpose is to ensure that the backlash in the transmission chain has been completely eliminated before the worktable begins actual displacement.

[0067] The preset low speed refers to the speed at which the servo motor is driven when performing backlash compensation motion. This speed is typically low to ensure the smoothness of the compensation process and the accuracy of the judgment. The synchronous sampling period refers to maintaining a consistent data acquisition time interval between the position signal of the worktable and the load current signal of the servo motor to ensure the temporal correspondence of the data and facilitate subsequent composite judgment.

[0068] Position signals are typically provided by position sensors such as linear encoders or grating rulers, reflecting the actual position information of the worktable. Load current signals reflect the load borne by the servo motor when driving the worktable, and their changes are closely related to the force state of the mechanical system.

[0069] The preset duration window refers to a preset time period used to observe changes in the position signal and load current signal during composite judgment. The preset position threshold refers to a reference value used in composite judgment to determine whether the change in the position signal along the command direction has reached the expected value.

[0070] The preset current threshold is a reference value used in composite judgment to determine whether the increase in load current signal reaches the expected level. The reference current value is the load current value of the servo motor in no-load or stable state before the backlash compensation motion begins, and serves as a reference for subsequent judgment of the increase in load current.

[0071] Dual closed-loop axial positioning control mode is a high-precision control strategy that typically includes a position loop and a speed loop. It uses feedback from sensors such as grating rulers and encoders to achieve precise control of the worktable position.

[0072] Specifically, when the axial movement direction is reversed, the servo motor is driven at a preset low speed to perform backlash compensation movement in order to eliminate mechanical backlash. For example, when the table of a CNC lathe needs to be moved precisely from one position to another, and the direction needs to be changed during the movement, the system will activate backlash compensation to ensure positioning accuracy. At this time, the servo motor will move slightly in the opposite direction at a preset, relatively small speed, such as 10 millimeters per minute. This low-speed movement helps the system overcome mechanical backlash more smoothly and provides stable conditions for subsequent signal acquisition and judgment.

[0073] During backlash compensation, the system acquires the position signal of the worktable and the load current signal of the servo motor at a synchronous sampling period. For example, a high-precision grating ruler can be used as a position sensor to collect real-time position data of the worktable at a frequency of once per millisecond. Simultaneously, the real-time load current data of the servo motor is collected at the same frequency of once per millisecond through the current sensor inside the servo driver. This synchronous sampling ensures the precise temporal correspondence between position changes and current changes, providing a reliable data foundation for subsequent composite judgments.

[0074] Next, a combined judgment is made based on the acquired position signal and load current signal. This combined judgment includes two main aspects: within a preset duration window, it is determined whether the change in the position signal along the command direction reaches a preset position threshold, and whether the increase in the load current signal relative to the reference current value before the backlash compensation movement reaches a preset current threshold. For example, the system sets a duration window of 50 milliseconds. Within this window, if the cumulative change in the position signal fed back by the grating ruler along the command direction (i.e., the direction of the backlash compensation movement) reaches the preset 0.005 mm position threshold, it indicates that the worktable may have begun to move. Simultaneously, if the load current signal of the servo motor increases by 0.2 amps relative to the reference current value before the compensation movement (e.g., 0.5 amps under no-load), reaching the preset 0.2 amp current threshold, it indicates that the motor is overcoming mechanical resistance and may have already contacted the other side of the ball screw. Both conditions must be met simultaneously.

[0075] When all conditions of the composite judgment are met simultaneously, the system determines that the mechanical backlash has been eliminated, terminates the backlash compensation motion, and switches to the normal dual-closed-loop axial positioning control mode. For example, if the conditions of the above position signal and load current signal are met simultaneously within a 50-millisecond window, the system will immediately stop the backlash compensation motion of the servo motor and switch its control mode to a high-precision dual-closed-loop control mode for accurate position tracking and positioning.

[0076] Conversely, if any one of the composite judgment conditions is not met, the system will maintain the backlash compensation movement and ignore intermittent small fluctuations originating solely from the position signal. For example, if the change in the position signal reaches a threshold within a 50-millisecond window, but the increase in the load current signal does not, or vice versa, the system will continue to perform the backlash compensation movement. Furthermore, if only intermittent small fluctuations occur in the position signal (e.g., due to grating ruler contamination), but the load current signal does not show a corresponding significant increase, the system will ignore these position signal fluctuations to avoid making incorrect judgments due to the unreliability of a single signal.

[0077] This application effectively solves the misjudgment problem caused by grating ruler signal fluctuations during gap compensation, which may be encountered in traditional methods, by introducing the load current signal of the servo motor as an auxiliary judgment basis and combining it with the position signal for composite judgment. Compared with the traditional method that relies solely on the position signal, the solution of this application has significant advantages. When the grating ruler signal is contaminated or interfered with, the traditional method is prone to misjudging intermittent small fluctuations as the actual displacement of the worktable, thus causing the gap compensation to end prematurely and introducing positioning deviation. This deviation will cause the dual closed-loop control system to exhibit a "creeping" phenomenon near the target position, seriously affecting the machining accuracy.

[0078] Furthermore, in another embodiment of this application, the composite judgment further includes:

[0079] S3100: Superimposes micro-amplitude high-frequency oscillation commands onto the servo motor and synchronously acquires the load current signal;

[0080] S3200: Performs high-pass filtering on the load current signal to obtain the filtered load current signal;

[0081] S3300: Performs spectrum analysis on the filtered load current signal to extract the current oscillation amplitude corresponding to the frequency of the micro-amplitude high-frequency oscillation command;

[0082] S3400: Determines whether the current oscillation amplitude continuously exceeds the preset rigid contact oscillation amplitude threshold.

[0083] Specifically, superimposing a micro-amplitude high-frequency oscillation command onto the servo motor means adding a periodic signal with a small amplitude and high frequency to the servo motor's control commands, in addition to the normal backlash compensation motion command. The purpose of this command is to actively excite the mechanical transmission chain, causing it to generate a detectable dynamic response during the backlash elimination process. Synchronously acquiring the load current signal means that while superimposing the oscillation command, the actual load current of the servo motor is acquired with the same sampling period or synchronization clock to ensure the time correspondence between the oscillation command and the current response.

[0084] High-pass filtering of the load current signal aims to separate the response component caused by the micro-amplitude high-frequency oscillation command from the original load current signal, while suppressing low-frequency moving current, DC bias, and other low-frequency noise. This yields the filtered load current signal, which primarily reflects the dynamic characteristics of the mechanical system under high-frequency excitation. In practical applications, spectral analysis of the filtered load current signal, such as using Fast Fourier Transform (FFT), can accurately extract the current oscillation amplitude corresponding to the frequency of the micro-amplitude high-frequency oscillation command. This amplitude is a key indicator for measuring the rigidity response of the mechanical system under specific high-frequency excitation.

