Positioning accuracy control system and method for CNC mold machining

Abstract

The application discloses a positioning precision control system for CNC die machining and belongs to the technical field of numerical control machine tools. The system comprises a configuration module, a data acquisition module and a timing control module. The configuration module constructs a multi-dimensional motion space based on a motion instruction set and a spatial position coordinate set, and extracts action characteristic parameters; the data acquisition module synchronously acquires die working state parameters; the timing control module establishes a mapping relationship between the action characteristic parameters and the working state parameters, and limits to obtain an action state coupling matrix. The system calculates and predicts adjustment parameters in the multi-dimensional motion space and generates an offset vector, and if the space-time boundary constraint is exceeded, a correction offset vector is generated, and predictive position compensation is performed. Through multi-dimensional motion space mapping and dynamic compensation, the positioning error caused by thermal deformation is reduced, and the machining precision and stability are improved.

Description

Technical Field

[0001] This invention relates to the field of numerical control technology, specifically to a graphic calibration system and method for CNC mold processing. Background Technology

[0002] With the increasing demand for precision molds in modern manufacturing, CNC machine tools are being used more and more widely in the mold processing field, and their positioning accuracy has become a key factor affecting the quality of mold processing. During CNC mold processing, thermal deformation of the machine tool and the mold is one of the main causes of positioning errors. As heat accumulates during processing, the machine tool structure and the mold itself undergo thermal expansion, leading to deviations between the actual processing position and the theoretical design position, severely affecting the dimensional accuracy and surface quality of the mold.

[0003] Currently, the temperature compensation methods commonly used in the industry have the following limitations: First, traditional compensation systems mostly use fixed temperature deformation models, which cannot dynamically adjust compensation parameters according to temperature changes during actual processing. Second, existing technologies lack a method to construct a multi-dimensional motion space based on motion command sets and spatial position coordinate sets, resulting in an inability to fully quantify the impact of motion characteristic parameters on the processing process, leading to significant differences in compensation effects under different operating conditions.

[0004] Secondly, existing technologies often only focus on the thermal deformation of the machine tool's main structure, neglecting the direct impact of the mold's own temperature changes on positioning accuracy. Most systems lack an effective motion-state coupling mechanism, failing to establish a precise mapping relationship between motion characteristic parameters and working state parameters, resulting in deviations between predicted compensation and actual needs.

[0005] Furthermore, existing CNC positioning control systems also have significant shortcomings in data processing. On the one hand, the systems typically use all historical data for model training, leading to inaccurate data from the early stages interfering with current predictions, especially during system startup; there is also a lack of an effective sliding window mechanism to limit the timeliness of the training dataset. On the other hand, single measurement errors can easily cause drastic fluctuations in compensation parameters, affecting system stability. Simultaneously, existing technologies lack adaptive adjustment mechanisms for different rates of temperature change, failing to respond promptly to rapid temperature changes or maintain parameter stability during stable system operation, and lacking the ability to dynamically adjust the smoothing coefficient.

[0006] In summary, existing technologies urgently need a CNC mold machining positioning accuracy control system capable of constructing a multi-dimensional motion space, dynamically updating model parameters, and possessing spatiotemporal boundary constraints and precise timing control capabilities to solve the aforementioned problems. Summary of the Invention

[0007] To address the above problems, this invention provides a positioning accuracy control system and method for CNC mold processing.

[0008] A first aspect of the present invention provides a positioning accuracy control system for CNC mold machining, comprising: The configuration module is configured to construct multiple multi-dimensional motion spaces based on the motion instruction set and spatial position coordinate set of the CNC machine tool, and extract motion feature parameters of the multi-dimensional motion spaces, including velocity vector magnitude distribution, acceleration rate of change, and trajectory curvature density. The data acquisition module is configured to synchronously acquire the working state parameters of the mold during the processing, including thermal data, position deviation signals, and processing cycle timestamp sequences, and limit the multi-dimensional state perception information. The timing control module is communicatively connected to the configuration module and the data acquisition module, and is configured to receive and preprocess the data uploaded by each module, establish a mapping relationship between motion feature parameters and working state parameters, reconstruct the first trajectory sequence of each axis of the CNC machine tool from the start time of the first processing cycle based on historical processing cycles, and send compensation control signals to each servo mechanism before the start of the second processing cycle according to the adjustment parameters output by the prediction model to perform predictive position compensation, wherein the predictive position compensation limits the starting compensation position of the motion trajectory of each axis.

