A mold wear early warning method, device, equipment and medium
By using second-order difference and nonlinear fitting to monitor the time series of burr height in real time, multi-level early warning prompts are generated, solving the problems of accuracy and timeliness in monitoring mold wear status and ensuring production safety and efficiency.
Patent Information
- Application Number
- CN202611379384.1
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2026-09-07
- Publication Date
- 2026-11-06
AI Technical Summary
Existing technologies have low accuracy and poor timeliness in monitoring mold wear, making it difficult to provide early warnings of quality and safety hazards caused by mold wear.
By acquiring product images, calculating the second-order difference and fitting residual of the burr height time series, and combining it with a nonlinear model, the wear state transformation is monitored in real time, generating multi-level early warning prompts.
It enables early warning of mold wear, avoids sudden quality accidents and equipment damage, and improves production safety and efficiency.
Smart Images
Figure CN123051925A_ABST
Abstract
Description
Technical Field
[0001] This application relates to the field of mold health status monitoring technology, specifically to a mold wear early warning method, device, equipment, and medium. Background Technology
[0002] In the metal stamping production process, dies are the core process equipment that determines product quality, production efficiency, and cost. During long-term, high-frequency service, the cutting edges of the punch and die inevitably wear due to cyclical impacts, shearing, and friction. With increasing stamping cycles, the cutting edges of the punch and die gradually dull, leading to a decline in the quality of the blanked section. One of the most significant characteristics is the gradual increase in burr height. Excessive burrs not only affect the dimensional accuracy and appearance quality of the product but may also cause difficulties in subsequent assembly, injuries to personnel, and in severe cases, even equipment jamming or electrical short circuits due to burr detachment, posing safety and quality hazards. To address the problem of monitoring die wear, relevant technologies typically employ methods such as periodic shutdown for manual inspection, measurement based on indirect physical quantities such as stamping force / sound / vibration, and machine vision surface inspection. However, these traditional methods suffer from low accuracy and poor timeliness. Summary of the Invention
[0003] In view of the shortcomings of the prior art described above, the purpose of this application is to provide a mold wear early warning method, device, equipment and medium to solve at least one of the above-mentioned defects.
[0004] Firstly, this application provides a mold wear early warning method, comprising: The image of the product produced by the same mold in each production cycle is obtained, and edge detection is performed on the image to obtain the actual outline of the product. The first burr height value is obtained based on the difference between the theoretical contour line of the product and the actual contour line of the product. The local rate of change of burr height over time is monitored in real time to identify the wear acceleration inflection point of the wear state transition, and the accelerated wear stage of the mold is determined based on the wear acceleration inflection point; wherein, the burr height over time includes the first burr height value stored in chronological order. A nonlinear fitting model was used to perform trend analysis on the second burr height value in the accelerated wear stage, and the trend analysis results were obtained. Wear and tear warnings are generated based on the trend analysis results.
[0005] This application constructs a burr height time series and uses inflection point monitoring and stage division to divide the wear process into an initial wear stage and an accelerated wear stage. For the accelerated wear stage, a nonlinear model is used for fitting, thereby enabling the microscopic deterioration trend of the wear state to be captured in advance before the mold experiences severe wear, which basically avoids sudden quality accidents and mold damage, and significantly advances the warning period.
[0006] In one embodiment, the method for determining the wear acceleration inflection point includes: Calculate the second difference of the first burr height value corresponding to the initial time period in the burr height time series; when the second difference of multiple consecutive first burr height values is greater than the first preset threshold, one of the multiple consecutive first burr height values is taken as the wear acceleration inflection point.
[0007] This application utilizes the sensitivity of second-order difference to wear acceleration to quickly respond to sudden changes in wear rate, thereby achieving real-time and reliable detection of wear acceleration inflection points.
[0008] In one embodiment of this application, the method for determining the wear acceleration inflection point includes: Linear regression fitting was performed on the first burr height value corresponding to the initial time period in the burr height time series to establish an initial linear model; Based on the initial linear model, the fitting residuals of multiple consecutive first burr height values are calculated to form a fitting residual sequence. When the fitted residual sequence is monotonically increasing and the fitted residual of multiple consecutive first burr height values is greater than the second preset threshold, one of the multiple consecutive first burr height values is taken as the wear acceleration inflection point.
[0009] Based on the initial linear model, this application achieves real-time and robust detection of wear acceleration inflection points by combining the monotonicity of the fitted residual sequence with a threshold criterion.
[0010] In one embodiment of this application, a nonlinear fitting model is used to perform trend analysis on the second burr height value in the accelerated wear stage, including: The second burr height value of the accelerated wear stage is simultaneously input into a preset exponential growth model and a power function model for parallel fitting; Calculate the first goodness of fit of the exponential growth model and the second goodness of fit of the power function model, respectively; Based on the magnitudes of the first and second goodness-of-fit scores, the target nonlinear fitting model is determined. Trend analysis is performed on the time series of burr height based on the target nonlinear fitting model.
[0011] This application utilizes a dual-model parallel fitting and dynamic optimization mechanism to adaptively fit the nonlinear degradation law under different working conditions, thereby improving the accuracy of mold wear early warning.
