A numerical control rotary table precision detection method and system

Abstract

The present application relates to the technical field of precision detection of numerical control machine tools, and discloses a numerical control rotary table precision detection method, comprising the following steps: S1, temperature sensors are arranged at heat response sensitive points of an axle seat, a transmission pair shell and a driving motor of a numerical control rotary table, and a displacement sensing link is established between an end of a main shaft and a rotary table reference element; S2, a control device controls the numerical control rotary table to execute a symmetric measurement track of forward rotation, reverse rotation and forward rotation again under a load working condition, so as to generate a distinguishable observation sequence with geometric symmetry constraint. By constructing a multi-source sensing link and executing a symmetric measurement track under a load working condition, in-situ capture of dynamic temperature rise and load fluctuation is realized, high unification of a detection environment and a working environment is achieved, real restoration of dynamic performance of a machine tool is realized, and the deficiencies of inconsistency between a traditional quasi-static detection result and a machining state and low re-inspection value are solved.

Description

A method and system for detecting the accuracy of a CNC rotary table Technical Field

[0001] This invention relates to the field of CNC machine tool precision testing technology, specifically to a CNC rotary table precision testing method and system. Background Technology

[0002] In high-precision machining scenarios of five-axis machining centers, the CNC rotary table, as a core functional component, directly determines the machining quality of complex parts through its spatial motion accuracy. With the ever-increasing demands for machining accuracy in fields such as aerospace and precision mold making, how to obtain the dynamic spatial error of the CNC rotary table under actual machining conditions and accurately compensate for it based on error patterns has become a highly challenging and novel technical problem in this field.

[0003] Currently, the industry typically uses methods such as dual ballbar testing, R-test wireless ballbar detection, or laser interferometers to perform static or quasi-static testing of turntable accuracy. These methods establish a specific geometric connection between the turntable and the spindle, and use sensors to capture the relative displacement fluctuations of the turntable during rotation in order to calculate the geometric error parameters of the turntable.

[0004] In existing technologies, most inspection tasks are performed after the machine tool has stopped, is under light load, operates at low speed, or has reached thermal steady state, resulting in quasi-static geometric errors. However, due to temperature rise, load, and speed variations during actual machining, these factors can cause instantaneous drift and dynamic runout of the axis, leading to inconsistencies between the inspection results and the actual machining conditions. This disconnect significantly reduces the re-inspection value of the inspection data, making it difficult to accurately reflect the dynamic performance of the machine tool during operation. Therefore, this invention proposes a CNC rotary table accuracy inspection method and system to solve the aforementioned problems. Summary of the Invention

[0005] To address the shortcomings of existing technologies, this invention provides a method and system for detecting the accuracy of CNC rotary tables, thereby solving the problems mentioned in the background section.

[0006] To achieve the above objectives, the present invention provides the following technical solution: a method for detecting the accuracy of a CNC rotary table, comprising:

[0007] S1. Temperature sensors are deployed on the spindle seat, transmission pair housing and drive motor thermal response sensitive points of the CNC rotary table, and a displacement sensing link is established between the spindle end and the rotary table reference component.

[0008] S2 controls the deployment of a CNC rotary table to perform symmetrical measurement trajectories of forward rotation, reverse rotation, and re-forward rotation under load conditions, generating a resolvable observation sequence with geometric symmetry constraints;

[0009] S3 uses the sampling trigger module to acquire temperature data, displacement data, and internal feedback angle data of the CNC turntable during the operation of the symmetrical measurement trajectory, and performs spatiotemporal synchronization alignment processing.

[0010] S4 resamples and maps the displacement data after spatiotemporal synchronization and alignment with the internal feedback angle data, and extracts the synchronization error component that is periodically repeated by the rotation angle.

[0011] S5. Construct a physical constraint error decoupling model that includes synchronization error components. Use the displacement difference between forward and reverse rotation to eliminate sensor system bias and isolate geometric inherent error, dynamic temperature drift component and load sensitive component.

[0012] S6. Substitute the dynamic temperature drift component, the load-sensitive component, and the real-time collected current rotation speed, current load, and current temperature rise gradient into the uncertainty calculation formula to obtain the dynamic spatial error uncertainty window.

[0013] S7 compares the dynamic spatial error uncertainty window with the preset tolerance threshold and generates an inspection report containing compensation values ​​and processing window suggestions based on the comparison results.

[0014] Preferably, the deployment of thermal response sensitive points in S1 refers to selecting the support bearing of the bearing seat, the meshing zone boundary of the transmission pair housing, and the winding housing of the drive motor as temperature acquisition reference points based on the heat source distribution characteristics inside the CNC rotary table, in order to capture thermal gradient source data that causes instantaneous axis drift.

[0015] The establishment of the displacement sensing link in S1 refers to arranging at least three sets of non-contact displacement sensors in an orthogonal configuration with a 90-degree angle between them on the outer circumferential surface and the top surface of the turntable reference component, and constructing a five-degree-of-freedom spatial original observation matrix containing tilt error terms and translation error terms by collecting radial runout data and axial movement data.

[0016] Preferably, the geometric symmetry constraint in S2 refers to utilizing the physical property that the geometric errors of the forward rotation observation sequence and the reverse rotation observation sequence at the same angular position are mirror images of each other. By performing superposition averaging and difference calculations on the distinguishable observation sequences, the physical separation of the synchronization drift term and the noise term during the rotation process is achieved.

