Five-axis linkage numerical control machining dynamic error compensation system based on industrial control system

Abstract

The invention belongs to the technical field of five-axis linkage numerical control machining, and particularly relates to a five-axis linkage numerical control machining dynamic error compensation system based on an industrial control system, which comprises a split-axis error detection module, a dynamic compensation model construction module, an industrial control real-time compensation execution module, an RTCP dynamic calibration unit and a self-adaptive learning optimization module. According to the invention, a real-time residual error monitoring and feedback mechanism is introduced to establish data connection with an original adaptive optimization module, so that parameter adaptive adjustment based on an actual compensation effect is realized, and the adaptability and compensation precision of the system under a complex dynamic processing condition are enhanced.

Description

Technical Field

[0001] The present invention belongs to the technical field of five-axis linkage numerical control machining, and in particular relates to a five-axis linkage numerical control machining dynamic error compensation system based on an industrial control system. Background Art

[0002] Five-axis CNC machining is a technology that achieves efficient and high-precision machining of workpieces with complex geometric shapes by simultaneously controlling the coordinated movement of five coordinate axes (usually three linear axes X, Y, and Z and two rotary axes A, B / C). As the manufacturing industry develops towards high-end and precision, five-axis CNC machining technology has been widely used in aerospace, mold manufacturing, precision parts processing and other fields due to its high efficiency and high precision. However, during high-speed and high-precision machining, factors such as thermal deformation, load deformation, geometric errors and tool wear of the machine tool structure will cause the machining trajectory to deviate from the ideal path, resulting in dynamic machining errors, which seriously affect the machining accuracy and surface quality of the parts. Therefore, how to effectively compensate for the dynamic errors in the five-axis CNC machining process is a key technical issue to improve machining accuracy and meet the needs of high-end manufacturing.

[0003] Problems with existing technologies: At present, the error compensation methods for five-axis linkage CNC machining mainly include offline error compensation and online error compensation. Offline error compensation is usually performed before machining and is corrected based on the pre-measured static error model of the machine tool. However, it is difficult to effectively compensate for dynamic errors during the machining process, such as thermal drift and deformation caused by cutting force. Online error compensation methods attempt to detect errors and compensate them in real time during the machining process. Common methods include sensor measurement-based compensation and model prediction-based compensation. For example, high-precision measuring equipment such as laser interferometers and ballbars are used for error detection, and compensation is performed through the control system. Other methods predict error changes based on the thermal characteristics or mechanical models of the machine tool for compensation. However, existing technologies still have limitations in dealing with the coupled effects of multi-source dynamic errors, nonlinear characteristics, and real-time performance, making it difficult to achieve high-precision and high-robust dynamic error compensation. Summary of the Invention

[0004] The purpose of the present invention is to provide a five-axis linkage CNC machining dynamic error compensation system based on an industrial control system, which can establish a data connection between the original adaptive optimization module by introducing a real-time residual error monitoring and feedback mechanism, thereby realizing parameter adaptive adjustment based on the actual compensation effect, and enhancing the system's adaptability and compensation accuracy under complex dynamic machining conditions.

[0005] The technical solutions adopted by the present invention are as follows: The five-axis linkage CNC machining dynamic error compensation system based on the industrial control system includes: Split-axis error detection module: A hybrid detection network consisting of a grating ruler, infrared thermal imager, cutting force sensor, and contact probe collects dynamic error data for each axis and generates an independent error model for each axis based on the improved DH method; Dynamic compensation model construction module: This module uses the S-type function and inverse Jacobian matrix decoupling algorithm to generate a single-axis compensation sub-model. It uses an LSTM neural network to predict time-varying errors and combines them with a tensor decomposition algorithm to fuse multi-physics field errors to form a global dynamic compensation vector for output to the five-axis linkage interpolator. Industrial control real-time compensation execution module: Embedded in the CNC system PLC controller, it synchronously outputs independent compensation instructions for each axis through a high-speed communication protocol. The rotation axis compensation instruction includes the axis direction error correction amount and dynamically adjusts the servo gain based on the sliding mode control algorithm. RTCP dynamic calibration unit: Integrates the contact probe online measurement system with the standardized workpiece, corrects the tool center point trajectory deviation in real time, and dynamically updates the RTCP compensation parameters; Adaptive learning optimization module: Based on the online reinforcement learning controller, with minimization of contour error as the objective function, it dynamically adjusts the compensation weight coefficients of each axis through the deep Q network, and prioritizes the optimization of the compensation strategy for high-frequency dynamic errors.