[0012] In one embodiment of this application, generating a wear warning alert based on the trend analysis results includes: Based on the trend characteristic parameters of the target nonlinear fitting model and the burr height threshold, the remaining life of the mold is predicted to obtain the predicted value of the remaining life of the mold. A wear warning is generated based on the predicted remaining life of the mold, the trend characteristic parameters, the wear acceleration inflection point, and the second burr height value.
[0013] This application extracts trend characteristic parameters from a nonlinear model, combines them with preset thresholds to infer remaining lifespan, and generates multi-dimensional early warning prompts, achieving quantitative assessment and visual management of mold wear status. This method avoids the subjectivity of manual experience-based judgment, provides a scientific basis for mold replacement on the production line, and effectively reduces the risk of downtime due to sudden mold failure.
[0014] In one embodiment of this application, a wear warning is generated based on the predicted remaining life of the mold, the trend characteristic parameters, the wear acceleration inflection point, and the second burr height value, including: If the wear acceleration inflection point is detected, a Level 1 warning alert will be generated; If the trend characteristic parameter is greater than the third preset threshold, or the predicted value of the remaining life of the mold is lower than the fourth preset threshold, a level 2 action warning prompt will be generated. If the second burr height value exceeds the quality threshold, or the predicted remaining life of the mold reaches the fifth preset threshold, a level three emergency warning will be generated.
[0015] Secondly, this application provides a mold wear early warning device, comprising: The detection module is used to acquire images of products produced by the same mold in each production cycle, and to perform edge detection on the images to obtain the actual outline of the product. An extraction module is used to obtain a first burr height value based on the theoretical contour line and the actual contour line of the product. The wear stage determination module is used to monitor the local change rate of the burr height time series in real time to identify the wear acceleration inflection point of the wear state transition, and determine the accelerated wear stage of the mold based on the wear acceleration inflection point; wherein, the burr height time series includes the first burr height value stored in chronological order. The trend analysis module is used to perform trend analysis on the second burr height value in the accelerated wear stage using a nonlinear fitting model, and obtain the trend analysis results. The early warning generation module is used to generate wear warnings based on the trend analysis results.
[0016] Thirdly, this application provides a processing equipment equipped with the aforementioned mold wear early warning device.
[0017] Fourthly, this application provides an electronic device, comprising: One or more processors; and A memory for storing one or more programs that, when executed by one or more processors, cause the processors to implement the method.
[0018] Fifthly, this application provides a machine-readable medium having instructions stored thereon that, when executed by one or more processors, cause the processors to perform the method described thereon.
[0019] It should be understood that the above general description and the following detailed description are exemplary and explanatory only, and do not limit this application. Attached Figure Description
[0020] To more clearly illustrate the technical solutions of the embodiments of this application, the accompanying drawings used in the description of the embodiments of this application will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.
[0021] Figure 1 This is a schematic flowchart of a mold wear early warning method according to an exemplary embodiment of this application; Figure 2 This is a flowchart illustrating a method for determining the wear acceleration inflection point in an exemplary embodiment of this application; Figure 3 This is a schematic diagram of burr height time series and inflection point detection in an exemplary embodiment of this application; Figure 4 This is a schematic diagram of an exemplary embodiment of the present application, illustrating the trend analysis of a time series of burr height using a nonlinear fitting model. Figure 5 This is a schematic diagram of an exemplary embodiment of the present application illustrating the process of generating wear warning prompts based on trend analysis results; Figure 6 This is a schematic diagram of a mold wear early warning device according to an exemplary embodiment of this application; Figure 7 A schematic diagram of a computer system suitable for implementing the memory of the embodiments of this application is shown. Detailed Implementation
[0022] The embodiments of this application will be described below with reference to the accompanying drawings and preferred embodiments. Those skilled in the art can easily understand other advantages and effects of this application from the content disclosed in this specification. This application can also be implemented or applied through other different specific embodiments, and various details in this specification can also be modified or changed based on different viewpoints and applications without departing from the spirit of this application. It should be understood that the preferred embodiments are only for illustrating this application and are not intended to limit the scope of protection of this application.
[0023] It should be noted that the illustrations provided in the following embodiments are only schematic representations of the basic concept of this application. Therefore, the drawings only show the components related to this application and are not drawn according to the actual number, shape and size of the components in the actual implementation. In the actual implementation, the shape, quantity and proportion of each component can be arbitrarily changed, and the layout of the components may also be more complex.
[0024] In the following description, numerous details are explored to provide a more thorough explanation of embodiments of the present application. However, it will be apparent to those skilled in the art that embodiments of the present application may be practiced without these specific details. In other embodiments, well-known structures and devices are shown in block diagram form rather than in detail to avoid obscuring embodiments of the present application.
[0025] Please see Figure 1 , Figure 1 This is a schematic flowchart of a mold wear early warning method according to an exemplary embodiment of this application. The mold wear early warning method includes at least steps S110 to S160, which are described in detail below: Step S110: Obtain images of products produced by the same mold in each production cycle; In this step, at a fixed station on the production line, an industrial camera is used to capture images of the critical edge areas of the products produced in each production cycle, thus obtaining product images. The fixed station is typically deployed on the exit side of the stamping production line. When the processing equipment completes a stamping operation and the robotic arm delivers the workpiece to the designated position on the conveyor belt, a trigger signal is sent to control the industrial camera and light source to flash synchronously, acquiring high-definition images online.