[0009] As a preferred embodiment, the working state parameters include thermal data, position deviation signals, and processing cycle timestamp sequences. The thermal data includes first thermal data of the pre-defined heat dissipation area of ​​the mold and second thermal data of the heat concentration area obtained from sampling. The position deviation signal includes a deviation value containing XYZ three-axis components generated by measuring the spatial vector deviation between the actual position and the target position of the mold, used to calibrate the position coordinate set of the multi-dimensional motion space. The processing cycle timestamp sequence includes the timestamp of a single cycle recorded in response to the processing start pulse and completion pulse issued by the CNC control system, providing reference cycle data for the timing control module. Furthermore, the first thermal data, second thermal data, position deviation signals, and processing cycle timestamp sequences are weighted and fused to determine a comprehensive evaluation index of the mold's working state.

[0010] As a preferred approach, the configuration module constructs the multidimensional motion space, specifically including the following steps: defining the time axis, spatial coordinate axis, and command logic axis as the base dimensions of the multidimensional motion space; mapping the velocity vector magnitude distribution, acceleration rate of change, and trajectory curvature density to the base dimensions to generate spatial feature vectors; calculating the correlation coefficient between the spatial feature vectors and the working state parameters to construct a parameter mapping relationship, wherein the parameter mapping relationship defines the motion state coupling matrix, which is used to quantify the influence weight of the motion feature parameters on the mold processing parameters.

[0011] As a preferred approach, the timing control module performs predictive adjustment, specifically including the following steps: establishing a predictive model, associating and mapping the spatial feature vector with the thermal data within the multidimensional motion space, calculating predictive adjustment parameters, and the predictive model defining the thermal deformation mapping of the motion space; when the thermal data is higher than a first preset threshold, generating an offset vector on the spatial coordinate axis of the multidimensional motion space according to the predictive adjustment parameters, so that the CNC machine tool reaches the compensation position before the start of the second machining cycle; the compensation position is the starting point of the motion trajectory of reducing the control time delay and thermal deformation compensation of each axis, represented by the set of control signal values ​​of each servo mechanism at the predicted timestamp, and the limiting condition of the compensation position is that the position completion timestamp plus the control delay of the servo mechanism is less than or equal to the start timestamp of the next action.

[0012] As a preferred approach, when the control signal value of the first servo mechanism reaches the value representing the target position in the first trajectory sequence, the timing control module dynamically adjusts the control signal value of the servo mechanism to the target value set based on the synchronization constraints of the time axis and spatial coordinate axis of the multi-dimensional motion space. The target value set limits the position time constraint interval, ensuring that the position completion timestamp plus the control delay of the servo mechanism is less than or equal to the start timestamp of the next action.

[0013] As a preferred approach, after each processing cycle is completed, the timing control module adds the newly acquired multidimensional motion space trajectory samples and actual position deviation data to the training dataset, updates the parameters of the motion state coupling matrix through a recursive linear regression algorithm, and uses a sliding window mechanism to limit the training dataset to retain only the multidimensional motion space segments corresponding to the most recent preset number of processing cycles. When the amount of data in the window reaches the preset number, the earliest acquired multidimensional motion space segment is automatically removed.

[0014] In a preferred embodiment, the instruction logic axis is configured to represent the logical sequence of the machining process, mapping the roughing stage, semi-finishing stage, and finishing stage to a first logic interval, a second logic interval, and a third logic interval on the instruction logic axis, respectively. Within the multi-dimensional motion space, different dimension weight coefficients are assigned to the first logic interval, the second logic interval, and the third logic interval. These dimension weight coefficients are used to adjust the contribution of the spatial feature vector to the motion state coupling matrix at different machining stages.

[0015] As a preferred embodiment, the multidimensional motion space is further defined by spatiotemporal boundary constraints, which are jointly defined by the maximum allowable span of the time axis and the maximum allowable displacement of the spatial coordinate axis. When the offset vector generated by the timing control module exceeds the spatiotemporal boundary constraints, the offset vector is projected onto the boundary surface of the spatiotemporal boundary constraints to generate a corrected offset vector. The corrected offset vector defines a safe compensation path to ensure that the motion trajectory of the CNC machine tool is always within the reachable area of ​​the multidimensional motion space.