[0006] The split-axis error detection module covers the full motion range of the five-axis linkage through the combined structure of the spiral groove and the inclined plane of the standardized workpiece, separates the axial direction error of the rotation axis and the positioning error of the translation axis, and generates an error separation database.

[0007] The split-axis error detection module includes a dual-channel data fusion submodule, wherein: The first channel uses the Kalman filter algorithm to separate high-frequency vibration errors based on the data from the grating scale and cutting force sensor; The second channel uses finite element analysis to reconstruct the thermal error field and geometric error field based on the infrared thermal imager and contact probe data; The data fusion layer integrates the dual-channel data through the weighted Hausdorff distance algorithm to generate a high-resolution error separation database.

[0008] The dynamic compensation model construction module includes a multi-physics field tensor decomposition submodule, specifically including: The thermal error, force error and motion error are expressed as third-order tensor forms; The Tucker decomposition algorithm is used to extract the principal components of the error, and the first three eigenmodes are retained as the compensation vector basis; Combining the kinematic constraints of the five-axis linkage interpolator, compensation instructions are generated.

[0009] The industrial control real-time compensation execution module adopts a partitioned and hierarchical compensation architecture: Core compensation area: Directly correct the axis direction error of the rotating shaft based on the sliding mode control algorithm; Secondary compensation area: adjust the translation axis servo gain through the fuzzy PID controller; Communication redundancy area: Dual-channel EtherCAT protocol is used to transmit compensation instructions.

[0010] The RTCP dynamic calibration unit includes a feature area adaptive selection module, specifically including: Ballbar data collaborative calibration submodule: collects rotation axis motion error data in real time through the ballbar and generates error spectrum diagram; LS-SVM dynamic fitting submodule: uses the least squares support vector machine to perform nonlinear fitting of the tool center point trajectory deviation, where the kernel function type is dynamically selected based on the ballbar error spectrum; Feature area weight allocation logic: The spiral groove and inclined plane areas of the standardized workpiece are pre-set with calibration weight coefficients. The touch probe dynamically adjusts the weight allocation ratio based on the curvature radius of the current machining path and the ballbar error spectrum. The formula is: ,in, is the calibration weight of the i-th feature region, is the standard deviation of the regional error of the ballbar measurement, is the inverse of the path curvature radius; Collaborative compensation triggering mechanism: When the LS-SVM fitting residual exceeds a preset threshold, the ballbar data is automatically triggered to be fed back to the dynamic compensation model building module to update the global dynamic compensation vector.

[0011] The adaptive learning optimization module includes a multi-objective reinforcement learning controller, specifically including: The objective function is the weighted sum of minimizing contour error and maximizing machining efficiency; The reward function of the deep Q network includes a tool wear rate constraint term; A transfer learning strategy is adopted to transfer the compensation strategy of the verified workpiece to the new workpiece processing scenario.

[0012] The dynamic error compensation method for five-axis CNC machining based on the industrial control system includes the following steps: Construct a distributed sensor model to obtain the dynamic error data of each axis in the five-axis linkage, and establish an independent error model for each axis based on the improved DH method; The inverse Jacobian matrix is used to decouple static error and dynamic error, and separate the coupling effects of synchronous error and asynchronous error; For the nonlinear error of the rotating axis, an S-type function is used to fit the error curve, and an LSTM neural network is combined to predict the time-varying error caused by cutting force fluctuations and temperature drift, generate feedforward compensation instructions, and fuse multi-physics field errors through a tensor decomposition algorithm to form a global dynamic compensation vector. The industrial control system decomposes the compensation vector into independent compensation instructions for each axis and sends them synchronously to the servo drive. The rotation axis compensation instruction includes the axis direction error correction value, and dynamically adjusts the servo gain based on the sliding mode control algorithm to suppress high-frequency vibration errors. The tool center point trajectory deviation is calibrated collaboratively using a contact probe online measurement system and a standardized workpiece. The RTCP compensation parameters are dynamically updated based on the ballbar calibration data. Based on the online reinforcement learning controller, the compensation weight coefficient of each axis is dynamically adjusted. The minimization of contour error is taken as the objective function. The compensation strategy is optimized in real time through the deep Q network to adapt to sudden changes in working conditions.