[0017] The symmetrical measurement trajectory in S2 refers to controlling the CNC turntable to perform rotary motion with a preset stepped speed sequence under load conditions, and triggering the displacement sensing link to perform multiple reciprocating micro-motion observations at each angle measurement point, in order to capture the nonlinear dynamic jumping characteristics caused by uneven meshing stiffness of the transmission pair.

[0018] Preferably, the spatiotemporal synchronization alignment process in S3 refers to synchronizing the displacement data and the internal feedback angle data to the same clock reference at the microsecond level through a hard trigger signal, so as to ensure that the measurement point and the motion coordinate correspond one-to-one.

[0019] The spatiotemporal synchronization alignment process includes using the sampling trigger module to perform resampling processing on the temperature data based on sampling frequency compensation, and using an interpolation algorithm to eliminate the thermal inertia hysteresis of the temperature sensor, so that the temperature data of each thermal response sensitive point and the displacement data at the same physical moment are phase aligned on the time axis.

[0020] Preferably, the extraction of synchronization error components in S4 refers to identifying harmonic components in the rotary signal through spectrum analysis, retaining signals whose frequency is an integer multiple of the rotary frequency of the CNC turntable, and filtering out asynchronous fluctuation components caused by environmental vibration.

[0021] The extraction of synchronization error components in S4 refers to using low-pass filtering and angle domain averaging techniques to remove non-periodic high-frequency components, projecting the resampled and mapped data onto the polar coordinate system, and extracting the low-order harmonic components reflecting the eccentricity of the CNC rotary table installation and the eccentricity of the worm gear machining through Fourier series decomposition.

[0022] Preferably, the physical constraint error decoupling model in S5 is as follows:

[0023] ,

[0024] in, To account for spatial errors, This is a geometrically inherent error base map. This refers to the dynamic temperature drift coefficient. For the temperature rise gradient at the sensitive point, For load sensitivity coefficient, This represents the equivalent load change. Let t be the rotation angle and t be the running time;

[0025] The geometric inherent error refers to the static geometric deviation curve that does not change over time, extracted by the forward and reverse symmetrical measurement trajectory when the CNC rotary table is in a preset thermal steady state and under no-load conditions.

[0026] The dynamic temperature drift component refers to the real-time monitoring of the temperature rise correlation characteristics between the bearing and the transmission pair housing, calculating the instantaneous drift of the dynamic spatial error due to the temperature field evolution, and establishing a temperature hysteresis correction term for error compensation.

[0027] The load-sensitive component refers to the identification of the elastic deformation characteristics of the CNC rotary table structure caused by gravity load by comparing the displacement offset caused by workpieces of different masses at the same angular position.

[0028] Preferably, the error decoupling model under physical constraints in S5 includes a spatial coordinate transformation matrix, which maps the geometric inherent error, the dynamic temperature drift component and the load-sensitive component to the machine tool coordinate system where the machining tip is located, thereby realizing the correlation transformation from the rotation accuracy of the CNC rotary table itself to the actual forming accuracy of the workpiece.

[0029] Preferably, the formula for calculating the dynamic spatial error uncertainty window in S6 is:

[0030] ,

[0031] Where U is the comprehensive uncertainty of dynamic space error, u r To measure the repeatability standard deviation, u c For the forward and reverse consistent residuals, u s For the traceability error of the reference component;

[0032] The forward and reverse consistency residual u c It refers to the maximum deviation of the spatial trajectory between the forward rotation measurement curve and the reverse rotation measurement curve after the offset is eliminated by the sensor system under the same ambient temperature and load conditions. It is used to quantitatively characterize the dynamic drift limit caused by the inconsistency of the thermal response of the drive motor and the transmission pair.

[0033] Preferably, the output machining window in S7 is suggested to be the effective continuous machining time for maintaining the CNC rotary table error within the allowable tolerance range based on the evolution trend of the dynamic temperature drift component.

[0034] The output machining window suggestion includes using the load-sensitive component to perform a mechanical coupling evaluation of the current workpiece quality, and combining the expansion rate of the dynamic spatial error uncertainty window to generate a process parameter suggestion table for the CNC system, which includes the upper limit of the optimal cutting feed rate and the expected shutdown cooling time.

[0035] A CNC rotary table accuracy testing system, the CNC rotary table accuracy testing system comprising:

[0036] The multi-source sensing module is used to acquire temperature data of thermal response sensitive points, displacement data of the spindle end, and angle data inside the turntable;

[0037] The synchronization triggering module is used to provide a microsecond-level clock reference for the multi-source sensing module to achieve spatiotemporal synchronization alignment between data items.

[0038] The trajectory control module is used to drive the CNC rotary table to perform a symmetrical measurement trajectory of forward rotation, reverse rotation and re-forward rotation under load conditions;

[0039] The error decoupling engine module is used to extract geometrically inherent errors, dynamic temperature drift components, and load-sensitive components based on the physical constraint error decoupling model.

[0040] The intelligent decision-making module is used to calculate the dynamic spatial error uncertainty window and generate a test report that includes compensation values ​​and processing window suggestions.