[0016] A second aspect of the present invention provides a method for controlling the positioning accuracy of CNC mold processing, comprising the following steps: Multiple multi-dimensional motion spaces are constructed based on the motion instruction set and spatial position coordinate set of a CNC machine tool. The time axis, spatial coordinate axis, and instruction logic axis are defined as the base dimensions of these multi-dimensional motion spaces, and motion feature parameters are extracted. Simultaneously, the working state parameters of the mold during processing are acquired, including thermal data, position deviation signals, and processing cycle timestamp sequences, thus defining multi-dimensional state perception information. The motion feature parameters are mapped to the base dimensions to generate spatial feature vectors, and the correlation coefficient between the spatial feature vectors and the working state parameters is calculated to construct a parameter mapping relationship, defining the motion state coupling matrix. Within the multi-dimensional motion space, the spatial feature vectors are associated and mapped with the thermal data to calculate predictive adjustment parameters. When the thermal data exceeds a first preset threshold, an offset vector is generated on the spatial coordinate axis of the multidimensional motion space according to the prediction adjustment parameters. If the offset vector exceeds the spatiotemporal boundary constraint of the multidimensional motion space, the offset vector is projected onto the boundary surface of the spatiotemporal boundary constraint to generate a corrected offset vector, and the compensation position is determined based on the offset vector or the corrected offset vector. The compensated control signal is sent to each servo mechanism, and based on the synchronization constraint of the time axis and spatial coordinate axis of the multidimensional motion space, it is ensured that the position completion timestamp plus the control delay is less than or equal to the start timestamp of the next action. After each processing cycle is completed, the newly collected multidimensional motion space trajectory samples are added to the training dataset, and the parameters of the action state coupling matrix are updated to optimize the prediction accuracy of subsequent cycles.

[0017] Compared with the prior art, the present invention has the following advantages: This system constructs a multi-dimensional motion space and establishes a mapping relationship between motion characteristic parameters and working state parameters. It can calculate and predict adjustment parameters based on real-time temperature data, thus compensating the servo mechanism in advance. Compared with traditional fixed compensation models, the dynamic compensation mechanism of this invention significantly improves the quality of mold processing.

[0018] A sliding window mechanism and linear regression algorithm are used to dynamically update the prediction model parameters, enabling the system to continuously optimize the compensation strategy based on the latest processed data. This mechanism avoids interference from historical data with current predictions, especially in the early stages of system operation, by retaining only the most recently preset number of processed cycle data, ensuring the timeliness and accuracy of the prediction model.

[0019] By setting spatiotemporal boundary constraints and minimum / maximum adjustment thresholds, the system can ensure the necessary compensation effect while preventing system instability caused by overcompensation. When the compensation vector exceeds the constraints, the system automatically generates a safe compensation path, effectively maintaining the stability of the system and avoiding the "overcompensation-undercompensation" oscillation phenomenon.

[0020] A weighted moving average method is used to smooth the predicted adjustment parameters, significantly reducing the impact of single measurement errors on the system. The system can dynamically adjust the smoothing coefficient according to the operating status: it responds quickly to temperature changes in the initial stage of startup and improves parameter stability after stabilization, achieving an optimal balance between response speed and stability.

[0021] By comprehensively considering the control delay of each servo mechanism, the system can ensure the precise matching of the position completion time and the start time of the next action based on the synchronization constraints of the time axis and the spatial coordinate axis. This effectively avoids positioning errors caused by timing mismatch and improves processing efficiency and accuracy. Attached Figure Description

[0022] The present invention will be further described with reference to the accompanying drawings, but the embodiments in the drawings do not constitute any limitation on the present invention. For those skilled in the art, other drawings can be obtained based on the following drawings without creative effort.

[0023] Figure 1 This is a schematic diagram of the system provided in an embodiment of the present invention. Detailed Implementation

[0024] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0025] This invention provides a positioning accuracy control system for CNC mold machining, such as... Figure 1As shown, the system is set up in a CNC machine tool environment with preset machining parameters. The CNC machine tool environment can be a specific machining center or a CNC device containing multi-axis linkage mechanisms. The preset machining parameters include the number of machine tool motion axes, maximum feed rate, acceleration limits, and mold material properties. These data are used to define the foundation and safety boundaries of the multi-dimensional motion space. The system mainly includes a configuration module, a data acquisition module, and a timing control module.