[0013] According to another aspect of an embodiment of the present invention, an electronic device is provided, comprising a memory and a processor; the memory is used to store a program; and the processor executes the program to implement any one of the aforementioned methods.

[0014] According to another aspect of the embodiments of the present invention, a computer-readable storage medium is provided, wherein the storage medium stores a computer program, and when the computer program is executed by a processor, any one of the aforementioned methods is implemented.

[0015] According to yet another aspect of an embodiment of the present invention, a computer program product is provided, including a computer program, wherein when the computer program is executed by a processor, any one of the aforementioned methods is implemented.

[0016] The technical effects achieved by the present invention are: The present invention realizes adaptive parameter adjustment based on actual compensation effect by introducing a real-time residual error monitoring and feedback mechanism and establishing a data connection with the original adaptive optimization module. In addition, by refining the multi-source data acquisition type and enhancing the execution method of the compensation instruction of the industrial control system, the adaptability and compensation accuracy of the system under complex dynamic processing conditions are further enhanced, and the dynamic error compensation accuracy and stability in the five-axis linkage CNC processing process are improved, especially in processing complex and rapidly changing dynamic errors. The dynamic error compensation accuracy and stability in the five-axis linkage CNC processing process are improved, especially in processing complex and rapidly changing dynamic errors. It provides timely, reliable and comprehensive input data for subsequent modeling, ensures the accuracy of error modeling, and thus enhances the effectiveness of compensation.

[0017] The present invention can effectively deal with multi-source and coupled dynamic errors during the machining process, and improve the accuracy and stability of dynamic error compensation. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] Figure 1 It is a structural block diagram of the five-axis linkage CNC machining dynamic error compensation system of the present invention; Figure 2 It is a structural diagram of the partitioned and hierarchical compensation architecture of the present invention; Figure 3 It is a flow chart of the dynamic error compensation method of five-axis linkage CNC machining in the present invention. DETAILED DESCRIPTION

[0019] In order to make the purpose and advantages of the present invention more clearly understood, the present invention is described in detail below with reference to the following examples. It should be understood that the following text is only used to describe one or more specific embodiments of the present invention and does not strictly limit the scope of protection of the present invention.

[0020] It should be noted that the terms "first", "second", etc. in the description and claims of the present invention and the above-mentioned drawings are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that the numbers used in this way can be interchanged where appropriate, so that the embodiments of the present invention described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions. For example, a process, method, system, product or device that includes a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0021] According to an embodiment of the present invention, a method embodiment of a five-axis linkage CNC machining dynamic error compensation method based on an industrial control system is provided. It should be noted that the steps shown in the flowchart of the accompanying drawings can be executed in a computer system such as a set of computer executable instructions, and although a logical order is shown in the flowchart, in some cases, the steps shown or described can be executed in an order different from that shown here.