[0085] Determining whether the current oscillation amplitude continuously exceeds a preset rigid contact oscillation amplitude threshold involves comparing the extracted current oscillation amplitude with a pre-set threshold. When the ball screw nut assembly transitions from a loose state with gaps to a rigid contact state, the overall stiffness of the mechanical system increases significantly, resulting in a noticeable increase in the load current oscillation amplitude generated by the servo motor under the same micro-amplitude high-frequency oscillation command. Therefore, this threshold is set as a critical value that can distinguish between the gap state and the rigid contact state. Continuously exceeding this threshold indicates that the mechanical system has stably reached a rigid contact state, avoiding misjudgments caused by instantaneous noise or accidental fluctuations.

[0086] This application's solution, by introducing micro-amplitude high-frequency oscillation commands and analyzing the load current response of the servo motor, enables more accurate detection of the elimination of mechanical backlash. When the ball screw nut pair is in a gap state, due to the looseness, the oscillation amplitude of the micro-amplitude high-frequency oscillation command on the servo motor's load current is relatively small because the energy is absorbed or dissipated by the gap. However, once the ball screw nut pair achieves rigid contact, the stiffness of the mechanical transmission chain increases significantly, and the system's response characteristics to high-frequency excitation change, resulting in a significant increase in the load current oscillation amplitude of the servo motor under the same micro-amplitude high-frequency oscillation command. High-pass filtering effectively isolates this high-frequency response, and its amplitude is accurately extracted through spectrum analysis. Comparing this amplitude with a preset rigid contact oscillation amplitude threshold provides a more sensitive and reliable basis for judging the state of mechanical rigid contact. This method effectively compensates for the shortcomings of relying solely on position changes and load current increases, especially in the critical state where the gap is about to be eliminated, providing a clearer signal distinction.

[0087] In some preferred embodiments, as a specific implementation method: assuming that during the backlash compensation motion, a micro-amplitude high-frequency oscillation command with a frequency of 200Hz and an amplitude of 0.05A is superimposed on the servo motor. Simultaneously, the load current signal of the servo motor is synchronously acquired at a sampling frequency of 1kHz. Subsequently, the acquired load current signal is subjected to high-pass filtering, for example, using a digital high-pass filter with a cutoff frequency of 150Hz, to filter out low-frequency motion current and DC components. The filtered signal is then subjected to Fast Fourier Transform (FFT) for spectral analysis to extract the current oscillation amplitude corresponding to the 200Hz frequency component. When the backlash is not eliminated, this amplitude may fluctuate around 0.01A; however, when the ball screw nut pair achieves rigid contact, due to the increase in mechanical stiffness, this amplitude will rapidly rise and stabilize above 0.03A. At this time, the preset rigid contact oscillation amplitude threshold can be set to 0.025A. When the detected current oscillation amplitude continuously exceeds 0.025A, for example, if it is higher than this threshold for 10 consecutive sampling cycles, it can be determined that the mechanical backlash has been eliminated, thereby terminating the backlash compensation motion and switching to the normal dual-closed-loop axial positioning control mode. This judgment mechanism based on high-frequency response can effectively distinguish between the true rigid contact state of the mechanical system and the instantaneous signal caused only by friction or slight collision, ensuring the accuracy of backlash elimination.

[0088] Furthermore, in another embodiment of this application, S3400 includes:

[0089] S3410: Before executing the backlash compensation motion, a micro-amplitude high-frequency oscillation command is superimposed on the servo motor, and the load current signal of the servo motor is acquired synchronously.

[0090] S3420: Performs high-pass filtering on the load current signal to obtain the filtered load current signal, and performs spectrum analysis on the filtered load current signal to obtain the current oscillation amplitude corresponding to the frequency of the micro-amplitude high-frequency oscillation command.

[0091] S3430: The current oscillation amplitude is calibrated according to the machine tool operating parameters, and the rigid contact oscillation amplitude threshold is adjusted to the threshold under the current working conditions for composite judgment.

[0092] Specifically, before the backlash compensation motion begins, a micro-amplitude high-frequency oscillation command is superimposed onto the servo motor to excite the inherent vibration response of the mechanical system under rigid contact conditions. The synchronously acquired load current signal will contain a current oscillation component corresponding to the frequency of this oscillation command. High-pass filtering effectively removes the DC component and low-frequency noise, making the current oscillation amplitude corresponding to the frequency of the micro-amplitude high-frequency oscillation command more prominent. Subsequently, spectral analysis of the filtered load current signal can accurately extract the current oscillation amplitude corresponding to the frequency of the micro-amplitude high-frequency oscillation command. Machine tool operating parameters can be understood as various environmental or working condition parameters that affect the mechanical rigidity characteristics of the machine tool, such as ambient temperature, cumulative machine tool running time, cumulative axial travel, spindle speed, and feed rate—at least one of these. Changes in these parameters directly or indirectly affect the rigidity of mechanical transmission chains such as ball screw pairs and bearings, thereby altering the current oscillation amplitude generated during rigid contact. In practical applications, calibrating the current oscillation amplitude based on machine tool operating parameters refers to correcting the measured current oscillation amplitude according to the actual operating state of the machine tool, in order to eliminate or reduce measurement errors or deviations caused by changes in operating parameters. For example, when the ambient temperature rises, mechanical components may undergo thermal expansion, leading to a slight decrease in rigidity. In this case, the measured current oscillation amplitude needs to be appropriately corrected. Furthermore, adjusting the rigid contact oscillation amplitude threshold to the threshold under the current operating conditions aims to enable the judgment threshold to dynamically adapt to the actual working state of the machine tool. This means that the threshold is no longer a fixed constant, but is dynamically calculated and updated based on the real-time monitored machine tool operating parameters, thereby ensuring that under different operating conditions, the threshold can accurately represent the current oscillation amplitude characteristics when the mechanical system reaches a rigid contact state.

[0093] This application's solution calibrates the current oscillation amplitude by introducing machine tool operating parameters and dynamically adjusts the rigid contact oscillation amplitude threshold, effectively solving the problem of inaccurate judgment under varying operating conditions caused by traditional fixed thresholds. Specifically, when machine tool operating parameters change, such as increased temperature or wear, the rigidity of the mechanical system changes accordingly, leading to changes in the current oscillation amplitude generated during rigid contact. By acquiring and analyzing the load current signal before the backlash compensation motion and calibrating the obtained current oscillation amplitude in conjunction with machine tool operating parameters, the true rigidity state of the mechanical system under the current operating conditions can be more accurately reflected. Based on this, the rigid contact oscillation amplitude threshold is dynamically adjusted to the actual threshold under the current operating conditions, ensuring a high degree of match between the threshold used for composite judgment and the actual physical state of the machine tool. This avoids misjudgments or omissions caused by changes in environment or operating conditions, making the judgment of backlash elimination more accurate and reliable.

