A multi-workpiece coordinate system-based automatic machining control method and system

Abstract

The application provides a kind of automatic machining control method and system based on multi-workpiece coordinate system, comprising: importing batch workpiece unified machining process model, constructing global reference coordinate system and standardized multi-workpiece coordinate system group and storing original offset parameter, carrying out full-station in-machine reference detection and double-layer cross consensus verification according to preset period, obtaining station deviation verification result, identifying and structuring deviation event information, no deviation then cyclically executing reference detection, calling multi-period historical deviation event for time series and space joint analysis, extracting coordinate system offset evolution law and determining multi-level offset risk level, adaptively correcting workpiece coordinate system offset parameter according to risk level, updating NC machining tool path and issuing machining instruction in linkage, completing multi-station workpiece batch automation machining, effectively solving the problems of low accuracy of traditional reference correction, fuzzy deviation traceability and single compensation method, significantly improving the reference stability and machining consistency of multi-station batch machining.

Description

Technical Field

[0001] This invention relates to the field of automatic machining control technology for CNC machine tools, and more specifically, to an automatic machining control method and system based on multiple workpiece coordinate systems. Background Technology

[0002] Multi-station CNC machine tools are widely used in precision manufacturing for batch processing. The accuracy of the workpiece coordinate system directly affects the product qualification rate. Affected by fixture deformation, machine tool thermal deformation, worktable wear, and clamping errors, the reference of each station is prone to offset. Failure to correct in time will cause batch workpieces to be scrapped due to out-of-tolerance dimensions.

[0003] Existing multi-station machine tool reference calibration schemes mostly adopt manual fixed-point sampling calibration mode, which can only detect reference deviation at a single point once. It is impossible to carry out periodic and continuous reference detection for all stations, and it is difficult to simultaneously obtain the difference between the real-time reference measured coordinates and the original offset parameters of each station. It cannot achieve full-station two-layer cross consensus verification, and it cannot distinguish between the two types of deviation sources: local offset of a single workpiece and systematic offset of the worktable. The accuracy of deviation identification is low.

[0004] Meanwhile, existing solutions only save single-cycle deviation data, lack a multi-cycle deviation event archiving mechanism, lack the ability to perform joint temporal and spatial analysis, cannot extract offset evolution characteristics or generate workstation-specific offset patterns, rely solely on static deviation levels to determine risks, have limited correction methods, and lack adaptability. Summary of the Invention

[0005] In view of the aforementioned problems, and in conjunction with the first aspect of the present invention, embodiments of the present invention provide an automatic machining control method based on a multi-workpiece coordinate system, comprising the following steps: S1. Import the unified processing technology model of batch workpieces, construct a global reference coordinate system, collect the reference calibration feature points of workpieces at multiple workstations on the workbench, construct a standardized multi-workpiece coordinate system group based on the global reference coordinate system and calibration feature points, and store the original offset parameters of each workpiece coordinate system. S2. Start the on-machine reference detection process according to the preset cycle, and simultaneously collect the current reference calibration feature points of all workpieces at all workstations. Collect the measured coordinates of each point feature. Perform a two-layer cross consensus verification based on the original offset parameters of the global reference coordinate system and each workpiece coordinate system to obtain the workstation deviation verification results. S3. Based on the workstation deviation verification results, perform event identification. If a deviation event is identified, extract the workstation basic information, offset quantization level, and coordinate system correctable parameters from the workstation deviation verification results, integrate the above information into a deviation event, and store it. If there are no deviation events, proceed directly to the next cycle of in-machine reference detection process; S4. Retrieve all historical deviation events stored in step S3, conduct joint analysis of temporal evolution and workstation spatial distribution based on all historical deviation events, and extract the workpiece coordinate system offset evolution law; determine the workpiece coordinate system offset risk level of each workstation based on the evolution law, and adaptively correct the corresponding workpiece coordinate system offset parameters according to the risk level, update the NC machining toolpath in linkage, and issue automatic machining control commands to complete batch machining.

[0006] Furthermore, embodiments of the present invention also provide an automatic machining control system based on a multi-workpiece coordinate system, used to implement the above-described method, including: The datum construction module is used to import the machining process model, generate a global datum coordinate system and a standardized multi-workpiece coordinate system group, and store the original offset parameters of each workpiece coordinate system. The detection and verification module, along with the periodic detection and consensus verification module, is used to periodically collect the measured coordinates of the workstation benchmark calibration feature points, perform two-layer cross-consensus verification, and output the workstation deviation verification results. The event storage module and the deviation event acquisition and storage module are used to identify, archive and store various deviation event information. When there is no deviation, it directly jumps to the next cycle of benchmark detection. The evolution correction module, offset evolution analysis and adaptive correction module are used to analyze historical deviation events to obtain offset evolution rules, determine the offset risk level and correct the coordinate system offset in stages, update the NC machining toolpath, and issue machining commands to complete batch machining.

[0007] Compared with the prior art, the present invention has the following beneficial effects: Step S1 establishes a unified reference framework by importing a unified processing technology model for batch workpieces. It uses machine tool reference balls and contact probes to collect reference calibration feature points of each station, and uniformly constructs a global reference coordinate system and a standardized multi-workpiece coordinate system group and stores the original offset parameters. This provides unified original reference data for the two-layer cross-consensus verification of all stations, ensuring that the coordinate reference of multi-station workpieces is unified and traceable.

[0008] Step S2 sets up periodic automatic on-machine reference detection, synchronously collects real-time reference measured coordinates of all workstations, completes double-layer cross consensus verification through longitudinal coordinate difference comparison and transverse workstation consistency comparison, distinguishes three types of results: no deviation, local offset, and systematic offset, comprehensively identifies reference deviations of different causes, and improves the coverage and accuracy of deviation identification.

[0009] Step S3 continuously identifies and archives complete deviation event information for each round of workstation deviation verification results, classifying them into minor, moderate, and severe deviations and attributes that can or cannot be corrected; when there is no deviation, it directly skips storage and enters the next round of detection, which not only completely retains the historical data of the full cycle of benchmark deviation, but also reduces the invalid storage overhead in the scenario of no deviation, providing complete data support for subsequent multi-dimensional evolution analysis.

