Method and system for correcting and positioning double-axis numerical control machine tool based on alignment platform

Abstract

The invention relates to the technical field of numerical control machine tool control, in particular to a method and system for correcting and positioning a double-shaft numerical control machine tool based on an alignment platform. The method comprises the following steps that an alignment platform is installed on the double-shaft numerical control machine tool, the alignment platform comprises a displacement measuring unit and an angle adjusting unit, the displacement measuring unit is used for collecting position information of the end of a main shaft, and the angle adjusting unit is used for multi-degree-of-freedom pose adjustment; controlling the double-shaft numerical control machine tool to execute a preset test motion trail, and acquiring actual position data of the end part of the main shaft through a displacement measurement unit to obtain an actual position data set; and extracting a theoretical position data set corresponding to the preset test movement track from a numerical control system of the double-shaft numerical control machine tool. According to the method, the problem that the asymmetric positioning deviation generated by heat-force coupling in complex curved surface machining of a double-shaft machine tool is difficult to trace and correct in real time is fundamentally solved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of numerical control machine tool control, and particularly relates to a method and system for realizing correction positioning of a double-shaft numerical control machine tool based on a positioning platform. BACKGROUND

[0002] In the field of precision machining, especially in the machining process of small precision components such as micro-electronic lead frame stamping and micro-connector injection molding, double-shaft numerical control machine tools have long faced the problem of difficulty in tracing dynamic errors. Looking back at its development history, in the early days, positioning calibration was only performed by static rulers and laser interferometers. With the surge in the demand for machining precision and efficiency of micro parts, the machine tool needs to maintain micron-level precision under continuous interpolation conditions. However, under actual machining conditions, when the machine tool performs complex trajectory interpolation motion, due to the coupling effect of spindle thermal elongation and guide rail pitch angle, an asymmetric positioning deviation that is difficult to measure directly by a laser interferometer will be generated in the workpiece coordinate system. This deviation is easily averaged and distributed to each motion axis in the traditional single-axis calibration mode, resulting in a mismatch between precision compensation data and actual error sources. SUMMARY

[0003] Therefore, it is necessary to provide a method and system for realizing correction positioning of a double-shaft numerical control machine tool based on a positioning platform to solve at least one of the above technical problems.

[0004] To achieve the above-mentioned purpose, a method for realizing correction positioning of a double-shaft numerical control machine tool based on a positioning platform comprises the following steps: Step S1: installing a positioning platform on a double-shaft numerical control machine tool, the positioning platform comprising a displacement measurement unit and an angle adjustment unit, wherein the displacement measurement unit is used to collect position information of the end of the spindle, and the angle adjustment unit is used for multi-degree-of-freedom pose adjustment; Step S2: controlling the double-shaft numerical control machine tool to perform a preset test motion trajectory, and simultaneously collecting actual position data of the end of the spindle by the displacement measurement unit to obtain an actual position data set; Step S3: extracting a theoretical position data set corresponding to the preset test motion trajectory from the numerical control system of the double-shaft numerical control machine tool; and calculating a position deviation vector set according to the actual position data set and the theoretical position data set; Step S4: based on the position deviation vector set, identifying a thermal deviation component caused by spindle thermal elongation and an angle deviation component caused by guide rail pitch angle through error separation processing, and synthesizing a coupled error vector; Step S5: controlling the angle adjustment unit of the positioning platform to perform real-time pose compensation according to the coupled error vector.

[0005] Preferably, the present invention also provides a system for calibration and positioning of a dual-axis CNC machine tool based on a alignment platform, used to execute the method for calibration and positioning of a dual-axis CNC machine tool based on a alignment platform as described above. The system for calibration and positioning of a dual-axis CNC machine tool based on a alignment platform includes: The benchmark establishment module is used to install a positioning platform on a dual-axis CNC machine tool. The positioning platform includes a displacement measurement unit and an angle adjustment unit. The displacement measurement unit is used to collect position information at the end of the spindle, and the angle adjustment unit is used for multi-degree-of-freedom pose adjustment. The position acquisition module is used to control the dual-axis CNC machine tool to execute a preset test motion trajectory, and at the same time, it acquires the actual position data of the spindle end through the displacement measurement unit to obtain the actual position dataset; The deviation calculation module is used to extract the theoretical position dataset corresponding to the preset test motion trajectory from the CNC system of a dual-axis CNC machine tool; and to calculate the position deviation vector set based on the actual position dataset and the theoretical position dataset. The error separation module is used to identify the thermal deviation component caused by the spindle thermal extension and the angular deviation component caused by the guide rail pitch angle based on the position deviation vector set, and synthesize the coupled error vector. The pose compensation module is used to control the angle adjustment unit of the alignment platform to perform real-time pose compensation based on the coupling error vector.

[0006] The beneficial effects of this invention are as follows: On the one hand, by using a closed loop of measurement-modeling-separation-compensation, the traditional post-event, static, single-axis calibration is upgraded to in-event, dynamic, coupled calibration. A high-precision alignment platform directly captures the true position and pose of the spindle end in the workpiece coordinate system, making the asymmetric deviations that cannot be identified by the laser interferometer explicit. Thus, without adding additional sensors, the complete observation of the error source is achieved, improving the completeness of the error model.

[0007] On the other hand, by using an error separation algorithm to project the position deviation vector set axially and radially, and combining the spindle housing temperature sequence and the guide rail pitch angle sequence for temporal-spatial correlation, the two physical sources of thermal extension and angular deformation are decoupled from the total deviation, forming an interpretable and predictable coupled error vector. This avoids the source-matching mismatch caused by the average distribution of errors in traditional single-axis calibration, and makes the compensation amount correspond one-to-one with the physical error, greatly improving the calibration accuracy.