[0026] The acquired images are typically color images containing RGB three channels. A series of preprocessing operations are performed on the images to eliminate environmental noise such as mechanical vibration, uneven lighting, and oil stain reflection in the industrial site, and to significantly enhance the burr characteristics.
[0027] Step S120: Based on the difference between the theoretical outline of the product and the actual outline of the product, obtain the first burr height value; In this step, an edge detection algorithm is invoked to perform pixel-level coarse edge localization within the Region of Interest (ROI). To overcome the limitations of the camera's physical pixel resolution, a sub-pixel precision edge localization algorithm can be employed. By performing polynomial fitting or spline interpolation on the grayscale values in the neighborhood of the coarsely located edge points, sub-pixel-level edge coordinates are accurately calculated, thereby precisely locating the actual contour line of the stamped part. Simultaneously, based on the standard boundary of the processing equipment in its initial, wear-free state during service, the theoretical contour line of the stamped part is determined. The vertical distance from each point on the actual contour line to the theoretical contour line is calculated, and the maximum value among these distances is defined as the burr height value for that sampling. h i That is, the height value of the first burr; where, i This is the current sampling sequence number, corresponding to the cumulative number of production strokes or time points.
[0028] Step S130: Real-time monitoring of the local change rate of the burr height time series to identify the wear acceleration inflection point of the wear state transition, and determination of the accelerated wear stage of the mold based on the wear acceleration inflection point; wherein, the burr height time series includes the first burr height value stored in chronological order. Please see Figure 3 , Figure 3 This is a schematic diagram of glitch height time series and inflection point detection in an exemplary embodiment of this application; from Figure 3 It can be seen that the degradation curve of the mold throughout its service life exhibits distinct phased physical characteristics. In the initial wear stage: burr height fluctuates smoothly and increases extremely slowly, almost linearly, corresponding to the relatively flat straight line on the left side of the graph; once the inflection point (point A) is crossed, i.e., in the accelerated wear stage: burr height growth accelerates significantly (the curve becomes steeper), corresponding to the rapidly rising curve on the right side of the graph. In the mold's lifespan, initial wear is slow and controllable; once it enters the accelerated stage, the remaining lifespan is rapidly shortened, requiring timely warning or mold replacement.
[0029] For example, the transition point of the physical state can be accurately identified by monitoring the local rate of change of the spur height time series in real time (i.e., the changing trend of the first and second derivatives).
[0030] As a specific implementation method, the second-order difference (i.e., the growth acceleration) of multiple consecutive first burr height values is calculated, or the residual change trend of the linear fitting model in the initial stage is monitored. When the second-order difference value is detected to be significantly greater than a first preset threshold, or the fitting residual of the linear model continues to increase and exceeds a second preset threshold, it is determined that the mold wear has crossed the critical state and entered the accelerated degradation period. This state transition point is marked as the wear acceleration inflection point. In this step, once the wear acceleration inflection point is identified, the accelerated wear stage is determined based on this wear acceleration inflection point (the stage after the wear acceleration inflection point is defined as the accelerated wear stage).
[0031] For the burr height time series, after extracting the first burr height value of the current stamped part, it is not treated in isolation. Instead, it is treated as the latest burr height value and appended to the historical burr height data queue according to the production time sequence (or the cumulative stamping number sequence), thereby dynamically constructing and updating the burr height time series. ,in, This represents the current cumulative total number of samples. While constructing the sequence, this sequence is associated and bound with the current mold unique identifier and production batch information, and stored in a structured database. The mold unique identifier is used to uniquely identify the currently serving mold number, ensuring the continuity and consistency of historical degradation data; the production batch information is used to identify the current production work order and batch switchover point.
[0032] Step S140: A nonlinear fitting model is used to perform trend analysis on the second burr height value in the accelerated wear stage to obtain the trend analysis results. Once the mold wear is determined to have entered the accelerated wear stage, the original linear model can no longer accurately describe the degradation pattern of the data. At this point, the system automatically switches to a nonlinear fitting model library, inputting the second burr height value within the accelerated wear stage into the nonlinear fitting model for dynamic fitting to complete trend analysis. To adaptively fit the accelerated wear patterns under different working conditions and mold materials, an exponential growth model and a power function model are used for parallel fitting. By comparing the determination coefficients of the two, the model with higher goodness of fit and best conforming to the current actual degradation physical laws is dynamically selected as the preferred nonlinear prediction model.
[0033] Step S150: Generate a wear warning prompt based on the trend analysis results.
[0034] In this step, the trend analysis results may include trend characteristic parameters, predicted mold remaining life, etc. Wear warning prompts are generated based on the trend characteristic parameters and predicted mold remaining life.
[0035] In one embodiment, the acquired image is preprocessed, including grayscale conversion, filtering and noise reduction, image enhancement, and edge ROI region extraction.
[0036] Grayscale conversion involves converting the acquired three-channel color image into a single-channel grayscale image to remove redundant color information and significantly reduce the computational cost of subsequent algorithms.