[0022] like Figure 1 As shown in FIG, the five-axis linkage CNC machining dynamic error compensation system based on the industrial control system includes: Split-axis error detection module: A hybrid detection network consisting of a grating ruler, infrared thermal imager, cutting force sensor, and contact probe collects dynamic error data for each axis and generates an independent error model for each axis based on the improved DH method; Dynamic compensation model construction module: This module uses the S-type function and inverse Jacobian matrix decoupling algorithm to generate a single-axis compensation sub-model. It uses an LSTM neural network to predict time-varying errors and combines them with a tensor decomposition algorithm to fuse multi-physics field errors to form a global dynamic compensation vector for output to the five-axis linkage interpolator. Industrial control real-time compensation execution module: Embedded in the CNC system PLC controller, it synchronously outputs independent compensation instructions for each axis through a high-speed communication protocol. The rotation axis compensation instruction includes the axis direction error correction amount and dynamically adjusts the servo gain based on the sliding mode control algorithm. RTCP dynamic calibration unit: Integrates the contact probe online measurement system with the standardized workpiece, corrects the tool center point trajectory deviation in real time, and dynamically updates the RTCP compensation parameters; Adaptive learning optimization module: Based on the online reinforcement learning controller, with minimization of contour error as the objective function, it dynamically adjusts the compensation weight coefficients of each axis through the deep Q network, and prioritizes the optimization of the compensation strategy for high-frequency dynamic errors.

[0023] According to the above, the improved DH method introduces temperature drift coefficient and dynamic weight adjustment; Furthermore, a temperature-related correction term is added to the DH parameters, such as , where α is the thermal expansion coefficient and ΔT is the temperature difference, which is used to dynamically correct the axis offset caused by thermal error; During the dynamic weight adjustment process, the coupling weight of the DH parameters is adjusted in real time according to the cutting force sensor data. For example, when the servo gain changes, the compensation coefficient of the connecting rod length parameter is updated.

[0024] According to the above, the real-time compensation problem of thermal drift and force deformation is solved through the dynamic coupling correction of temperature and cutting force to improve the robustness of the model.

[0025] Furthermore, in combination with the kinematic constraints of the five-axis linkage interpolator, according to the ISO230-2 compatibility rules, the first three eigenmodes after Tucker decomposition are mapped into compensation instructions that comply with the ISO230-2 standard. For example, the principal components are converted into displacement corrections for the X / Y / Z axes and angle compensation values for the rotation axes, so as to achieve direct mapping between tensor decomposition and the standard, ensure the universality and verifiability of the compensation instructions, and avoid compatibility issues between the algorithm and hardware.

[0026] It should be further explained that the split-axis error detection module covers the full motion range of the five-axis linkage through the combined structure of the spiral groove and inclined plane of the standardized workpiece, separates the axial direction error of the rotation axis and the positioning error of the translation axis, and generates an error separation database.

[0027] It should be further explained that the split-axis error detection module includes a dual-channel data fusion submodule, in which: The first channel uses the Kalman filter algorithm to separate high-frequency vibration errors based on the data from the grating scale and cutting force sensor; The second channel uses finite element analysis to reconstruct the thermal error field and geometric error field based on the infrared thermal imager and contact probe data; The data fusion layer integrates the dual-channel data through the weighted Hausdorff distance algorithm to generate a high-resolution error separation database.

[0028] It should be further explained that the dynamic compensation model construction module contains a multi-physics field tensor decomposition sub-module, which specifically includes: The thermal error, force error and motion error are expressed as third-order tensor forms; The Tucker decomposition algorithm is used to extract the principal components of the error, and the first three eigenmodes are retained as the compensation vector basis; Combining the kinematic constraints of the five-axis linkage interpolator, compensation instructions are generated.

[0029] It should be further explained that the industrial control real-time compensation execution module adopts a partitioned and hierarchical compensation architecture. Please refer to Figure 2 : Core compensation area: Directly correct the axis direction error of the rotating shaft based on the sliding mode control algorithm; Secondary compensation area: adjust the translation axis servo gain through the fuzzy PID controller; Communication redundancy area: Dual-channel EtherCAT protocol is used to transmit compensation instructions.

[0030] Furthermore, the priority switching condition between sliding mode control and fuzzy PID is that in the partitioned and hierarchical compensation architecture, when the axial direction error of the rotating shaft is ≥Xμm (X is a positive number less than or equal to 10), the core compensation zone priority is triggered, and the sliding mode control algorithm intervenes with a correction period of 1ms. Otherwise, the fuzzy PID of the secondary compensation zone adjusts the servo gain of the translational axis with a period of 5ms.