[0008] On the other hand, by applying the coupling error reverse vector in real time through the 3 translation + 2 rotation degrees of freedom of the alignment platform, micron-level dynamic compensation is completed under continuous interpolation conditions. Without stopping the machine, the contour misalignment and multi-step cumulative error in the processing of micro-components can be reduced from ≥5μm to 2μm. This fundamentally solves the problem that it is difficult to trace and correct the asymmetric positioning deviation caused by thermo-mechanical coupling in high-precision, high-speed processing of dual-axis machine tools. Attached Figure Description

[0009] Other features, objects, and advantages of the present invention will become more apparent from the following detailed description taken in conjunction with the accompanying drawings: Fig. 1 A flowchart illustrating the steps of a method for correcting and positioning a dual-axis CNC machine tool based on a alignment platform is shown in one embodiment.

[0010] Fig. 2 A detailed flowchart of step S3 of one embodiment is shown.

[0011] Fig. 3 A diagram of a dual-axis CNC machine tool according to an embodiment is shown. Detailed Implementation

[0012] The technical method of the present invention will now be clearly and completely described with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of the present invention. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without inventive effort are within the scope of protection of the present invention.

[0013] Furthermore, the accompanying drawings are merely illustrative of the invention and are not necessarily drawn to scale. The same reference numerals in the drawings denote the same or similar parts, and therefore repeated descriptions of them will be omitted. Some block diagrams shown in the drawings are functional entities and do not necessarily correspond to physically or logically independent entities. These functional entities can be implemented in software, in one or more hardware modules or integrated circuits, or in different network and / or processor methods and / or microcontroller methods.

[0014] It should be understood that although the terms "first," "second," etc., may be used herein to describe various units, these units should not be limited by these terms. These terms are used merely to distinguish one unit from another. For example, without departing from the scope of the exemplary embodiments, a first unit may be referred to as a second unit, and similarly, a second unit may be referred to as a first unit. The term "and / or" as used herein includes any and all combinations of one or more of the associated listed items.

[0015] To achieve the above objectives, please refer to Figs. 1 to 3 This invention provides a method for calibration and positioning of a dual-axis CNC machine tool based on a alignment platform, comprising the following steps: Step S1: Install a positioning platform on a dual-axis CNC machine tool. The positioning platform includes a displacement measurement unit and an angle adjustment unit. The displacement measurement unit is used to collect position information of the spindle end, and the angle adjustment unit is used for multi-degree-of-freedom pose adjustment. Step S2: Control the dual-axis CNC machine tool to execute the preset test motion trajectory, and at the same time collect the actual position data of the spindle end through the displacement measurement unit to obtain the actual position dataset; Step S3: Extract the theoretical position dataset corresponding to the preset test motion trajectory from the CNC system of the dual-axis CNC machine tool; calculate the position deviation vector set based on the actual position dataset and the theoretical position dataset; Step S4: Based on the position deviation vector set, identify the thermal deviation component caused by the spindle thermal extension and the angular deviation component caused by the guide rail pitch angle through error separation processing, and synthesize the coupled error vector; Step S5: Control the angle adjustment unit of the alignment platform to perform real-time pose compensation based on the coupling error vector.

[0016] In this embodiment, please refer to Fig. 3 The dual-axis CNC machine tool is equipped with a worktable 101, on which a alignment platform 102 is installed. A displacement measuring unit 103 is provided on the alignment platform 102 corresponding to the end of the spindle, which is used to detect the displacement of the end of the spindle. An angle adjustment unit 104 is connected to the alignment platform 102, and the angle adjustment unit 104 is linked with the displacement measuring unit 103 through a control signal.

[0017] Preferably, the installation of the alignment platform on the dual-axis CNC machine tool in step S1 includes: The alignment platform is fixed to the worktable of a dual-axis CNC machine tool using a rigid connector, and the relative positional relationship between the alignment platform and the machine tool coordinate system is calibrated using a laser tracker to establish the transformation matrix between the platform coordinate system and the machine tool coordinate system. In this embodiment, four M12 high-strength stainless steel rigid connectors are evenly arranged on the lower surface of the alignment platform. The lower ends of the connectors directly mate with the T-slots of the dual-axis CNC machine tool's worktable. Furthermore, a laser tracker is mounted 2.5m above the machine tool, allowing the tracker's beam to cover all target points within the maximum travel range of the worktable. First, the coordinates of the centers of the three reference holes on the worktable are collected in the machine tool coordinate system as machine tool reference points. Then, the coordinates of the corresponding three points are collected on a 1.5-inch SMR target ball attached to the platform surface as platform reference points. The rotation and translation amounts between the two sets of point sets are automatically solved using the instrument's built-in SPA algorithm, generating a 4×4 homogeneous transformation matrix, thereby completing the initial alignment from the platform coordinate system to the machine tool coordinate system.

[0018] In this case, please refer to the dual-axis CNC machine tool worktable. Fig. 3 .

[0019] In one implementation of this embodiment, it is assumed that the coordinates of the machine tool reference point are respectively ( , , ), ( , , ), ( , , The platform reference point coordinates are ( , , ), ( , , ), ( , , The transformation matrix calculated by the laser tracker software is denoted as... ;Will The data is written to the non-volatile memory area of ​​the CNC system, and a coordinate transformation validity flag is set; thereafter, the CNC system calls up the data each time the displacement measurement unit outputs a set of initial position data. Perform real-time coordinate transformation to generate an initial position dataset in the machine tool coordinate system.

[0020] By using a transformation matrix, the initial position data collected by the displacement measurement unit of the alignment platform is transformed to the machine tool coordinate system, generating an initial position dataset in the machine tool coordinate system. In this embodiment, the displacement measurement unit of the alignment platform consists of three capacitive displacement sensors with a resolution of 0.1µm. The sensors are arranged in an equilateral triangle to read the local coordinates of the spindle end relative to the platform in real time. After the laser tracker completes the initial alignment, it triggers a coordinate transformation command, substituting the local coordinate values ​​(u,v,w) measured by the three sensors at the same sampling time into the transformation matrix. The three-dimensional coordinates (U, V, W) in the machine tool coordinate system are obtained; 200 sets of data are continuously collected for 200ms, and outliers other than 3σ are removed and averaged to form the initial position dataset in the machine tool coordinate system, denoted as . This dataset contains the initial offsets of the spindle end face center in the X, Y, and Z directions.