[0026] The configuration module is configured to construct multiple multi-dimensional motion spaces based on the motion instruction set and spatial coordinate set of the CNC machine tool. Specifically, the multi-dimensional motion space is an abstract mathematical model used to comprehensively represent the motion state of the CNC machine tool during machining. It is not a physically existing space, but rather maps the physical motion, time process, and technological logic of the machine tool into a unified data space, allowing the system to simultaneously consider the combined impact of position, speed, and technological stages on machining accuracy. The system first reads the CNC program of the CNC machine tool, extracts the motion instructions and coordinate data, and then establishes multiple coordinate axes as the framework of this space. The collected real-time motion data is then filled into this framework. The motion state at each machining moment corresponds to a point in this space, and the entire machining process forms a trajectory in this space. The configuration module defines the time axis, spatial coordinate axis, and instruction logic axis as the base dimensions of the multi-dimensional motion space. The time axis represents the passage of time during the machining process, obtained by reading the timestamp signal from the system's internal clock or controller, recording the moment each action occurs. The spatial coordinate axis represents the actual physical position of each axis of the machine tool, obtained by reading feedback data from the machine tool's servo encoder or position commands from the CNC system, recording the physical coordinate value at each moment. The instruction logic axis represents the logical sequence of the machining process, obtained by parsing code segments in the CNC program, such as identifying which code segments belong to roughing and which belong to finishing, and mapping them to different intervals on the logic axis. The configuration module extracts motion feature parameters from the multi-dimensional motion space, including velocity vector magnitude distribution, acceleration rate of change, and trajectory curvature density. Regarding the acquisition process of these parameters, the velocity vector magnitude distribution is obtained by calculating the change in position per unit time, and then statistically analyzing the distribution of multiple velocity values ​​over a period of time to observe whether the velocity is constant or fluctuating; the acceleration change rate is calculated by comparing the acceleration values ​​in adjacent time periods to determine the magnitude of the change; if the acceleration value changes frequently and with a large magnitude, the parameter value is higher; the trajectory curvature density is calculated by analyzing the shape of the path formed by consecutive position points to determine the degree of path curvature, and the number of curved points within a unit length of path is counted; the greater the number, the higher the density. These parameters together constitute the spatial feature vector describing the motion state of the machine tool.

[0027] To improve the model's adaptability to different processing stages, the instruction logic axis is configured to represent the logical sequence of the processing flow, mapping the roughing, semi-finishing, and finishing stages to the first, second, and third logical intervals on the instruction logic axis, respectively. Within the multi-dimensional motion space, different dimensional weight coefficients are assigned to the first, second, and third logical intervals. For example, the finishing stage requires higher positional accuracy, so its spatial coordinate axis weight coefficient can be higher than that of the roughing stage; the roughing stage requires higher efficiency, so its time axis weight coefficient can be relatively higher. The dimensional weight coefficients are used to adjust the contribution of spatial feature vectors to the motion-state coupling matrix at different processing stages. Specifically, the calculation method involves multiplying the feature vectors within each logical interval by their corresponding weight coefficients and then concatenating or weighted summing them to generate staged spatial feature vectors. This allows the system to focus more on the motion characteristics of key processing stages, making the generated compensation strategy more targeted.

[0028] The data acquisition module is configured to synchronously acquire the working status parameters of the mold during the processing. These parameters include thermal data, position deviation signals, and processing cycle timestamp sequences, thus defining multi-dimensional state perception information. The working status parameters are indicators reflecting the actual physical state of the mold and the processing environment, directly affecting the final processing accuracy. The thermal data includes first thermal data from the mold's preset heat dissipation area and second thermal data from the heat concentration area, obtained by directly reading temperature values ​​from temperature sensors installed in different areas of the mold (such as the heat dissipation area and the heat concentration area). The first thermal data reflects the overall heat dissipation of the mold, while the second thermal data reflects the temperature of local hot spots. The position deviation signal includes a deviation value containing XYZ three-axis components generated by measuring the spatial vector deviation between the actual position and the target position of the mold. This deviation value is used to calibrate the position coordinate set in the multi-dimensional motion space. It can be obtained through non-contact measurement using a high-precision laser interferometer or a grating ruler, by comparing the actual position with the target position issued by the CNC system and calculating the difference. The machining cycle timestamp sequence includes the timestamp of a single cycle recorded in response to the machining start pulse and completion pulse issued by the CNC control system. This provides reference cycle data for the timing control module, obtained by monitoring the start and end signals issued by the CNC control system and recording the specific time point of each signal occurrence. Furthermore, the first thermal data, second thermal data, position deviation signal, and machining cycle timestamp sequence are weighted and fused to determine a comprehensive evaluation index of the mold's working state. The weighted fusion process can employ principal component analysis or expert experience weighting to ensure that each parameter contributes reasonably to the comprehensive evaluation index.