[0010] Step S4 retrieves all historical deviation events for joint temporal and workstation spatial analysis, extracts two types of dimensional evolution features to generate workstation-specific offset evolution patterns; based on the evolution features, it classifies four gradient risk levels, matches corresponding static, first-order, and second-order compensation strategies and critical workstation alarm locking mechanisms, adaptively corrects coordinate system offsets and dynamically adjusts the detection cycle; finally, based on the updated coordinate system, it reconstructs the NC machining toolpath and automatically issues machining commands, realizing full automation of the reference deviation identification, graded compensation, toolpath update, and batch machining process, continuously maintaining long-term machining accuracy across multiple workstations, and avoiding batch workpieces from exceeding tolerances and being scrapped. Attached Figure Description

[0011] Figure 1 This is a flowchart of the steps of an automatic machining control method based on a multi-workpiece coordinate system according to the present invention; Figure 2 This invention relates to an automatic machining control system based on a multi-workpiece coordinate system. Detailed Implementation

[0012] The present invention will now be described in detail with reference to the accompanying drawings. Figure 1 This is a flowchart of the steps of an automatic machining control method based on a multi-workpiece coordinate system according to the present invention. The following is a detailed description of the automatic machining control method based on a multi-workpiece coordinate system.

[0013] Step S1: Import the unified machining process model for batch workpieces, construct a global reference coordinate system, collect the reference calibration feature points of workpieces at multiple workstations on the workbench, construct a standardized multi-workpiece coordinate system group based on the global reference coordinate system and the calibration feature points, and store the original offset parameters of each workpiece coordinate system; specifically, step S1 includes steps S11-S14: Step S11: Load the unified processing technology model for batch workpieces and filter all effective processing stations on the machine tool worktable.

[0014] Among them, the unified processing technology model for batch workpieces is a standardized process parameter model established in advance according to the processing requirements of batch workpieces. This process model includes at least five types of core process parameters: workpiece shape parameters, workpiece datum feature parameters, worktable station layout parameters, batch processing procedure parameters, and workpiece theoretical coordinate offset.

[0015] Specifically, the workpiece shape parameters include the workpiece's length, width, and height dimensions, blank allowance, and coordinates of shape feature points; the workpiece reference feature parameters are the workpiece's preset positioning reference holes, reference surfaces, and reference edge features used for coordinate system calibration; the worktable station layout parameters include the total number of worktable stations, the row and column spacing of each station, the theoretical coordinates of the station center, and the station number sequence; the batch processing process parameters include the process sequence of roughing, semi-finishing, and finishing, the processing feed parameters, and the spindle speed parameters; the theoretical coordinate offset is the preset standard offset coordinate value of the workpiece at each station relative to the machine tool's mechanical origin.

[0016] By loading the complete process model described above, the processing reference logic and station layout rules for batch workpieces can be unified. Then, based on the station layout parameters in the process model, idle workstations, faulty workstations, and invalid workstations that exceed the workpiece processing stroke are eliminated. Only valid processing workstations that meet the clamping conditions and processing stroke conditions are selected, thus defining the effective processing area for subsequent multi-station unified reference calibration.

[0017] Step S12: Perform full-domain on-machine detection using the machine tool standard reference sphere, fit and construct a global reference coordinate system, and lock the theoretical reference parameters of the global reference.

[0018] Among them, the machine tool standard reference ball is a standard part specially used for machine tool factory calibration. It has a fixed standard ball diameter, high roundness and high dimensional accuracy, and is the standard carrier for machine tool mechanical reference traceability; the whole-domain on-machine detection is a reference acquisition method that controls the machine tool probe to traverse the whole-domain detection points of the worktable.

[0019] The specific detection and fitting process is as follows: the standard reference ball of the machine tool is fixed at the reference installation position of the worktable, the inherent mechanical origin of the machine tool is used as the initial reference zero point, the contact probe is controlled to perform multi-point full-domain contact detection on the standard reference ball, and multiple sets of measured coordinates of the center of the reference ball are collected; the coordinates of the center of the reference ball are fitted and corrected by the least squares method to eliminate the random error of a single detection, and a global reference coordinate system covering all effective positions of the worktable is accurately fitted and generated.

[0020] The global reference coordinate system uses the machine tool's mechanical origin as the coordinate origin, constructs three orthogonal reference coordinate axes of X, Y, and Z, locks the three-axis theoretical zero point coordinates, three-axis theoretical vector directions, and global reference calibration characteristic theoretical coordinates of the global reference coordinate system, and establishes a unique, unified, and traceable overall machining reference for the entire worktable, serving as the top-level reference basis for the workpiece coordinate system of all workstations.

[0021] Step S13: Based on the global reference coordinate system, the multi-station reference features are detected synchronously by the probe, and the station-specific workpiece coordinate system is generated one by one, and the standardized multi-workpiece coordinate system group is obtained by summarizing.

[0022] Among them, the reference calibration feature points are the fixed reference acquisition points preset by the workpiece process at each station. They are the core positioning feature points for the construction of the workpiece coordinate system. Each effective station is preset with no less than 3 non-collinear reference calibration feature points to ensure the uniqueness and orthogonality of the coordinate system construction.

[0023] The specific construction process is as follows: After the machine tool completes the construction of the global reference coordinate system, it controls the contact probe to automatically touch and detect the reference calibration feature points of the workpieces in all effective stations according to the preset detection path, and collects the real-time measured reference coordinates of each reference calibration feature point in each station; the measured reference coordinates of all stations are uniformly mapped to the constructed global reference coordinate system to complete the unified registration of global coordinates.

[0024] For each valid workstation, based on the registration coordinates of all reference calibration feature points of that workstation, an independent workpiece coordinate system is fitted to generate that workstation, which uniquely corresponds to the workpiece clamping reference and machining coordinate reference of a single workstation. After traversing all valid workstations to complete the construction of the single workstation coordinate system, all workstation-specific workpiece coordinate systems are summarized and integrated according to the worktable workstation arrangement order to form a standardized multi-workpiece coordinate system group that covers the entire machining area, is independent of each other, and has a unified traceable global reference.