[0021] Based on the initial position dataset, calculate the initial alignment error between the alignment platform and the theoretical coordinates of the machine tool; In this embodiment, the initial alignment error is quantized using Euclidean distance: Read After obtaining the three-dimensional coordinates, the spatial deviation vector is obtained by subtracting them from the theoretical coordinates given by the CAM model one by one. The magnitude of the spatial deviation vector is then calculated as the alignment error value. The system is set to continuously collect data 10 times, with an interval of 50ms between each time. The 10 magnitude values ​​are sorted and the maximum and minimum values ​​are removed. The average of the remaining 8 values ​​is taken to obtain the final alignment error. If the final alignment error is less than or equal to the preset alignment threshold, it is judged as qualified; otherwise, recalibration is performed.

[0022] It is important to note that CAM model refers to Computer-Aided Manufacturing model, which is the theoretical toolpath data or tool position file generated by CAM software based on part design drawings and used to control the machining path of machine tools. It contains the theoretical sequence of coordinate points that the tool should traverse during machining and serves as the reference for the CNC system to execute machining commands.

[0023] When the initial alignment error exceeds the preset alignment threshold, adjust the installation position of the alignment platform and re-perform the calibration steps until the initial alignment error meets the preset alignment threshold.

[0024] In this embodiment, when the initial alignment error exceeds a preset alignment threshold, the CNC system outputs an adjustment vector ( X, Y, Z) guides the operator to move the alignment platform slightly along the X, Y, and Z axes using a handheld terminal. The amount of movement is automatically calculated by the system based on the error vector and visualized on the terminal using arrows and numerical values. After each movement, the system re-collects 10 sets of coordinates and calculates the new alignment error until the error is lower than the threshold for the first time, completing the iterative convergence. A maximum of 5 cycles are allowed; otherwise, the system will stop and report an installation reference abnormality.

[0025] Preferably, step S2 includes: According to the preset test motion trajectory, the dual-axis CNC machine tool is controlled to perform compound motion in two stages. The preset test motion trajectory includes a three-dimensional spiral trajectory and a rectangular grid trajectory. In this embodiment, the preset test motion trajectory is divided into two continuously executed stages: the first stage is a three-dimensional spiral trajectory, used to induce axial thermal extension error of the spindle; the second stage is a rectangular grid trajectory, used to induce radial error caused by the guide rail pitch angle. There is no pause between the two stages, and the transition section uses a 5mm circular arc for smooth connection. The CNC system loads both trajectory segments at once via G-code and forms a continuous interpolation data stream in the internal buffer.

[0026] In one implementation of this embodiment, the preset first spacing value (spiral pitch) is set to 8mm, the preset first radius value is set to 25mm, the spiral has 3 turns, and the total stroke is about 235mm; the preset second spacing value (grid row spacing and column spacing) is set to 10mm, the grid range is 100mm×100mm, and there are a total of 11×11=121 nodes; the machine tool first completes the spiral ascent at a feed speed of 2000mm / min, and then completes the entire grid in a zigzag pattern at a feed speed of 1500mm / min, with a total time of about 90s; the two trajectories need to start from the same coordinate origin in the CNC system.

[0027] In the first stage of the compound motion, a three-dimensional spiral trajectory is executed, with the pitch of the spiral trajectory being a preset first spacing value and the radius being a preset first radius value. In this embodiment, the three-dimensional helical trajectory is generated by the helical interpolation function G02.3 / G03.3 built into the CNC system. The parameters include radius, pitch, total number of turns, and direction of rotation. Before trajectory generation, the CNC system automatically checks whether the product of radius and pitch exceeds the maximum rotation diameter of the worktable. If it exceeds the limit, an illegal helical parameter prompt will pop up and the program will be terminated.

[0028] The second stage of the composite motion executes a rectangular grid trajectory, where the row spacing and column spacing of the rectangular grid trajectory are both preset second spacing values; In this embodiment, the rectangular grid trajectory adopts a zigzag tool path strategy; the grid boundary is implemented through a G-code framework loop (FOR-NEXT), with the loop variables corresponding to the X-axis and Y-axis coordinates respectively, and the step size being the spacing value; at each node, the CNC system inserts a macro program call instruction to trigger the displacement measurement unit to collect position data once, and at the same time records the current interpolation cycle counter value as a timestamp.

[0029] In one implementation of this embodiment, the row and column spacing are both set to 10mm. The loop variable X increments by 10mm from -50mm to +50mm, and the loop variable Y has the same range. X increments in odd-numbered rows and decrements in even-numbered rows, forming a zigzag path. Macro program m198 P100 (P100 represents the acquisition sequence number) is inserted at each node. When the CNC system reaches the node, the spindle end face coordinates (X,Y,Z) and the cycle counter value are recorded immediately.

[0030] During trajectory execution, the actual position data of the spindle end is collected by the displacement measurement unit at a frequency no less than the preset sampling frequency, and the timestamp corresponding to each collection point is recorded. In one implementation of this embodiment, the preset sampling frequency is set to no less than 500Hz (i.e., 2ms period). The CNC system reads the sensor buffer register within the interrupt to obtain the current (u,v,w) local coordinates, and then calls the transformation matrix to obtain the machine tool coordinates (U,V,W). At the same time, the current interpolation period number is written to the end of the data packet as a timestamp.

[0031] The trajectory deviation between the actual motion trajectory and the preset test motion trajectory is calculated based on the actual location data. When the trajectory deviation exceeds the preset trajectory tolerance, the preset test motion trajectory is re-executed.

