Auxiliary positioning method and system in multi-process switching of precision mold

Abstract

The invention provides an auxiliary positioning method and system in multi-process switching of a precision mold, and relates to the technical field of precision machining.The method comprises the steps that process reference information of a mold in a current process is collected, and a three-dimensional mold grid model is constructed; performing grid comparison and offset calculation by using the three-dimensional mold grid model on the basis of the positioning reference data to obtain real-time offset information; taking elimination of the real-time offset information as a target, performing primary position adjustment on the mold to obtain a primary corrected mold; and according to the historical positioning error of the next adjacent working procedure, secondary position adjustment is conducted on the primary correction mold, and mold machining of the next adjacent working procedure is executed. The technical problems of mold positioning error accumulation and machining precision reduction caused by lack of a real-time pose offset sensing and historical error compensation mechanism in the prior art are solved, the mold multi-process switching positioning precision is remarkably improved, the stability and reliability of multi-process switching are enhanced, and the mold machining quality is improved.

Description

Technical Field

[0001] This application relates to the field of precision machining technology, and particularly to an auxiliary positioning method and system in multi-process switching of precision molds. Background Art

[0002] In the field of precision manufacturing, mold processing usually involves continuous switching of multiple processes, such as rough machining, finish machining, surface treatment, and assembly debugging. To ensure that the mold maintains a unified machining reference and clamping accuracy between different processes, the multi-process switching positioning technology of the mold has become a key link to ensure machining quality and efficiency.

[0003] Currently, the positioning methods for multi-process switching of precision molds mostly rely on fixed reference blocks, mechanical positioning pins, or tooling fixtures, etc., and the mold is realigned manually or semi-automatically. Although such methods have simple structures and low costs, they have significant problems such as limited positioning accuracy, low operation efficiency, and inability to detect and compensate errors in a timely manner. Especially during the mold switching process, affected by factors such as temperature changes, clamping stress, and micro-deformation of the mold itself, the pose of the mold is prone to small offsets, and traditional methods are difficult to perceive such errors in a timely manner, resulting in the gradual amplification of errors in subsequent processes and affecting the overall machining accuracy. Summary of the Invention

[0004] This application provides an auxiliary positioning method and system in multi-process switching of precision molds, which solves the technical problem that the existing technology lacks a real-time pose offset perception and historical error compensation mechanism for the mold during the multi-process switching process, resulting in the accumulation of mold positioning errors and the decline of machining accuracy, and achieves the technical effects of significantly improving the positioning accuracy of multi-process switching of the mold, enhancing the stability and reliability of multi-process switching, and improving the machining quality of the mold.

[0005] In view of the above problems, on the one hand, this application provides an auxiliary positioning method in multi-process switching of precision molds, and the method includes: when machining the mold, collecting the process reference information of the mold in the current process and establishing positioning reference data; when the mold enters the process switching stage, using a multi-source sensing device to collect information about the mold and constructing a three-dimensional mold grid model; using the three-dimensional mold grid model for grid comparison and offset calculation based on the positioning reference data to obtain real-time offset information; aiming at eliminating the real-time offset information, controlling an auxiliary positioning device to perform a primary position adjustment on the mold to obtain a once-corrected mold; and performing a secondary position adjustment on the once-corrected mold through a multi-axis adjustment device according to the historical positioning error of the next adjacent process to complete the mold position correction and execute the mold machining of the next adjacent process.

[0006] On the other hand, the present application also provides an auxiliary positioning system in the multi-process switching of precision molds. The system includes: a positioning reference construction module, which is used to collect the process reference information of the mold in the current process during mold processing and establish positioning reference data; a mold information collection module, which is used to collect information on the mold using multi-source sensing devices when the mold enters the process switching stage and construct a three-dimensional mold grid model; an offset calculation module, which is used to perform grid comparison and offset calculation using the three-dimensional mold grid model based on the positioning reference data to obtain real-time offset information; a primary correction module, which is used to control the auxiliary positioning device to perform a primary position adjustment on the mold with the goal of eliminating the real-time offset information to obtain a primarily corrected mold; a secondary correction module, which is used to perform a secondary position adjustment on the primarily corrected mold through a multi-axis adjustment device according to the historical positioning error of the next adjacent process, complete the mold position correction, and perform the mold processing of the next adjacent process.

[0007] One or more technical solutions provided in the present application have at least the following beneficial effects: By collecting the process reference information of the mold in the current process during mold processing and establishing positioning reference data, an accurate reference standard is provided for positioning in subsequent process switching, ensuring the consistency and continuity of data between multiple processes. When the mold enters the process switching stage, multi-source sensing devices are used to collect information on the mold and construct a three-dimensional mold grid model to truly reflect the actual position and morphology of the mold after switching, providing a detailed data basis for subsequent comparison and calculation. Based on the positioning reference data, grid comparison and offset calculation are performed using the three-dimensional mold grid model to realize the quantitative analysis of the difference between the actual state of the mold and the theoretical reference, obtain accurate real-time offset information, and provide a decision basis for subsequent positioning correction. With the goal of eliminating the real-time offset information, the auxiliary positioning device is controlled to perform a primary position adjustment on the mold to quickly correct the main deviation direction of the mold, achieve preliminary accuracy recovery, and obtain a primarily corrected mold; according to the historical positioning error of the next adjacent process, a secondary position adjustment is performed on the primarily corrected mold through a multi-axis adjustment device, complete the mold position correction, improve the stability and feedforward compensation ability of overall positioning, ensure the consistency of the starting point of the next process, and then perform the mold processing of the next adjacent process.

[0008] In summary, the present application realizes high-precision auxiliary positioning during the multi-process switching of precision molds by integrating a dual mechanism of real-time perception and historical data-driven. The solution uses process reference data and a three-dimensional mesh model to calculate the real-time offset of the mold pose, and through a two-stage position adjustment strategy of current offset correction and historical error compensation, it significantly improves the positioning consistency and compensation accuracy during the mold process connection. It not only effectively avoids the problem of error accumulation caused by insufficient perception ability of traditional methods, but also enhances the adaptability to small deformations and displacements under complex working conditions, ultimately achieving the technical effects of significantly improving the positioning accuracy of mold multi-process switching, enhancing the stability and reliability of multi-process switching, and improving the quality of mold processing.