[0094] In some preferred embodiments, the following specific example illustrates the situation:

[0095] Assume that before the backlash compensation motion of a CNC lathe, the system monitors an ambient temperature of 25℃ and a cumulative operating time of 1000 hours. At this time, a micro-amplitude high-frequency oscillation command with a frequency of 500Hz and an amplitude of 0.1V is superimposed on the servo motor. Simultaneously, the load current signal of the servo motor is acquired, and after high-pass filtering and spectrum analysis, the current oscillation amplitude corresponding to the 500Hz frequency is found to be 0.05A. Based on preset temperature-rigidity correction models and operating time-rigidity correction models, the system calculates that under the current operating conditions, this 0.05A current oscillation amplitude needs to be calibrated upwards by 5%, i.e., the calibrated current oscillation amplitude is 0.0525A. Simultaneously, based on these machine tool operating parameters, the system dynamically calculates and adjusts the rigid contact oscillation amplitude threshold. For example, under standard operating conditions (20℃, new machine), this threshold might be set to 0.06A. However, under the current operating conditions of 25℃ and 1000 hours, the system calculates a comprehensive rigidity correction factor based on the preset parameter influence weight set, and adjusts the threshold to 0.058A accordingly. When the amplitude of the calibrated current oscillation continuously exceeds 0.058A during subsequent backlash compensation movements, it can be determined that the mechanical backlash has been eliminated, thus terminating the backlash compensation movement and switching to the normal dual-closed-loop axial positioning control mode. This dynamic adjustment mechanism ensures that the backlash elimination judgment remains accurate and reliable even when the actual operating state of the machine tool changes.

[0096] Furthermore, in another embodiment of this application, S3430 includes:

[0097] S3431: Monitor machine tool operating parameters, including at least one of ambient temperature, cumulative running time, cumulative axial travel, spindle speed, and feed rate;

[0098] S3432: Calculate the comprehensive rigidity correction factor based on machine tool operating parameters and a preset set of parameter influence weights;

[0099] S3433: Dynamically adjust the threshold value of rigid contact oscillation based on the current oscillation amplitude and the comprehensive rigidity correction factor.

[0100] Specifically, machine tool operating parameters refer to various measurable physical quantities or operating indicators that affect the rigidity of a mechanical system. Real-time monitoring of these parameters is crucial for accurately assessing the current operating condition of the machine tool. For example, ambient temperature affects the thermal expansion of materials and the viscosity of lubricating oil, thereby altering mechanical clearances and frictional characteristics; cumulative running time or cumulative axial travel can reflect the degree of wear on mechanical components, which can lead to increased clearances or decreased rigidity; spindle speed and feed rate directly affect cutting forces, vibrations, and bearing loads, all of which alter the dynamic rigidity of the system. By monitoring at least one of ambient temperature, cumulative running time, cumulative axial travel, spindle speed, and feed rate, comprehensive information affecting the rigidity of the machine tool can be obtained.

[0101] The comprehensive rigidity correction factor is a quantifiable index of the current rigidity state of a machine tool. This factor is calculated by combining multiple machine tool operating parameters with a preset set of parameter influence weights. The parameter influence weight set characterizes the relative importance or sensitivity of different operating parameters to the rigidity state of the machine tool. For example, under certain operating conditions, the influence of spindle speed on rigidity may be much greater than that of ambient temperature; in this case, the weight corresponding to spindle speed will be higher. By weighting and combining each operating parameter with its corresponding weight, a correction factor that comprehensively reflects the overall rigidity change of the machine tool under the current operating conditions can be obtained.

[0102] In practical applications, the dynamic adjustment of the rigid contact oscillation amplitude threshold is based on the currently acquired current oscillation amplitude and the calculated comprehensive rigidity correction factor. The current oscillation amplitude directly reflects the mechanical system's response to micro-amplitude high-frequency oscillation commands and is the direct basis for judging rigid contact. The comprehensive rigidity correction factor provides background information for calibrating this amplitude. By combining the current oscillation amplitude with the comprehensive rigidity correction factor, the preset rigid contact oscillation amplitude threshold can be adjusted in real time and adaptively. The purpose of this adjustment is to ensure that the threshold used accurately reflects the actual mechanical contact characteristics of the machine tool under complex working conditions such as current ambient temperature, wear level, and load conditions, thereby avoiding misjudgment or missed judgment due to inappropriate threshold.

[0103] This application's solution achieves dynamic adjustment of the rigid contact oscillation amplitude threshold by introducing the monitoring of multiple machine tool operating parameters and calculating a comprehensive rigidity correction factor based on these parameters and preset weights. Its working principle lies in the fact that traditional fixed thresholds or simple calibration thresholds are difficult to adapt to the complex mechanical characteristic changes of machine tools under different operating conditions. For example, when the ambient temperature rises, mechanical components may undergo thermal expansion, leading to a decrease in actual clearance or an increase in friction. Using the threshold set for low temperatures in this case may result in misjudgment. By monitoring parameters such as ambient temperature, cumulative running time, cumulative axial travel, spindle speed, and feed rate, the system can perceive changes in the machine tool's operating conditions in real time. These parameters are input into a calculation model based on a preset parameter influence weight set, generating a comprehensive rigidity correction factor. This correction factor quantifies the comprehensive impact of the current operating conditions on the rigidity of the mechanical system. Finally, the currently measured current oscillation amplitude is combined with this comprehensive rigidity correction factor to dynamically correct the rigid contact oscillation amplitude threshold. This dynamic correction mechanism makes the threshold no longer static, but adaptively adjustable according to the actual operating state of the machine tool, thereby ensuring that the determined mechanical backlash elimination point has higher accuracy and reliability under any working condition.

[0104] In some preferred embodiments, the following specific example illustrates the situation:

[0105] Suppose that during the operation of a CNC lathe, the system monitors an increase in ambient temperature from 20°C to 35°C, while the spindle speed increases from 1000 rpm to 5000 rpm. In a preset parameter influence weight set, ambient temperature has a weight of 0.3, and spindle speed has a weight of 0.5. The system first calculates the correction component for rigidity based on these parameter changes: ambient temperature (e.g., the rigidity correction factor increases by 0.01 for every 1°C increase in temperature) and spindle speed (e.g., the rigidity correction factor increases by 0.02 for every 1000 rpm increase in speed). Then, based on these correction components and their respective weights, a comprehensive rigidity correction factor is calculated. For example, if the initial rigid contact oscillation amplitude threshold is X, the calculated comprehensive rigidity correction factor under the combined effects of temperature and speed increases is 1.1. At this point, the system combines the current current oscillation amplitude with this correction factor to dynamically adjust the rigid contact oscillation amplitude threshold to X*1.1 to adapt to the actual mechanical rigidity state under the current high temperature and high speed conditions. This dynamic adjustment ensures that the system can accurately determine whether mechanical backlash has been eliminated even when operating conditions change significantly, thus guaranteeing the accuracy and stability of axial positioning.