[0041] This invention provides a method and system for detecting the accuracy of a CNC rotary table. It has the following beneficial effects:

[0042] 1. This invention achieves in-situ capture of dynamic temperature rise and load fluctuation by constructing a multi-source sensing link and executing symmetrical measurement trajectories under load conditions, thereby achieving a high degree of uniformity between the detection environment and the working environment, realizing the true restoration of the dynamic performance of the machine tool, and solving the shortcomings of traditional quasi-static detection results being inconsistent with the processing state and having low re-inspection value.

[0043] 2. This invention employs a physical constraint error decoupling model and spatiotemporal synchronization alignment technology. Through mirror analysis of multidimensional observation sequences, it achieves precise separation of geometric errors, dynamic temperature drift, and load deformation, thereby achieving quantitative decomposition of comprehensive errors, accurate tracing of error causes, and solving the shortcomings of instantaneous axis drift caused by multiple dynamic factors that are difficult to diagnose in a targeted manner.

[0044] 3. This invention employs uncertainty window calculation and processing window prediction technology, combined with real-time operating parameters to dynamically generate compensation values ​​and process suggestions, achieving closed-loop guidance of production decisions from detection data, realizing a leap from single-precision evaluation to intelligent process planning, and solving the shortcomings of existing technologies that cannot predict the reliability of equipment continuous operation under complex conditions. Attached Figure Description

[0045] Figure 1 is a flowchart of the present invention;

[0046] Figure 2 is a system framework diagram of the present invention. Detailed Implementation

[0047] To enable those skilled in the art to understand the present invention, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are some, but not all, of the embodiments of the present invention. Other embodiments obtained by those skilled in the art based on the embodiments of the present invention without creative effort should fall within the scope of protection of the present invention.

[0048] The present invention will now be described in detail with reference to the accompanying drawings:

[0049] Please refer to Figures 1 and 2. An embodiment of the present invention provides a method for detecting the accuracy of a CNC rotary table, comprising:

[0050] S1. Temperature sensors are deployed on the spindle seat, transmission pair housing and drive motor thermal response sensitive points of the CNC rotary table, and a displacement sensing link is established between the spindle end and the rotary table reference component.

[0051] The deployment of thermal response sensitive points in S1 refers to selecting the support bearing of the shaft seat, the meshing zone boundary of the transmission pair housing, and the winding housing of the drive motor as temperature acquisition reference points based on the heat source distribution characteristics inside the CNC rotary table, in order to capture thermal gradient source data that causes instantaneous axis drift.

[0052] The establishment of the displacement sensing link in S1 refers to the arrangement of at least three sets of non-contact displacement sensors in an orthogonal form with a 90-degree angle between each other on the outer circumferential surface and the top surface of the turntable reference component, and the construction of a five-degree-of-freedom spatial original observation matrix containing tilt error terms and translation error terms by collecting radial runout data and axial movement data.

[0053] S2 controls the deployment of a CNC rotary table to perform symmetrical measurement trajectories of forward rotation, reverse rotation, and re-forward rotation under load conditions, generating a resolvable observation sequence with geometric symmetry constraints;

[0054] The geometric symmetry constraint in S2 refers to the physical property that the geometric errors of the forward rotation observation sequence and the reverse rotation observation sequence are mirror images of each other at the same angular position. By performing superposition averaging and difference calculations on the distinguishable observation sequences, the physical separation of the synchronous drift term and noise term during the rotation process can be achieved.

[0055] The symmetrical measurement trajectory in S2 refers to the control of the CNC turntable to perform rotary motion with a preset stepped speed sequence under load conditions, and triggering the displacement sensing link at each angle measurement point to perform multiple reciprocating micro-motion observations, in order to capture the nonlinear dynamic jumping characteristics caused by uneven meshing stiffness of the transmission pair.

[0056] S3 uses the sampling trigger module to acquire temperature data, displacement data, and internal feedback angle data of the CNC turntable during the operation of the symmetrical measurement trajectory, and performs spatiotemporal synchronization alignment processing.

[0057] The spatiotemporal synchronization alignment process in S3 refers to synchronizing displacement data and internal feedback angle data to the same clock reference at the microsecond level through a hard trigger signal, ensuring that the measurement point and the motion coordinate correspond one-to-one.

[0058] The spatiotemporal synchronization alignment process includes using the sampling trigger module to perform resampling processing on the temperature data based on sampling frequency compensation, and using the interpolation algorithm to eliminate the thermal inertia hysteresis of the temperature sensor, so that the temperature data of each thermal response sensitive point and the displacement data at the same physical moment are phase aligned on the time axis.

[0059] S4 resamples and maps the displacement data after spatiotemporal synchronization and alignment with the internal feedback angle data, and extracts the synchronization error component that is periodically repeated by the rotation angle.

[0060] Extracting synchronization error components in S4 refers to identifying harmonic components in the rotary signal through spectrum analysis, retaining signals whose frequencies are integer multiples of the CNC rotary table's rotation frequency, and filtering out asynchronous fluctuation components caused by environmental vibrations.

[0061] Extracting synchronization error components in S4 refers to using low-pass filtering and angle domain averaging techniques to remove non-periodic high-frequency components, projecting the resampled and mapped data onto the polar coordinate system, and extracting low-order harmonic components reflecting the eccentricity of CNC rotary table installation and worm gear machining through Fourier series decomposition.

[0062] S5. Construct a physical constraint error decoupling model that includes synchronization error components. Use the displacement difference between forward and reverse rotation to eliminate sensor system bias and isolate geometric inherent error, dynamic temperature drift component and load sensitive component.