[0037] Filtering and noise reduction is a crucial step in eliminating high-frequency noise in images. As a specific implementation method, filtering and noise reduction employs Gaussian filtering, median filtering, or bilateral filtering algorithms. Bilateral filtering, in particular, effectively preserves edge information while filtering out noise, avoiding the edge blurring problems caused by traditional filtering algorithms. When using Gaussian filtering, fixed filtering parameters are not used; instead, the size of the Gaussian kernel is adaptively adjusted according to the image resolution. When the image resolution is high, the size of the Gaussian kernel is adaptively increased to ensure that a large range of random noise is effectively smoothed; when the image resolution is low, the size of the Gaussian kernel is adaptively decreased to avoid over-smoothing that would erase microscopic burr features. The standard deviation of the Gaussian kernel is positively correlated with the image resolution; through dynamic adjustment, an optimal balance is achieved between smoothing noise and preserving the microscopic burr details of the product edges.
[0038] Image enhancement uses histogram equalization or contrast-limited adaptive histogram equalization (CLAHE) algorithms to locally improve the contrast of the product edge area, making the gray-scale transition between burrs and the background steeper, which is beneficial to improving the sub-pixel positioning accuracy of edge detection.
[0039] Edge ROI extraction is based on the prior geometric position of the workpiece, automatically cropping out the local region containing the edge to be measured from the image. By limiting subsequent edge detection and height calculation to the ROI region, not only can stray interference from non-detection areas of the workpiece be completely eliminated, but the algorithm's running time can also be compressed to the millisecond level, fully meeting the high-speed cycle requirements of stamping production lines.
[0040] In one embodiment, the method for determining the wear acceleration inflection point includes: calculating the second difference of the first burr height value corresponding to the initial time period in the burr height time series; when the second difference of multiple consecutive first burr height values is greater than a first preset threshold, one of the multiple consecutive first burr height values is taken as the wear acceleration inflection point.
[0041] The initial time period refers to a continuous sampling interval after the mold is put into use and before wear has accelerated (e.g., the first 20-30 first burr height values).
[0042] The initial time period can be a part of the initial wear stage. The transition from the initial wear stage to the accelerated wear stage of the mold is mainly caused by surface microstructure instability and fatigue crack generation, which is the most critical critical state transition point in mold life prediction. In order to accurately capture this physical transition before the burr height increases significantly, the embodiments of this application propose a second-order difference verification mechanism.
[0043] Second-order difference Mathematically, this represents the acceleration of the spike's rapid increase. In discrete time series, the first difference... The second difference represents the increase in burr height (i.e., wear rate) between two adjacent samples. This represents the rate of change of the growth (i.e., the wear acceleration). and These are the first burr height values from the previous and the two previous samples, respectively.
[0044] In the initial wear stage, since the wear rate is constant, the first-order difference remains essentially constant, therefore the second-order difference value... It fluctuates slightly near zero. Once the microstructure of the die cutting edge becomes unstable, the wear rate begins to rise sharply, and the second-order difference... It will rise rapidly.
[0045] Meanwhile, to prevent instantaneous exceedances of the second-order difference due to sporadic noise, continuous... The second-order difference values of each first burr height value (e.g., 5 first burr height values) all satisfy the following: (First preset threshold) Determine that the mold wear has entered an accelerated state, and record one of the multiple consecutive first burr height values as the wear acceleration inflection point.
[0046] As a specific implementation method, the first of a series of first burr height values is recorded as the wear acceleration inflection point.
[0047] This application utilizes the sensitivity of second-order difference to wear acceleration to quickly respond to sudden changes in wear rate, thereby achieving real-time and reliable detection of wear acceleration inflection points.
[0048] Please see Figure 2 , Figure 2 This is a flowchart illustrating a method for determining the wear acceleration inflection point according to an exemplary embodiment of this application. The method for determining the wear acceleration inflection point includes at least steps S210 to S230, which are described in detail below: Step S210: Perform linear regression fitting on the first burr height value corresponding to the initial time period in the burr height time series to establish an initial linear model; Linear regression was performed using the least squares method on the first spur height value within the initial time period to obtain a linear model: This model is used to describe the linear growth law of burr height during the normal wear stage.
[0049] Where α is the slope (characterizing the initial wear rate) and β is the intercept (representing the initial burr height baseline). t This is the time sequence number.
[0050] As a specific implementation method, after the initial linear model fitting is completed, the determination coefficients of the linear model can be calculated. To evaluate the quality of the fit and the linearity of the data, the coefficient of determination is used. The calculation method is as follows:
[0051] in, For the first The first burr height value of the next sample. For the linear model in the th Predicted value of spur height at each time point. This is the average of the height values of all first glitches within the current sliding window. This represents the total number of burr height values within the sliding window. (Determination coefficient) The closer the value is to 1, the more linearly the data follows a distribution, and the higher the data stability. The linear model established at this stage will serve as a benchmark for subsequent judgments on whether the mold has undergone a qualitative change in its wear state.
[0052] In the initial fitting, a sliding window is used for data extraction, with the window size set to 50 to 200 production cycles. If the window is too small (e.g., less than 50), it is highly susceptible to interference from accidental noise in a single measurement, leading to an inconsistency in the fitted slope. Severe fluctuations can trigger false alarms; if the window is too large (e.g., greater than 200), it will cause a serious lag in the response to changes in mold wear status and increase the calculation time.
[0053] As production continues and the sliding window moves along the time axis, each time a new glitch height value is added for a new production cycle, the oldest historical data point within the current window is automatically removed, and linear regression fitting is immediately re-performed. This first-in-first-out queue mechanism dynamically updates the slope of the linear model. and intercept It tracks the latest trend of mold wear in real time with extremely low computational overhead, effectively filtering out the interference of long-term cumulative drift on the current state assessment.