[0106] Furthermore, in another embodiment of this application, the axial positioning control method for a CNC lathe further includes:

[0107] S6000: Before performing the backlash compensation motion, the actual backlash value and friction torque parameter of the ball screw are measured by micro-amplitude detection motion. Based on the actual backlash value and friction torque parameter, a backlash crossing optimization trajectory is generated. The backlash compensation motion is completed along the backlash crossing optimization trajectory. After the backlash compensation motion is completed, the system smoothly switches to the normal dual closed-loop axial positioning control mode.

[0108] Specifically, micro-amplitude detection motion refers to applying a very small displacement command to the servo motor before the backlash compensation motion begins, causing the worktable to perform a tiny reciprocating motion within the backlash range. During this process, the critical point at which the ball screw contacts the other side can be precisely captured, thereby measuring the actual backlash value of the ball screw. Simultaneously, by monitoring the current or torque response of the servo motor, the frictional torque parameters of the ball screw during the backlash crossing process can be evaluated, such as static friction torque and dynamic friction torque.

[0109] The process of generating an optimized clearance crossing trajectory based on the actual clearance value and friction torque parameters can be understood as dynamically planning a motion trajectory best suited to the current working conditions based on the actually measured mechanical characteristics. For example, this trajectory could be a non-linear velocity curve, crossing the clearance at a relatively high speed and then smoothly decelerating just before contact to avoid impact and ensure precise contact. The goal is to minimize compensation time and reduce mechanical wear while ensuring clearance elimination.

[0110] In practical applications, the backlash compensation motion is achieved by traversing an optimized trajectory along the gap. Specifically, this means that the servo motor no longer simply moves at a preset low speed, but strictly follows a pre-generated optimized trajectory. For example, the control system precisely controls the servo motor output based on the speed and acceleration commands in the optimized trajectory, enabling the worktable to smoothly and efficiently traverse the mechanical backlash along the planned path. After the backlash compensation motion is completed, a smooth switch to the normal dual-closed-loop axial positioning control mode is achieved. This ensures that after the backlash is eliminated, the machine tool can seamlessly enter a high-precision positioning and machining state, avoiding transient shocks or positional deviations caused by control mode switching, thereby guaranteeing machining accuracy and machine tool stability.

[0111] The proposed solution introduces a micro-amplitude probe motion before the backlash compensation motion, enabling precise measurement of the actual backlash value and frictional torque parameters of the ball screw. It is precisely this real-time, accurate mechanical characteristic data that allows the system to dynamically generate an optimized backlash crossing trajectory based on the current operating conditions. This optimized trajectory fully considers the backlash size and frictional characteristics, allowing it to cross the backlash with the most suitable speed and acceleration curves, avoiding the inefficiency or impact problems that may arise from traditional fixed low-speed compensation. By performing compensation motion along this optimized trajectory, not only can the mechanical backlash be eliminated more quickly and smoothly, but also, after compensation, a smooth switching mechanism ensures seamless transition of control modes, effectively avoiding transient disturbances caused by mode switching, thereby guaranteeing the accuracy and stability of subsequent positioning control.

[0112] In some preferred embodiments, the following specific example illustrates the situation:

[0113] Suppose that after a CNC lathe has been running for a long time, the mechanical backlash and friction characteristics of its ball screw may change slightly. When a reversal of the axial motion is required, the control system first triggers a micro-amplitude detection motion. For example, the servo motor is instructed to perform a small reciprocating motion within a range of ±0.01mm at a speed of 0.1mm / s, while a high-precision encoder records the position of the worktable and a current sensor records the load current of the servo motor. By analyzing the abrupt changes in the position and current signals, the system accurately measures the current actual backlash of the ball screw to be 0.05mm, and there is a static friction torque of 0.5Nm when crossing the backlash.

[0114] Based on these measurement data, the control system dynamically generates an optimized clearance crossing trajectory. For example, this trajectory might be designed as follows: accelerating through the initial 0.02mm of the clearance at a speed of 0.5mm / s; maintaining the speed at 0.5mm / s for the next 0.02mm; and smoothly decelerating to 0.1mm / s just before contact in the final 0.01mm to ensure flexible contact. Subsequently, the servo motor drives the worktable to perform reverse clearance compensation motion strictly according to this optimized trajectory. Once the worktable completes the clearance crossing according to the optimized trajectory and reaches the preset contact state, the control system immediately and smoothly switches to the normal dual-closed-loop axial positioning control mode, enabling the machine tool to quickly and accurately enter the next machining command. The entire process is efficient and shock-free.

[0115] Furthermore, in another embodiment of this application, S3432 includes:

[0116] A1000: Before performing backlash compensation motion, calculate multiple independent rigidity correction components based on machine tool operating parameters and parameter influence weight set;

[0117] A2000: Dynamically adjusts nonlinear combination coefficients based on machine tool operating parameters;

[0118] A3000: The adjusted nonlinear combination coefficients are used to perform a nonlinear combination on multiple independent rigid correction components to obtain a comprehensive rigid correction factor. The nonlinear combination is an adaptive nonlinear combination.

[0119] Specifically, multiple independent rigidity correction components refer to correction values ​​calculated separately for different machine tool operating parameters (such as ambient temperature, cumulative running time, cumulative axial travel, spindle speed, and feed rate), reflecting the degree of influence of each parameter on mechanical rigidity. For example, a temperature correction component can be calculated for ambient temperature, and a speed correction component can be calculated for spindle speed. The purpose is to decompose the complex combined effect into sub-items that can be independently analyzed and calculated. The nonlinear combination coefficients can be understood as parameters used to adjust the interaction strength between different rigidity correction components. These coefficients are not fixed but dynamically adjusted based on machine tool operating parameters to adapt to different working conditions. The purpose is to more accurately simulate the nonlinear coupling effect of each operating parameter on mechanical rigidity. In practical applications, adaptive nonlinear combination refers to a nonlinear function that can automatically adjust the combination method and / or combination coefficients based on real-time or historical operating data. For example, machine learning algorithms such as neural networks, fuzzy logic, and support vector machines can be used, or parameters can be optimized based on a preset nonlinear function model (such as polynomials, exponential functions, etc.) to achieve nonlinear combination of multiple rigidity correction components. Its purpose is to ensure that the comprehensive stiffness correction factor can accurately reflect the actual mechanical stiffness state under the current complex working conditions.