[0063] The physical constraint error decoupling model in S5 is:

[0064] ,

[0065] in, To account for spatial errors, This is a geometrically inherent error base map. This refers to the dynamic temperature drift coefficient. For the temperature rise gradient at the sensitive point, For load sensitivity coefficient, This represents the equivalent load change. Let t be the rotation angle and t be the running time;

[0066] Geometric inherent error refers to the static geometric deviation curve that does not change over time, extracted by measuring the forward and reverse symmetrical trajectories of the CNC rotary table under a preset thermal steady state and no load conditions.

[0067] The dynamic temperature drift component refers to the real-time monitoring of the temperature rise correlation characteristics between the bearing and the transmission pair housing, calculating the instantaneous drift of the dynamic spatial error due to the evolution of the temperature field, and establishing a temperature hysteresis correction term for error compensation.

[0068] The load-sensitive component refers to the identification of the elastic deformation characteristics of the CNC rotary table structure caused by gravity load by comparing the displacement offset caused by workpieces of different masses at the same angular position.

[0069] The error decoupling model under physical constraints in S5 includes a spatial coordinate transformation matrix. By mapping the geometric inherent error, dynamic temperature drift component and load-sensitive component to the machine tool coordinate system where the machining tip is located, the correlation transformation from the rotation accuracy of the CNC rotary table itself to the actual forming accuracy of the workpiece is realized.

[0070] S5. Substitute the dynamic temperature drift component, the load-sensitive component, and the real-time collected current rotation speed, current load, and current temperature rise gradient into the uncertainty calculation formula to obtain the dynamic spatial error uncertainty window.

[0071] The formula for calculating the dynamic spatial error uncertainty window in S6 is: Where U is the comprehensive uncertainty of dynamic space error, u r To measure the repeatability standard deviation, u c For the forward and reverse consistent residuals, u s For the traceability error of the reference component;

[0072] Forward and reverse consistency residual u c It refers to the maximum deviation of the spatial trajectory between the forward rotation measurement curve and the reverse rotation measurement curve after being eliminated by the bias of the sensor system under the same ambient temperature and load conditions. It is used to quantitatively characterize the dynamic drift limit caused by the inconsistency of the thermal response of the drive motor and the transmission pair.

[0073] S7 compares the dynamic spatial error uncertainty window with the preset tolerance threshold and generates an inspection report containing compensation values ​​and processing window suggestions based on the comparison results.

[0074] The output machining window suggestion in S7 refers to the effective continuous machining time for maintaining the CNC rotary table error within the allowable tolerance range based on the evolution trend of dynamic temperature drift components.

[0075] The output machining window suggestion includes a mechanical coupling evaluation of the current workpiece quality using load-sensitive components, combined with the expansion rate of the dynamic spatial error uncertainty window, to generate a process parameter suggestion table for the CNC system that includes the upper limit of the optimal cutting feed rate and the expected shutdown cooling time.

[0076] A CNC rotary table accuracy testing system, comprising:

[0077] The multi-source sensing module is used to acquire temperature data of thermal response sensitive points, displacement data of the spindle end, and angle data inside the turntable;

[0078] The synchronization triggering module is used to provide a microsecond-level clock reference for the multi-source sensing module to achieve spatiotemporal synchronization and alignment between data items.

[0079] The trajectory control module is used to drive the CNC rotary table to perform symmetrical measurement trajectories of forward rotation, reverse rotation, and re-forward rotation under load conditions.

[0080] The error decoupling engine module is used to extract geometrically inherent errors, dynamic temperature drift components, and load-sensitive components based on the physical constraint error decoupling model.

[0081] The intelligent decision-making module is used to calculate the dynamic spatial error uncertainty window and generate a test report that includes compensation values ​​and processing window suggestions.

[0082] The multi-source sensing module constructs a comprehensive monitoring network at the turntable, which can simultaneously collect temperature information reflecting structural thermal changes, spatial displacement deviation at the end of the main shaft, and real-time angular coordinates of the rotation axis, providing the system with detailed raw physical state data.

[0083] The synchronous triggering module establishes a strict microsecond-level timing standard to ensure that all sensor data acquisition times remain completely consistent under the global clock, eliminating the risk of data phase misalignment caused by inconsistent sampling frequencies or communication link fluctuations.

[0084] The trajectory control module drives the turntable to execute specific symmetrical motion sequences, constructing a geometric observation field with self-calibration characteristics at the physical level, and using the kinematic mirror principle to achieve physical suppression of sensor system bias and environmental background noise.

[0085] The error decoupling engine module relies on a rigorous physical constraint mechanism to perform in-depth dimensional decomposition of the acquired hybrid observation matrix, successfully separating the intertwined geometric accuracy defects, thermally induced spatial drift, and load-sensitive elastic deformation.

[0086] The intelligent decision-making module enables the logical transition from massive amounts of testing data to production process instructions. By calculating the uncertainty window and comparing tolerance thresholds, it can endow the testing report with forward-looking predictive capabilities and auxiliary decision-making value.

[0087] Thermally sensitive points refer to the physical locations within the turntable's internal structure that are most significantly affected by temperature changes and most prone to thermal deformation or thermal stress due to thermal expansion and contraction. These points are typically located along heat conduction paths or in areas with low heat capacity.