[0054] Step S220: Based on the initial linear model, calculate the fitting residuals of multiple consecutive first burr height values to form a fitting residual sequence; For each first spur height value collected after the initial time period H k Calculate the fitting residuals based on the linear model. e k .
[0055]
[0056] All calculated residuals are arranged in chronological order to form the fitted residual sequence. .
[0057] Step S230: When the fitted residual sequence is monotonically increasing and the fitted residual of multiple consecutive first burr height values is greater than the second preset threshold, one of the multiple consecutive first burr height values is taken as the wear acceleration inflection point.
[0058] The second preset threshold can be set based on the standard deviation of the residuals within the initial time period. The specific determination rule is as follows: Monotonically increasing: For continuous L Each burr height value (e.g., L ≥3), satisfying ; Furthermore, the residual of each first burr height value is greater than the second preset threshold.
[0059] When the above conditions are met for the first time, it will continue L The first of the initial burr height values is identified as the wear acceleration inflection point. This wear acceleration inflection point marks the end of the initial linear wear phase and the beginning of the accelerated wear phase.
[0060] Based on the initial linear model, this application achieves real-time and robust detection of wear acceleration inflection points by combining the monotonicity of the fitted residual sequence with a threshold criterion.
[0061] Please see Figure 4 , Figure 4 This is a schematic flowchart illustrating an exemplary embodiment of this application, demonstrating the use of a nonlinear fitting model to perform trend analysis on the second burr height value during the accelerated wear stage. The trend analysis of the second burr height value during the accelerated wear stage using a nonlinear fitting model includes at least steps S410 to S430, detailed below: Step S410: The second burr height value of the accelerated wear stage is simultaneously input into the preset exponential growth model and power function model for parallel fitting; Once it is determined that the die wear has passed the wear acceleration inflection point and entered the accelerated wear stage, the burr height will exhibit a strong nonlinear growth. However, for different die materials and different stamping conditions, the subsequent nonlinear acceleration law varies physically. Therefore, a single nonlinear model cannot adaptively fit all conditions, easily leading to significant deviations in later trend fitting and remaining life prediction. Based on this, this embodiment provides a dual-model parallel fitting and dynamic optimization mechanism. After identifying the accelerated wear stage, dual-thread parallel computation is initiated, using the nonlinear least squares method to solve for the undetermined coefficients of the exponential growth model and the power function model, respectively.
[0062] Step S420: Calculate the first goodness of fit of the exponential growth model and the second goodness of fit of the power function model, respectively. For the exponential growth model:
[0063] By taking the logarithm of both sides of the equation, it can be transformed into a linear relationship. Then, the fitting coefficients are solved using the linear least squares method. and exponential growth rate parameter .
[0064] For the power function model:
[0065] Similarly, by taking the logarithm, it is transformed into Then, the fitting coefficients can be solved. Power parameter .
[0066] After solving for the parameters, the goodness of fit of the two models during the accelerated wear stage is calculated, i.e., the first goodness of fit of the exponential growth model. (Determination coefficient) and the second goodness of fit of the power function model (Determination coefficient).
[0067] Step S430: Determine the target nonlinear fitting model based on the magnitudes of the first and second goodness-of-fit scores.
[0068] By comparing the values of the first and second goodness-of-fit scores in real time, the model with the larger goodness-of-fit score is dynamically selected as the target nonlinear prediction model for the current working condition.
[0069] Step S440: Perform trend analysis on the burr height time series based on the target nonlinear fitting model.
[0070] This application utilizes a dual-model parallel fitting and dynamic optimization mechanism to adaptively fit the nonlinear degradation law under different working conditions, thereby improving the accuracy of mold wear early warning.
[0071] Please see Figure 5 , Figure 5 This is a schematic flowchart illustrating an exemplary embodiment of this application of generating a wear warning alert based on trend analysis results. The generation of the wear warning alert based on trend analysis results includes at least steps S510 to S520, which are described in detail below: Step S510: The trend characteristic parameters of the target nonlinear fitting model and the burr height threshold are used to predict the remaining life of the mold, and the predicted value of the remaining life of the mold is obtained. After determining the target nonlinear fitting model, the characteristic parameters of the target nonlinear fitting model are extracted. Trend characteristic parameters refer to key values extracted from the preferred nonlinear model that can quantitatively describe the wear acceleration trend. If the preferred model is an exponential growth model... The trend characteristic parameter is the exponential growth rate. (Relative growth rate per unit time); if the preferred model is a power function model The trend characteristic parameter is the power exponent. (Reflects the degree of nonlinearity of wear acceleration).
[0072] As a specific implementation method, after completing the nonlinear fitting, the data is directly read from the fitting results. or The values of are recorded, along with their confidence intervals or standard errors, to assess the reliability of the parameter estimates.
[0073] Wherein, the burr height threshold is a preset quality threshold (denoted as ). Predicting the remaining life of the mold is done through inverse function derivation, predicting when the burr height will reach the quality threshold. Cumulative number of times .
[0074] If the preferred model is an exponential growth model, then the cumulative number of impulses when the quality threshold is reached is predicted. The derivation formula for the inverse function is:
[0075] If the preferred model is a power function model, then the cumulative number of impulses at which the quality threshold is reached is predicted. The derivation formula for the inverse function is:
[0076] Finally, the remaining lifespan of the mold is calculated. That is, the predicted mold life:
[0077] in, This represents the current cumulative number of attempts.