[0029] The timing control module is connected to the configuration module and the data acquisition module. It is configured to receive and preprocess data uploaded by each module, establishing a mapping relationship between motion characteristic parameters and working state parameters. Specifically, the timing control module calculates the correlation coefficient between spatial feature vectors and working state parameters, constructing a parameter mapping relationship. This parameter mapping relationship defines the motion state coupling matrix. The motion state coupling matrix is ​​a data structure used to store the correlation between motion characteristic parameters and working state parameters. It quantifies the influence weight of the machine tool's motion mode (e.g., high speed, sharp turns) on the mold state (e.g., heating, deformation). The acquisition process is as follows: the system collects motion characteristic parameter and corresponding working state parameter data from historical processing cycles. Through statistical analysis, it seeks the changing patterns between the two, such as analyzing whether thermal data increases with increasing trajectory curvature density. These analyzed correlation strength values ​​are then filled into the matrix to form the coupling matrix. As new data is added, the values ​​in the matrix are continuously adjusted to match the latest patterns. This matrix can reflect, for example, the correlation strength between "high acceleration change rate" and "local heat concentration."

[0030] The timing control module performs predictive adjustment, specifically including the following steps: establishing a predictive model, associating spatial feature vectors with thermal data in a multi-dimensional motion space, calculating predictive adjustment parameters, and using the predictive model to define the thermal deformation mapping in the motion space. When the thermal data exceeds a first preset threshold, an offset vector is generated on the spatial coordinate axes of the multi-dimensional motion space based on the predictive adjustment parameters. The offset vector is a theoretical compensation amount calculated by the system based on the predictive model to compensate for thermal deformation and time delay, including direction and magnitude. It is obtained by the predictive model outputting a suggested position adjustment amount based on the current thermal data and motion characteristics. This ensures that the CNC machine tool reaches the compensation position before the start of the second machining cycle. The compensation position is the starting point of the motion trajectory where the sum of the reduction in control time delay and thermal deformation compensation for each axis is obtained, represented by the set of control signal values ​​of each servo mechanism at the predicted timestamp. The condition for the compensation position is that the position completion timestamp plus the control delay of the servo mechanism is less than or equal to the start timestamp of the next action.

[0031] To ensure machining safety, the multi-dimensional motion space is also constrained by spatiotemporal boundary constraints. These constraints are safety limitations set by the system to ensure that compensation operations do not exceed the machine tool's physical capabilities or safety limits. They include the maximum permissible span in time and the maximum permissible displacement in space. The acquisition process is as follows: the system reads the machine tool's technical specifications, such as the maximum travel range and the maximum permissible machining cycle time, and sets these physical limits as boundaries. When the offset vector generated by the timing control module exceeds the spatiotemporal boundary constraints, the offset vector is projected onto the boundary surface of the spatiotemporal boundary constraints to generate a corrected offset vector. The corrected offset vector is the adjusted safety compensation amount when the theoretical compensation amount exceeds the safety range. Its acquisition process is as follows: the system first determines whether the endpoint of the offset vector exceeds the range defined by the spatiotemporal boundary constraints. If it does, the system forcibly moves the endpoint of the vector to the boundary surface and recalculates the vector from the starting point to that point on the boundary surface as the corrected offset vector. The corrected offset vector limits the safe compensation path, ensuring that the CNC machine tool's motion trajectory always remains within the reachable area of ​​the multi-dimensional motion space. This mechanism prevents the risk of machine tool overtravel or collision due to overcompensation.

[0032] Regarding timing synchronization control, when the control signal value of the first servo mechanism reaches the value representing the target position in the first trajectory sequence, the timing control module dynamically adjusts the control signal value of the servo mechanism to the target value set based on the synchronization constraints of the time axis and spatial coordinate axis of the multi-dimensional motion space. The target value set limits the position-time constraint interval, which is the time range that ensures the correct timing of the machine tool's motion, guaranteeing that one action is completed before the next action begins, avoiding conflicts. The acquisition process is as follows: the system records the expected completion time of the current action and adds the inherent response delay time of the servo mechanism to obtain a latest completion time. The system compares this time with the planned start time of the next action; the time difference between the two is the constraint interval. The system adjusts the control signal to ensure that the actual completion time falls within this interval, thereby eliminating positioning deviations caused by timing conflicts in the multi-dimensional motion space.