[0088] The displacement sensing link is a closed-loop measurement system composed of a series of high-precision non-contact sensors and rigid reference components. Its design purpose is to capture the relative spatial displacement changes between the spindle and the turntable in real time and accurately, so as to provide a data basis for error compensation.

[0089] Symmetrical measurement trajectory is a high-precision motion control strategy that uses the physical mirror symmetry of the motion process to compensate for the inherent bias or zero-position error in the measurement system by controlling the turntable to perform forward and reverse rotations respectively.

[0090] Spatiotemporal synchronization alignment refers to the process of matching and fusing multi-source monitoring data from different sensors with different sampling frequencies and different physical properties, through a unified high-precision time reference and a common spatial coordinate system, to achieve millisecond-level time accuracy and spatial consistency.

[0091] Resampling mapping is a technique that transforms measurement data that was originally collected in a time series and was unstable due to speed fluctuations into stable data with turntable rotation angle as the independent variable through interpolation or fitting methods.

[0092] Synchronization error components refer to those error components that are strongly correlated with the rotation angle of the turntable, repeat in each rotation cycle and change in a regular manner. They mainly reflect the inherent manufacturing precision and fit clearance of the turntable mechanical parts during processing and assembly.

[0093] The physical constraint error decoupling model is a mathematical mapping system based on the kinematic principles and thermodynamic behavior laws of the turntable system. Its core function is to decompose the mixed errors coupled together in actual measurement into independent error components with clear physical causes.

[0094] The dynamic spatial error uncertainty window refers to the statistical interval that characterizes the reliability and confidence level of the measurement results after quantitatively evaluating the fluctuation range and probability distribution of spatial positioning error under the complex and ever-changing thermo-mechanical coupling dynamic working conditions of the turntable.

[0095] The recommended machining window is a safe and efficient optimal operating time range that ensures the machining accuracy of parts is always within the design tolerance range, based on modeling and predicting the evolution of turntable error with time, temperature and operating conditions.

[0096] By deploying high-precision sensors at the sensitive points of the shaft and motor thermal response and establishing a stable and reliable displacement sensing link, the S1 system can construct a comprehensive and responsive neural network for monitoring the machine tool's operating status. This system can capture in real time the minute physical deformations caused by heat generated during cutting and changes in ambient temperature, achieving dynamic perception of the machine tool's thermal evolution. This multi-dimensional, multi-point sensing layout significantly enhances the system's adaptability to complex working conditions, completely overcoming the limitations of traditional detection methods that ignore internal thermal dynamic changes in the machine tool. It provides original, accurate, and continuous data support for the real-time capture and compensation control of high-precision dynamic drift, enabling the detection system to deeply understand and respond to the complex physical environment of the machining site.

[0097] S2 employs a symmetrical measurement trajectory design for forward and reverse rotation under load conditions, which can fully utilize the kinematic mirror principle to generate observation data sequences with self-calibration capabilities. This method effectively eliminates systematic errors introduced by sensor installation deviations and the measurement link itself through the symmetrical characteristics of geometric errors in forward and reverse motion. This trajectory design highly replicates the actual stress state during processing at the physical level, ensuring that the measurement results accurately reflect the dynamic stiffness changes and motion accuracy performance of the turntable under load conditions. It overcomes the shortcomings of traditional static detection methods in effectively assessing structural deformation caused by load, significantly improving the reliability and applicability of condition monitoring.

[0098] S3 achieves microsecond-level high-precision spatiotemporal synchronization and phase compensation through hardware trigger signals, completely eliminating the asynchronous risks and thermal inertia hysteresis effects caused by timing differences in data acquisition from various types of sensors. This mechanism ensures that displacement, angle, and temperature physical quantities are acquired and logically correlated at strictly consistent physical moments, guaranteeing the rigor and consistency of the overall data topology. This highly synchronized processing method effectively avoids the risk of misjudgment caused by sampling time deviations, laying a solid and reliable data foundation for obtaining high-fidelity, high-timeliness instantaneous dynamic response curves under extreme conditions such as high-speed rotation.

[0099] S4 transforms the originally chaotic time-domain signal into angular-domain features with clear periodicity by resampling and mapping the original spatiotemporally synchronized data and extracting synchronization error components. Simultaneously, combined with advanced signal processing techniques such as spectrum analysis, it can accurately identify and effectively filter out noise components introduced by irrelevant factors such as environmental vibration and electromagnetic interference. This processing strategy significantly improves the overall signal-to-noise ratio, enabling the clear extraction and identification of manufacturing precision deviations and mechanical wear characteristics present in CNC rotary tables. This provides highly pure and well-structured analytical samples for accurately separating stable geometric error components from complex mixed signals.

[0100] By constructing an error decoupling model with physical constraints, the S5 can effectively eliminate sensor bias based on the difference between forward and reverse displacement measurements. Furthermore, it performs multi-dimensional component attribution and mathematical decomposition of the mixed geometric, thermal, and load-induced errors. This deep-level error decoupling capability enables technicians to accurately grasp the specific weight of each error source's impact on the overall system accuracy. It successfully overcomes the technical bottleneck of multiple dynamic error factors being mutually coupled and difficult to quantify, providing a clear, scientific, and quantifiable mathematical basis for developing targeted accuracy compensation strategies and optimizing machine tool performance.