[0031] By determining the high-frequency error threshold mentioned above, the threshold determination realizes multi-time-scale collaborative compensation, improves the dynamic response speed, and shortens the high-frequency vibration error suppression time by at least 40%.

[0032] The experimental results are used to verify the improvement effect of the partitioned hierarchical compensation architecture on the high-frequency vibration error suppression time: Experimental comparison items Error type Suppression time / ms This improvement plan Improved suppression time / ms Traditional PID control High frequency vibration error (≥10μm) 25 Partitioned and hierarchical compensation architecture (sliding mode + fuzzy PID) 14 Fuzzy PID control trajectory deviation 18 Zoning and grading compensation structure 10 Single sliding mode control Thermal drift error 16 Zoning and grading compensation structure 9 Multi-timescale collaborative compensation (simulation) Servo hysteresis error 20 Zoning and grading compensation structure 11 Actual processing scenario (spiral groove processing) Tool center point trajectory deviation 30 Feature area adaptive calibration + partition and graded compensation 17 As shown in the table, the partitioned and hierarchical compensation architecture has achieved optimization in high-frequency vibration error suppression, multi-time scale collaborative response, and adaptability to actual processing scenarios.

[0033] It should be further explained that the RTCP dynamic calibration unit includes a feature area adaptive selection module, which specifically includes: Ballbar data collaborative calibration submodule: collects rotation axis motion error data in real time through the ballbar and generates error spectrum diagram; LS-SVM dynamic fitting submodule: uses the least squares support vector machine to perform nonlinear fitting of the tool center point trajectory deviation, where the kernel function type is dynamically selected based on the ballbar error spectrum; Feature area weight allocation logic: The spiral groove and inclined plane areas of the standardized workpiece are pre-set with calibration weight coefficients. The touch probe dynamically adjusts the weight allocation ratio based on the curvature radius of the current machining path and the ballbar error spectrum. The formula is: ,in, is the calibration weight of the i-th feature region, is the standard deviation of the regional error of the ballbar measurement, is the inverse of the path curvature radius; Collaborative compensation triggering mechanism: When the LS-SVM fitting residual exceeds a preset threshold, the ballbar data is automatically triggered to be fed back to the dynamic compensation model building module to update the global dynamic compensation vector.

[0034] Furthermore, if the proportion of high-frequency components in the ballbar error spectrum exceeds a threshold, where the threshold is the proportion of high-frequency components and the minimum is 60%, the RBF kernel is switched to enhance the nonlinear fitting capability. Otherwise, the polynomial kernel is used to reduce the computational complexity.

[0035] Through dynamic adjustment of the kernel function and feedback of ballbar data, a closed-loop calibration process is formed to overcome the limitations of the fixed kernel function of LS-SVM.

[0036] Furthermore, in five-axis linkage machining, the impact mechanisms of high-frequency errors (such as servo vibration and sudden changes in cutting force) and low-frequency errors (such as thermal drift and geometric errors) on machining accuracy are different. When the high-frequency component accounts for more than 60%, the error fluctuates violently and the nonlinear characteristics are obviously prominent. A kernel function with strong nonlinear fitting capabilities (such as the RBF kernel) is required. Low-frequency-dominated scenarios (such as those dominated by thermal errors) are suitable for polynomial kernels with higher computational efficiency.

[0037] Among them, the 60% threshold setting is the result of systematic optimization based on the error spectrum characteristic analysis and kernel function performance matching. It solves the problem that the fixed kernel function cannot take into account both high-frequency error fitting accuracy and computational efficiency, and achieves a breakthrough in the dynamic calibration algorithm.

[0038] Furthermore, the RBF kernel has higher fitting accuracy when processing nonlinear and highly volatile data, but the computational complexity is higher; the polynomial kernel is suitable for smoothing the error field, but is sensitive to high-frequency noise. By using a 60% threshold division, an optimal balance can be achieved between fitting accuracy and computational efficiency.