[0009] The above description is only an overview of the technical solution of the present application. In order to be able to understand the technical means of the present application more clearly, it can be implemented according to the content of the specification. And in order to make the above and other purposes, features and advantages of the present application more obvious and understandable, the following specifically illustrates the specific embodiments of the present application. BRIEF DESCRIPTION OF THE DRAWINGS

[0010] Figure 1 It is a schematic flow chart of the auxiliary positioning method in the multi-process switching of precision molds provided by the embodiment of the present application.

[0011] Figure 2 It is a schematic flow chart of constructing a three-dimensional mold mesh model in the auxiliary positioning method in the multi-process switching of precision molds provided by the embodiment of the present application.

[0012] Figure 3 It is a schematic structural diagram of the auxiliary positioning system in the multi-process switching of precision molds provided by the embodiment of the present application.

[0013] Description of reference numerals: positioning reference construction module 10, mold information acquisition module 20, offset calculation module 30, primary correction module 40, secondary correction module 50. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0014] The embodiment of the present application provides an auxiliary positioning method and system in the multi-process switching of precision molds, which solves the technical problem in the prior art that due to the lack of real-time pose offset perception and historical error compensation mechanism during the multi-process switching of molds, the mold positioning error accumulates and the processing accuracy decreases, and achieves the technical effects of significantly improving the positioning accuracy of mold multi-process switching, enhancing the stability and reliability of multi-process switching, and improving the quality of mold processing.

[0015] Embodiment 1, as Figure 1 shown, the embodiment of the present application provides an auxiliary positioning method in the multi-process switching of precision molds, and the method includes: Step S100: During mold processing, collect the process reference information of the mold in the current process and establish positioning reference data.

[0016] Specifically, the process reference information refers to the geometric elements and parameters based on which the machining or clamping position of the mold is determined in the current process, usually including reference points, reference planes, reference lines, and machining and clamping process parameters. In the initial machining stage of the mold, through the sensor system installed on the numerical control equipment or inspection table, the key reference features of the mold in the current process are collected in real time, including geometric information such as reference planes (such as the bottom surface or clamping surface), reference lines (such as the symmetry center line), and reference points (such as the center of the positioning hole). At the same time, a machining model is established by combining machining parameters (such as tool path, rotational speed, feed) and clamping parameters (such as clamping force, clamping position). On this basis, a local coordinate system of the mold is constructed with the collected reference geometric information, and its posture is fine-tuned according to the process parameters, so as to generate positioning reference data with high-precision description ability, realize the accurate capture and standardized expression of the position and posture of the mold in the initial machining stage, provide a high-quality reference basis for the position and posture comparison and correction in the subsequent multi-process switching process, and significantly improve the accuracy of subsequent positioning compensation. For example, in the first milling process of a certain automotive mold, an XY plane reference is established with the centers of four positioning holes as reference points, and then the fixture positioning edge is used as the Z-direction reference plane. The machining parameters are a feed of 300 mm / min, a spindle speed of 2000 rpm, and a clamping force of 1200 N. Finally, a positioning reference coordinate system including 6D position and posture description is established.

[0017] Step S200: When the mold enters the process switching stage, use multi-source sensing devices to collect information about the mold and construct a three-dimensional mold mesh model.

[0018] Specifically, before the end of the current process and before preparing to enter the next process, use multi-source sensing devices deployed in the workshop measurement area or around the machining platform to collect information about the mold. The multi-source sensing devices include image sensors, laser scanners, structured light devices, etc. The image sensor captures photos of the mold surface texture and features from multiple angles; the laser scanner obtains point cloud data through beam scanning, covering the entire shape of the mold. The contour of the image data is extracted and the feature points are located, and then fused with the point cloud data to fit and form a complete three-dimensional mold surface model. Subsequently, according to the process accuracy requirements, the grid size is set, and the three-dimensional surface is meshed to generate a three-dimensional mesh model with continuous structure for subsequent comparison and analysis. Through multi-source perception and high-precision three-dimensional modeling, the real state of the mold during process switching can be fully restored, providing a digital basis for subsequent position and posture deviation analysis and dynamic correction, and ensuring the comprehensiveness and accuracy of error identification.

[0019] Step S300: Using the positioning reference data as a benchmark, perform grid comparison and offset calculation with the 3D die grid model to obtain real-time offset information.

[0020] Further, the real-time offset information includes coordinate position offset and attitude angle deviation.

[0021] Specifically, take the positioning reference data established in Step S100 as the target template, and take the 3D die grid model obtained in Step S200 as the current state. Perform point-to-point comparison on the two through the ICP (Iterative Closest Point) algorithm. During the comparison process, calculate the spatial offset of each grid point and summarize it into real-time offset information including coordinate position offset and attitude angle deviation. Among them, the coordinate position offset includes the displacement changes in the three-dimensional space (X, Y, Z directions) of the die, and the attitude angle deviation includes the changes in the rotation attitude of the die (pitch angle, yaw angle, roll angle), which can be integrated and represented as six-degree-of-freedom offset parameters: three position offsets (ΔX, ΔY, ΔZ) and three attitude angle deviations (ΔPitch, ΔYaw, ΔRoll). Among them, ΔX is the displacement deviation in the X-axis direction, ΔY is the displacement deviation in the Y-axis direction, ΔZ is the displacement deviation in the Z-axis direction, ΔPitch is the pitch angle deviation, ΔYaw is the yaw angle deviation, and ΔRoll is the roll angle deviation. This step realizes the high-precision detection and parametric expression of the current pose error of the die, provides a clear quantitative target for subsequent automatic position adjustment, and greatly improves the error response speed and compensation accuracy during the process switching.

[0022] Step S400: Aiming to eliminate the real-time offset information, control the auxiliary positioning device to perform a primary position adjustment on the die to obtain a once-corrected die.