[0101] S6 extends traditionally single, static detection results into dynamic interval assessments with a certain confidence level by substituting dynamic error components and real-time operating parameters into a rigorous uncertainty calculation formula. This method comprehensively considers the impact of measurement repeatability, system consistency, and result traceability on the final conclusion, achieving a leap from point estimation to interval estimation. This dual verification mechanism based on statistical principles and physical laws provides a quantifiable assessment method for the operational reliability of machine tools under different speeds and temperature rise gradients, enabling the detection system to dynamically adjust the expected accuracy and tolerance range according to real-time operating conditions, greatly enhancing its adaptability and practicality in complex and ever-changing industrial environments.

[0102] S7 achieves a closed-loop function, from basic accuracy assessment to high-level process support decision-making, by comparing the calculated uncertainty window with the predetermined tolerance threshold in real time and automatically outputting machining window suggestions and compensation strategies. This mechanism provides operators with specific compensation values ​​and predictions of the equipment's sustainable and stable operation time, effectively guiding the CNC system to optimize cutting parameters and machining paths in real time, avoiding workpiece scrap and quality defects caused by excessive temperature drift or structural deformation. This data-driven intelligent decision support significantly improves the consistency of machining quality and greatly enhances the utilization efficiency of machine tools and overall production benefits.

[0103] Example 1: Continuous high-speed cutting scenario for thin-walled aerospace parts

[0104] In this embodiment, the CNC rotary table is continuously operating at high speed. This simulation scenario aims to fully verify the effectiveness of the method proposed in this invention in solving the problem of the disconnect between the actual processing state and the inspection state.

[0105] Before performing the machining of thin-walled aerospace aluminum alloy components on a certain type of high-precision five-axis CNC machining center, in strict accordance with the technical solution in S1, high-response temperature sensors were tightly installed on the outer ring surface of the turntable bearing support and the key temperature measurement points of the drive motor housing; at the same time, three sets of high-precision eddy current displacement sensors were arranged in a spatially orthogonal manner around the turntable reference sphere to construct a multi-dimensional observation field that can capture dynamic changes in all directions.

[0106] The CNC rotary table is controlled to perform a symmetrical rotational motion trajectory, including forward and reverse rotation, at a base speed of 60 revolutions per minute under simulated real cutting load conditions. During this process, the drive motor generates a large amount of heat due to continuous high-speed operation, and monitoring data shows that the temperature in the bearing area rises by a cumulative 15 degrees Celsius within 30 minutes.

[0107] By utilizing the specially designed synchronous triggering module of S3, the instantaneous data stream of the sensor chain is captured in real time. Furthermore, an advanced interpolation algorithm is used to dynamically correct the measurement lag caused by thermal inertia of the temperature sensor, ensuring that the collected thermal gradient source data and the axial drift recorded by the displacement sensing system can be precisely aligned at the millisecond level in the time dimension.

[0108] The system accurately identifies the instantaneous tilt error of the axis caused by the time-varying evolution of the temperature field based on real-time data streams. It successfully transforms the instantaneous thermal drift phenomenon, which cannot be observed under quasi-static conditions by traditional methods, into a quantifiable and compensable dynamic error component. Ultimately, the detection results are highly consistent with the thermal performance in the actual high-speed machining process, effectively eliminating the long-standing technical defect of the disconnect between the detection state and the machining state caused by differences in working conditions.

[0109] Example 2: For heavy-duty, multi-angle machining scenarios of large molds

[0110] In this embodiment, the CNC rotary table is subjected to a large change in load and performs a complex multi-angle motion trajectory. This scenario is mainly used to verify the decoupling and separation capability of the present invention for multi-source coupling errors.

[0111] A large mold steel blank specimen weighing 500 kg was installed on the turntable. According to the measurement procedure specified in S2, a forward and reverse symmetrical scanning trajectory containing small amplitude reciprocating motion was executed to capture the nonlinear jumping characteristics exhibited by the transmission pair under different force angles.

[0112] The system uses the high-frequency sampling module of S4 to acquire a mixed displacement signal that includes manufacturing deviations, structural deformation caused by gravity, and external environmental vibration interference. It uses spectrum analysis technology to filter out the runout component that is not synchronized with the motion, and uses the angle domain averaging algorithm to extract low-order harmonic components in order to accurately identify the machining eccentricity error term of the worm gear pair.

[0113] Substituting the observed time series data into the physical constraint error decoupling mathematical model established by S5, the symmetrical difference between the forward and reverse displacement vectors is used to effectively eliminate the sensor's own installation bias error, thereby completely separating and analyzing the coupling relationship between the elastic structural deformation caused by heavy load and the inherent geometric accuracy error of the turntable.

[0114] By mapping each independent error component to the actual tool tip position through a spatial coordinate transformation matrix, it is possible to clearly distinguish which error components originate from load sensitivity effects and which belong to the inherent geometric errors of the system. This enables accurate tracing and attribution of complex coupled error sources, and successfully solves engineering problems that are difficult to diagnose effectively in the context of intertwined dynamic factors.

[0115] Example 3: Long-term online monitoring of precision optical lenses

[0116] In this embodiment, the processing task places extremely high demands on the accuracy and stability of the equipment during long-term operation. This scenario is mainly used to verify the application value of the present invention in processing window prediction and intelligent decision-making closed loop.

[0117] In a high-stability workshop with constant temperature and humidity, the CNC rotary table is continuously monitored online for 8 hours. Based on the method proposed in S6, the comprehensive uncertainty of dynamic spatial error is calculated in real time.