[0025] Step S14: Bind and store the original reference parameters of the workpiece coordinate system of each station, and establish a two-layer cross-consensus verification original reference database.

[0026] The reference parameters that need to be bound and stored specifically include four types of core parameters: the original three-axis offset values ​​of each workpiece coordinate system relative to the global reference coordinate system, the workstation number, the workpiece model, and the coordinates of the reference point.

[0027] Specifically, the original three-axis offset values ​​are the standard offsets of the X, Y, and Z axes of the single-station workpiece coordinate system relative to the theoretical zero point of the global reference coordinate system, which are the core reference values ​​for subsequent deviation comparisons; the station number is the unique identification code of each effective station on the workbench, used to distinguish the reference data of different stations; the workpiece model is the model parameter of the workpiece being processed in the current batch, used to match the reference process requirements of the corresponding workpiece; and the reference point coordinates are the theoretical standard coordinates of the calibration feature points used to construct the coordinate system of each station.

[0028] The above four types of parameters are bound and archived to the machine tool CNC system one by one, serving as fixed original reference data for subsequent S2 steps of double-layer cross-consensus verification, station deviation comparison, and offset deviation identification, ensuring the data uniformity and traceability accuracy of the full-cycle reference verification.

[0029] Step S2: Start the on-machine reference detection process according to the preset cycle, synchronously collect the current reference calibration feature points of all workpieces at all workstations, and collect the measured coordinates of each point feature; perform a two-layer cross consensus verification based on the original offset parameters of the global reference coordinate system and each workpiece coordinate system to obtain the workstation deviation verification results. Specifically, step S2 includes steps S21-S24: Step S21: Preset a fixed reference detection cycle, trigger synchronous reference detection at all workstations at the cycle node, and collect the real-time measured reference coordinates of each workstation.

[0030] In this step, the preset detection cycle is a time trigger parameter that the user has pre-programmed into the CNC system based on the batch processing cycle, processing accuracy requirements, and machine tool operating conditions. The unit is minutes. It is used to realize periodic online benchmark monitoring of the processing process, which is different from the traditional single-start calibration mode.

[0031] The in-machine reference detection process is an in-situ reference detection process that does not stop the machine tool or disassemble the workpiece. Automatic point acquisition is completed by the machine tool's built-in contact probe. The synchronous acquisition refers to the machine tool triggering the reference feature acquisition of all effective stations in parallel at one time according to the multi-station detection logic, rather than acquiring each station sequentially. This ensures that the measured coordinates of all stations belong to the reference state under the same processing condition at the same time, avoiding verification errors caused by timing deviations.

[0032] For example: In this embodiment, the preset detection cycle is 15 minutes under batch precision machining conditions. The machine tool automatically triggers a global reference detection task every 15 minutes. The system calls the contact probe to traverse all effective workstations from workstation 1 to workstation 7 at once, and collects the real-time coordinates of the three preset reference calibration feature points of each workstation to obtain the real-time measured reference coordinate dataset of each workstation.

[0033] Step S22: Construct a two-layer cross-consensus verification mechanism, defining two levels of verification logic: vertical benchmark comparison and horizontal workstation cross-comparison.

[0034] This invention innovatively sets up a two-layer cross-consensus verification mechanism, including a vertical benchmark comparison level and a horizontal workstation cross-comparison level. The two levels of verification are executed simultaneously and mutually corroborate each other, so as to accurately distinguish the source of deviation.

[0035] The longitudinal benchmark comparison is a single-station self-comparison verification logic. The original offset parameters of the workpiece coordinate system of each station are archived and stored in step S1 as the absolute benchmark. The difference calculation is performed between the current real-time measured coordinates of the same station and the historical standard original offset parameters of the station to identify the benchmark offset of a single station. The cross-comparison of the horizontal workstations is a multi-workstation comparison and verification logic. Based on the rigid installation characteristics of the same workbench and the principle of consistency of the clamping process of the same batch of workpieces, the measured coordinate data of all workstations at the current moment are compared for consistency, which is used to determine whether the deviation is an independent anomaly of a single workstation or an overall systemic anomaly of the workbench.

[0036] Step S23: Perform longitudinal benchmark comparison and transverse workstation cross-comparison to obtain the double-layer comparison difference and consistency results.

[0037] The longitudinal benchmark comparison process is as follows: For any valid workstation, retrieve the original offset parameter that is uniquely corresponding to the workstation, calculate the three-axis deviation difference between the measured coordinates of the X, Y, and Z axes and the original standard coordinates, and obtain the three-axis offset of the single workstation, which is used to characterize the degree of offset of the workstation's own benchmark.

[0038] The cross-comparison process of horizontal workstations is as follows: extract the three-axis measured coordinates of all valid workstations in the current cycle, calculate the mean and dispersion of the coordinate data of each workstation, determine whether the offset trend of each workstation is consistent, whether the offset direction is the same, and whether the offset magnitude is close, and output the coordinate consistency evaluation results between workstations.

[0039] For example: In a certain periodic detection, the measured deviation of the X-axis of station 1 is +0.022mm and the measured deviation of the Y-axis is +0.020mm. The deviations of the three axes of the other stations 2 to 7 are all less than 0.003mm. The lateral dispersion is large and the station consistency is poor, which is judged as an independent anomaly of a single station. If all stations from station 1 to station 7 synchronously produce a deviation of +0.018mm to +0.022mm in the same direction on the X-axis and the deviation trend is highly uniform, it is judged as an overall deviation trend of the worktable.

[0040] Step S24: Based on the two-layer cross-consensus comparison results, output the three types of standardized workstation deviation verification results in a hierarchical manner.

[0041] This step determines the deviation type by coupling the longitudinal difference result and the lateral consistency result, and classifies the results into three unique verification outputs: no deviation result, single workpiece local offset result, and worktable system offset result. This achieves accurate classification of deviation types and provides a standardized judgment basis for subsequent S3 deviation event identification.