[0023] Specifically, the auxiliary positioning device refers to a mechanical or electronically controlled structure used to precisely adjust the position of the die, such as pneumatic positioning pins, six-degree-of-freedom flexible positioning platforms, servo control mechanisms, etc. After receiving the offset information output in Step S300, calculate the adjustment path according to the quality and structural characteristics of the die, and drive the auxiliary positioning device to perform position correction. The auxiliary positioning device performs small displacements or attitude adjustments on the die according to the set path to ensure that the die enters the error tolerance range (for example, within ±0.05 mm, ±0.1°). After the adjustment is completed, mark the current die state as the "once-corrected die" for subsequent more refined historical error correction, reduce the complexity of subsequent fine-tuning, and improve the stability of the die switching efficiency and correction effect.

[0024] Step S500: According to the historical positioning error of the next adjacent process, perform a secondary position adjustment on the once-corrected die through a multi-axis adjustment device to complete the die position correction and perform the die processing of the next adjacent process.

[0025] Specifically, historical positioning error is the high-frequency offset information obtained from the positioning adjustment records of the same or similar molds in similar processes in the past. The multi-axis adjustment device is a positioning platform or robot with multiple degrees of freedom adjustment capabilities, which can achieve high-precision spatial posture adjustment.

[0026] To further improve the positioning accuracy, after one correction, the historical positioning error data of similar molds in the adjacent process stored on the industrial Internet platform is consulted, and an error prediction model is established in combination with the current mold geometry deviation. According to the offset trend output by the model, a small range of high-precision fine-tuning (such as ±0.01mm) is performed through a multi-axis adjustment device, and the mold posture is finally brought close to the theoretical optimal state. After the adjustment is completed, the mold is locked in the new position and enters the processing flow of the next process. This step introduces historical data feedback and a high-precision multi-axis adjustment mechanism to make the positioning strategy adaptive and forward-looking, effectively prevent error accumulation, and improve the continuity and processing accuracy of process connection.

[0027] Furthermore, step S100 includes: Step S110: collecting the reference points, reference planes, reference lines and process parameters of the mold in the current process as the process reference information, wherein the process parameters include processing parameters and clamping parameters.

[0028] Step S120: establishing a reference coordinate system and a target attitude angle of the mold based on the reference points, reference planes and reference lines.

[0029] Step S130: fine-tuning the reference coordinate system and the target attitude angle according to the processing parameters and the clamping parameters to obtain the positioning reference data.

[0030] Specifically, in the initial process of the mold, the geometric reference elements such as the reference points, reference planes, and reference lines of the mold in this process, as well as process parameters such as processing parameters and clamping parameters are collected by connecting to the machine tool control system, sensor interface or process design system (such as CAM software). Among them, the processing parameters include the parameters required for CNC processing such as feed speed, spindle speed, and cutting path, and the clamping parameters include the clamping force, clamping method, clamping position and other process conditions used to fix the mold. For example, the center of the positioning hole, the clamping plane, etc. are identified by using visual sensors and probes; the processing speed, tool trajectory, etc. in the G code are read by the controller; the fixture monitoring system records the clamping force and contact point position in real time. This information together constitutes the complete process benchmark information for this process.

[0031] According to the collected reference geometric elements, a logical coordinate system is determined in the mold model. For example, set the reference point as the origin of the coordinates, the reference plane defines the Z-axis direction, the reference line defines the X-axis direction, automatically deduce the Y-axis and complete the right-hand construction of the coordinate system. At the same time, by comparing the designed posture of the model, generate the ideal target posture angles required for the current process (such as 0° yaw, 0° roll, -3° pitch), transform the discrete geometric features into a standardized coordinate system and posture definition, realize the quantifiable expression of the mold state, and provide a stable reference for offset detection and automatic correction.

[0032] Due to machining errors, clamping deformations or mold loading deviations, there may be slight offsets in the original reference coordinate system and posture angles. Therefore, combine machining parameters (such as the direction of the spindle thrust, the tool contact area) and clamping parameters (the direction of the force applied at the clamping point, the stiffness of the fixture) for simulation calculations, generate a fine-tuning vector, and apply it to the original reference coordinate system to dynamically correct the reference coordinate system and the target posture angles, obtain the positioning reference data, realize the adaptive correction mechanism that combines the ideal posture and the actual working conditions, improve the fitness and usability of the reference data, and thus enhance the accuracy and robustness of the subsequent position recognition and correction actions of the mold.

[0033] Furthermore, as Figure 2 shown, step S200 includes: Step S210: After the processing of the current process is completed, use an image sensor to collect images of the mold at a preset acquisition angle to obtain a set of mold images, and perform an omnidirectional scan of the mold with a laser scanner to obtain a point cloud set of the mold.

[0034] Step S220: Extract contour lines according to the set of mold images, and fuse and construct a three-dimensional mold contour.

[0035] Step S230: Use the three-dimensional mold contour to perform point cloud fitting on the point cloud set of the mold to construct a three-dimensional mold model.

[0036] Step S240: Perform mesh division on the three-dimensional mold model according to a preset mesh size to obtain the three-dimensional mold mesh model.

[0037] Specifically, after the processing of the current process is completed, start the image acquisition and scanning process. First, under the scheduling of the control system, the image sensor collects images around the mold at a preset angle (for example, rotates every 30°) to generate a multi-view image set; subsequently, the laser scanner performs high-precision laser ranging on the entire surface of the mold to obtain the original point cloud set of the mold, providing a basis for subsequent three-dimensional model construction.

[0038] Call image processing algorithms (such as Canny edge detection, Sobel operator, depth edge recognition model) to analyze the mold image set and extract contour feature lines. Convert these contour lines into three-dimensional coordinates through camera calibration data (such as stereo vision reconstruction, beam triangulation, etc.), and then form a preliminary three-dimensional contour of the mold. These contours can be used as structural boundary references in the subsequent fusion process with point cloud data.

[0039] Use the obtained three-dimensional contour as a structural constraint to perform surface fitting and topological reconstruction on the obtained point cloud set, convert the mold point cloud into a continuous surface model, construct a three-dimensional mold model, realize the transformation of the mold surface from discrete points to a continuous surface, ensure that the model has good visualization effects and applicability for subsequent analysis, and improve the expression accuracy of the mold model.