[0032] In this embodiment, the spatial deviation vector is obtained by subtracting the current theoretical coordinates from the collected actual coordinates point by point. The deviation calculation is performed in the background task. The maximum deviation value of each row of grid is calculated as the trajectory deviation after each row is completed. If the maximum deviation exceeds the preset trajectory tolerance of 0.020mm, the feed is paused, a trajectory deviation dialog box pops up, and the position of the deviation node is highlighted, allowing the operator to choose to rerun this segment or continue running. If the operator chooses to rerun, the machine tool will automatically return to the starting point of the row and re-execute the trajectory segment.

[0033] Preferably, step S3 includes: Step S31: Obtain the theoretical position dataset through the CNC system; In this embodiment, after loading the preset test motion trajectory file, the CNC system parses the tool position information segment by segment in the internal interpolation buffer and generates a theoretical coordinate stream in real time according to the 2ms interpolation cycle. The coordinate stream is written into the power-off holding register in rows, with each row containing the theoretical coordinates of the X, Y, and Z axes and the corresponding interpolation cycle number, forming a theoretical position dataset.

[0034] Step S32: Based on the theoretical position dataset and the actual position data, calculate the Euclidean space distance deviation value at each position point, and calculate the direction angle deviation value between the theoretical direction vector and the actual motion direction; In this embodiment, after obtaining each set of actual coordinates, the theoretical coordinates corresponding to the same cycle number are called; the difference between the two sets of coordinates is calculated axis by axis to obtain the three-dimensional difference value, and then the sum of squares is accumulated. Furthermore, the floating-point square root instruction is called to obtain the spatial straight-line distance, which is the Euclidean spatial distance deviation value of the point.

[0035] Most importantly, the theoretical direction vector is obtained based on the theoretical coordinates of the current point and the next point. The actual direction vector is then obtained in the same way. The built-in vector dot product and magnitude command is then called to calculate the angle between the two vectors, which is the direction angle deviation value of the segment.

[0036] Step S33: Based on the Euclidean distance deviation value and the direction angle deviation value, construct a set of position deviation vectors, where each position deviation vector contains both magnitude and direction feature information.

[0037] In this embodiment, after obtaining the distance deviation value and direction angle deviation value of each point, the distance deviation is projected along the direction based on the theoretical direction vector to obtain the three-dimensional components of the deviation vector; the angle deviation is converted into a small rotation vector perpendicular to the theoretical direction, and then superimposed with the distance projection vector to form a position deviation vector that simultaneously contains linear offset and angle offset; the deviation vectors of all nodes are stored in a dynamic array in the order of acquisition sequence, which is the position deviation vector set.

[0038] Preferably, step S4 includes: The axial deviation component along the main axis and the radial deviation component perpendicular to the main axis are separated from the position deviation vector set. The axial deviation component is calculated by the axial projection of the main axis, and the radial deviation component is calculated by the radial plane projection of the main axis. In this embodiment, after the CNC system reads the position deviation vector set, it first determines the instantaneous direction unit vector of the spindle axis in the machine tool coordinate system based on the current spindle angle encoder value. Then, it projects each deviation vector onto the direction of the unit vector to obtain the axial deviation component. The remaining part is then projected onto the radial plane perpendicular to the spindle axis to obtain the radial deviation component.

[0039] The spindle housing temperature sequence is acquired in real time, the axial deviation component is correlated with the spindle housing temperature sequence in time, and the thermal deviation component caused by the thermal extension of the spindle is extracted. In this embodiment, the CNC system collects the instantaneous temperature of three PT100 temperature sensors on the surface of the spindle housing at a frequency of 1Hz via RS-485 bus, and forms a spindle housing temperature sequence by averaging the data. The UTC timestamp of each temperature point is recorded synchronously and aligned with the timestamp of the axial deviation array A[]. Data with a time difference exceeding 0.5s will be automatically interpolated. After alignment, the temperature change rate is calculated using a sliding window method. When the temperature rise within the window is greater than the preset temperature rise threshold (e.g., 0.3℃) and the axial deviation increment is greater than the preset deviation increment (e.g., 0.005mm), it is determined that the deviation is mainly caused by thermal extension and is extracted as a thermal deviation component.

[0040] The guide rail tilt angle sequence is collected in real time by the guide rail tilt angle sensor. The radial deviation component is spatially correlated with the guide rail tilt angle sequence, and the angular deviation component caused by the guide rail tilt angle is extracted. In this embodiment, a 0.001° resolution mEMS tilt sensor is arranged on each side of the machine tool worktable to measure the change in guide rail pitch angle in real time. The CNC system reads the average value of the two sensors at a frequency of 100Hz to form a pitch angle sequence, which is aligned with the acquisition time of the radial deviation array R[]. Then, the radial deviation is grouped according to the node position and linearly fitted with the pitch angle value at the corresponding position. The segment with a fitting slope greater than the preset slope (e.g., 0.002mm / °) and a correlation coefficient higher than the preset coefficient threshold (e.g., 0.85) is considered to be mainly affected by the guide rail angle, and its fitting value is extracted as the angle deviation component.

[0041] The thermal deviation component and the angular deviation component are vector-synthesized according to a preset spatial geometric relationship to form a coupled error vector.

[0042] In this embodiment, the thermal deviation component is directly accumulated along the axial direction of the main shaft, and the angular deviation component acts along the pitch direction of the guide rail. The two are approximately perpendicular in space and are formed by linear superposition to form a coupling error vector. The synthesized vector contains three-dimensional components and action direction labels and is stored in the coupling error array C[].

[0043] Preferably, after step S1, an unloaded test is performed using a positioning platform to collect background vibration data and generate a vibration compensation table.

