An information execution management system for mold manufacturing

Abstract

The present invention relates to the technical field of information execution management, and specifically, to an information execution management system for mold manufacturing. It includes a blockchain storage unit, a visual recognition and extraction unit, a deviation evaluation unit, and a condition simulation unit. On the one hand, the present invention can accurately reflect the geometric shape of the mold, reduce manual intervention, and improve the recognition accuracy and efficiency. On the other hand, when the deviation value does not match the standard deviation threshold, the condition simulation unit draws a three-dimensional structure diagram of the mold for simulation, which is beneficial to simulating the working state of the mold in a virtual environment, discovering potential problems in advance, and avoiding mold failure caused by design defects. At the same time, if the simulation result is a normal use result, the standard deviation threshold of the blockchain storage unit is updated, which is beneficial to reducing the scrap rate and improving production efficiency, forming a complete data closed-loop, from data collection to simulation result feedback, realizing the continuous optimization and iteration of data.

Description

Technical Field

[0001] The present invention relates to the technical field of information execution management, and more particularly to an information execution management system for mold manufacturing. Background Art

[0002] In the modern mold manufacturing process, staff first construct a three-dimensional model of the mold through computer-aided design (CAD) software, which determines the geometric shape, dimensional accuracy of the mold, and the selection of subsequent processing technologies. In the three-dimensional model, according to product requirements and production processes, key parameters of the mold (such as inclined plane angle, thickness, hole diameter, etc.) are determined, and digital manufacturing technologies (such as CNC machining or 3D printing) are used to produce the mold. However, after manufacturing, due to the influence of various factors (material shrinkage, machining tolerance, assembly error, environmental factors), the manufactured mold may still have slight deviations. However, key parameters such as the inclined plane and thickness of the mold usually rely on manual measurement or simple automated detection means, resulting in the following problems:

[0003] Insufficient accuracy: It is difficult for the naked eye observation and simple measuring tools to accurately identify slight deviations;

[0004] Lack of error threshold management: Although some minor errors are within the allowable range and do not significantly affect the function of the mold and product quality, certain errors may cause the performance of the mold to decline under specific working conditions, but it can still be used by adjusting process parameters. Deviations beyond the error threshold will directly cause the mold to malfunction or even damage production equipment. However, the error threshold range for the same type of mold is not clear, resulting in molds with slight deviations being misjudged as unqualified, causing the molds to be directly discarded and resulting in waste of resources. In view of this, we propose an information execution management system for mold manufacturing. Summary of the Invention

[0005] The purpose of the present invention is to provide an information execution management system for mold manufacturing to solve the problems raised in the above background art.

[0006] To achieve the above purpose, the present invention provides an information execution management system for mold manufacturing, including a blockchain storage unit, a visual recognition and extraction unit, a deviation evaluation unit, and a condition simulation and simulation unit;

[0007] The blockchain storage unit is used to mine standard parameter values and standard deviation thresholds of various molds from known data, reflect the inclined plane angle, thickness, and hole diameter of the mold through the standard parameter values, and write them into the blockchain for management and storage;

[0008] The visual recognition and extraction unit is used to scan the current mold to obtain the three-dimensional point cloud data on the mold surface, which represents the geometric shape of the mold, identify and extract the key points of the mold, and calculate the current parameter values of the mold through the coordinate relationship between the key points, where:

[0009] The deviation evaluation unit is used to extract the standard parameter values of the blockchain storage unit and compare them with the current parameter values of the visual recognition and extraction unit, and calculate the deviation value. If the deviation value matches the standard deviation threshold, a normal signal is output. If the deviation value does not match the standard deviation threshold, a warning signal is output, and the current parameter values are extracted to the condition simulation unit, so that the condition simulation unit draws the three-dimensional structure diagram of the mold, and uses finite element analysis software in the virtual environment to calculate and predict the stress distribution and deformation of the mold under different working conditions, runs the simulation program, and outputs the comprehensive simulation data results. If it is a normal use result, the standard deviation threshold of the blockchain storage unit is updated. If it is an abnormal use result, the key points corresponding to the deviation value are located, and an adjustment strategy is generated.