[0042] The first type is the result without deviation: when the difference between the measured three-axis coordinates of all workstations and their original offset parameters is within the system's preset allowable deviation range, and the measured coordinates of all workstations are highly consistent with each other and the dispersion is lower than the threshold, then it is determined that the reference of all workstations on the current workbench is stable, and the result without deviation is output.

[0043] The second type is the result of local offset of a single workpiece: when the measured coordinates of a single station deviate from its original offset parameters and the three-axis difference exceeds the allowable range, while the measured coordinates of all other stations match the original parameters and the consistency between stations is good, the deviation exists only in an independent single station and is not affected by the overall working condition of the worktable, the result of local offset of a single workpiece is output. This type of deviation is mostly caused by micro-deformation of the fixture at a single station or loose clamping of a single workpiece.

[0044] The third category is the systematic offset result of the worktable: when multiple workstations deviate from the original offset parameters synchronously and in the same direction, and the offset magnitude and offset trend of each workstation are highly consistent, the data between workstations are well consistent but the overall offset occurs, and the output is the systematic offset result of the worktable. This type of deviation is mostly caused by global factors such as machine tool thermal deformation, overall stress drift of the worktable, and overall changes in ambient temperature.

[0045] Through the above-mentioned two-layer cross-consensus verification, single-point individual deviations and global system deviations can be accurately distinguished, solving the technical defects of traditional single comparisons that cannot identify the source of deviations, have low verification accuracy, and have a high misjudgment rate. This provides accurate and traceable original deviation data for subsequent time-series evolution analysis and risk classification correction.

[0046] Step S3: Based on the workstation deviation verification results, perform event identification. If a deviation event is identified, extract the workstation basic information, offset quantization level, and coordinate system correctable parameters from the workstation deviation verification results, integrate the above information into a deviation event, and store it. If there is no deviation event, proceed directly to the on-machine reference detection process of the next cycle. Specifically, step S3 includes steps S31-S34: Step S31: Continuously read the multi-cycle workstation deviation verification results to complete the standardized deviation event identification and judgment.

[0047] This step continuously receives the results of each round of dual-layer cross-consensus verification output from step S2, and performs a comprehensive judgment on the workbench status of each cycle of detection. This invention uniformly classifies the two types of abnormal verification results output from step S2—single workpiece local offset results and workbench systemic offset results—as valid deviation events, and classifies the results without deviation as normal working conditions, thereby achieving a binary standardized distinction of working conditions.

[0048] Unlike traditional methods that only identify numerical deviations and cannot distinguish the causes of anomalies, this step, by binding the result types of dual-layer verification, can accurately distinguish individual anomalies caused by single-station clamping and fixture deformation, as well as system anomalies caused by machine tool thermal drift and overall stress changes on the worktable, providing a preliminary classification basis for subsequent differentiated data analysis.

[0049] For example: The system continuously reads the detection and verification data every 15 minutes. If the output of this round shows a local offset of a single workpiece at station 1 or a systematic offset in the same direction between station 1 and station 7, the deviation event identification process is triggered. If the output of this round shows no deviation results across the entire range, the current processing condition is determined to be stable.

[0050] Step S32: For the identified deviation events, analyze and extract multi-dimensional information to construct basic workstation information.

[0051] The basic information of the workstation is a structured dataset used to uniquely trace the occurrence of deviations, specifically including four core fields: event timestamp, abnormal workstation number, corresponding workpiece coordinate system, and reference calibration feature point range.

[0052] The event timestamp is used to record the detection cycle time corresponding to the deviation event, accurately locating the timing node where the deviation occurred; the abnormal workstation number is the unique identification code of the valid workstation of the workbench, used to lock the physical workstation where the abnormality occurred; the corresponding workpiece coordinate system is the workstation-specific coordinate system constructed and stored in step S1, used to bind the reference system corresponding to the deviation; the reference calibration feature point range is the set of workpiece reference feature acquisition points that triggered the deviation, used to locate the specific reference area where the offset occurred.

[0053] By combining multiple dimensions of fields, the timing of each deviation event can be traced, the workstation can be located, and the benchmark can be matched, thus eliminating the problems of ambiguous deviation data and lack of traceability.

[0054] Step S33: Divide the offset difference into multiple offset quantization levels and generate coordinate system correctable parameters.

[0055] This step uses the triaxial difference between the measured coordinates obtained in step S2 and the original offset parameters stored in step S1 as the basis for judgment. Combined with the system's preset allowable coordinate deviation threshold, it realizes the quantification of the degree of offset and the determination of its correctability.

[0056] The preset coordinate allowable deviation threshold is the maximum reference offset tolerance value preset according to the workpiece machining accuracy level. In this embodiment, the preset allowable deviation threshold is 0.01mm in the precision batch machining scenario.

[0057] Based on the magnitude of the three-axis offset difference, the offset level is divided into three quantization levels: slight offset, moderate offset, and severe offset. When the offset difference is less than or equal to 0.01mm, it is judged as a slight offset; When the offset difference is greater than 0.01mm and less than or equal to 0.03mm, it is judged as a moderate offset; When the offset difference is greater than 0.03mm, it is judged as a severe offset.

[0058] Simultaneously, the correctability is determined by combining the preset allowable coordinate deviations: slight and moderate offsets are within the adjustable range of the machine tool coordinate system's dynamic compensation and are determined to be correctable offsets; severe offsets exceed the machine tool's dynamic compensation adjustment limits and are determined to be uncorrectable offsets. Finally, the offset quantification level and the correctability determination results are integrated to generate correctable parameters for the structured coordinate system.

[0059] For example: If the X-axis offset difference of station 2 is 0.02mm and the Y-axis offset difference is 0.021mm during a certain cycle, and the difference is greater than 0.01mm and less than 0.03mm, it is judged as a moderately correctable offset; if the three-axis offset difference of a certain station reaches 0.04mm, it is judged as a severe uncorrectable offset.

[0060] Step S34: Bind and archive the multi-dimensional parameters to complete the structured storage of deviation events.