[0033] To achieve system self-evolution, after each processing cycle, the timing control module adds newly acquired multidimensional motion space trajectory samples and actual position deviation data to the training dataset, and updates the parameters of the action-state coupling matrix using a recursive linear regression algorithm. The multidimensional motion space trajectory samples include the spatial feature vector sequence within that cycle. In the early stages of system operation, a sliding window mechanism is used to limit the training dataset to retain only the multidimensional motion space segments corresponding to the most recent preset number of processing cycles, avoiding interference from historical data in current predictions. The sliding window mechanism is a data management strategy used to ensure that model training uses only the freshest and most relevant data, preventing outdated data from interfering with current predictions. The acquisition process is as follows: the system sets a fixed-capacity data storage area (window). Each time a new processing cycle is completed, the system puts newly generated data into the window. If the amount of data in the window reaches a preset upper limit, the system automatically removes the earliest stored data set to make room for new data. This ensures that the window always retains data from the most recent few processing cycles. Furthermore, when the amount of data in the window reaches a preset number, the earliest acquired multidimensional motion space segment is automatically removed. The preset number can be set according to processing stability, for example, the most recent 50 processing cycles. This mechanism ensures the timeliness and accuracy of the prediction model, enabling the system to maintain good adaptability under different operating conditions.

[0034] This invention also provides a positioning accuracy control method for CNC mold machining, applied to a CNC machine tool environment with preset machining parameters. The method includes the following steps: constructing multiple multi-dimensional motion spaces based on the motion command set and spatial position coordinate set of the CNC machine tool; defining the time axis, spatial coordinate axis, and command logic axis as the base dimensions of the multi-dimensional motion spaces; extracting motion feature parameters of the multi-dimensional motion spaces; synchronously acquiring the working state parameters of the mold during machining, including thermal data, position deviation signals, and machining cycle timestamp sequences, thus defining multi-dimensional state perception information; mapping the motion feature parameters to the base dimensions to generate spatial feature vectors, and calculating the correlation coefficient between the spatial feature vectors and the working state parameters to construct a parameter mapping relationship, thus defining the motion state coupling matrix; and associating and mapping the spatial feature vectors with the thermal data within the multi-dimensional motion space to calculate predictions. The parameters are adjusted such that when the thermal data exceeds a first preset threshold, an offset vector is generated on the spatial coordinate axis of the multidimensional motion space based on the predicted adjustment parameters. If the offset vector exceeds the spatiotemporal boundary constraints of the multidimensional motion space, the offset vector is projected onto the boundary surface of the spatiotemporal boundary constraints to generate a corrected offset vector, and the compensation position is determined based on the offset vector or the corrected offset vector. The compensated control signal is sent to each servo mechanism, and based on the synchronization constraints of the time axis and spatial coordinate axis of the multidimensional motion space, it is ensured that the position completion timestamp plus the control delay is less than or equal to the start timestamp of the next action. After each processing cycle is completed, the newly acquired multidimensional motion space trajectory samples are added to the training dataset, and the parameters of the action state coupling matrix are updated to optimize the prediction accuracy of subsequent cycles. The specific implementation details of this method can be referred to the description of the above system embodiment, especially the construction logic of the multidimensional motion space, the projection mechanism of the spatiotemporal boundary constraints, and the model update mechanism based on the sliding window, all of which are applicable to this method embodiment.

[0035] The foregoing description and accompanying drawings fully illustrate embodiments of this disclosure to enable those skilled in the art to practice them. Other embodiments may include structural, logical, electrical, procedural, and other changes. The embodiments represent only possible variations. Individual components and functions are optional unless explicitly required, and the order of operations may vary. Parts and features of some embodiments may be included in or replace parts and features of other embodiments. Moreover, the terminology used in this application is for describing embodiments only and is not intended to limit the claims. As used in the description of embodiments and claims, the singular forms “a,” “an,” and “the” are intended to equally include the plural forms unless the context clearly indicates otherwise. Similarly, the term “and / or” as used in this application means including one or more of the associated listed items and all possible combinations thereof. Additionally, when used in this application, the term "comprise" and its variations "comprises" and / or "comprising" refer to the presence of stated features, integrals, steps, operations, elements, and / or components, but do not exclude the presence or addition of one or more other features, integrals, steps, operations, elements, components, and / or groups thereof. Without further limitations, an element defined by the phrase "comprises a..." does not exclude the presence of other identical elements in the process, method, or apparatus that includes said element. In this document, each embodiment may focus on the differences from other embodiments, and similar or identical parts between embodiments can be referred to mutually. For methods, products, etc., disclosed in the embodiments, if they correspond to the method section disclosed in the embodiments, the relevant parts can be referred to the description of the method section.