[0010] As a further improvement of this technical solution, the blockchain storage unit includes a data mining module and a smart contract module;

[0011] The data mining module is used to collect sufficient historical data, scale the numerical features in the historical data to the same magnitude, and use clustering analysis to find the common features between similar molds to form the standard parameter values of the same type of mold;

[0012] The smart contract module is used to create and write smart contracts to define the rules and processes for parameter storage, enable authorized and authenticated users to upload new parameter data, and each upload requires multiple confirmations, and package the standard parameter data into blocks and add them to the blockchain through the consensus mechanism.

[0013] As a further improvement of this technical solution, the visual recognition and extraction unit includes a three-dimensional scanner, a normalization module, a key point recognition module, and a parameter analysis module;

[0014] The three-dimensional scanner is used to emit laser beams to irradiate the object surface, measure the time difference of the laser reflected back at multiple angles, calculate the three-dimensional coordinates of the points on the object surface, and form three-dimensional point cloud data with multiple three-dimensional coordinates;

[0015] The normalization module is used to preprocess the three-dimensional point cloud data output by the three-dimensional scanner and align the point cloud data from different perspectives to the same coordinate system;

[0016] The key point recognition module is used to extract the features of key points in the same coordinate system by using a deep learning model, and identify the corner points of the mold and the edge points of the mold as key points;

[0017] The parameter analysis module is used to calculate vectors based on the coordinates between key points, and represent the inclined surface angle, thickness, and hole diameter parameters of the mold through the included angles between vectors and the distances between points.

[0018] As a further improvement of this technical solution, the key point recognition module uses a three-dimensional corner detection algorithm to recognize the corner points of the mold, including the following steps:

[0019] The three-dimensional coordinates of each key point are represented as: , where represents the th point, recognize the three-dimensional intensity gradient in the area around the key point, calculate the three-dimensional covariance matrix in the area around the point, and recognize the corner point according to the product of the eigenvalues;

[0020] The key point recognition module uses a three-dimensional edge detection algorithm to recognize the edge points of the mold, including the following steps:

[0021] By calculating the three-dimensional gradient in the area around the point, judge whether it is an edge point according to the magnitude of the gradient. When the magnitude exceeds the threshold, it is determined as an edge point.

[0022] As a further improvement of this technical solution, the deviation evaluation unit includes a synchronization comparison module and a preliminary warning module;

[0023] The synchronization comparison module is used to read the point cloud data of the standard parameter values and the point cloud data of the current parameter values in the blockchain, align them to the same coordinate system, use the feature matching algorithm to identify and match the key points to ensure that the coordinates at the same position are correctly corresponding, and according to the inclined surface angle, thickness, and hole diameter parameters output from the parameter analysis module for the matching key points of the standard mold and the current mold in sequence, use the numerical comparison algorithm to calculate the deviation value of the same category of data;

[0024] The preliminary warning module is used to retrieve the standard deviation threshold from the blockchain storage unit, and again use the numerical comparison algorithm to calculate whether the deviation value matches the standard deviation threshold. If not, send an initial warning signal.

[0025] As a further improvement of this technical solution, the synchronization comparison module further includes a multi-data evaluation algorithm. The multi-data evaluation algorithm is used to calculate the Euclidean distance for each pair of matching points and statistically analyze the distance distribution of all matching points;

[0026] Output , where is the Euclidean distance of the th pair of matching points, is the set distance threshold, is the indicator function, is the total number of matching points;

[0027] If the matching rate exceeds the preset threshold, it is considered that the deviation is within the acceptable range, and a "normal signal" is output;

[0028] Otherwise, calculate , and output a warning signal to indicate the matching point with the largest specific deviation value.