[0061] This step binds the basic station information extracted in step S32, the offset quantization level generated in step S33, and the coordinate system correctable parameters to the deviation events identified in this round, forming complete single-cycle structured deviation event data, and archives and stores it in the CNC system's time-series storage area.

[0062] Each deviation event data has an independent time series label, workstation label, offset level label, and correctable attribute label, ensuring a complete and well-organized data source for subsequent multi-cycle time series traversal, spatial distribution statistics, and evolution pattern analysis, and avoiding analysis failures caused by data loss or parameter mismatch.

[0063] Step S35: Execute cycle jump logic for unbiased operating conditions to achieve normalized and efficient detection loop.

[0064] The current round of workstation deviation verification results are read in real time. If no deviation event of local offset of a single workpiece or systematic offset of the worktable is identified in this round, that is, the reference offset of all workstations is within the preset allowable deviation range and the coordinate consistency between workstations is good, then it is determined that the reference state of the workpiece coordinate system of all workstations is stable and there is no need to generate and store deviation event data.

[0065] The system keeps the current coordinate system parameters unchanged, enters the standby timing state, waits for the preset detection cycle to end, and then directly jumps to execute the on-machine reference detection process in step S2 to enter the next round of reference verification cycle.

[0066] This logic can effectively avoid invalid data storage and redundant calculations under normal operating conditions, reduce system computational overhead, and ensure the efficiency and continuity of benchmark monitoring in batch processing.

[0067] S4. Retrieve all historical deviation events stored in step S3, conduct joint analysis of temporal evolution and workstation spatial distribution based on all historical deviation events, and extract the workpiece coordinate system offset evolution law; determine the workpiece coordinate system offset risk level of each workstation based on the evolution law, and adaptively correct the corresponding workpiece coordinate system offset parameters according to the risk level, update the NC machining toolpath in linkage, and issue automatic machining control commands to complete batch machining. Specifically, step S4 includes steps S41, S42, S43, and S44: Step S41: Retrieve historical deviation events from multiple cycles, conduct joint analysis of the temporal dimension and the workstation space dimension, and extract the workpiece coordinate system offset evolution law; This step retrieves the complete multi-cycle deviation event records archived and saved in step S3, and performs quantitative analysis from the time sequence dimension and workstation space dimension. All features are judged using fixed quantitative standards to ensure that the solution is reproducible and the technical content is fully disclosed.

[0068] Define basic parameters: Let the total number of continuous detection cycles be N, and the cycle number i = 1, 2, ..., N; the effective station number of the worktable be k = 1, 2, ..., K; the difference between the measured coordinates of the k-th station in the i-th cycle and the original offset triaxial coordinates is _____. , , Single-station overall offset E is used to uniformly characterize the overall size of the single-cycle reference offset; the system presets a reference allowable offset threshold. In this embodiment, .

[0069] In the time series dimension, the frequency of deviation exceeding the tolerance and the periodic change range of the deviation value for each workstation are statistically analyzed over multiple consecutive periods. The long-term deviation change trend of the workstation is extracted and divided into three types of time series evolution characteristics based on the frequency and the rate of deviation growth: Stable low-frequency offset: The frequency of workstation deviations is low, and the overall offset over multiple cycles does not increase significantly; Progressive acceleration of deviation: The workstation continuously exceeds the tolerance for multiple consecutive cycles, and the deviation amount in each cycle continues to increase; Sudden large deviation: The baseline state of the workstation is stable in the early stage, but a large value deviation far exceeding the threshold suddenly occurs in a single cycle, without a slow and gradual change process.

[0070] For example: In this embodiment, after 10 consecutive detection cycles, 8 cycles of station 3 showed out-of-tolerance, and the offset value continued to increase cycle by cycle. It was determined that the time sequence evolution characteristic of this station was a gradual acceleration offset.

[0071] In terms of workstation space, the number, distribution location, and clustering degree of all abnormal workstations were statistically analyzed. Spatial distribution clustering features were extracted and divided into three categories of standardized spatial distribution features: Random and scattered offset across the entire domain: Abnormal workstations are randomly scattered across the workbench, with no adjacent clustering pattern; Localized clustering offset at fixed workstations: The abnormality is stable and occurs in a single or a few adjacent workstations over a long period of time, while the baseline of the remaining workstations is normal. Full-table continuous diffusion offset: Most workstations on the worktable shift synchronously, with the offset direction and offset value of each workstation being nearly identical.

[0072] By matching and combining the temporal evolution characteristics of a single workstation with the spatial distribution characteristics of the workstation, a workpiece coordinate system offset evolution law specific to that workstation is generated, providing a complete quantitative basis for subsequent risk level determination.

[0073] Step S42: Based on the migration evolution pattern, perform a four-level gradient migration risk level determination; This step pre-sets four levels of offset risk judgment criteria from low to high within the system. Combined with the evolutionary features extracted in step S41 and the offset quantification level divided in step S3, the offset level classification criteria are fixed: slight offset 0 < E ≤ 0.01 mm, moderate offset 0.01 mm < E ≤ 0.03 mm, and severe offset E > 0.03 mm.

[0074] Level 1 Judgment Criteria: The workstation has only stable low-frequency offset and random scattered offset across the entire area, and the offset quantification level is slight, which is judged as Level 1; Level 2 Judgment Criteria: The workstation has stable low-frequency offset, and the fixed workstation has localized concentrated offset. The offset quantification level is slight or moderate, and it is judged as Level 2. Level 3 Judgment Criteria: If a workstation exhibits gradual acceleration of offset or localized clustering of offset at a fixed workstation, and the offset quantification level is moderate, it is judged as Level 3. Level 4 Judgment Criteria: A sudden and significant shift occurs at the workstation, or a continuous and widespread shift occurs across the entire worktable. The shift quantification level is severe or uncorrectable, and the shift is judged as Level 4.

[0075] For example: The single-cycle comprehensive offset of station 5 reaches 0.035mm, and all stations on the worktable offset synchronously in the same direction. At the same time, it belongs to sudden large offset and continuous spread offset of the entire worktable. Therefore, this station is judged to be a level 4 risk station.