[0036] Those skilled in the art will recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented using electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods for each specific application to achieve the described functions, but such implementation should not be considered beyond the scope of the embodiments of this disclosure. Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working processes of the described devices, apparatuses, and units can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here.

[0037] The flowcharts and block diagrams in the accompanying drawings illustrate the architecture, function, and operation of possible implementations of apparatus, methods, and computer program products according to embodiments of the present disclosure. In this regard, each block in a flowchart or block diagram may represent a module, segment, or portion of code containing one or more executable instructions for implementing a specified logical function. In some alternative implementations, the functions indicated in the blocks may occur in a different order than those indicated in the drawings. For example, two consecutive blocks may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. In the descriptions corresponding to the flowcharts and block diagrams in the accompanying drawings, the operations or steps corresponding to different blocks may also occur in a different order than those disclosed in the description; sometimes there is no specific order between different operations or steps. For example, two consecutive operations or steps may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. Each block in a block diagram and / or flowchart, and combinations of blocks in a block diagram and / or flowchart, can be implemented using a dedicated hardware-based device that performs the specified function or action, or using a combination of dedicated hardware and computer instructions.

Claims

1. A positioning accuracy control system for CNC mold machining, characterized in that, include: The configuration module is configured to construct multiple multi-dimensional motion spaces based on the motion instruction set and spatial position coordinate set of the CNC machine tool, and extract motion feature parameters of the multi-dimensional motion spaces, including velocity vector magnitude distribution, acceleration rate of change, and trajectory curvature density. The data acquisition module is configured to synchronously acquire the working state parameters of the mold during the processing, including thermal data, position deviation signals, and processing cycle timestamp sequences, and limit the multi-dimensional state perception information. The timing control module is communicatively connected to the configuration module and the data acquisition module, and is configured to receive and preprocess the data uploaded by each module, establish a mapping relationship between motion feature parameters and working state parameters, reconstruct the first trajectory sequence of each axis of the CNC machine tool from the start time of the first processing cycle based on historical processing cycles, and send compensation control signals to each servo mechanism before the start of the second processing cycle according to the adjustment parameters output by the prediction model to perform predictive position compensation, wherein the predictive position compensation limits the starting compensation position of the motion trajectory of each axis.

2. The positioning accuracy control system for CNC mold processing according to claim 1, characterized in that, The working status parameters include thermal data, position deviation signals, and processing cycle timestamp sequences. The thermal data includes first thermal data of the pre-defined heat dissipation area of ​​the mold and second thermal data of the heat concentration area obtained from sampling. The position deviation signal includes a deviation value containing XYZ three-axis components generated by measuring the spatial vector deviation between the actual position and the target position of the mold, used to calibrate the position coordinate set of the multi-dimensional motion space. The processing cycle timestamp sequence includes the timestamp of a single cycle recorded in response to the processing start pulse and completion pulse issued by the CNC control system, providing reference cycle data for the timing control module. Furthermore, the first thermal data, second thermal data, position deviation signals, and processing cycle timestamp sequences are weighted and fused to determine a comprehensive evaluation index of the mold's working status.

3. The positioning accuracy control system for CNC mold processing according to claim 1, characterized in that, The configuration module constructs the multidimensional motion space, specifically including the following steps: defining the time axis, spatial coordinate axis, and command logic axis as the base dimensions of the multidimensional motion space; mapping the velocity vector magnitude distribution, acceleration rate of change, and trajectory curvature density to the base dimensions to generate spatial feature vectors; calculating the correlation coefficient between the spatial feature vectors and the working state parameters to construct a parameter mapping relationship, wherein the parameter mapping relationship limits the obtained motion state coupling matrix, which is used to quantify the influence weight of motion feature parameters on mold processing parameters.

4. The positioning accuracy control system for CNC mold processing according to claim 3, characterized in that, The timing control module performs predictive adjustment, specifically including the following steps: establishing a predictive model, associating and mapping the spatial feature vector with the thermal data in the multi-dimensional motion space, calculating predictive adjustment parameters, and the predictive model constraining the thermal deformation mapping of the motion space; when the thermal data is higher than a first preset threshold, generating an offset vector on the spatial coordinate axis of the multi-dimensional motion space according to the predictive adjustment parameters, so that the CNC machine tool reaches the compensation position before the start of the second machining cycle; the compensation position is the starting point of the motion trajectory of each axis reducing the control time delay and the total thermal deformation compensation amount, represented by the set of control signal values ​​of each servo mechanism at the predicted timestamp, and the limiting condition of the compensation position is that the position completion timestamp plus the control delay of the servo mechanism is less than or equal to the start timestamp of the next action.