[0039] Based on the above, when the high-frequency component is ≥60%, the local approximation characteristics of the RBF kernel can more accurately capture the sudden deviation of the tool center point (TCP) trajectory (such as micron-level jumps caused by cutting force impact), thereby reducing the LS-SVM fitting residual. In low-frequency-dominated scenarios (high frequency <60%), the use of a polynomial kernel can reduce the amount of calculation (such as the number of floating-point operations) by about 40%, avoiding the degradation of real-time performance due to over-reliance on the RBF kernel, and meeting the millisecond-level compensation requirements of the industrial control system. By triggering the kernel function switching by the threshold, the system can dynamically adapt to changes in machining conditions (such as tool wear and material differences), avoiding the "overfitting" or "underfitting" problems caused by a single kernel function, and improving the compensation robustness.

[0040] It should be further explained that the adaptive learning optimization module includes a multi-objective reinforcement learning controller, which specifically includes: The objective function is the weighted sum of minimizing contour error and maximizing machining efficiency; The reward function of the deep Q network includes a tool wear rate constraint term; A transfer learning strategy is adopted to transfer the compensation strategy of the verified workpiece to the new workpiece processing scenario.

[0041] Among them, in the multi-objective reinforcement learning controller, the cross-workpiece compensation parameter migration rules are set, and the compensation weight coefficients of the verified workpieces are used as prior knowledge. The Bayesian optimization algorithm is used to quickly adapt to the new workpiece processing scenarios. For example, when the tool material changes, the reward function weight of the deep Q network is automatically adjusted. Through transfer learning, the debugging time of new workpiece processing is shortened, thereby improving production efficiency.

[0042] It should be noted that the above modules can be implemented through software or hardware. For the latter, it can be implemented in the following ways, but not limited to: the above modules are all located in the same processor; or the above modules are located in different processors in any combination.

[0043] Please refer to Figure 3 ,The dynamic error compensation method for five-axis CNC machining based on the industrial control system includes the following steps: S1. Build a distributed sensor model to obtain the dynamic error data of each axis in the five-axis linkage, and establish an independent error model for each axis based on the improved DH method; S2. Use the inverse Jacobian matrix to decouple static error and dynamic error, and separate the coupling effects of synchronous error and asynchronous error; S3: For nonlinear errors in the rotation axis (such as pitch compensation error), an S-shaped function is used to fit the error curve. Combined with the LSTM neural network, it predicts the time-varying errors caused by cutting force fluctuations and temperature drift, generates feedforward compensation instructions, and fuses multi-physics field errors through a tensor decomposition algorithm to form a global dynamic compensation vector. S4. The industrial control system decomposes the compensation vector into independent compensation instructions for each axis and sends them synchronously to the servo drive. The rotation axis compensation instruction includes the axis direction error correction amount and dynamically adjusts the servo gain based on the sliding mode control algorithm to suppress high-frequency vibration errors. S5. Use the contact probe online measurement system and the standardized workpiece (combination of spiral grooves and inclined planes) to calibrate the tool center point trajectory deviation, and dynamically update the RTCP compensation parameters based on the ballbar calibration data. S6. Based on the online reinforcement learning controller, the compensation weight coefficient of each axis is dynamically adjusted, and the minimization of the contour error is used as the objective function. The compensation strategy is optimized in real time through the deep Q network to adapt to sudden changes in working conditions.

[0044] In step S1, the distributed sensor model includes but is not limited to a grating scale, an infrared thermal imager, a cutting force sensor and a contact probe, which respectively collects the dynamic error data of each axis, such as thermal error, geometric error, and servo error. The temperature drift coefficient is introduced by improving the DH method to dynamically correct the axis offset caused by the thermal error, and the compensation coefficient of the connecting rod length parameter is adjusted through the cutting force sensor data.

[0045] In step S2, the inverse Jacobian matrix is used to separate the synchronous error (such as servo gain mismatch) and the asynchronous error (such as the reverse clearance of the rotating axis) to solve the compensation lag problem caused by multi-source error coupling, and further combine the real-time sensor data to achieve dynamic decoupling.

[0046] In step S3, the tensor decomposition algorithm expresses the thermal error, force error and motion error as a third-order tensor, extracts the principal components through Tucker decomposition and generates a global dynamic compensation vector.