[0044] In this embodiment, after the alignment platform completes rigid installation and initial alignment, the CNC system switches to no-load calibration mode, automatically disabling the spindle coolant and tool compensation functions. The platform displacement measurement unit continuously collects the micro-displacement of the spindle end in a static state at a preset sampling rate, with a collection time set to 10 seconds, corresponding to 20,000 data points. The system simultaneously records the ambient temperature, humidity, and machine tool servo enable status as background vibration raw data packets. The data acquisition task is handled by the embedded DSP (Digital Signal Processor) built into the alignment platform. The DSP uses 24-bit... The ADC synchronously reads analog signals from three displacement sensors, with an internal hardware FIFO buffer depth of 1024 points, triggering DMA transfer when half-full. During acquisition, the DSP synchronously records the instantaneous values ​​of the machine tool servo current. If the current of any axis suddenly changes by more than 2% of the rated value, the time period is marked as an electrical disturbance and a time tag is inserted into the data packet header for subsequent rejection or weighting. The CNC system calls the built-in FFT library to perform 8192-point segmented spectrum estimation on the cleaned vibration data, with an overlap rate of 50%, using the Hanning window function, and a frequency resolution of approximately 0.24Hz. The system extracts 20 The maximum amplitude and corresponding frequency of the low-frequency band below Hz are recorded, and the power frequency peak at 50Hz and its harmonics are recorded as background vibration characteristic spectral lines. The characteristic data is written into a structure array, and each element contains frequency, amplitude, and phase information. A vibration compensation table is generated based on the characteristic spectral lines: for each frequency to be compensated, a phase-amplitude two-dimensional table is established in the range of 0–360° with a step size of 30°. The amplitude in the table is taken as the measured value and the direction is taken as the opposite phase. The version number, timestamp, and temperature value are written in the header of the compensation table to prevent the misuse of old tables. The table capacity is fixed at 128 records. When the capacity is exceeded, adjacent frequency bands with the same amplitude are merged.

[0045] Preferably, step S5 includes: The coupling error vector is spatially decomposed according to the adjustable degrees of freedom of the alignment platform to obtain three translational degrees of freedom adjustment amounts and two rotational degrees of freedom adjustment amounts. The translational degrees of freedom adjustment amounts include linear displacement compensation values ​​of the X-axis, Y-axis and Z-axis, and the rotational degrees of freedom adjustment amounts include angular displacement compensation values ​​about the X-axis and Y-axis. In this embodiment, the coupling error vector is first projected onto the X, Y, and Z axes of the alignment platform coordinate system to obtain three translational degree of freedom compensation values; then, with the platform rotation center as the origin, the minute rotation angles of the error vector around the X and Y axes are calculated to obtain two rotational degree of freedom compensation values; the five compensation values ​​are written into the structure, with limit values ​​of ±10µm and ±0.05° respectively. Exceeding the limit triggers a compensation saturation warning to prevent over-driving of the piezoelectric ceramic or angle motor.

[0046] In one implementation of this embodiment, the coupling error vector is (0.008,0,0.015)mm, which is decomposed into 8µm in the X direction, 0µm in the Y direction, and 15µm in the Z direction, with a rotation angle of 0.02° around the X direction and a rotation angle of −0.03° around the Y direction. All five values ​​are written into the structure and labeled with node number 90. After amplitude limiting judgment, none of them exceed the limit.

[0047] The corresponding control signal sequence is generated based on the adjustment amount of each degree of freedom. The translational degree of freedom adjustment amount is converted into the voltage control signal of the piezoelectric ceramic drive platform, and the rotational degree of freedom adjustment amount is converted into the current control signal of the angle adjustment unit. In this embodiment, the translation amount is converted into a piezoelectric ceramic driving voltage by a 16-bit DAC: 0µm corresponds to 0V and 10µm corresponds to 100V. The conversion coefficient in this embodiment is fixed in the chip EEPROM. The rotation amount is output as 0–200mA current by a 12-bit current-type DAC to drive the voice coil motor to achieve micro-rotation. The CNC system packages the five signals into a CAN message at a frequency of 1kHz.

[0048] In one implementation of this embodiment, 8µm in the X direction corresponds to a voltage of 80V, 15µm in the Z direction corresponds to 150V, 0.02° around the X direction corresponds to a current of 40mA, and −0.03° around the Y direction corresponds to −60mA. The five signals are written into the same 8-byte data field of the CAN frame and sent within 1ms. After receiving the data, the DSP on the platform side (the DSP installed on the side of the piezoelectric ceramic drive platform) sends back the frame sequence number for confirmation. If there is no confirmation within 2ms, the CNC system automatically retransmits the data.

[0049] Obtain the vibration compensation table; In this embodiment, the CNC system first reads the vibration compensation table in each control cycle and obtains the real-time vibration compensation amount based on the current phase interpolation. The interpolation result is split into five micro-compensation values ​​according to the degrees of freedom, which are linearly superimposed with the macro-compensation value obtained by error decomposition to form the final compensation instruction. The version number of the compensation table and the temperature label are compared in real time. If the temperature change exceeds ±2℃, the table temperature drift is indicated and the no-load update process is automatically triggered.

[0050] The control signal sequence is sent to the angle adjustment unit of the alignment platform to dynamically adjust the control signal. According to the vibration compensation table, the alignment platform is controlled to perform pose fine-tuning at a frequency not lower than the preset compensation frequency, and pose compensation is completed within each preset control cycle.

[0051] In this embodiment, the DSP on the positioning platform runs a position-current dual closed-loop PID at a frequency of 2kHz. After receiving the compensation command, the speed feedforward is calculated first through the position loop, and then the drive signal is output through the current loop. The PID parameters are divided into three levels: large signal level, medium signal level, and micro signal level. The CNC system automatically switches according to the absolute value of the compensation amount. The proportional gain of the micro signal level is reduced by 50% to prevent small overshoot below 1µm.