[0078] Step S520: Generate a wear warning prompt based on the predicted remaining life of the mold, trend characteristic parameters, wear acceleration inflection point, and second burr height value.
[0079] The wear warning system is a comprehensive output that provides operators or automated systems with an intuitive assessment of the mold's health status. Warnings can be displayed on a human-machine interface, triggered by audible and visual alarms, or transmitted to a remote monitoring system to facilitate timely maintenance.
[0080] This application extracts trend characteristic parameters from a nonlinear model, combines them with preset thresholds to infer remaining lifespan, and generates multi-dimensional early warning prompts, achieving quantitative assessment and visual management of mold wear status. This method avoids the subjectivity of manual experience-based judgment, provides a scientific basis for mold replacement on the production line, and effectively reduces the risk of downtime due to sudden mold failure.
[0081] In one embodiment, a wear warning is generated based on the predicted remaining life of the mold, trend characteristic parameters, wear acceleration inflection point, and second burr height value, including: If a wear acceleration inflection point is detected, a Level 1 warning is generated; if the trend characteristic parameter is greater than the third preset threshold, or the predicted remaining life of the mold is lower than the fourth preset threshold, a Level 2 action warning is generated; if the second burr height exceeds the quality threshold, or the predicted remaining life of the mold reaches the fifth preset threshold, a Level 3 emergency warning is generated.
[0082] In practice, traditional mold early warning systems often rely on a single, fixed threshold for judgment, immediately halting production once burrs exceed the limit. This not only leads to frequent unplanned shutdowns of the production line, severely disrupting continuous production cycles, but may also result in large quantities of scrap due to delayed detection. Therefore, this application provides a multi-level early warning mechanism. Specifically, it includes: During operation in the initial wear stage, a level one warning is triggered once a potential wear acceleration inflection point is detected. At this time, the actual extracted first burr height value is still far below the quality threshold. The product quality is fully qualified. However, the triggering of the Level 1 warning can promptly remind process personnel that the mold has ended its stable wear period and begun to enter the wear transition period, which requires close attention in subsequent production.
[0083] During the accelerated wear phase, the trend characteristic parameters (exponential growth rate) of the preferred nonlinear prediction model are continuously evaluated. or power index ) and the predicted remaining life of the mold .
[0084] If the trend characteristic parameter is greater than the third preset threshold (indicating that wear is accelerating rapidly) (for example, λ is greater than the preset threshold), , p Greater than the preset threshold ), or the predicted remaining life of the mold. Below the preset number of safety periods (like When a certain number of cycles occur, a Level 2 action warning is triggered. During this stage, the system communicates with the MES system via industrial Ethernet to automatically push planned mold repair time windows. This allows production schedulers to utilize shift changes, maintenance, or mold replacement breaks to schedule production in advance and prepare mold parts, achieving predictive maintenance and largely avoiding unplanned downtime.
[0085] When the height of the second burr has exceeded the quality threshold or the predicted remaining lifespan Reaching or falling below the preset number of dangerous cycles (like Strike, and When this occurs, the highest level, Level 3, emergency warning is triggered. At this stage, it indicates that the local condition of the die cutting edge has deteriorated rapidly, and a large number of scrap products or die damage may occur at any time. An emergency stop signal is immediately issued, and the stamping line is stopped via the emergency stop interlock interface, thus preventing the generation of more defective products to a certain extent.
[0086] It should be understood that the sequence number of each step in the above embodiments does not imply the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of this application.
[0087] Figure 6 This is a schematic diagram of a mold wear early warning device according to an exemplary embodiment of this application. Figure 6 As shown, a mold wear early warning device includes: The detection module 610 is used to acquire images of products produced by the same mold in each production cycle, and to perform edge detection on the images to obtain the actual outline of the product. Extraction module 620 is used to obtain a first burr height value based on the difference between the theoretical contour line of the product and the actual contour line of the product. Wear stage determination module 630 is used to monitor the local change rate of burr height time series in real time to identify the wear acceleration inflection point of wear state transition, and determine the accelerated wear stage of the mold based on the wear acceleration inflection point; wherein, the burr height time series includes a first burr height value stored in chronological order. The trend analysis module 640 is used to perform trend analysis on the second burr height value in the accelerated wear stage using a nonlinear fitting model, and obtain the trend analysis results. The early warning generation module 650 is used to generate wear warnings based on the trend analysis results.
[0088] In one embodiment, the method for determining the wear acceleration inflection point includes: Calculate the second difference of the first spur height value corresponding to the initial time period in the spur height time series; If the second-order difference of multiple consecutive first burr height values is greater than the first preset threshold, then one of the multiple consecutive first burr height values will be taken as the wear acceleration inflection point.
[0089] In one embodiment, the wear stage determination module is further configured to perform linear regression fitting on the first burr height value corresponding to the initial time period in the burr height time series to establish an initial linear model; based on the initial linear model, calculate the fitting residuals of multiple consecutive first burr height values to form a fitting residual sequence; when the fitting residual sequence monotonically increases and the fitting residuals of multiple consecutive first burr height values are greater than a second preset threshold, then one of the multiple consecutive first burr height values is taken as the wear acceleration inflection point.