[0047] In step S4, the core compensation area corrects the axial direction error of the rotating axis based on the sliding mode control algorithm, the secondary compensation area adjusts the servo gain of the translational axis through fuzzy PID, and the communication redundancy area adopts the dual-channel EtherCAT protocol to ensure the reliability of command transmission. Multi-time scale collaborative compensation is achieved through partitioned and hierarchical design to improve the dynamic response speed.

[0048] In step S6, the objective function is the weighted sum of minimizing the contour error and maximizing the machining efficiency. The reward function of the deep Q network includes a tool wear rate constraint term, and the compensation strategy of the verified workpiece is migrated to the new workpiece machining scenario through the transfer learning strategy, and the debugging time of the new workpiece machining is shortened through transfer learning.

[0049] According to another aspect of an embodiment of the present invention, an electronic device is provided. The electronic device includes a memory and a processor; the memory is used to store programs; and the processor executes the programs to implement any one of the aforementioned methods.

[0050] According to another aspect of an embodiment of the present invention, a computer-readable storage medium is provided. The storage medium stores a computer program. When the computer program is executed by a processor, any one of the aforementioned methods is implemented.

[0051] According to yet another aspect of an embodiment of the present invention, a computer program product is provided, including a computer program, which implements any of the aforementioned methods when executed by a processor.

[0052] The foregoing is merely a preferred embodiment of the present invention. It should be noted that those skilled in the art may make various improvements and modifications without departing from the principles of the present invention, and such improvements and modifications are also within the scope of protection of the present invention. Structures, devices, and operating methods not specifically described or explained herein shall, unless otherwise specified or limited, be implemented in accordance with conventional means in the art.

Claims

1. The five-axis linkage CNC machining dynamic error compensation system based on the industrial control system is characterized by: include: Split-axis error detection module: A hybrid detection network consisting of a grating ruler, infrared thermal imager, cutting force sensor, and contact probe collects dynamic error data for each axis and generates an independent error model for each axis based on the improved DH method; Dynamic compensation model construction module: This module uses the S-type function and inverse Jacobian matrix decoupling algorithm to generate a single-axis compensation sub-model. It uses an LSTM neural network to predict time-varying errors and combines them with a tensor decomposition algorithm to fuse multi-physics field errors to form a global dynamic compensation vector for output to the five-axis linkage interpolator. Industrial control real-time compensation execution module: Embedded in the CNC system PLC controller, it synchronously outputs independent compensation instructions for each axis through a high-speed communication protocol. The rotation axis compensation instruction includes the axis direction error correction amount and dynamically adjusts the servo gain based on the sliding mode control algorithm. RTCP dynamic calibration unit: Integrates the contact probe online measurement system with the standardized workpiece, corrects the tool center point trajectory deviation in real time, and dynamically updates the RTCP compensation parameters; Adaptive learning optimization module: Based on the online reinforcement learning controller, with minimization of contour error as the objective function, it dynamically adjusts the compensation weight coefficients of each axis through the deep Q network, and prioritizes the optimization of the compensation strategy for high-frequency dynamic errors.

2. The five-axis linkage CNC machining dynamic error compensation system based on the industrial control system according to claim 1 is characterized in that: The split-axis error detection module covers the full motion range of the five-axis linkage through the combined structure of the spiral groove and the inclined plane of the standardized workpiece, separates the axial direction error of the rotation axis and the positioning error of the translation axis, and generates an error separation database.

3. The five-axis linkage CNC machining dynamic error compensation system based on the industrial control system according to claim 1 is characterized in that: The split-axis error detection module includes a dual-channel data fusion submodule, wherein: The first channel uses the Kalman filter algorithm to separate high-frequency vibration errors based on the data from the grating scale and cutting force sensor; The second channel uses finite element analysis to reconstruct the thermal error field and geometric error field based on the infrared thermal imager and contact probe data; The data fusion layer integrates the dual-channel data through the weighted Hausdorff distance algorithm to generate a high-resolution error separation database.