[0040] According to the preset grid size parameters, use grid generation algorithms (such as Delaunay triangulation, octree segmentation, etc.) to divide the three-dimensional surface, and generate a three-dimensional mold grid model with equal density or adaptive density. Among them, the preset grid size is determined according to the actual machining state. This model has the characteristics of both structural lightweight and high precision, and is convenient for subsequent offset analysis, attitude comparison or simulation verification. Through the meshing process, the three-dimensional mold model obtains a standardized structural form, and can efficiently participate in subsequent calculation processes such as comparison, error analysis and attitude adjustment, significantly improving the system operation efficiency and analysis accuracy.

[0041] Further, step S220 includes: Step S221: Use the sample mold image set and sample mold contour set of the same type of mold to train the convolutional neural network until convergence to obtain a contour line extractor.

[0042] Step S222: Use the contour line extractor to extract contour lines from multiple mold images in the mold image set respectively to obtain multiple mold contours.

[0043] Step S223: Based on the preset acquisition angle, perform feature registration and smoothing repair according to the multiple mold contours to construct a three-dimensional mold contour.

[0044] Specifically, the sample die image set is a multi-angle image data of known dies of the same type collected in advance. The sample die contour set is a set of die edge lines manually marked or automatically extracted, serving as training labels. Using the supervised learning method, a large number of images of the same type of dies and their corresponding contour labels are used as training data to train a convolutional neural network model to learn the edge structure features in the images, and a contour line extractor is obtained. An example of the training process is as follows: First, collect and organize the sample die image set of the same type of dies and the corresponding contour line annotation images to construct training sample pairs. Among them, each sample includes die images from multiple preset acquisition angles and their corresponding contour line annotation maps. The annotation process is to draw pixel-level contour edges through image annotation software and uniformly convert them into binary mask images. To enhance the generalization ability of the model, data augmentation operations including rotation, flipping, brightness perturbation, blurring, and affine transformation are performed on the image samples to expand the sample quantity and visual diversity. Subsequently, a convolutional neural network model with U-Net as the backbone structure is constructed. The input is the RGB die image, and the output is the contour probability map. In the downsampling part of the network, ResNet-18 is used as the feature extraction backbone, and in the upsampling part, a multi-layer transposed convolution module is used for resolution recovery. During the training process, a weighted combination of binary cross-entropy loss (BCE), Dice loss, and edge structure loss is used as the total loss function to enhance the recognition accuracy of the model for contour edges. The model training uses the Adam optimizer, the initial learning rate is set to 0.001, the number of samples selected for one training is 8, the number of training epochs is 100, and the training set and validation set are divided in a ratio of 8:2. The model performance is monitored through metrics such as intersection over union, F1 score, and edge offset error. When the performance metrics of the model on the validation set tend to converge and stabilize, the weight file of the finally trained model is saved as the contour line extractor for subsequent batch contour extraction tasks of die images.

[0045] Input the die image set collected by the image sensor into the contour line extractor, perform edge detection, filtering, and contour localization operations on each image respectively, and output a contour map that matches the image size. The set of contour maps corresponding to all images constitutes multiple die contours, preparing for subsequent 3D modeling.

[0046] Utilize the preset acquisition angles during image shooting to perform 3D reconstruction projection on the contour lines at different angles. Subsequently, methods such as the ICP (Iterative Closest Point) registration algorithm and feature point matching are used to spatially align the contours from each perspective. Perform curvature filtering and smooth interpolation on the edge lines with defects, burrs, and breakpoints to achieve contour repair, and finally combine them into a continuous and complete 3D contour framework to realize the unified expression of die multi-angle image contours in 3D space, generate a 3D contour framework with continuous structure and controllable accuracy, and provide high-quality geometric guiding information for subsequent point cloud fitting.

[0047] Further, step S230 includes: Step S231: Using the three-dimensional mold contour, perform a first random fitting on the mold point cloud set to obtain a first point cloud fitting result, and calculate the ratio of the number of points falling into the first point cloud fitting result to the total number of points in the mold point cloud set to obtain a first fitting accuracy.

[0048] Step S232: Again using the three-dimensional mold contour, perform a second random fitting on the mold point cloud set to obtain a second point cloud fitting result and a second fitting accuracy.

[0049] Step S233: Perform iterative fitting until a preset number of fittings is reached. Output the point cloud fitting result with the maximum fitting accuracy as the optimal point cloud fitting result, and construct a three-dimensional mold model based on the optimal point cloud fitting result.

[0050] Specifically, before fitting the mold point cloud set, use the already constructed three-dimensional mold contour as a reference model and adopt the Random Sample Consensus algorithm (RANSAC) for preliminary point cloud fitting. The specific operation is as follows: Randomly select several points in the mold point cloud as initial fitting seed points, and use these seed points to generate a preliminary plane or surface fitting model. Then, calculate the distance between each point in all the mold point clouds and the fitting model, and determine whether it meets the preset tolerance threshold. The points that meet the conditions are considered to be the points falling into the fitting result, and the number of these points is recorded. Calculate the ratio of the number of points falling into the fitting result to the total number of points in the mold point cloud set to obtain the accuracy of the first fitting. The higher this ratio, the better the fitting effect and the higher the accuracy.

[0051] After the first fitting, to further improve the fitting accuracy, use the three-dimensional mold contour to perform a second fitting on the point cloud again. At this time, adopt the same Random Sample Consensus algorithm (RANSAC) as the first time, but the initially selected seed points and tolerance values are adjusted to further refine the fitting result. The new fitting model will generate a new point cloud fitting result and recalculate the fitting accuracy. Similarly, the standard for calculating the fitting accuracy is the ratio of the number of points falling into the fitting result to the total number of points. If the accuracy of the second fitting result is higher than that of the first time, then the point cloud fitting result generated at this time is the second fitting result.

[0052] Iterative fitting is performed according to the above steps until the preset maximum number of fitting times is reached. For each point cloud fitting result, the fitting accuracy needs to be calculated. When the number of iterations reaches the preset limit, the fitting result with the highest fitting accuracy is selected as the optimal point cloud fitting result. Finally, a three-dimensional mold model is constructed based on this optimal point cloud fitting result to ensure that the geometric shape of the model matches the point cloud data of the actual mold to the greatest extent, providing a high-quality model basis for subsequent operations such as mesh generation and positioning analysis.