[0076] Step S43: Adaptively adjust the workpiece coordinate system offset parameters according to the four-level risk level; Level 1 station: Maintains the original offset parameters of the three axes of the current workpiece coordinate system unchanged, and does not perform any parameter compensation or correction operations.

[0077] Level 2 station: The three-axis difference between the measured coordinates of the current cycle and the original offset is used as the compensation amount. The three-axis offset parameters are corrected by first-order static correction. The new offset after correction = the original standard offset + the measured deviation value of the corresponding axis, which directly offsets the reference offset generated in the current cycle.

[0078] For example: The measured deviation of the X-axis at station 2 is 0.02mm. After correction, the offset value of the X-axis workpiece coordinate system is equal to the original X-axis standard offset plus 0.02mm, thus completing the static deviation compensation.

[0079] Level 3 station: Combining the frequency change trend of the station's multi-cycle offset and the offset growth rate of each cycle, a second-order dynamic compensation offset is generated to complete the synchronous correction of the three-axis offset; at the same time, the subsequent on-machine reference detection cycle is shortened, the original cycle duration is multiplied by an adjustment coefficient of less than 1, the reference monitoring frequency is increased, and the continuous accumulation of offset is suppressed.

[0080] For example: The original detection cycle was 15 minutes. Based on the increase in the offset of the workstation, the adjustment coefficient was set to 0.8, and the detection cycle was shortened to 12 minutes after the update.

[0081] Level 4 station: Lock all parameters of the current workpiece coordinate system, pause the automatic machining process of the station and output equipment alarm prompts; wait for the staff to manually complete the tooling and datum re-inspection and correction, re-collect the station datum calibration feature points, refit and update the original offset parameters of the station coordinate system before machining can resume.

[0082] Step S44: Update the NC machining toolpath, issue automatic machining control commands, and complete batch machining. First, the three-axis offset parameters after all workstations have been corrected are synchronously written into the internal storage area of ​​the CNC system, overwriting the original historical coordinate system offset data in the system and uniformly updating the system's reference parameters.

[0083] Secondly, the pre-stored batch workpiece general NC machining main program is retrieved. Based on the updated multi-station coordinate system offset parameters, the tool's three-dimensional motion trajectory is reconstructed station by station. The actual machining trajectory coordinates of the tool = theoretical basic trajectory coordinates + station-corrected three-axis offset values, generating an NC machining toolpath adapted to the new coordinate system.

[0084] Finally, according to the preset workstation arrangement on the workbench, the corresponding toolpath machining control commands for each workstation are issued sequentially in time, driving the machine tool to perform cutting operations and complete the continuous batch automated machining of multiple workpieces.

[0085] For example, after station 2 completes the second-level and first-order compensation correction, the system automatically refits the tool advance and retraction and cutting trajectory of that station, without the need for manual program modification, and directly executes subsequent batch processing.

[0086] Figure 2This is a schematic diagram of an automatic machining control system based on a multi-workpiece coordinate system according to the present invention.

[0087] Specifically, an automatic machining control system based on a multi-workpiece coordinate system, used to implement the aforementioned automatic machining control method based on a multi-workpiece coordinate system, includes: The benchmark construction module is used to import a unified machining process model for batch workpieces, select effective machining stations on machine tools, fit and construct a global benchmark coordinate system, collect benchmark calibration feature points of each station and generate a standardized multi-workpiece coordinate system group, and bind and store the original offset parameters of each workpiece coordinate system.

[0088] The unified processing model for batch workpieces includes five core process parameters: workpiece dimensions, reference feature positions, workbench station layout, batch processing steps, and theoretical coordinate offsets. The global reference coordinate system uses the machine tool origin as the unique zero point and is obtained through global detection and fitting of a standard reference sphere, providing a unified traceability reference for all workstations on the entire workbench. The standardized multi-workpiece coordinate system group is a set of workstation-specific coordinate systems that are independently generated for each effective workstation and are all traceable to the global reference. The original offset parameters include the original offset values ​​of the X, Y, and Z axes of each workstation coordinate system relative to the global reference, the workstation number, the workpiece model, and the theoretical coordinates of the reference point, providing original reference data for subsequent two-layer cross-consensus verification.

[0089] The detection and verification module is used to periodically trigger on-machine reference detection according to a preset fixed detection cycle, and synchronously collect the real-time measured coordinates of the reference calibration feature points of all workstations. Through a two-layer cross consensus verification mechanism of longitudinal reference comparison and transverse workstation cross comparison, the module determines the type of workstation reference offset and outputs the standardized workstation deviation verification result.

[0090] The preset detection cycle is a timed trigger parameter preset according to the processing accuracy requirements, which adapts to the benchmark monitoring requirements of different precision processing conditions; synchronous acquisition is a multi-station parallel one-time detection and acquisition, ensuring that the measured coordinates of all stations belong to the same time sequence condition, avoiding verification errors caused by time sequence deviation; longitudinal benchmark comparison is a comparison of the difference between the real-time measured coordinates of a single station and its own original offset parameters, used to identify the benchmark drift problem of a single station; transverse station cross comparison is a comparison of the consistency of the measured coordinates of all stations, used to accurately distinguish between local anomalies of a single station and systemic anomalies of the worktable; finally, it stably outputs three types of standardized verification results: no deviation, local offset of a single workpiece, and systemic offset of the worktable, providing standardized data support for subsequent deviation event identification.

[0091] The event storage module is used to continuously read the deviation verification results of each round of workstations, identify two types of deviation events: local offset of a single workpiece and systematic offset of the worktable, extract and archive basic workstation information, offset quantization level, and coordinate system correctable parameters, and directly jump to the next cycle of benchmark detection process under the condition of no deviation.

[0092] The basic information of the workstation includes event timestamps, abnormal workstation numbers, corresponding workpiece coordinate systems, and reference calibration feature point ranges, enabling full-dimensional traceability of deviation events. The offset quantification level is divided into three levels: slight, moderate, and severe, based on the measured offset amount, using a fixed offset threshold as the basis for classification, thus achieving quantitative differentiation of the degree of offset. The coordinate system correctable parameters are used to accurately determine whether the current deviation is correctable or uncorrectable, adapting to subsequent graded correction logic. The module can achieve complete structured archiving of abnormal data and non-redundant storage under normal working conditions, effectively reducing system computational overhead and providing continuous and complete time-series data support for subsequent multi-dimensional offset evolution analysis.