[0029] As a further improvement of this technical solution, the condition simulation unit includes a parameter mapping module, a simulation operation module, and a parameter acquisition module;

[0030] The parameter mapping module is used to receive the point cloud data output by the 3D scanner, convert it into a format supported by the FEA software, map the geometric parameters to the physical properties in the FEA model, and generate an accurate 3D model based on the point cloud data;

[0031] The simulation operation module is used to discretize the 3D model into finite element meshes by applying an automatic mesh generation tool, and perform simulations under different loading conditions according to the actual application scenario, input the model, material properties, and loading conditions, and run the simulation program;

[0032] The parameter acquisition module is used to pre-define a set of theoretical standards, write a script to automatically compare the simulation results with the theoretical standards, determine whether the mold meets the usage requirements. If the simulation results show that the mold works normally, it feeds back to the smart contract module to update the standard deviation threshold in the blockchain. If the simulation results show that the mold works abnormally, it locates the key points corresponding to the deviation values and generates an adjustment strategy.

[0033] As a further improvement of this technical solution, the condition simulation unit further includes a deviation location module. The deviation location module is used to sort the point cloud data in the multi-data evaluation algorithm from largest to smallest based on the deviation range of the average deviation value, output the position information of the key points, and highlight the deviation area by the depth of color, and develop an interactive interface to allow the staff to click on the deviation points to view the detailed information.

[0034] Compared with the prior art, the beneficial effects of the present invention:

[0035] In the information execution management system for mold manufacturing, the blockchain storage unit first obtains known standard parameter values. After the mold is manufactured, the visual recognition extraction unit scans the current mold to obtain the three-dimensional point cloud data on the mold surface, representing the geometric shape of the mold. And through the coordinate relationship between key points, the current parameter values of the mold are calculated. On the one hand, it can accurately reflect the geometric shape of the mold, providing a reliable data basis for subsequent parameter calculation and deviation evaluation, reducing manual intervention, and improving the recognition accuracy and efficiency. On the other hand, the deviation evaluation unit calculates the deviation value. When the deviation value does not match the standard deviation threshold, the current parameter values are extracted to the condition simulation unit, enabling the condition simulation unit to draw the three-dimensional structure diagram of the mold for simulation. This is beneficial for simulating the working state of the mold in a virtual environment, discovering potential problems in advance, and avoiding mold failure caused by design defects. At the same time, if the simulation result is a normal usage result, the standard deviation threshold of the blockchain storage unit is updated, which is beneficial for reducing the scrap rate and improving production efficiency, forming a complete data closed-loop, from data collection to simulation result feedback, realizing the continuous optimization and iteration of data. BRIEF DESCRIPTION OF THE DRAWINGS

[0036] Figure 1 It is a block diagram of the overall structural principle of the present invention.

[0037] The meanings of the reference numerals in the figure are as follows:

[0038] 100, blockchain storage unit; 110, data mining module; 120, intelligent contract module;

[0039] 200, visual recognition extraction unit; 210, three-dimensional scanner; 220, normalization module; 230, key point recognition module; 240, parameter analysis module;

[0040] 300, deviation evaluation unit; 310, synchronous comparison module; 320, preliminary warning module;

[0041] 400, condition simulation unit; 410, parameter mapping module; 420, simulation operation module; 430, parameter acquisition module. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0042] The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention. Embodiment

[0043] Please refer to Figure 1As shown in the figure, this embodiment provides an information execution management system for mold manufacturing, including a blockchain storage unit 100, a visual recognition extraction unit 200, a deviation evaluation unit 300, and a condition simulation unit 400;