[0090] In one embodiment, the wear stage determination module is further configured to simultaneously input the second burr height value of the accelerated wear stage into a preset exponential growth model and a power function model for parallel fitting; calculate the first goodness of fit of the exponential growth model and the second goodness of fit of the power function model respectively; determine the target nonlinear fitting model based on the magnitude of the first goodness of fit and the second goodness of fit; and perform trend analysis on the burr height time series based on the target nonlinear fitting model.
[0091] In one embodiment, the warning prompt generation module is used to extract the trend feature parameters of the target nonlinear fitting model; predict the remaining life of the mold based on the trend feature parameters and the burr height threshold to obtain the predicted value of the remaining life of the mold; and generate a wear warning prompt based on the predicted value of the remaining life of the mold, the trend feature parameters, the wear acceleration inflection point, and the second burr height value.
[0092] In one embodiment, the warning generation module is used to generate a level one attention warning if a wear acceleration inflection point is detected; generate a level two action warning if the trend characteristic parameter is greater than a third preset threshold or the predicted value of the remaining life of the mold is lower than a fourth preset threshold; and generate a level three emergency warning if the second burr height value exceeds a quality threshold or the predicted value of the remaining life of the mold reaches a fifth preset threshold.
[0093] It should be noted that the mold wear early warning device and the mold wear early warning method provided in the above embodiments belong to the same concept. The specific operation methods of each module and unit have been described in detail in the method embodiments and will not be repeated here. In practical applications, the mold wear early warning device provided in the above embodiments can be assigned to different functional modules as needed, that is, the internal structure of the device can be divided into different functional modules to complete all or part of the functions described above. This is not a limitation here.
[0094] An embodiment of this application also provides a processing device equipped with a mold wear early warning device.
[0095] Embodiments of this application also provide an electronic device, including: one or more processors; and a memory for storing one or more programs, which, when executed by one or more processors, cause the memory to implement the mold wear warning method described in the above embodiments.
[0096] Embodiments of this application also provide a machine-readable medium having instructions stored thereon that, when executed by one or more processors, cause the processors to perform the mold wear warning method described in the above embodiments.
[0097] Figure 7 A schematic diagram of a computer system suitable for implementing the memory of embodiments of this application is shown. It should be noted that... Figure 7 The computer system with the memory shown is merely an example and should not impose any limitation on the functionality and scope of use of the embodiments of this application.
[0098] like Figure 7 As shown, the computer system 700 includes a Central Processing Unit (CPU) 710, which can perform various appropriate actions and processes, such as executing the methods described in the above embodiments, based on a program stored in Read-Only Memory (ROM) 720 or a program loaded from storage into Random Access Memory (RAM) 730. The RAM also stores various programs and data required for system operation. The CPU, ROM, and RAM are interconnected via a bus. An Input / Output (I / O) interface 750 is also connected to the bus 740.
[0099] The following components are connected to the I / O interface: an input section 760 including a keyboard, mouse, etc.; an output section 770 including a cathode ray tube (CRT), liquid crystal display (LCD), and speakers, etc.; a storage section 780 including a hard disk, etc.; and a communication section 790 including a network interface card such as a LAN (Local Area Network) card and a modem, etc. The communication section performs communication processing via a network such as the Internet. A drive 7100 is also connected to the I / O interface as needed. Removable media, such as disks, optical disks, magneto-optical disks, semiconductor memories, etc., are installed on the drive as needed so that computer programs read from them can be installed into the storage section as needed.
[0100] Specifically, according to embodiments of this application, the processes described above with reference to the flowcharts can be implemented as computer software programs. For example, embodiments of this application include a computer program product comprising a computer program carried on a computer-readable medium, the computer program including methods for performing processes. Figure 1 The computer program for the mold wear early warning method is shown. In such an embodiment, the computer program can be downloaded and installed from a network via a communication component, and / or installed from a removable medium 7110. When the computer program is executed by the central processing unit (CPU), it performs various functions defined in the system of this application.
[0101] It should be noted that the computer-readable medium shown in the embodiments of this application can be a computer-readable signal medium or a computer-readable storage medium, or any combination of the two. A computer-readable storage medium can be, for example, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination thereof. More specific examples of a computer-readable storage medium may include, but are not limited to: an electrical connection having one or more wires, a portable computer disk, a hard disk, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM), flash memory, optical fiber, portable compact disc read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination thereof. In this application, a computer-readable signal medium may include a data signal propagated in baseband or as part of a carrier wave, carrying a computer-readable computer program. Such propagated data signals can take various forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination thereof. Computer-readable signal media can also be any computer-readable medium other than computer-readable storage media, which can send, propagate, or transmit a program for use by or in connection with an instruction execution system, apparatus, or device. The computer program contained on the computer-readable medium can be transmitted using any suitable medium, including but not limited to wireless, wired, etc., or any suitable combination thereof.
[0102] The flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of systems, methods, and computer program products according to various embodiments of this application. Each block in a flowchart or block diagram may represent a module, segment, or portion of code, which contains one or more executable instructions for implementing a specified logical function. It should also be noted that 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 consecutively indicated blocks may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each block in a block diagram or flowchart, and combinations of blocks in a block diagram or flowchart, may be implemented using a dedicated hardware-based system that performs the specified function or operation, or using a combination of dedicated hardware and computer instructions.