[0093] The evolution correction module is used to retrieve historical deviation events from multiple cycles, jointly analyze and extract the coordinate system offset evolution law from the temporal and spatial dimensions, determine the four-level offset risk level, adaptively and grade the workpiece coordinate system offset parameters according to the risk level, dynamically update the NC machining toolpath, and issue automatic machining control commands to complete multi-station batch automatic machining.

[0094] Among them, the temporal dimension can extract three types of evolution features: stable low-frequency offset, gradual acceleration offset, and sudden large offset, accurately representing the temporal change trend of the workstation reference offset; the spatial dimension can extract three types of distribution features: global random scattered offset, fixed workstation local clustered offset, and full-worktable continuous diffusion offset, effectively distinguishing the spatial causes of the deviation; the four-level risk level is divided into level one, level two, level three, and level four from low to high, corresponding to four graded correction strategies: no correction required, first-order static compensation, second-order dynamic compensation + periodic encryption, and lock alarm re-inspection, respectively, to achieve accurate matching between offset risk and correction method; the module can reconstruct the tool motion trajectory based on the corrected coordinate system parameters, automatically update the NC machining toolpath adapted to the reference parameters, and issue machining commands in a time-sharing manner, realizing the closed-loop control of the entire process of reference deviation identification, risk assessment, adaptive correction, and automatic machining, ensuring the accuracy and stability of multi-workstation batch machining.

[0095] It should be understood that the term "and / or" in this article is merely a description of the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent three cases: A alone, A and B simultaneously, and B alone. A and B can be singular or plural. Additionally, the character " / " in this article generally indicates an "or" relationship between the preceding and following related objects, but it can also represent an "and / or" relationship. Please refer to the context for a more accurate understanding.

[0096] It should be understood that, in the embodiments of the present invention, the order of the above-mentioned process numbers does not imply the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present invention.

[0097] The above-described embodiments are only used to illustrate the technical solutions of the present invention, and are not intended to limit it. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention, and should all be included within the protection scope of the present invention.

Claims

1. An automatic machining control method based on a multi-workpiece coordinate system, characterized in that, Includes the following steps: S1. Import the unified processing technology model of batch workpieces, construct a global reference coordinate system, collect the reference calibration feature points of workpieces at multiple workstations on the workbench, construct a standardized multi-workpiece coordinate system group based on the global reference coordinate system and calibration feature points, and store the original offset parameters of each workpiece coordinate system. S2. Start the on-machine reference detection process according to the preset cycle, and simultaneously collect the current reference calibration feature points of all workpieces at all workstations, and collect the measured coordinates of the features at each point. A two-layer cross-consensus verification is performed based on the original offset parameters of the global reference coordinate system and each workpiece coordinate system to obtain the workstation deviation verification result. S3. Based on the workstation deviation verification results, perform event identification. If a deviation event is identified, extract the workstation basic information, offset quantization level, and coordinate system correctable parameters from the workstation deviation verification results, integrate the above information into a deviation event, and store it. If there are no deviation events, proceed directly to the next cycle of in-machine reference detection process; S4. Retrieve all historical deviation events stored in step S3, conduct joint analysis of temporal evolution and workstation spatial distribution based on all historical deviation events, and extract the workpiece coordinate system offset evolution law; determine the workpiece coordinate system offset risk level of each workstation based on the evolution law, and adaptively correct the corresponding workpiece coordinate system offset parameters according to the risk level, update the NC machining toolpath in linkage, and issue automatic machining control commands to complete batch machining.

2. The automatic machining control method based on a multi-workpiece coordinate system according to claim 1, step S1 specifically includes: Load a unified machining process model for batch workpieces, including workpiece shape, datum features, station layout, machining process, and theoretical coordinate offset, and filter all effective machining stations on the worktable. The machine tool standard reference ball is used to perform full-domain on-machine detection. A global reference coordinate system is generated by fitting the machine tool mechanical origin, and the three-axis theoretical zero point and the reference calibration feature theoretical coordinate of the global reference coordinate system are locked. By synchronously detecting the preset reference calibration feature points of the workpiece at each station using the machine tool contact probe, the measured reference coordinates are mapped to the global reference coordinate system, and a dedicated workpiece coordinate system for each station is generated one by one, which are then summarized to form a standardized multi-workpiece coordinate system group. The original three-axis offset values ​​of each workpiece coordinate system relative to the global reference coordinate system, the workstation number, the workpiece model, and the reference point coordinates are bound and stored as the original reference data for subsequent two-layer cross-consensus verification.

3. The automatic machining control method based on a multi-workpiece coordinate system according to claim 1, characterized in that, Step S2 specifically includes: A fixed detection cycle is pre-configured. After the cycle node is reached, the machine tool probe is controlled to collect the current reference calibration feature points of all workpieces at all workstations in parallel and synchronously to obtain the real-time measured reference coordinates of each workstation. The two-layer cross-consensus verification is divided into two levels: vertical benchmark comparison and horizontal workstation cross-comparison. Longitudinal benchmark comparison involves comparing the difference between the real-time measured coordinates of a single workstation and the original offset parameters of the workpiece coordinate system at that workstation. Cross-comparison of horizontal workstations involves comparing the actual measured coordinates of all workstations under the same workbench to ensure consistency. Based on the results of the two-layer cross-consensus comparison, three types of verification outputs are divided: the measured coordinates of all workstations match the original offset and are consistent with each other between workstations, and the output is a result without deviation. The measured coordinates of a single workstation deviate from their original offset, while the parameters of other workstations are unified, and the output is the local offset result of a single workpiece. Multiple workstations synchronously deviate from the original offset in the same direction, and the offset trends between workstations are consistent, outputting the systematic offset results of the worktable.