[0053] Furthermore, the method for generating the preset mesh size described in step S240 includes: Step S240-1: During the processing of the current process, monitor and obtain the processing parameter sequence and clamping parameter sequence.

[0054] Step S240-2: Based on the standard processing parameters and standard clamping parameters of the current process, perform deviation analysis on the processing parameter sequence and clamping parameter sequence to obtain the processing parameter deviation amplitude sequence and clamping parameter deviation amplitude sequence.

[0055] Step S240-3: Calculate the mean value and fluctuation analysis of the processing parameter deviation amplitude sequence, and output the mean value of the processing parameter deviation amplitude and the processing parameter fluctuation coefficient, where the fluctuation coefficient is the ratio of the standard deviation to the mean value.

[0056] Step S240-4: Calculate the mean value and fluctuation analysis of the clamping parameter deviation amplitude sequence, and output the mean value of the clamping parameter deviation amplitude and the clamping parameter fluctuation coefficient.

[0057] Step S240-5: Calculate the current processing deviation scale by weighted calculation based on the mean value of the processing parameter deviation amplitude, the processing parameter fluctuation coefficient, the mean value of the clamping parameter deviation amplitude, and the clamping parameter fluctuation coefficient.

[0058] Step S240-6: Calculate the ratio of the preset processing deviation scale of the current process to the current processing deviation scale, and multiply it by the initial mesh size to obtain the preset mesh size.

[0059] Specifically, during the processing of the current process, continuously monitor and collect various processing parameters and clamping parameters. These parameters include information such as cutting speed, feed speed, tool position, workpiece position, clamping force, etc. These parameter sequences are recorded in real time through sensors, control systems, or data acquisition devices, providing the necessary raw data for subsequent deviation analysis.

[0060] After obtaining the processing parameter sequence and clamping parameter sequence of the current process, taking the standard processing parameters and standard clamping parameters of the current process as the benchmark, perform deviation analysis on the actual processing parameter sequence and clamping parameter sequence. Specifically, calculate the deviation amplitude of each processing parameter and clamping parameter relative to the standard value, and generate a processing parameter deviation amplitude sequence and a clamping parameter deviation amplitude sequence. These deviation amplitude data can be used to reflect the instability or error sources in the processing process.

[0061] Next, perform mean calculation and fluctuation analysis on the processing parameter deviation amplitude sequence. First, obtain the mean of the processing parameter deviation, that is, the average degree of deviation. Then, by calculating the standard deviation of the deviation sequence, obtain the fluctuation coefficient. The fluctuation coefficient formula is: Fluctuation coefficient = Standard deviation / Mean. The fluctuation coefficient is used to describe the fluctuation amplitude of the processing parameters, that is, the degree of change of the parameters in the actual processing process. The smaller the mean of the processing parameter deviation amplitude, the lower the fluctuation coefficient, indicating that the processing process is more stable and the error is smaller.

[0062] Similarly, perform mean calculation and fluctuation analysis on the clamping parameter deviation amplitude sequence. Obtain the mean of the clamping parameter deviation and calculate its fluctuation coefficient. The deviation analysis of the clamping parameters can reveal the possible errors and unstable factors in the clamping process. Similarly, the smaller the mean of the deviation amplitude, the smaller the fluctuation coefficient, indicating that the stability of the clamping process is higher.

[0063] Perform weighted calculation based on the mean of the processing parameter deviation amplitude, the processing parameter fluctuation coefficient, the mean of the clamping parameter deviation amplitude, and the clamping parameter fluctuation coefficient to obtain the processing deviation scale of the current process. This processing deviation scale comprehensively considers the influence of all deviation amplitudes and fluctuation coefficients and can accurately reflect the overall situation of the actual processing errors in the process. Generally speaking, the smaller the deviation and the smaller the fluctuation, the smaller the processing deviation scale.

[0064] Finally, calculate the ratio of the preset processing deviation scale of the current process to the actual processing deviation scale. This ratio reflects the deviation between the processing accuracy of the current process and the preset standard. Multiply this ratio by the initially set grid size to finally obtain the preset grid size. If the deviation is large, the grid size needs to be reduced to improve the accuracy of grid division and ensure higher processing accuracy. On the contrary, if the deviation is small, the grid size can be appropriately increased to optimize the calculation efficiency. Among them, the initial grid size is the grid division size set according to the mold size and the initial accuracy requirements, serving as the benchmark for adjusting the preset grid size.

[0065] Through the above steps, the preset grid size can be dynamically adjusted according to the actual deviation conditions of the processing parameters and clamping parameters, ensuring that the mold positioning and error correction in the processing process can achieve the required accuracy and stability.

[0066] Further, step S500 includes: Step S510: Use the three-dimensional mold grid model to perform grid comparison and offset calculation, obtain the geometric shape deviation of the mold in the current process, and expand the geometric shape deviation according to a preset deviation tolerance interval to obtain a geometric shape deviation interval.

[0067] Step S520: With the geometric shape deviation interval and the next adjacent process as retrieval constraints, retrieve the processing information of similar molds based on the industrial Internet to determine a set of sample historical positioning errors.

[0068] Step S530: Analyze and determine the historical positioning error according to the set of sample historical positioning errors, and with the goal of eliminating the historical positioning error, perform a secondary position adjustment on the once-corrected mold through a multi-axis adjustment device.

[0069] Specifically, the preset deviation tolerance interval is an allowable deviation range set according to process requirements and experience, and is used to expand the geometric shape deviation to form a geometric shape deviation interval. Use the currently generated three-dimensional mold grid model to perform grid comparison and offset calculation. By comparing the actual geometric shape of the mold in the current process with the preset geometric shape (usually a standard model or a reference model), calculate the geometric shape deviation of the mold, including dimensional deviation, angular deviation, etc. According to the preset deviation tolerance interval, expand the geometric shape deviation to form a geometric shape deviation interval. This expansion is to cover the possible error range, ensuring that even if the deviation fluctuates, it can still be tolerated and corresponding compensation can be made. In this way, it can be ensured that when the mold is subjected to secondary adjustment, it can still be maintained within a suitable tolerance range, avoiding other problems caused by excessive correction.