[0120] This application's solution effectively solves the problem of insufficient accuracy in rigidity correction factor calculation caused by the difficulty in accurately capturing the complex nonlinear coupling relationship between parameters in traditional methods by decomposing the calculation of the comprehensive rigidity correction factor into multiple independent rigidity correction components and introducing nonlinear combination coefficients based on dynamic adjustment of machine tool operating parameters, as well as adopting an adaptive nonlinear combination method. Specifically, by calculating multiple independent rigidity correction components, the individual influence of each operating parameter on mechanical rigidity can be quantified more precisely; while the nonlinear combination coefficients based on dynamic adjustment of machine tool operating parameters allow these independent components to be combined in a way that is more in line with physical laws according to the actual working conditions, avoiding the limitations of fixed weights or linear combinations; finally, the introduction of adaptive nonlinear combination further enhances the system's adaptability to complex working conditions and the accuracy of the correction factor, ensuring that the comprehensive rigidity correction factor can accurately reflect the actual mechanical state under the current working conditions.

[0121] In some preferred embodiments, the following specific example illustrates the situation:

[0122] Assume the machine tool operating parameters include ambient temperature, cumulative running time, and spindle speed. Before performing backlash compensation, the system can calculate three independent rigidity correction components based on these parameters: a temperature correction component, a time correction component, and a speed correction component. For example, the temperature correction component could be a function inversely proportional to the ambient temperature, the time correction component could be a function directly proportional to the cumulative running time (reflecting wear), and the speed correction component could be a function related to the spindle speed (reflecting vibration effects). Furthermore, based on the current machine tool operating parameters, the system can dynamically adjust the nonlinear combination coefficients. For example, a neural network-based model can be preset, taking ambient temperature, cumulative running time, and spindle speed as inputs and outputting three nonlinear combination coefficients. These coefficients reflect the relative weights and nonlinear coupling relationships of the effects of temperature, time, and speed on overall rigidity under the current operating conditions. Finally, the adjusted nonlinear combination coefficients are used to adaptively and nonlinearly combine the multiple independent rigidity correction components to obtain a comprehensive rigidity correction factor. This adaptive nonlinear combination method can more accurately capture the complex interactions between different operating parameters, thereby enabling the comprehensive rigidity correction factor to accurately reflect the actual mechanical state under the current working conditions, providing a more reliable basis for the dynamic adjustment of the subsequent rigid contact oscillation amplitude threshold.

[0123] Furthermore, in another embodiment of this application, A2000 includes:

[0124] A2100: Calculate the dynamic influence of each operating parameter on the nonlinear combination coefficient based on the machine tool operating parameters;

[0125] A2200: Based on the degree of dynamic influence, the nonlinear combination coefficients between multiple rigid correction components are adjusted in real time to reflect the true coupling relationship between the rigid correction components under the current working conditions.

[0126] Specifically, "calculating the dynamic influence of each operating parameter on the nonlinear combination coefficients" refers to conducting in-depth analysis of monitored machine tool operating parameters such as ambient temperature, cumulative operating time, cumulative axial travel, spindle speed, and feed rate to quantify the specific contribution and influence weight of each parameter to the nonlinear combination coefficients under the current operating conditions. This can be achieved using pre-established mathematical models, machine learning algorithms (e.g., regression analysis, neural networks), or rule bases based on expert experience, with the aim of obtaining a dynamic weight set that accurately reflects the influence of each parameter.

[0127] The phrase "adjusting the nonlinear combination coefficients between multiple rigidity correction components in real time based on the degree of dynamic influence" can be understood as follows: once the degree of dynamic influence of each operating parameter is obtained, the system will update and optimize the nonlinear combination coefficients used to combine the various rigidity correction components in real time according to these degrees of dynamic influence. For example, if the influence of a certain operating parameter (such as spindle speed) on mechanical rigidity is significantly enhanced under the current operating conditions, the weight or mode of action of the rigidity correction component related to that parameter in the nonlinear combination will be adjusted accordingly to ensure that its representation in the comprehensive rigidity correction factor is more accurate.

[0128] In practical applications, "reflecting the true coupling relationship between various rigidity correction components under the current operating conditions" specifically refers to using the calculation and real-time adjustment of the aforementioned dynamic influence level to ensure that the nonlinear combination coefficients can accurately simulate and express the interaction and dependence between different rigidity correction components under the current actual operating conditions. For example, under high-temperature and high-speed operating conditions, the rigidity changes caused by thermal deformation and those caused by dynamic loads may have specific synergistic or antagonistic effects. This scheme aims to adjust the combination coefficients to ensure that these complex coupling relationships are realistically and accurately reflected in the comprehensive rigidity correction factor.

[0129] This application's solution addresses the problem of inaccurate parameter coupling relationships in traditional methods when handling complex operating conditions by introducing the calculation of the dynamic influence of each operating parameter and adjusting the nonlinear combination coefficients in real time based on this calculation. It is precisely because the dynamic contribution of each machine tool operating parameter to the nonlinear combination coefficients can be accurately quantified, and the combination of rigid correction components adjusted accordingly, that the comprehensive rigid correction factor can more realistically and precisely reflect the actual mechanical state under the current operating conditions. This mechanism ensures that the nonlinear combination coefficients can adaptively adjust when machine tool operating parameters change, thereby avoiding compensation errors caused by insufficient understanding of parameter coupling relationships.

[0130] In some preferred embodiments, the following specific example illustrates the situation:

[0131] Assume that during the machining process of a CNC lathe, machine tool operating parameters such as ambient temperature, spindle speed, and feed rate continuously change. To accurately adjust the nonlinear combination coefficients, the following method can be used:

[0132] First, a machine learning model, such as a multilayer perceptron neural network, is established based on historical operating data and expert knowledge. The model's input includes current machine tool operating parameters (such as ambient temperature, spindle speed, and feed rate), and its output is the dynamic influence of each operating parameter on the nonlinear combination coefficients. For example, when the ambient temperature rises, the model may output a higher influence value, indicating that temperature should be more sensitive to adjustments in the nonlinear combination coefficients.

[0133] Secondly, during the backlash compensation process, the system monitors the machine tool's operating parameters in real time and inputs these parameters into a pre-trained machine learning model. Based on the current operating parameters, the model calculates the dynamic influence of ambient temperature, spindle speed, feed rate, and other factors on the nonlinear combination coefficients.