4. The five-axis linkage CNC machining dynamic error compensation system based on the industrial control system according to claim 1 is characterized in that: The dynamic compensation model construction module includes a multi-physics field tensor decomposition submodule, specifically including: The thermal error, force error and motion error are expressed as third-order tensor forms; The Tucker decomposition algorithm is used to extract the principal components of the error, and the first three eigenmodes are retained as the compensation vector basis; Combining the kinematic constraints of the five-axis linkage interpolator, compensation instructions are generated.

5. The five-axis linkage CNC machining dynamic error compensation system based on the industrial control system according to claim 1 is characterized in that: The industrial control real-time compensation execution module adopts a partitioned and hierarchical compensation architecture: Core compensation area: Directly correct the axis direction error of the rotating shaft based on the sliding mode control algorithm; Secondary compensation area: adjust the translation axis servo gain through the fuzzy PID controller; Communication redundancy area: Dual-channel EtherCAT protocol is used to transmit compensation instructions.

6. The five-axis linkage CNC machining dynamic error compensation system based on the industrial control system according to claim 1 is characterized in that: The RTCP dynamic calibration unit includes a feature area adaptive selection module, specifically including: Ballbar data collaborative calibration submodule: collects rotation axis motion error data in real time through the ballbar and generates error spectrum diagram; LS-SVM dynamic fitting submodule: uses the least squares support vector machine to perform nonlinear fitting of the tool center point trajectory deviation, where the kernel function type is dynamically selected based on the ballbar error spectrum; Feature area weight allocation logic: The spiral groove and inclined plane areas of the standardized workpiece are pre-set with calibration weight coefficients. The touch probe dynamically adjusts the weight allocation ratio based on the curvature radius of the current machining path and the ballbar error spectrum. The formula is: ,in, is the calibration weight of the i-th feature region, is the standard deviation of the regional error of the ballbar measurement, is the inverse of the path curvature radius; Collaborative compensation triggering mechanism: When the LS-SVM fitting residual exceeds a preset threshold, the ballbar data is automatically triggered to be fed back to the dynamic compensation model building module to update the global dynamic compensation vector.

7. The five-axis CNC machining dynamic error compensation system based on the industrial control system according to claim 1 is characterized in that: The adaptive learning optimization module includes a multi-objective reinforcement learning controller, specifically including: The objective function is the weighted sum of minimizing contour error and maximizing machining efficiency; The reward function of the deep Q network includes a tool wear rate constraint term; A transfer learning strategy is adopted to transfer the compensation strategy of the verified workpiece to the new workpiece processing scenario.

8. A five-axis CNC machining dynamic error compensation method based on an industrial control system, using any system as claimed in claims 1 to 7, characterized in that: The steps include: Construct a distributed sensor model to obtain the dynamic error data of each axis in the five-axis linkage, and establish an independent error model for each axis based on the improved DH method; The inverse Jacobian matrix is used to decouple static error and dynamic error, and separate the coupling effects of synchronous error and asynchronous error; For the nonlinear error of the rotating axis, an S-type function is used to fit the error curve, and an LSTM neural network is combined to predict the time-varying error caused by cutting force fluctuations and temperature drift, generate feedforward compensation instructions, and fuse multi-physics field errors through a tensor decomposition algorithm to form a global dynamic compensation vector. The industrial control system decomposes the compensation vector into independent compensation instructions for each axis and sends them synchronously to the servo drive. The rotation axis compensation instruction includes the axis direction error correction value, and dynamically adjusts the servo gain based on the sliding mode control algorithm to suppress high-frequency vibration errors. The tool center point trajectory deviation is calibrated collaboratively using a contact probe online measurement system and a standardized workpiece. The RTCP compensation parameters are dynamically updated based on the ballbar calibration data. Based on the online reinforcement learning controller, the compensation weight coefficient of each axis is dynamically adjusted. The minimization of contour error is taken as the objective function. The compensation strategy is optimized in real time through the deep Q network to adapt to sudden changes in working conditions.

9. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, the method according to claim 8 is implemented.

10. A computer program product comprising a computer program, characterized in that When the computer program is executed by a processor, the method according to claim 8 is implemented.

Images