[0070] Use the obtained geometric shape deviation interval as a retrieval constraint, and combined with the processing requirements of the next adjacent process, use the industrial Internet platform to retrieve the processing information of similar molds, and query the processing data of similar molds under similar processes. From these retrieved historical data, screen out a set of sample historical positioning errors. This error set contains the positioning error information experienced by other molds in the same or similar processes. By analyzing these historical positioning errors, reference data can be provided for the secondary adjustment of the current mold.

[0071] Perform statistical analysis on the data in the set of sample historical positioning errors to determine the historical positioning error. On this basis, set a correction plan with the goal of eliminating the historical positioning error, and perform a secondary position adjustment. Use a multi-axis adjustment device to perform a reverse adjustment on the mold (i.e., pre-compensation), and preset a compensation amount before the process to eliminate the upcoming error, so as to reduce the accumulation of errors and effectively improve the processing accuracy of the next process.

[0072] Furthermore, step S530 includes: Step S531: Obtain the sample historical positioning error set, where the sample historical positioning error includes X-axis offset data, Y-axis offset data, Z-axis offset data, pitch angle deviation, yaw angle deviation, and roll angle deviation.

[0073] Step S532: Randomly select the first sample historical positioning error. If the proportion of any deviation value in the first sample historical positioning error that is less than the preset deviation threshold compared to other similar data in the sample historical positioning error set is greater than the preset proportion, then set the first sample historical positioning error as the high-frequency historical positioning error, and sequentially screen to obtain the high-frequency historical positioning error set.

[0074] Step S533: Calculate the mean value of the high-frequency historical positioning error set to obtain the historical positioning error.

[0075] Specifically, retrieve the positioning data of the same or similar molds in the same or similar processes to obtain the sample historical positioning error set. Each sample historical positioning error data includes X-axis offset data, Y-axis offset data, Z-axis offset data, pitch angle deviation, yaw angle deviation, and roll angle deviation. Among them, the X-axis offset data is the position offset of the mold in the X-axis direction; the Y-axis offset data is the position offset of the mold in the Y-axis direction; the Z-axis offset data is the position offset of the mold in the Z-axis direction; the pitch angle deviation is the inclination angle deviation of the mold around the X-axis direction; the yaw angle deviation is the rotation angle deviation of the mold around the Z-axis direction; the roll angle deviation is the rotation angle deviation of the mold around the Y-axis direction.

[0076] Randomly select the first sample historical positioning error from the sample historical positioning error set. Then, compare the deviation value of this first sample historical positioning error with the deviation amplitude in other similar data. If, among the deviation values of this sample, the proportion of the deviation amplitude in any direction that is less than the preset deviation threshold is greater than the preset proportion (for example, 80% of the deviation values are less than the set threshold), then consider this error sample as the high-frequency historical positioning error, that is, this error sample is a common positioning error. In the specific operation process, the deviation value between the first X-axis offset data of the first sample historical positioning error and other X-axis offset data in the sample historical positioning error set can be calculated first, and the proportion of the data less than the preset deviation threshold can be counted. If it is greater than the preset proportion, then continue to calculate the proportions of the Y-axis offset data, Z-axis offset data, pitch angle deviation, yaw angle deviation, and roll angle deviation in sequence; if all are greater than the preset proportion, then set the first sample historical positioning error as the high-frequency historical positioning error. Repeat the above screening process to screen out a series of high-frequency historical positioning error sets from the sample historical positioning error set, and these errors will be used as the reference data for the correction target.

[0077] Calculate the mean value of the filtered high-frequency historical positioning error set to obtain a representative historical positioning error value, which can reflect the average level of positioning error in the machining process of the same type of mold. This historical positioning error value provides an accurate target value for the secondary position adjustment of the primary corrected mold through the multi-axis adjustment device, helping to more precisely eliminate the historical positioning error and improve the machining accuracy of the mold.

[0078] In summary, the auxiliary positioning method in the multi-process switching of precision molds provided by the embodiments of the present application has the following beneficial effects: The embodiments of the present application achieve high-precision auxiliary positioning of precision molds during the multi-process switching process by integrating the dual mechanisms of real-time perception and historical data-driven. The real-time offset calculation of the mold pose is realized by using the process reference data and the three-dimensional grid model, and the dual-stage position adjustment strategy of current offset correction and historical error compensation significantly improves the positioning consistency and compensation accuracy during the mold process connection. It not only effectively avoids the error accumulation problem caused by the insufficient perception ability of traditional methods, but also enhances the adaptability to small deformations and displacements under complex working conditions, and finally achieves the technical effects of significantly improving the positioning accuracy of mold multi-process switching, enhancing the stability and reliability of multi-process switching, and improving the machining quality of the mold.

[0079] Embodiment 2, as Figure 3 shown, based on the same inventive concept as the foregoing Embodiment 1, the embodiments of the present application provide an auxiliary positioning system in the multi-process switching of precision molds, and the system includes: A positioning reference construction module 10, configured to collect the process reference information of the mold under the current process during mold machining and establish positioning reference data.

[0080] A mold information acquisition module 20, configured to use multi-source sensing devices to collect information about the mold and construct a three-dimensional mold grid model when the mold enters the process switching stage.

[0081] An offset calculation module 30, configured to use the three-dimensional mold grid model to perform grid comparison and offset calculation based on the positioning reference data to obtain real-time offset information.

[0082] Further, the real-time offset information includes coordinate position offset and attitude angle deviation.

[0083] A primary correction module 40, configured to control the auxiliary positioning device to perform a primary position adjustment on the mold with the goal of eliminating the real-time offset information to obtain a primary corrected mold.

[0084] The secondary correction module 50 is used to perform secondary position adjustment on the primary correction die through a multi-axis adjustment device according to the historical positioning error of the next adjacent process, complete the die position correction, and perform the die machining of the next adjacent process.

[0085] Furthermore, the positioning reference construction module 10 of the embodiment of the present application is further used to perform the following steps: Collect the reference points, reference planes, reference lines and process parameters of the die in the current process as the process reference information, where the process parameters include machining parameters and clamping parameters; establish the reference coordinate system and target attitude angle of the die based on the reference points, reference planes and reference lines; fine-tune the reference coordinate system and target attitude angle according to the machining parameters and clamping parameters to obtain the positioning reference data.