[0052] Most importantly, the pose compensation process also includes: The actual adjustment amount of each degree of freedom of the positioning platform is monitored in real time, and the actual adjustment amount is compared with the preset theoretical adjustment amount. When the comparison deviation exceeds the preset tolerance, the proportional coefficient, integral coefficient and derivative coefficient of the PID control algorithm are automatically adjusted, and the compensation amount sequence is recalculated until the deviation between the actual adjustment amount and the preset theoretical adjustment amount meets the preset tolerance requirement.

[0053] In this embodiment, the alignment platform reads feedback from five sensors after each compensation cycle: three capacitive displacement sensors for actual X, Y, and Z displacements, and two mEMS tilt sensors for actual rotation angles around X and Y. The feedback values, along with the theoretical adjustment values, are written into a circular array with a depth of 200 groups. The CNC system calculates the absolute difference every 10ms. If any degree of freedom difference exceeds 0.02µm or 0.001°, a deviation anomaly flag is triggered, illuminating a yellow warning light and initiating the PID self-tuning subroutine. Tolerance thresholds are divided into micrometer and micro-angle levels: linear degree of freedom tolerance 0.02µm, rotational degree of freedom tolerance 0.001°. These thresholds are stored in EEPROM and can be modified via an authorized interface. A restart is required after modification to prevent on-site misoperation. The CNC system uses "OR" logic for judgment; if any degree of freedom exceeds the tolerance, PID adjustment is initiated. The judgment cycle is related to the sensor... The PID self-tuning uses an incremental trial-and-error method: the proportional coefficient Kp is increased by 10% first, and the deviation trend is observed within 5 cycles. If the absolute value of the deviation decreases, the gain is retained; otherwise, it is rolled back and the gain is reduced by 10%. The integral coefficient Ki increases in 5% increments when the steady-state error is greater than 0.01µm, and the derivative coefficient Kd decreases in 2% increments when an oscillation trend appears. Each coefficient modification is immediately written to RAm and the DSP on the platform side is updated. After the coefficient is updated, the system immediately recalculates the compensation sequence for the next 100 cycles with the new PID parameters: the theoretical adjustment remains unchanged, and the actual output value is iterated in a closed loop according to the new coefficients until the deviation is lower than the tolerance and there is no rebound for 3 consecutive cycles, which indicates convergence. After convergence, the system automatically turns off the yellow light and records the convergence cycle number and the final deviation value. If convergence is not achieved within 100 cycles, the system stops and reports PID self-tuning failure, prompting manual intervention.

[0054] Preferably, step S5 further includes: After the initial calibration is completed, the dual-axis CNC machine tool is controlled again to execute the preset test motion trajectory and new actual position data is collected. In this embodiment, after the initial pose compensation is completed, the CNC system automatically enters the retest subroutine: it eliminates manual intervention and re-drives the machine tool according to the original spiral + rectangular grid dual-stage trajectory, with all parameters being completely consistent with the initial test; the alignment platform collects the spindle end coordinates in real time with the same 2kHz sampling rate and the same sensor configuration to form a new actual position dataset.

[0055] Based on the newly acquired actual location dataset and theoretical location dataset, the updated location deviation vector set is recalculated. In this embodiment, after the retest data is uploaded to the host computer, the CNC system calls the same theoretical dataset and calculates the new Euclidean distance deviation and the new direction angle deviation node by node. The calculation process is the same as the first time, generating an updated position deviation vector set. Each vector still contains magnitude, direction, timestamp, and temperature label.

[0056] Calculate the average magnitude of all position deviation vectors in the updated position deviation vector set; In this embodiment, the CNC system reads the updated position deviation vector set, extracts the straight-line distance of the vector space line by line, sums them up and divides them by the total number of nodes to obtain the updated average module length. In the module length calculation, 64-bit double-precision floating point is used, and the result is retained to three decimal places in micrometers. If the average module length is less than the preset precision threshold of 0.010mm, it is judged to be qualified; otherwise, it is marked that it needs to be iterated again.

[0057] The correction result is determined based on the comparison between the average modulus and the preset accuracy threshold.

[0058] Preferably, determining the correction result based on the comparison between the average modulus and the preset accuracy threshold includes: The average module length is compared with the preset accuracy threshold. When the average module length is less than the preset accuracy threshold, the correction and positioning are determined to be complete, and the final correction result is output. In this embodiment, when the CNC system performs a power-on self-test, it reads the preset accuracy threshold of 0.010mm into the RAM and performs a floating-point comparison with the updated average module length. If the average module length is less than the threshold, it immediately sets the correction completion flag, locks the alignment platform compensation parameters, stops the trajectory retest, and pops up a dialog box on the human-machine interface to display the correction and positioning completion and the final average module length value.

[0059] In one implementation of this embodiment, the average mold length is calculated to be 0.006 mm. After comparing 0.006 mm with 0.010 mm, it is determined to be qualified. The green light on the interface lights up and indicates that the average mold length is 6µm, and the calibration is completed. After the operator clicks to confirm, the system generates the final calibration result, which includes the date, average mold length, maximum single-point deviation, and a snapshot of the compensation parameters.

[0060] When the average modulus is greater than or equal to the preset accuracy threshold, the correction process is repeated, and the average modulus of the position deviation vector set is recalculated after each iteration. In this embodiment, if the average module length is greater than or equal to the threshold, the CNC system sets the iteration flag, automatically returns to step S2, drives the machine tool to re-execute the same spiral + grid trajectory, collects new actual position data, and recalculates the updated deviation vector set and average module length. Before each iteration, the current compensation parameters are written to the iteration log. The iteration process does not require manual intervention and is allowed up to 5 times to prevent infinite loops.

[0061] In one implementation of this embodiment, when the average mold length exceeds the threshold of 0.012 mm for the first time, the CNC system displays a "Entering the second iteration" prompt, the machine tool automatically returns to zero and re-runs the trajectory; after the second retest, the average mold length drops to 0.009 mm, the system immediately determines that it is qualified, terminates the iteration, and saves the corresponding compensation parameters.