[0134] Finally, based on these calculated dynamic influence levels, the system adjusts the nonlinear combination coefficients between multiple rigidity correction components in real time (e.g., one component reflects the rigidity change caused by temperature, and another reflects the rigidity change caused by spindle speed). For example, if the model indicates that under the current high-temperature, high-speed operating conditions, the weight of the spindle speed's influence on rigidity should be increased, while the weight of the ambient temperature's influence should be slightly decreased, the nonlinear combination coefficients will be adjusted accordingly. This adjustment ensures that the comprehensive rigidity correction factor can accurately reflect the complex interactions and coupling relationships between different physical factors (reflected through the rigidity correction components) under the current operating conditions, thus providing a more accurate basis for the dynamic adjustment of the rigid contact oscillation amplitude threshold.

[0135] Furthermore, in another embodiment of this application, A2100 includes:

[0136] A2110: To address the measurement noise of machine tool operating parameters, the monitored operating parameters are sampled at multiple points and processed by moving average to obtain the processed operating parameters;

[0137] A2120: To address the measurement delay of each operating parameter, the processed operating parameters are timestamped and synchronized to obtain calibrated operating parameters;

[0138] A2130: Based on the calibrated operating parameters, calculate the dynamic influence of each operating parameter on the nonlinear combination coefficients.

[0139] Specifically, regarding measurement noise of machine tool operating parameters, multi-point sampling and moving average processing of the monitored operating parameters can effectively suppress the influence of random noise on the measurement results. Multi-point sampling refers to performing multiple consecutive measurements on the same parameter within a short period of time to obtain a set of data points. Moving average processing involves averaging these consecutive sampling points. For example, using a moving average filter, the data curve is smoothed by averaging the data over a certain time window, reducing the amplitude of instantaneous noise and obtaining more stable and reliable processed operating parameters. The aim is to improve the signal-to-noise ratio of the operating parameter data, providing more accurate input for subsequent calculations.

[0140] Specifically, to address the measurement delays of various operating parameters, timestamp calibration and synchronization are performed on the processed operating parameters to ensure time consistency between different operating parameters. Measurement delays may originate from sensor response time, data acquisition system processing speed, or differences in data transmission paths. Timestamp calibration involves attaching a precise timestamp to each measurement data point to record its acquisition time. Synchronization processing involves aligning the data of different parameters based on these timestamps, for example, through interpolation or resampling, so that all relevant operating parameter data can be accurately acquired and used at the same point in time, resulting in calibrated operating parameters. The purpose is to eliminate errors caused by asynchronous data acquisition and transmission, ensuring that the parameter data used when calculating the degree of dynamic influence are time-matched.

[0141] In practical applications, processed operating parameters refer to machine tool operating parameter data after noise suppression, such as ambient temperature, cumulative running time, cumulative axial travel, spindle speed, and feed rate after moving average filtering. Calibrated operating parameters refer to operating parameter data that, based on processed operating parameters, have undergone further timestamp calibration and synchronization processing to ensure time consistency.

[0142] This application's solution improves the quality of machine tool operating parameter data from the source by performing noise suppression and time synchronization processing. Specifically, multi-point sampling and moving average processing effectively filter out random noise generated during the measurement process, making the acquired operating parameter data smoother and more accurate, avoiding misjudgments caused by noise. Simultaneously, timestamp calibration and synchronization processing solve the measurement delay problem that may exist between different sensors or data channels, ensuring that when calculating the dynamic influence of each operating parameter on the nonlinear combination coefficients, all relevant operating parameter data can accurately reflect the machine tool's operating condition at the same moment. It is precisely because of the provision of high-precision, highly synchronized operating parameter data that the subsequent calculation of the dynamic influence of each operating parameter is more accurate, thereby enabling more precise adjustment of the nonlinear combination coefficients to truly reflect the coupling relationship between the rigid correction components under the current operating condition.

[0143] In some preferred embodiments, the following specific example illustrates the situation:

[0144] Suppose we need to monitor three machine tool operating parameters: ambient temperature, spindle speed, and feed rate. To address the measurement noise of ambient temperature, we can configure the temperature sensor to sample every 100 milliseconds, continuously collecting 10 data points. Then, we can perform a moving average processing on these 10 data points, for example, calculating their arithmetic mean, to obtain a smoothed ambient temperature value.

[0145] For spindle speed and feed rate measurements, there may be microsecond-level delays due to the possibility of acquisition by different encoders or sensors and different data transmission paths. Therefore, a high-precision timestamp is appended to each speed and feed rate data point during data acquisition. Before calculating the dynamic impact of each operating parameter, the system synchronizes the spindle speed and feed rate data based on these timestamps. For example, using the sampling time of ambient temperature data as a reference, methods such as linear interpolation or nearest-neighbor interpolation are used to adjust the spindle speed and feed rate data to the same time point as the ambient temperature data, thus obtaining calibrated operating parameters.

[0146] Based on these calibrated operating parameters that have undergone noise suppression and time synchronization processing, the system can more accurately calculate the dynamic influence of ambient temperature, spindle speed, and feed rate on the nonlinear combination coefficients, thereby guiding the real-time adjustment of the nonlinear combination coefficients.

[0147] Furthermore, in another embodiment of this application, A2130 includes:

[0148] A2131: Based on historical operating data, a trend model is constructed to predict the changing trends of machine tool operating parameters and obtain the operating parameter trends;

[0149] A2132: Combine the operating parameter trends with the calibrated operating parameters and current oscillation amplitude to generate a prediction correction factor;

[0150] A2133: Based on the prediction correction factor, feedforward compensation is performed to correct the dynamic influence of each operating parameter on the nonlinear combination coefficient, so as to compensate for the dynamic deviation caused by the lag of each operating parameter.

[0151] Specifically, building trend models based on historical operating data refers to the system continuously collecting and storing historical data on the operating parameters of CNC lathes under different working conditions, such as ambient temperature, cumulative operating time, cumulative axial travel, spindle speed, and feed rate. This historical data can be used to train various predictive models, such as time series models (e.g., ARIMA, exponential smoothing), machine learning models (e.g., support vector machines, neural networks), or statistical regression models. Through these models, the changing trends of machine tool operating parameters over a future period can be predicted, thus obtaining the operating parameter trends. The purpose is to anticipate the dynamic changes in machine tool operating parameters in advance, providing a basis for subsequent compensation.

[0152] Specifically, the predicted correction factor is generated by combining the trend of operating parameters with the calibrated operating parameters and the current oscillation amplitude. This means that the predicted trend of operating parameters is not directly used for correction, but is comprehensively considered in conjunction with the calibrated operating parameters processed for noise and delay at the current moment, as well as the real-time acquired current oscillation amplitude. For example, these three types of information can be fused using methods such as weighted averaging, fuzzy logic, or neural networks to generate a predicted correction factor that reflects future dynamic changes and the current actual state. This predicted correction factor aims to quantify the deviation in the degree of dynamic impact that may be caused by parameter lag.