[0086] Furthermore, the die information acquisition module 20 of the embodiment of the present application is further used to perform the following steps: After the machining of the current process is completed, use an image sensor to collect images of the die at a preset acquisition angle to obtain a die image set, and perform an omnidirectional scan of the die through a laser scanner to obtain a die point cloud set; extract contour lines from the die image set and fuse them to construct a three-dimensional die contour; use the three-dimensional die contour to perform point cloud fitting on the die point cloud set to construct a three-dimensional die model; perform mesh division on the three-dimensional die model according to a preset mesh size to obtain the three-dimensional die mesh model.

[0087] Furthermore, the die information acquisition module 20 of the embodiment of the present application is further used to perform the following steps: Use the sample die image set and sample die contour set of the same type of die to train the convolutional neural network until convergence to obtain a contour line extractor; use the contour line extractor to extract contour lines from multiple die images in the die image set to obtain multiple die contours; based on the preset acquisition angle, perform feature registration and smoothing repair according to the multiple die contours to construct a three-dimensional die contour.

[0088] Furthermore, the die information acquisition module 20 of the embodiment of the present application is further used to perform the following steps: Use the three-dimensional die contour to perform a first random fitting on the die point cloud set to obtain a first point cloud fitting result, and calculate the ratio of the number of point clouds falling into the first point cloud fitting result to the total number of point clouds in the die point cloud set to obtain a first fitting accuracy; use the three-dimensional die contour again to perform a second random fitting on the die point cloud set to obtain a second point cloud fitting result and a second fitting accuracy; perform iterative fitting until a preset number of fittings is reached, output the point cloud fitting result with the maximum fitting accuracy as the optimal point cloud fitting result, and construct a three-dimensional die model according to the optimal point cloud fitting result.

[0089] Further, the system according to the embodiment of the present application is further configured to perform the following steps: During the processing of the current process, monitor and obtain the processing parameter sequence and the clamping parameter sequence; based on the standard processing parameters and standard clamping parameters of the current process, perform deviation analysis on the processing parameter sequence and the clamping parameter sequence to obtain the processing parameter deviation amplitude sequence and the clamping parameter deviation amplitude sequence; perform mean calculation and fluctuation analysis on the processing parameter deviation amplitude sequence, and output the mean value of the processing parameter deviation amplitude and the processing parameter fluctuation coefficient, where the fluctuation coefficient is the ratio of the standard deviation to the mean value; perform mean calculation and fluctuation analysis on the clamping parameter deviation amplitude sequence, and output the mean value of the clamping parameter deviation amplitude and the clamping parameter fluctuation coefficient; calculate the current processing deviation scale by weighted calculation based on the mean value of the processing parameter deviation amplitude, the processing parameter fluctuation coefficient, the mean value of the clamping parameter deviation amplitude, and the clamping parameter fluctuation coefficient; calculate the ratio of the preset processing deviation scale of the current process to the current processing deviation scale, and multiply it by the initial grid size to obtain the preset grid size.

[0090] Further, the secondary correction module 50 according to the embodiment of the present application is further configured to perform the following steps: Use the three-dimensional mold grid model to perform grid comparison and offset calculation, obtain the geometric shape deviation of the mold in the current process, and expand the geometric shape deviation according to the preset deviation tolerance interval to obtain the geometric shape deviation interval; use the geometric shape deviation interval and the next adjacent process as the retrieval constraints, retrieve the processing information of similar molds based on the industrial Internet, and determine the sample historical positioning error set; analyze and determine the historical positioning error according to the sample historical positioning error set, and take eliminating the historical positioning error as the goal, and perform secondary position adjustment on the primary corrected mold through the multi-axis adjustment device.

[0091] Further, the secondary correction module 50 according to the embodiment of the present application is further configured to perform the following steps: Obtain the sample historical positioning error set, where the sample historical positioning error includes X-axis offset data, Y-axis offset data, Z-axis offset data, pitch angle deviation, yaw angle deviation, and roll angle deviation; randomly select the first sample historical positioning error, if the proportion of any deviation value in the first sample historical positioning error that is less than the preset deviation threshold compared with other similar data in the sample historical positioning error set is greater than the preset proportion, then set the first sample historical positioning error as the high-frequency historical positioning error, and sequentially screen to obtain the high-frequency historical positioning error set; perform mean calculation on the high-frequency historical positioning error set to obtain the historical positioning error.

[0092] Through the foregoing detailed description of the auxiliary positioning method in the multi-process switching of precision molds, those skilled in the art can clearly understand the auxiliary positioning system in the multi-process switching of precision molds in this embodiment. For the system disclosed in Embodiment 2, since it corresponds to the method disclosed in Embodiment 1, it has corresponding functional modules and beneficial effects. For the relevant parts, reference can be made to the description in the method section.

[0093] The foregoing description of the disclosed embodiments enables those skilled in the art to implement or use the present application. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein can be implemented in other embodiments without departing from the spirit or scope of the present application. Therefore, the present application will not be limited to the embodiments shown herein, but rather to the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. An auxiliary positioning method in multi-process switching of precision molds, characterized in that, The method includes: When machining a mold, collecting the process reference information of the mold in the current process and establishing positioning reference data; When the mold enters the process switching stage, using a multi-source sensing device to collect information about the mold and constructing a three-dimensional mold mesh model; Based on the positioning reference data, using the three-dimensional mold mesh model to perform mesh comparison and offset calculation to obtain real-time offset information; Aiming to eliminate the real-time offset information, controlling an auxiliary positioning device to perform a primary position adjustment on the mold to obtain a once-corrected mold; According to the historical positioning error of the next adjacent process, using a multi-axis adjustment device to perform a secondary position adjustment on the once-corrected mold to complete the mold position correction and execute the mold machining of the next adjacent process.

2. The auxiliary positioning method in the multi-process switching of the precision mold according to claim 1, wherein, Collecting the process reference information of the mold in the current process and establishing positioning reference data includes: Collecting the reference points, reference planes, reference lines and process parameters of the mold in the current process as the process reference information, where the process parameters include machining parameters and clamping parameters; Based on the reference points, reference planes and reference lines, establishing a reference coordinate system and a target attitude angle of the mold; Fine-tuning the reference coordinate system and the target attitude angle according to the machining parameters and clamping parameters to obtain the positioning reference data.