[0103] The units described in the embodiments of this application can be implemented in software or hardware, and the described units can also be located in a processor. The names of these units do not necessarily limit the specific unit itself.
[0104] Another aspect of this application provides a computer-readable storage medium storing a computer program that, when executed by a computer's processor, causes the computer to perform the aforementioned mold wear warning method. This computer-readable storage medium may be included in the memory described in the above embodiments, or it may exist independently without being assembled into that memory.
[0105] Another aspect of this application provides a computer program product or computer program including computer instructions stored in a computer-readable storage medium. A processor of a computer device reads the computer instructions from the computer-readable storage medium and executes the computer instructions, causing the computer device to perform the mold wear early warning method provided in the various embodiments described above.
[0106] The above embodiments are merely illustrative of the principles and effects of this application and are not intended to limit this application. Any person skilled in the art can modify or alter the above embodiments without departing from the spirit and scope of this application. Therefore, all equivalent modifications or alterations made by those skilled in the art without departing from the spirit and technical concept disclosed in this application should still be covered by the claims of this application.
Claims
1. A mold wear early warning method characterized by, include: The image of the product produced by the same mold in each production cycle is obtained, and edge detection is performed on the image to obtain the actual outline of the product. The first burr height value is obtained based on the difference between the theoretical contour line of the product and the actual contour line of the product. The local rate of change of burr height over time is monitored in real time to identify the wear acceleration inflection point of the wear state transition, and the accelerated wear stage of the mold is determined based on the wear acceleration inflection point; wherein, the burr height over time includes the first burr height value stored in chronological order. A nonlinear fitting model was used to perform trend analysis on the second burr height value in the accelerated wear stage, and the trend analysis results were obtained. Wear and tear warnings are generated based on the trend analysis results.
2. The mold wear early warning method of claim 1, wherein, Methods for determining the wear acceleration inflection point include: Calculate the second difference of the first spur height value corresponding to the initial time period in the spur height time series; If the second-order difference of multiple consecutive first burr height values is greater than the first preset threshold, then one of the multiple consecutive first burr height values will be taken as the wear acceleration inflection point.
3. The mold wear early warning method of claim 1, wherein, Methods for determining the wear acceleration inflection point include: Linear regression fitting was performed on the first burr height value corresponding to the initial time period in the burr height time series to establish an initial linear model; Based on the initial linear model, the fitting residuals of multiple consecutive first burr height values are calculated to form a fitting residual sequence. When the fitted residual sequence is monotonically increasing and the fitted residual of multiple consecutive first burr height values is greater than the second preset threshold, one of the multiple consecutive first burr height values is taken as the wear acceleration inflection point.
4. The mold wear early warning method of claim 1, wherein A nonlinear fitting model was used to analyze the trend of the second burr height value in the accelerated wear stage, including: The second burr height value of the accelerated wear stage is simultaneously input into a preset exponential growth model and a power function model for parallel fitting; Calculate the first goodness of fit of the exponential growth model and the second goodness of fit of the power function model, respectively; Based on the magnitudes of the first and second goodness-of-fit scores, the target nonlinear fitting model is determined. Trend analysis is performed on the time series of burr height based on the target nonlinear fitting model.
5. The mold wear early warning method of claim 4, wherein, Based on the trend analysis results, a wear warning alert is generated, including: Based on the trend characteristic parameters of the target nonlinear fitting model and the burr height threshold, the remaining life of the mold is predicted to obtain the predicted value of the remaining life of the mold. A wear warning is generated based on the predicted remaining life of the mold, the trend characteristic parameters, the wear acceleration inflection point, and the second burr height value.
6. The mold wear early warning method of claim 5, wherein, Based on the predicted remaining life of the mold, the trend characteristic parameters, the wear acceleration inflection point, and the second burr height value, a wear warning is generated, including: If the wear acceleration inflection point is detected, a Level 1 warning alert will be generated; If the trend characteristic parameter is greater than the third preset threshold, or the predicted value of the remaining life of the mold is lower than the fourth preset threshold, a level 2 action warning prompt will be generated. If the second burr height value exceeds the quality threshold, or the predicted remaining life of the mold reaches the fifth preset threshold, a level three emergency warning will be generated.
7. A die wear early warning device characterized by, include: The detection module is used to acquire images of products produced by the same mold in each production cycle, and to perform edge detection on the images to obtain the actual outline of the product. An extraction module is used to obtain a first burr height value based on the difference between the theoretical contour line of the product and the actual contour line of the product. The wear stage determination module is used to monitor the local change rate of the burr height time series in real time to identify the wear acceleration inflection point of the wear state transition, and determine the accelerated wear stage of the mold based on the wear acceleration inflection point; wherein, the burr height time series includes the first burr height value stored in chronological order. The trend analysis module is used to perform trend analysis on the second burr height value in the accelerated wear stage using a nonlinear fitting model, and obtain the trend analysis results. The early warning generation module is used to generate wear warnings based on the trend analysis results.
8. A processing apparatus characterized by comprising: The mold wear early warning device as described in claim 7 is provided.
9. An electronic device, comprising: include: One or more processors; and A memory for storing one or more programs, which, when executed by one or more processors, cause the processors to implement the method as described in any one of claims 1 to 6.
10. A machine-readable medium, characterized in that, It stores instructions that, when executed by one or more processors, cause the processors to perform the method as described in any one of claims 1 to 6.