[0118] The system automatically statistically analyzes the trend of the standard deviation of repeatability measurement and the consistency residual of forward and reverse rotation over time. It finds that during long-term operation, the difference in the thermal response characteristics of the transmission pair leads to a nonlinear expansion trend in the comprehensive uncertainty range.

[0119] The S7's intelligent decision-making module compares the current uncertainty prediction window with a preset tolerance threshold of 0.005 mm in real time. When the system's prediction error will exceed the allowable range after 2 hours, it automatically generates a detailed process parameter adjustment suggestion table, including "reduce the feed rate by 15%" and "force the cooling program after 120 minutes".

[0120] By feeding the compensation values ​​back to the CNC system in real time and dynamically adjusting the machining parameter window, the traditional passive measurement is transformed into a prediction-based proactive prevention and control strategy, ultimately ensuring the quality consistency of the entire batch of machining. This fully demonstrates the significant engineering value of this invention in improving the overall utilization rate of equipment and ensuring forming accuracy.

[0121] Example 1 focuses on the thermal changes caused by high-speed continuous cutting. Through precise placement of thermal response sensitive points and thermal hysteresis compensation of the sampling frequency, this example successfully demonstrates that the method can transform the implicit temperature rise gradient into an explicit instantaneous drift component. The core technology lies in eliminating the observation phase difference between the time and angular domains, ensuring that the detection data can track the physical evolution of the CNC rotary table during high-speed operation in real time. The successful execution of this example signifies that the detection system has completely abandoned the outdated mode of cold-state detection guiding hot-state machining, achieving physical-level synchronization between the detection conditions and the actual machining state.

[0122] Example 2 addresses the problem of intertwined geometric and mechanical errors in heavy-duty multi-angle machining. By executing a symmetrical measurement trajectory with mirror characteristics and deconstructing the original five-degree-of-freedom observation matrix using a physical constraint decoupling model, this example verifies the system's accuracy in decomposing mixed error components. It can quantify and attribute manufacturing deviations of the worm gear, sensor installation biases, and elastic deformation caused by gravity loads, making the sources of each error clearly visible. The implementation of this example demonstrates the system's ability to extract pure physical features from chaotic signals, providing mathematical support for targeted compensation.

[0123] Example 3 focuses on reliability assessment and decision output in long-term online monitoring scenarios. Through analysis of the expansion rate of the dynamic spatial error uncertainty window, this example demonstrates the system's transformation from a data acquisition unit to a process decision-making engine. The closed-loop verification of this example showcases the significant potential of this invention in improving part processing consistency and equipment utilization.

[0124] Embodiments of the present invention have been presented and described. It will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to the embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. A method for detecting the accuracy of a CNC rotary table, characterized in that, include: S1. Deploy temperature sensors at the spindle bearings, transmission pair housings, and thermal response sensitive points of the drive motor of the CNC rotary table, and establish a displacement sensing link between the spindle end and the rotary table reference component; S2. Control the deployed CNC rotary table to execute symmetrical measurement trajectories of forward rotation, reverse rotation, and re-forward rotation under load conditions, generating a resolvable observation sequence with geometric symmetry constraints; S3. Use a sampling trigger module to acquire temperature data, displacement data, and internal feedback angle data of the CNC rotary table during the operation of the symmetrical measurement trajectory, and perform spatiotemporal synchronization alignment processing; S4. Resample and map the spatiotemporally synchronized displacement data and internal feedback angle data, and extract the synchronization error component that is periodically repeated by the rotation angle; S5. Construct a physical constraint error decoupling model containing the synchronization error component, use the displacement difference between forward and reverse rotation to eliminate sensor system bias, and isolate the geometrically inherent error, dynamic temperature drift component, and load-sensitive component; S6, Substitute the dynamic temperature drift component, load-sensitive component, and the real-time collected current rotation speed, current load, and current temperature rise gradient into the uncertainty calculation formula to obtain the dynamic spatial error uncertainty window; S7, Compare the dynamic spatial error uncertainty window with the preset tolerance threshold, and generate an inspection report containing compensation values ​​and processing window suggestions based on the comparison results.

2. The method for detecting the accuracy of a CNC rotary table according to claim 1, characterized in that, The deployment of thermal response sensitive points in S1 refers to selecting the support bearing of the shaft seat, the meshing zone boundary of the transmission pair housing, and the winding housing of the drive motor as temperature acquisition reference points based on the heat source distribution characteristics inside the CNC turntable, in order to capture thermal gradient source data that causes instantaneous axis drift; the establishment of displacement sensing links in S1 refers to arranging at least three sets of non-contact displacement sensors in an orthogonal form with a 90-degree angle between each other on the outer circumferential surface and the top surface of the turntable reference component, and constructing a five-degree-of-freedom spatial original observation matrix containing tilt error terms and translation error terms by collecting radial runout data and axial movement data.

3. The method for detecting the accuracy of a CNC rotary table according to claim 1, characterized in that, The geometric symmetry constraint in S2 refers to the physical characteristic that the geometric errors of the forward rotation observation sequence and the reverse rotation observation sequence at the same angular position are mirror images of each other. By performing superposition averaging and difference calculations on the distinguishable observation sequences, the physical separation of the synchronous drift term and the noise term during the rotation process is achieved. The symmetrical measurement trajectory in S2 refers to controlling the CNC turntable to perform rotational motion with a preset stepped speed sequence under load conditions, and triggering the displacement sensing link to perform multiple reciprocating micro-motion observations at each angular measurement point to capture the nonlinear dynamic jumping characteristics caused by uneven meshing stiffness of the transmission pair.