[0062] Record the trend of the average magnitude of the position deviation vector set during each iteration. When the average magnitude change rate of three consecutive iterations is less than the preset convergence threshold, the correction process is determined to have reached a stable state. In this embodiment, a circular array with a depth of 5 groups is established for the average modulus after each iteration, and the change rate of the most recent 3 groups is calculated: change rate = (current modulus - previous modulus) ÷ previous modulus × 100%; when the absolute value of the change rate for 3 consecutive times is less than the preset convergence threshold of 5%, the determination process is stable, the iteration is stopped, and the current result is output; the determination logic is written into the firmware to prevent misjudgment caused by occasional fluctuations.

[0063] When the number of iterations reaches the preset maximum value, the iteration process stops and the current optimal correction result is output.

[0064] In this embodiment, the maximum number of iterations is preset to 5. The counter increments before each return to S2. When it reaches 5 iterations, the machine is forcibly stopped, the iteration limit is reported, and the 5th result is output as the current best. Before stopping, the 5th mold length and corresponding parameters are packaged into the current best correction result. If the final mold length still exceeds the threshold, the interface pops up a red alarm to check mechanical looseness or ambient temperature, guiding manual intervention to prevent unqualified parts from flowing downstream.

[0065] Preferably, the present invention also provides a system for calibration and positioning of a dual-axis CNC machine tool based on a alignment platform, used to execute the method for calibration and positioning of a dual-axis CNC machine tool based on a alignment platform as described above. The system for calibration and positioning of a dual-axis CNC machine tool based on a alignment platform includes: The benchmark establishment module is used to install a positioning platform on a dual-axis CNC machine tool. The positioning platform includes a displacement measurement unit and an angle adjustment unit. The displacement measurement unit is used to collect position information at the end of the spindle, and the angle adjustment unit is used for multi-degree-of-freedom pose adjustment. The position acquisition module is used to control the dual-axis CNC machine tool to execute a preset test motion trajectory, and at the same time, it acquires the actual position data of the spindle end through the displacement measurement unit to obtain the actual position dataset; The deviation calculation module is used to extract the theoretical position dataset corresponding to the preset test motion trajectory from the CNC system of a dual-axis CNC machine tool; and to calculate the position deviation vector set based on the actual position dataset and the theoretical position dataset. The error separation module is used to identify the thermal deviation component caused by the spindle thermal extension and the angular deviation component caused by the guide rail pitch angle based on the position deviation vector set, and synthesize the coupled error vector. The pose compensation module is used to control the angle adjustment unit of the alignment platform to perform real-time pose compensation based on the coupling error vector.

[0066] Therefore, the embodiments should be considered exemplary and non-limiting in all respects, and the scope of the invention is defined by the appended claims rather than the foregoing description. Thus, all variations falling within the meaning and scope of the equivalents of the application are intended to be included within the invention.

[0067] The above description is merely a specific embodiment of the present invention, enabling those skilled in the art to understand or implement the invention. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the invention. Therefore, the present invention is not to be limited to the embodiments shown herein, but is to be accorded the widest scope consistent with the principles and novel features of the invention herein.

Claims

1. A method for calibration and positioning of a dual-axis CNC machine tool based on a alignment platform, characterized in that, Includes the following steps: Step S1: Install a positioning platform on a dual-axis CNC machine tool. The positioning platform includes a displacement measurement unit and an angle adjustment unit. The displacement measurement unit is used to collect position information of the spindle end, and the angle adjustment unit is used for multi-degree-of-freedom pose adjustment. Step S2: Control the dual-axis CNC machine tool to execute the preset test motion trajectory, and at the same time collect the actual position data of the spindle end through the displacement measurement unit to obtain the actual position dataset; Step S3: Extract the theoretical position dataset corresponding to the preset test motion trajectory from the CNC system of the dual-axis CNC machine tool; Calculate the position deviation vector set based on the actual position dataset and the theoretical position dataset; Step S4: Based on the position deviation vector set, identify the thermal deviation component caused by the spindle thermal extension and the angular deviation component caused by the guide rail pitch angle through error separation processing, and synthesize the coupled error vector; Step S5: Control the angle adjustment unit of the alignment platform to perform real-time pose compensation based on the coupling error vector.

2. The method for calibration and positioning of a dual-axis CNC machine tool based on a alignment platform according to claim 1, characterized in that, Step S1, which involves installing the alignment platform on the dual-axis CNC machine tool, includes: The alignment platform is fixed to the worktable of a dual-axis CNC machine tool using a rigid connector, and the relative positional relationship between the alignment platform and the machine tool coordinate system is calibrated using a laser tracker to establish the transformation matrix between the platform coordinate system and the machine tool coordinate system. By using a transformation matrix, the initial position data collected by the displacement measurement unit of the alignment platform is transformed to the machine tool coordinate system, generating an initial position dataset in the machine tool coordinate system. Based on the initial position dataset, calculate the initial alignment error between the alignment platform and the theoretical coordinates of the machine tool; When the initial alignment error exceeds the preset alignment threshold, adjust the installation position of the alignment platform and re-perform the calibration steps until the initial alignment error meets the preset alignment threshold.

3. The method for calibration and positioning of a dual-axis CNC machine tool based on a alignment platform according to claim 1, characterized in that, Step S2 includes: According to the preset test motion trajectory, the dual-axis CNC machine tool is controlled to perform compound motion in two stages. The preset test motion trajectory includes a three-dimensional spiral trajectory and a rectangular grid trajectory. In the first stage of the compound motion, a three-dimensional spiral trajectory is executed, with the pitch of the spiral trajectory being a preset first spacing value and the radius being a preset first radius value. The second stage of the composite motion executes a rectangular grid trajectory, where the row spacing and column spacing of the rectangular grid trajectory are both preset second spacing values; During trajectory execution, the actual position data of the spindle end is collected by the displacement measurement unit at a frequency no less than the preset sampling frequency, and the timestamp corresponding to each collection point is recorded. The trajectory deviation between the actual motion trajectory and the preset test motion trajectory is calculated based on the actual location data. When the trajectory deviation exceeds the preset trajectory tolerance, the preset test motion trajectory is re-executed.