3. The auxiliary positioning method in the multi-process switching of the precision mold according to claim 1, characterized in that, Using a multi-source sensing device to collect information about the mold and constructing a three-dimensional mold mesh model includes: After the machining in the current process is completed, using an image sensor to collect images of the mold at a preset collection angle to obtain a set of mold images, and using a laser scanner to perform an omnidirectional scan of the mold to obtain a point cloud set of the mold; Extracting contour lines according to the set of mold images and fusing them to construct a three-dimensional mold contour; Using the three-dimensional mold contour to perform point cloud fitting on the point cloud set of the mold to construct a three-dimensional mold model; Dividing the three-dimensional mold model into grids according to a preset grid size to obtain the three-dimensional mold mesh model.

4. The auxiliary positioning method in the multi-process switching of the precision mold according to claim 3, characterized in that Extracting contour lines according to the set of mold images and fusing them to construct a three-dimensional mold contour includes: Using a sample mold image set and a sample mold contour set of the same type of mold to train a convolutional neural network until convergence to obtain a contour line extractor; Using the contour line extractor to extract contour lines from multiple mold images in the set of mold images respectively to obtain multiple mold contours; Based on the preset collection angle, performing feature registration and smoothing repair according to the multiple mold contours to construct a three-dimensional mold contour.

5. The auxiliary positioning method in multi-process switching of precision molds according to claim 3, wherein Using the three-dimensional mold contour to perform point cloud fitting on the point cloud set of the mold to construct a three-dimensional mold model includes: Using the three-dimensional mold contour to perform a first random fitting on the point cloud set of the mold to obtain a first point cloud fitting result, and calculating the ratio of the number of point clouds falling into the first point cloud fitting result to the total number of point clouds in the point cloud set of the mold to obtain a first fitting accuracy; Using the three-dimensional mold contour again to perform a second random fitting on the point cloud set of the mold to obtain a second point cloud fitting result and a second fitting accuracy; Perform iterative fitting until the preset number of fitting times is reached. Output the point cloud fitting result with the maximum fitting accuracy as the optimal point cloud fitting result, and construct a three-dimensional mold model based on the optimal point cloud fitting result.

6. The auxiliary positioning method in multi-process switching of precision molds according to claim 3, characterized in that, The method for generating the preset grid size includes: During the processing of the current process, monitor and obtain the processing parameter sequence and the clamping parameter sequence; Based on the standard processing parameters and standard clamping parameters of the current process, perform deviation analysis on the processing parameter sequence and the clamping parameter sequence to obtain the processing parameter deviation amplitude sequence and the clamping parameter deviation amplitude sequence; Perform mean calculation and fluctuation analysis on the processing parameter deviation amplitude sequence, and output the mean of the processing parameter deviation amplitude and the processing parameter fluctuation coefficient, where the fluctuation coefficient is the ratio of the standard deviation to the mean; Perform mean calculation and fluctuation analysis on the clamping parameter deviation amplitude sequence, and output the mean of the clamping parameter deviation amplitude and the clamping parameter fluctuation coefficient; Weightedly calculate the current processing deviation scale based on the mean of the processing parameter deviation amplitude, the processing parameter fluctuation coefficient, the mean of the clamping parameter deviation amplitude, and the clamping parameter fluctuation coefficient; Calculate the ratio of the preset processing deviation scale of the current process to the current processing deviation scale, and multiply it by the initial grid size to obtain the preset grid size.

7. The auxiliary positioning method in the multi-process switching of the precision mold according to claim 1, wherein The real-time offset information includes coordinate position offset and attitude angle deviation.

8. The auxiliary positioning method in multi-process switching of precision molds according to claim 1, characterized in that According to the historical positioning error of the next adjacent process, perform secondary position adjustment on the primary corrected mold through a multi-axis adjustment device, including: Use the three-dimensional mold grid model for grid comparison and offset calculation to obtain the geometric shape deviation of the mold in the current process, and expand the geometric shape deviation according to the preset deviation tolerance interval to obtain the geometric shape deviation interval; Using the geometric shape deviation interval and the next adjacent process as retrieval constraints, retrieve the processing information of similar molds based on the industrial Internet to determine the sample historical positioning error set; Analyze and determine the historical positioning error according to the sample historical positioning error set, and perform secondary position adjustment on the primary corrected mold through a multi-axis adjustment device with the goal of eliminating the historical positioning error.

9. The auxiliary positioning method in the multi-process switching of the precision mold according to claim 8, characterized in that, Analyze and determine the historical positioning error according to the sample historical positioning error set, including: Obtain the sample historical positioning error set, where the sample historical positioning error includes X-axis offset data, Y-axis offset data, Z-axis offset data, pitch angle deviation, yaw angle deviation, and roll angle deviation; Randomly select the first sample historical positioning error. If the proportion of any deviation value in the first sample historical positioning error that is less than the preset deviation threshold compared to other similar data in the sample historical positioning error set is greater than the preset proportion, then set the first sample historical positioning error as the high-frequency historical positioning error, and sequentially screen to obtain the high-frequency historical positioning error set; Perform mean calculation on the high-frequency historical positioning error set to obtain the historical positioning error.

10. Auxiliary positioning system in multi-process switching of precision molds, characterized in that, The system is used to execute the auxiliary positioning method in the multi-process switching of precision molds according to any one of claims 1-9, including: A positioning reference construction module, which is used to collect the process reference information of the mold in the current process and establish positioning reference data during mold processing; The mold information acquisition module is used to collect mold information by using multi-source sensing devices when the mold enters the process switching stage, and construct a three-dimensional mold grid model; The offset calculation module is used to perform grid comparison and offset calculation by using the three-dimensional mold grid model based on the positioning reference data, and obtain real-time offset information; The primary correction module is used to control the auxiliary positioning device to perform a primary position adjustment on the mold with the goal of eliminating the real-time offset information, and obtain a primary corrected mold; The secondary correction module is used to perform a secondary position adjustment on the primary corrected mold through a multi-axis adjustment device according to the historical positioning error of the next adjacent process, complete the mold position correction, and perform the mold processing of the next adjacent process.

Images