4. The automatic machining control method based on a multi-workpiece coordinate system according to claim 1, characterized in that, Step S3 involves event identification of the workstation deviation verification results. If a deviation event is identified, the basic workstation information, offset quantization level, and coordinate system correctable parameters corresponding to the deviation event are collected and stored. Specifically, this includes: Continuously read the workstation deviation verification results of each round, and uniformly classify the local offset of a single workpiece and the systematic offset of the worktable as deviation events; For each identified deviation event, the event occurrence timestamp, abnormal workstation number, corresponding workpiece coordinate system, and reference calibration feature point range are extracted and combined to form the basic information of the workstation. Based on the difference between the measured coordinates and the original offset parameters, the offset is classified into three levels: slight, moderate and severe. Combined with the preset allowable coordinate deviation, it is determined whether the current deviation is a correctable offset or an uncorrectable offset, and the correctable parameters of the coordinate system are generated. The basic information of the workstation, the offset quantization level, the coordinate system correctable parameters, and the corresponding deviation events are bound and archived one by one.

5. An automatic machining control method based on a multi-workpiece coordinate system according to claim 4, characterized in that, If no deviation event occurs in step S3, proceed directly to the next cycle of in-machine reference detection process, which specifically includes: The system reads the deviation verification results of the current workstation in real time. If no deviation event of any kind, such as local offset of a single workpiece or systematic offset of the worktable, is identified, it is determined that the coordinate system reference of all workpieces at the current workstation is stable and no deviation event storage data is generated. After the preset detection cycle timer ends, it directly jumps to execute the on-machine reference detection process of S2.

6. The automatic machining control method based on a multi-workpiece coordinate system according to claim 4, characterized in that, Step S4 involves conducting a joint analysis of the temporal evolution and spatial distribution of historical deviation events and extracting the workpiece coordinate system offset evolution law. Specifically, this includes: Retrieve all deviation event records from archived storage across multiple consecutive periods, conduct joint analysis from the time-series dimension and the workstation space dimension, and extract corresponding features; The frequency variation trend of each workstation's shift over a period of time was extracted in the time-series dimension and divided into three types of time-series evolution characteristics: stable low-frequency shift, gradual acceleration shift, and sudden large shift. In the spatial dimension of the workstation, the distribution clustering characteristics of the deviation workstations are extracted and divided into three types of spatial distribution characteristics: global random scattered offset, fixed workstation local clustered offset, and whole workbench continuous diffusion offset. By combining the temporal evolution characteristics with the spatial distribution characteristics of the workstation, the offset evolution law of the workpiece coordinate system corresponding to the workstation is generated.

7. The automatic machining control method based on a multi-workpiece coordinate system according to claim 6, characterized in that, Step S4 determines the risk level of workpiece coordinate system offset at each workstation based on the evolution law, specifically including: The pre-defined four-level offset risk assessment criteria are as follows: Level 1, Level 2, Level 3, and Level 4, from lowest to highest. If a workstation only has stable low-frequency offsets, random scattered offsets across the entire domain, and the offset quantification level is slight, it is judged as Level 1; If there is a stable low-frequency offset at the workstation or a localized clustered offset at a fixed workstation, and the offset quantification level is slight or moderate, it is judged as level two. If a workstation exhibits progressively accelerating offset or localized concentrated offset at a fixed workstation, and the offset quantification level is moderate, it is judged as level three. If a workstation experiences a sudden and significant shift, or if the entire worktable experiences a continuous and widespread shift, and the shift quantification level is classified as severe or uncorrectable, it is determined to be Level 4.

8. The automatic machining control method based on a multi-workpiece coordinate system according to claim 7, characterized in that, Step S4 adaptively adjusts the corresponding workpiece coordinate system offset parameters according to the risk level, specifically including: Level 1 station: Maintains the original offset parameters of the three axes of the current workpiece coordinate system unchanged, and does not perform parameter correction operations; Secondary station: Generates first-order compensation offset based on the difference between the measured coordinates and the original offset parameters, and corrects the three-axis offset parameters of the workpiece coordinate system; Level 3 station: Combines the frequency change trend of each station's offset over a cycle to generate a second-order dynamic compensation offset, synchronously corrects the three-axis offset parameters of the workpiece coordinate system, and shortens the subsequent on-machine reference detection process. Level 4 station: Lock the current workpiece coordinate system, pause the processing at this station and output an alarm prompt. After manual re-inspection of the datum, re-sample the datum to calibrate the feature points and update the original offset parameters of the coordinate system.

9. The automatic machining control method based on a multi-workpiece coordinate system according to claim 8, characterized in that, Step S4, which involves updating the NC machining toolpath, issuing automatic machining control commands, and completing batch machining, specifically includes: The corrected coordinate system three-axis offset parameters of each workpiece are synchronously written into the CNC system storage area, overwriting the original coordinate system offset data. Retrieve the pre-stored NC machining main program for batch workpieces, and reconstruct the tool motion trajectory station by station based on the updated multi-workpiece coordinate system parameters to generate the NC machining toolpath adapted to the corrected coordinate system. Based on the workbench station layout, the corresponding toolpath machining control commands for each station are issued in time intervals to drive the machine tool to perform machining actions and complete the batch automatic machining of multiple workpieces.

10. An automatic machining control system based on a multi-workpiece coordinate system, used to implement the method described in any one of claims 1 to 9, characterized in that, include: The datum construction module is used to import the machining process model, generate a global datum coordinate system and a standardized multi-workpiece coordinate system group, and store the original offset parameters of each workpiece coordinate system. The detection and verification module, along with the periodic detection and consensus verification module, is used to periodically collect the measured coordinates of the workstation benchmark calibration feature points, perform two-layer cross-consensus verification, and output the workstation deviation verification results. The event storage module and the deviation event acquisition and storage module are used to identify, archive and store various deviation event information. When there is no deviation, it directly jumps to the next cycle of benchmark detection. The evolution correction module, offset evolution analysis and adaptive correction module are used to analyze historical deviation events to obtain offset evolution rules, determine the offset risk level and correct the coordinate system offset in stages, update the NC machining toolpath, and issue machining commands to complete batch machining.

Images