[0153] In practical applications, the dynamic influence of each operating parameter on the nonlinear combination coefficients is compensated by a feedforward compensation based on a prediction correction factor. This means that when calculating the dynamic influence of the nonlinear combination coefficients, the calculation no longer relies solely on the current calibrated operating parameters, but instead incorporates the prediction correction factor as a feedforward signal. For example, a term related to the prediction correction factor can be added to the formula for calculating the dynamic influence, or the initially calculated dynamic influence can be adjusted based on the prediction correction factor using lookup tables, mapping functions, etc. The purpose is to correct the dynamic influence before it actually occurs, thereby effectively compensating for dynamic deviations caused by the lag of each operating parameter.

[0154] The proposed solution constructs a trend model based on historical operating data and makes predictions, enabling early insight into the potential direction and magnitude of changes in machine tool operating parameters. This forward-looking predictive capability allows the system to generate a predictive correction factor based on the current calibrated operating parameters and current oscillation amplitude before actual parameter changes occur. This prediction correction factor, acting as a feedforward signal, is used to compensate for the dynamic influence of each operating parameter on the nonlinear combination coefficients. Through this feedforward compensation mechanism, the system effectively overcomes the dynamic deviation caused by the lag in operating parameters in traditional methods, ensuring that the adjustment of the nonlinear combination coefficients takes future trends into account before actual parameter changes occur, thereby improving the real-time performance and accuracy of dynamic influence calculation.

[0155] In some preferred embodiments, the following specific example illustrates the situation:

[0156] Assuming that after a CNC lathe has been running for a long time, the ambient temperature and spindle speed will gradually increase, and these changes will affect the rigidity of the mechanical system. In traditional solutions, even if these parameters are monitored and calibrated in real time, the inherent delays in sensor response, data transmission, and processing mean that the dynamic impact calculated by the system will always lag behind when the actual temperature or speed changes. To address this issue, this application implements the following: First, the system continuously records historical data of machine tool operating parameters such as ambient temperature and spindle speed over the past few hours or even days. Using this historical data, a trend model based on time series analysis can be constructed. For example, exponential smoothing or Kalman filtering can be used to predict the future trends of ambient temperature and spindle speed, thereby obtaining the operating parameter trends for the next 10 seconds.

[0157] Secondly, during backlash compensation, the system acquires real-time data on the calibrated ambient temperature, spindle speed, and current oscillation amplitude. This real-time data is then combined with previously predicted operating parameter trends. For example, a weighted function can be designed, assigning a certain weight to the predicted trend (e.g., 0.3), a higher weight to the current calibrated operating parameters (e.g., 0.5), and a remaining weight to the current oscillation amplitude (e.g., 0.2). This combined approach calculates a prediction correction factor. This correction factor reflects the potential deviation in the degree of dynamic impact caused by parameter lag under the current operating conditions.

[0158] Finally, when calculating the dynamic impact of each operating parameter on the nonlinear combination coefficients, the system incorporates this prediction correction factor as a feedforward. For example, if the prediction correction factor indicates that the future temperature will rise, leading to a decrease in mechanical stiffness, then when calculating the dynamic impact, even if the current calibrated temperature has not yet fully reflected the increased state, the system will pre-adjust the nonlinear combination coefficients to better reflect the lower stiffness state. Through this feedforward compensation correction, the system can pre-adjust the nonlinear combination coefficients before the actual temperature or speed changes are fully reflected, thereby effectively compensating for the dynamic deviations caused by the lag of each operating parameter, ensuring the accuracy of the gap elimination judgment and the stability of the control.

[0159] In the field of modern precision manufacturing, the axial positioning accuracy of CNC lathes is a key factor determining product quality. Traditional CNC lathes, when reversing the axial movement direction for backlash compensation, suffer from misjudgments due to intermittent fluctuations in the grating ruler signal. This causes the dual closed-loop control system to become unstable near the target position, exhibiting a continuous "creeping" phenomenon that severely affects machining accuracy.

[0160] Reference Figure 2This application further proposes an axial positioning control system for a CNC lathe, comprising:

[0161] Drive module 1 is used to drive the servo motor to perform backlash compensation motion at a preset low speed when the axial movement direction is reversed.

[0162] Acquisition module 2 is used to acquire the position signal of the worktable and the load current signal of the servo motor at a synchronous sampling period during the backlash compensation motion.

[0163] The judgment module 3 is used to make a composite judgment based on the position signal and the load current signal: within the preset duration window, it judges whether the change of the position signal along the command direction reaches the preset position threshold, and judges whether the increase of the load current signal relative to the reference current value before the backlash compensation movement reaches the preset current threshold.

[0164] Termination module 4 is used to determine that the mechanical backlash has been eliminated when all the composite judgment items are met at the same time, terminate the backlash compensation motion, and switch to the normal dual closed-loop axial positioning control mode.

[0165] Switching module 5 is used to maintain the backlash compensation motion when any one of the composite judgments is not satisfied.

[0166] The CNC lathe axial positioning control system of this application, through the coordinated operation of drive module 1, acquisition module 2, judgment module 3, termination module 4, and switching module 5, particularly the judgment module 3 employing a composite judgment mechanism of position signal and servo motor load current signal, effectively solves the misjudgment problem caused by grating ruler signal fluctuations that may occur during the backlash compensation process in traditional methods. Compared to traditional methods that rely solely on position signals, this system, when judging the elimination of mechanical backlash, not only considers the position change of the worktable but also introduces the load current signal reflecting the stress state of the mechanical system. When the mechanical backlash is truly eliminated and the ball screw transmission chain reaches rigid contact, the resistance that the servo motor needs to overcome increases significantly, resulting in a significant increase in load current. By simultaneously monitoring the change in position signal and the increase in load current signal through judgment module 3, and requiring both to simultaneously meet preset thresholds, this system can effectively distinguish between false displacement caused by grating ruler signal fluctuations and actual mechanical contact. Therefore, this system significantly improves the accuracy and robustness of gap elimination judgment, ensures the accuracy and stability of axial positioning of CNC lathes in complex industrial environments, effectively eliminates the "creeping" phenomenon of the worktable near the target position, and thus guarantees the quality of precision machining tasks.

[0167] The above are merely embodiments of this application and are not intended to limit the scope of protection of this application. Various modifications and variations can be made to this application by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the scope of protection of this application.