4. The method for calibration and positioning of a dual-axis CNC machine tool based on a alignment platform according to claim 1, characterized in that, Step S3 includes: Step S31: Obtain the theoretical position dataset through the CNC system; Step S32: Based on the theoretical position dataset and the actual position data, calculate the Euclidean space distance deviation value at each position point, and calculate the direction angle deviation value between the theoretical direction vector and the actual motion direction; Step S33: Based on the Euclidean distance deviation value and the direction angle deviation value, construct a set of position deviation vectors, where each position deviation vector contains both magnitude and direction feature information.

5. The method for calibration and positioning of a dual-axis CNC machine tool based on a alignment platform according to claim 1, characterized in that, Step S4 includes: The axial deviation component along the main axis and the radial deviation component perpendicular to the main axis are separated from the position deviation vector set. The axial deviation component is calculated by the axial projection of the main axis, and the radial deviation component is calculated by the radial plane projection of the main axis. The spindle housing temperature sequence is acquired in real time, the axial deviation component is correlated with the spindle housing temperature sequence in time, and the thermal deviation component caused by the thermal extension of the spindle is extracted. The guide rail tilt angle sequence is collected in real time by the guide rail tilt angle sensor. The radial deviation component is spatially correlated with the guide rail tilt angle sequence, and the angular deviation component caused by the guide rail tilt angle is extracted. The thermal deviation component and the angular deviation component are vector-synthesized according to a preset spatial geometric relationship to form a coupled error vector.

6. The method for calibration and positioning of a dual-axis CNC machine tool based on a alignment platform according to claim 1, characterized in that, After step S1, an unloaded test is performed using a positioning platform to collect background vibration data and generate a vibration compensation table.

7. The method for calibration and positioning of a dual-axis CNC machine tool based on a alignment platform according to claim 1, characterized in that, Step S5 includes: The coupling error vector is spatially decomposed according to the adjustable degrees of freedom of the alignment platform to obtain three translational degrees of freedom adjustment amounts and two rotational degrees of freedom adjustment amounts. The translational degrees of freedom adjustment amounts include linear displacement compensation values ​​of the X-axis, Y-axis and Z-axis, and the rotational degrees of freedom adjustment amounts include angular displacement compensation values ​​about the X-axis and Y-axis. The corresponding control signal sequence is generated based on the adjustment amount of each degree of freedom. The translational degree of freedom adjustment amount is converted into the voltage control signal of the piezoelectric ceramic drive platform, and the rotational degree of freedom adjustment amount is converted into the current control signal of the angle adjustment unit. Obtain the vibration compensation table; The control signal sequence is sent to the angle adjustment unit of the alignment platform to dynamically adjust the control signal. According to the vibration compensation table, the alignment platform is controlled to perform pose fine-tuning at a frequency not lower than the preset compensation frequency, and pose compensation is completed within each preset control cycle.

8. The method for calibration and positioning of a dual-axis CNC machine tool based on a alignment platform according to claim 1, characterized in that, Step S5 is followed by: After the initial calibration is completed, the dual-axis CNC machine tool is controlled again to execute the preset test motion trajectory and new actual position data is collected. Based on the newly acquired actual location dataset and theoretical location dataset, the updated location deviation vector set is recalculated. Calculate the average magnitude of all position deviation vectors in the updated position deviation vector set; The correction result is determined based on the comparison between the average modulus and the preset accuracy threshold.

9. The method for calibration and positioning of a dual-axis CNC machine tool based on a alignment platform according to claim 8, characterized in that, The correction results are determined based on the comparison between the average modulus and the preset accuracy threshold, including: The average module length is compared with the preset accuracy threshold. When the average module length is less than the preset accuracy threshold, the correction and positioning are determined to be complete, and the final correction result is output. When the average modulus is greater than or equal to the preset accuracy threshold, the correction process is repeated, and the average modulus of the position deviation vector set is recalculated after each iteration. Record the trend of the average magnitude of the position deviation vector set during each iteration. When the average magnitude change rate of three consecutive iterations is less than the preset convergence threshold, the correction process is determined to have reached a stable state. When the number of iterations reaches the preset maximum value, the iteration process stops and the current optimal correction result is output.

10. A system for calibration and positioning of a dual-axis CNC machine tool based on a alignment platform, characterized in that, For performing the method of dual-axis CNC machine tool calibration and positioning based on a alignment platform as described in claim 1, the system for dual-axis CNC machine tool calibration and positioning based on a alignment platform includes: The benchmark establishment module is used to install a positioning platform on a dual-axis CNC machine tool. The positioning platform includes a displacement measurement unit and an angle adjustment unit. The displacement measurement unit is used to collect position information at the end of the spindle, and the angle adjustment unit is used for multi-degree-of-freedom pose adjustment. The position acquisition module is used to control the dual-axis CNC machine tool to execute a preset test motion trajectory, and at the same time, it acquires the actual position data of the spindle end through the displacement measurement unit to obtain the actual position dataset; The deviation calculation module is used to extract the theoretical position dataset corresponding to the preset test motion trajectory from the CNC system of a dual-axis CNC machine tool; and to calculate the position deviation vector set based on the actual position dataset and the theoretical position dataset. The error separation module is used to identify the thermal deviation component caused by the spindle thermal extension and the angular deviation component caused by the guide rail pitch angle based on the position deviation vector set, and synthesize the coupled error vector. The pose compensation module is used to control the angle adjustment unit of the alignment platform to perform real-time pose compensation based on the coupling error vector.

Images