4. The method for detecting the accuracy of a CNC rotary table according to claim 1, characterized in that, The spatiotemporal synchronization alignment process in S3 refers to synchronizing the displacement data and the internal feedback angle data to the same clock reference at the microsecond level through a hard trigger signal, ensuring that the measurement point and the motion coordinate correspond one-to-one. The spatiotemporal synchronization alignment process includes using the sampling trigger module to perform resampling processing on the temperature data based on sampling frequency compensation, and using an interpolation algorithm to eliminate the thermal inertia hysteresis of the temperature sensor, so that the temperature data of each thermal response sensitive point and the displacement data at the same physical moment are phase aligned on the time axis.

5. The method for detecting the accuracy of a CNC rotary table according to claim 1, characterized in that, The extraction of synchronization error components in S4 refers to identifying harmonic components in the rotary signal through spectrum analysis, retaining signals whose frequencies are integer multiples of the rotary frequency of the CNC rotary table, and filtering out asynchronous fluctuation components caused by environmental vibrations. The extraction of synchronization error components in S4 also refers to using low-pass filtering and angle domain averaging techniques to remove non-periodic high-frequency components, projecting the resampled and mapped data onto the polar coordinate system, and extracting low-order harmonic components reflecting the eccentricity of the CNC rotary table installation and the eccentricity of the worm gear machining through Fourier series decomposition.

6. The method for detecting the accuracy of a CNC rotary table according to claim 1, characterized in that, The physical constraint error decoupling model in S5 is as follows: ,in, To account for spatial errors, This is a geometrically inherent error base map. This refers to the dynamic temperature drift coefficient. For the temperature rise gradient at the sensitive point, For load sensitivity coefficient, This represents the equivalent load change. Here, t represents the rotation angle, and t represents the running time. The geometrically inherent error refers to the static geometric deviation curve that does not change over time, extracted from the forward and reverse symmetrical measurement trajectory when the CNC rotary table is in a preset thermal steady state and under no-load conditions. The dynamic temperature drift component refers to the real-time monitoring of the temperature rise correlation characteristics between the bearing and the transmission pair housing, calculating the instantaneous drift of the dynamic spatial error due to the temperature field evolution, and establishing a temperature hysteresis correction term for error compensation. The load-sensitive component refers to the identification of the elastic deformation characteristics of the CNC rotary table structure caused by gravity load by comparing the displacement offset caused by workpieces of different masses at the same angular position.

7. The method for detecting the accuracy of a CNC rotary table according to claim 1, characterized in that, The error decoupling model under physical constraints in S5 includes a spatial coordinate transformation matrix. By mapping the geometric inherent error, the dynamic temperature drift component, and the load-sensitive component to the machine tool coordinate system where the machining tip is located, the correlation transformation from the rotation accuracy of the CNC rotary table itself to the actual forming accuracy of the workpiece is realized.

8. The method for detecting the accuracy of a CNC rotary table according to claim 1, characterized in that, The formula for calculating the dynamic spatial error uncertainty window in S6 is as follows: Where U is the comprehensive uncertainty of dynamic space error, u r To measure the repeatability standard deviation, u c For the forward and reverse consistent residuals, u s The reference component traceability error; the forward and reverse consistency residual u c It refers to the maximum deviation of the spatial trajectory between the forward rotation measurement curve and the reverse rotation measurement curve after the offset is eliminated by the sensor system under the same ambient temperature and load conditions. It is used to quantitatively characterize the dynamic drift limit caused by the inconsistency of the thermal response of the drive motor and the transmission pair.

9. The method for detecting the accuracy of a CNC rotary table according to claim 1, characterized in that, The output machining window suggestion in S7 refers to predicting the effective continuous machining time for the CNC rotary table error to remain within the allowable tolerance range based on the evolution trend of the dynamic temperature drift component. The output machining window suggestion includes using the load-sensitive component to perform a mechanical coupling evaluation of the current workpiece quality, and combining the expansion rate of the dynamic spatial error uncertainty window to generate a process parameter suggestion table for the CNC system that includes the upper limit of the optimal cutting feed rate and the expected shutdown cooling time.

10. A CNC rotary table accuracy testing system, comprising a CNC rotary table accuracy testing method according to any one of claims 1-9, characterized in that, The CNC rotary table accuracy detection system includes: a multi-source sensing module for acquiring temperature data of thermally sensitive points, displacement data of the spindle end, and angle data inside the rotary table; a synchronization triggering module for providing a microsecond-level clock reference for the multi-source sensing module to achieve spatiotemporal synchronization alignment between data items; a trajectory control module for driving the CNC rotary table to execute symmetrical measurement trajectories of forward rotation, reverse rotation, and re-forward rotation under load conditions; an error decoupling engine module for extracting geometrically inherent errors, dynamic temperature drift components, and load-sensitive components based on a physical constraint error decoupling model; and an intelligent decision-making module for calculating the dynamic spatial error uncertainty window and generating a detection report containing compensation values ​​and machining window suggestions.

Images