Claims

1. A method for axial positioning control of a CNC lathe, characterized in that, include: When the axial movement direction is reversed, the servo motor is driven to perform backlash compensation movement at a preset low speed; During the backlash compensation motion, the position signal of the worktable and the load current signal of the servo motor are acquired with a synchronous sampling period. Based on the position signal and the load current signal, a combined judgment is made: within a preset duration window, it is determined whether the change in the position signal along the command direction reaches a preset position threshold, and whether the increase in the load current signal relative to the reference current value before the backlash compensation movement reaches a preset current threshold. When all the conditions of the composite judgment are met at the same time, it is determined that the mechanical backlash has been eliminated, the reverse backlash compensation movement is terminated, and the normal dual closed-loop axial positioning control mode is switched. If any of the composite judgments is not satisfied, the reverse backlash compensation motion is maintained. The composite judgment also includes: A micro-amplitude high-frequency oscillation command is superimposed on the servo motor, and the load current signal is acquired synchronously. The load current signal is subjected to high-pass filtering to obtain the filtered load current signal; Spectral analysis is performed on the filtered load current signal to extract the current oscillation amplitude corresponding to the frequency of the micro-amplitude high-frequency oscillation command; Determine whether the current oscillation amplitude continuously exceeds the preset rigid contact oscillation amplitude threshold.

2. The axial positioning control method for CNC lathes according to claim 1, characterized in that, The step of determining whether the current oscillation amplitude continuously exceeds a preset rigid contact oscillation amplitude threshold includes: Before performing the backlash compensation motion, a micro-amplitude high-frequency oscillation command is superimposed on the servo motor, and the load current signal of the servo motor is acquired simultaneously. The load current signal is subjected to high-pass filtering to obtain a filtered load current signal, and the filtered load current signal is subjected to spectrum analysis to obtain the current oscillation amplitude corresponding to the frequency of the micro-amplitude high-frequency oscillation command. The current oscillation amplitude is calibrated according to the machine tool operating parameters, and the rigid contact oscillation amplitude threshold is adjusted to the threshold under the current working condition for the composite judgment.

3. The axial positioning control method for CNC lathes according to claim 2, characterized in that, The step of calibrating the current oscillation amplitude based on machine tool operating parameters and adjusting the rigid contact oscillation amplitude threshold accordingly to the threshold under the current operating conditions includes: The machine tool operating parameters are monitored, including at least one of ambient temperature, cumulative running time, cumulative axial travel, spindle speed, and feed rate; Based on the machine tool operating parameters and the preset parameter influence weight set, a comprehensive rigidity correction factor is calculated; The threshold value of the rigid contact oscillation is dynamically adjusted based on the current oscillation amplitude and the comprehensive rigidity correction factor.

4. The axial positioning control method for a CNC lathe according to claim 1, characterized in that, The method further includes: Before performing the backlash compensation motion, the actual backlash value and friction torque parameter of the ball screw are measured by micro-amplitude detection motion, and a backlash crossing optimization trajectory is generated based on the actual backlash value and friction torque parameter. The backlash compensation motion is completed along the backlash crossing optimization trajectory, and after the backlash compensation motion is completed, the system smoothly switches to the normal dual closed-loop axial positioning control mode.

5. The axial positioning control method for a CNC lathe according to claim 3, characterized in that, The calculation of the comprehensive rigidity correction factor based on the machine tool operating parameters and a preset set of parameter influence weights includes: Before performing the backlash compensation motion, multiple independent rigidity correction components are calculated based on the machine tool operating parameters and the parameter influence weight set; The nonlinear combination coefficients are dynamically adjusted based on the machine tool operating parameters. The adjusted nonlinear combination coefficients are used to perform a nonlinear combination on multiple independent rigid correction components to obtain the comprehensive rigid correction factor. The nonlinear combination is an adaptive nonlinear combination.

6. The axial positioning control method for a CNC lathe according to claim 5, characterized in that, The dynamic adjustment of the nonlinear combination coefficients based on the machine tool operating parameters includes: Based on the machine tool operating parameters, calculate the dynamic influence of each operating parameter on the nonlinear combination coefficients; Based on the dynamic influence level, the nonlinear combination coefficients between the multiple rigid correction components are adjusted in real time to reflect the true coupling relationship between the rigid correction components under the current working condition.

7. The axial positioning control method for a CNC lathe according to claim 6, characterized in that, The step of calculating the dynamic influence of each operating parameter on the nonlinear combination coefficients based on the machine tool operating parameters includes: To address the measurement noise of the machine tool operating parameters, the monitored operating parameters are sampled at multiple points and processed using a moving average to obtain the processed operating parameters. To address the measurement delay of each operating parameter, the processed operating parameters are timestamped and synchronized to obtain calibrated operating parameters. Based on the calibrated operating parameters, the dynamic influence of each operating parameter on the nonlinear combination coefficients is calculated.

8. The axial positioning control method for a CNC lathe according to claim 7, characterized in that, The calculation of the dynamic influence of each operating parameter on the nonlinear combination coefficients includes: A trend model is constructed based on historical operating data to predict the changing trends of the machine tool's operating parameters, thereby obtaining the operating parameter trends; The trend of the operating parameters is combined with the calibrated operating parameters and the current oscillation amplitude to generate a prediction correction factor; Based on the prediction correction factor, the dynamic influence of each operating parameter on the nonlinear combination coefficient is corrected by feedforward compensation to compensate for the dynamic deviation caused by the lag of each operating parameter.

9. An axial positioning control system for a CNC lathe, characterized in that, The method for axial positioning control of a CNC lathe as described in claim 1 includes: The drive module is used to drive the servo motor to perform backlash compensation motion at a preset low speed when the axial movement direction is reversed. The acquisition module is used to acquire the position signal of the worktable and the load current signal of the servo motor at a synchronous sampling period during the backlash compensation motion. The judgment module is used to perform a composite judgment based on the position signal and the load current signal: within a preset duration window, it judges whether the change in the position signal along the command direction reaches a preset position threshold, and whether the increase in the load current signal relative to the reference current value before the backlash compensation movement reaches a preset current threshold; it superimposes a micro-amplitude high-frequency oscillation command onto the servo motor and synchronously acquires the load current signal; it performs high-pass filtering on the load current signal to obtain a filtered load current signal; it performs spectrum analysis on the filtered load current signal to extract the current oscillation amplitude corresponding to the frequency of the micro-amplitude high-frequency oscillation command; and it judges whether the current oscillation amplitude continuously exceeds a preset rigid contact oscillation amplitude threshold. The termination module is used to determine that the mechanical backlash has been eliminated when all the composite judgment items are satisfied at the same time, terminate the backlash compensation movement, and switch to the normal dual closed-loop axial positioning control mode. The switching module is used to maintain the reverse backlash compensation movement when any one of the composite judgments is not satisfied.

Images