5. The positioning accuracy control system for CNC mold processing according to claim 4, characterized in that, When the control signal value of the first servo mechanism reaches the value representing the target position in the first trajectory sequence, the timing control module dynamically adjusts the control signal value of the servo mechanism to the target value set based on the synchronization constraints of the time axis and spatial coordinate axis of the multi-dimensional motion space. The target value set limits the position time constraint interval to ensure that the position completion timestamp plus the control delay of the servo mechanism is less than or equal to the start timestamp of the next action.

6. The positioning accuracy control system for CNC mold processing according to claim 4, characterized in that, After each processing cycle is completed, the timing control module adds the newly acquired multidimensional motion space trajectory samples and actual position deviation data to the training dataset, updates the parameters of the motion state coupling matrix through a recursive linear regression algorithm, and uses a sliding window mechanism to limit the training dataset to retain only the multidimensional motion space segments corresponding to the most recent preset number of processing cycles. When the amount of data in the window reaches the preset number, the earliest acquired multidimensional motion space segment is automatically removed.

7. The positioning accuracy control system for CNC mold processing according to claim 3, characterized in that, The instruction logic axis is configured to represent the logical sequence of the machining process, and the roughing stage, semi-finishing stage and finishing stage are mapped to the first logic interval, the second logic interval and the third logic interval on the instruction logic axis, respectively; in the multi-dimensional motion space, different dimension weight coefficients are assigned to the first logic interval, the second logic interval and the third logic interval, and the dimension weight coefficients are used to adjust the contribution of the spatial feature vector to the action state coupling matrix at different machining stages.

8. The positioning accuracy control system for CNC mold processing according to claim 4, characterized in that, The multidimensional motion space is also limited by spatiotemporal boundary constraints, which are defined by the maximum allowable span of the time axis and the maximum allowable displacement of the spatial coordinate axis. When the offset vector generated by the timing control module exceeds the spatiotemporal boundary constraints, the offset vector is projected onto the boundary surface of the spatiotemporal boundary constraints to generate a corrected offset vector. The corrected offset vector limits the safe compensation path to ensure that the motion trajectory of the CNC machine tool is always within the reachable area of ​​the multidimensional motion space.

9. A method for controlling the positioning accuracy of CNC mold machining, applied to a CNC machine tool environment with preset machining parameters, characterized in that, The steps include: constructing multiple multi-dimensional motion spaces based on the motion instruction set and spatial position coordinate set of the CNC machine tool; defining the time axis, spatial coordinate axis and instruction logic axis as the base dimensions of the multi-dimensional motion space; and extracting the motion feature parameters of the multi-dimensional motion space. Synchronously acquire the working state parameters of the mold during processing, including thermal data, position deviation signals, and processing cycle timestamp sequences, to obtain multi-dimensional state perception information; map the motion feature parameters to the base dimension to generate spatial feature vectors, and calculate the correlation coefficient between the spatial feature vectors and the working state parameters to construct a parameter mapping relationship, thus obtaining a motion state coupling matrix; associate and map the spatial feature vectors with the thermal data in the multi-dimensional motion space to calculate predictive adjustment parameters. When the thermal data is higher than a first preset threshold, generate a prediction adjustment parameter on the spatial coordinate axis of the multi-dimensional motion space according to the predictive adjustment parameter. Offset vector; if the offset vector exceeds the spatiotemporal boundary constraints of the multidimensional motion space, the offset vector is projected onto the boundary surface of the spatiotemporal boundary constraints to generate a corrected offset vector, and the compensation position is determined based on the offset vector or the corrected offset vector; the compensated control signal is sent to each servo mechanism, and based on the synchronization constraints of the time axis and spatial coordinate axis of the multidimensional motion space, it is ensured that the position completion timestamp plus the control delay is less than or equal to the start timestamp of the next action; after each processing cycle is completed, the newly acquired multidimensional motion space trajectory samples are added to the training dataset, and the parameters of the action state coupling matrix are updated to optimize the prediction accuracy of subsequent cycles.

Images