[0044] In traditional mold manufacturing, parameter data is usually stored in a centralized database, which has the risks of data tampering, loss, or leakage. The blockchain storage unit 100 is used to mine the standard parameter values and standard deviation thresholds of various molds from known data, reflect the mold inclined surface angle, thickness, and aperture through the standard parameter values, and write them into the blockchain for management and storage. Through blockchain technology, the historical records and change processes of all mold parameters can be traced and queried, increasing the transparency of the manufacturing process. Enterprises upstream and downstream of the supply chain can view mold parameters in real time, promoting collaboration and ensuring the security and credibility of data;

[0045] Therefore, the blockchain storage unit 100 includes a data mining module 110 and a smart contract module 120;

[0046] The data mining module 110 is used to collect sufficient historical data. The historical data includes records from previous manufacturing processes, quality inspection reports, or existing mold products that are already in normal use. By scanning a large number of mold products through the visual recognition extraction unit 200, three-dimensional point cloud data on the surface is obtained, representing the geometric shape of the mold, and key points are identified. The key points include the vertices, edge intersections, hole positions, etc. of the mold. Through the coordinate relationships between the key points, parameters such as the mold inclined surface angle, thickness, and aperture are calculated as standard parameter values to ensure that the data covers different types of molds and their key parameters (such as inclined surface angle, thickness, aperture, etc.). The collected data is cleaned to remove noise and outliers, ensuring the quality of the data, and the numerical features in the historical data are scaled to the same magnitude. Cluster analysis is used to find the common characteristics between similar molds, forming the standard parameter values of the same type of mold;

[0047] The smart contract module 120 is used to create and write smart contracts to define the rules and processes for parameter storage, enabling authorized authenticated users to upload new parameter data, and each upload requires multiple confirmations. The standard parameter data is packaged into blocks and added to the blockchain through a consensus mechanism (such as PoW, PoS, PBFT, etc.). Once the parameter data is successfully written into the blockchain, it needs to be effectively managed and queried. All parameter changes will be recorded, forming a clear timeline, which helps to trace the reasons and time points for any parameter changes. And once the parameter data is written into the blockchain, it cannot be tampered with, effectively preventing parameter errors caused by malicious modification or misoperation;

[0048] When these parameters need to be used, they can be queried through a blockchain browser or other dedicated tools. The query results can be directly used in subsequent simulation analysis, manufacturing process control, etc., enabling all participants to view the historical records of the parameters, increasing the transparency of the entire manufacturing process, and facilitating collaboration between upstream and downstream enterprises in the supply chain. For example, a certain automotive parts manufacturer has achieved transparent management and full traceability of mold design parameters by introducing blockchain technology, which can not only improve product quality but also optimize supply chain management and reduce operating costs.

[0049] Traditional mold inspection usually relies on manual measurement or simple two-dimensional image recognition, with low accuracy and slow efficiency. The visual recognition and extraction unit 200 is used to scan the current mold to obtain the three-dimensional point cloud data of the mold surface, which represents the geometric shape of the mold, identify and extract the key points of the mold, and calculate the current parameter values of the mold through the coordinate relationship between the key points. It can accurately reflect the geometric shape of the mold, provide a reliable data basis for subsequent parameter calculation and deviation evaluation, reduce manual intervention, and improve the recognition accuracy and efficiency.

[0050] Specifically, the visual recognition and extraction unit 200 includes a three-dimensional scanner 210, a normalization module 220, a key point recognition module 230, and a parameter analysis module 240.

[0051] The three-dimensional scanner 210 is used to emit laser beams to irradiate the surface of the object, measure the time difference of the laser reflected back at multiple angles, calculate the three-dimensional coordinates of the points on the object surface, and form three-dimensional point cloud data with multiple three-dimensional coordinates. In specific implementation, the mold is fixed in a stable position to ensure that the mold does not move during the scanning process. According to the size and surface characteristics of the mold, parameters such as the power, scanning range, and resolution of the laser beam are adjusted, and the mold is scanned multiple times from different angles to cover the entire mold surface. Each scan generates a set of point cloud data, that is, a large number of three-dimensional coordinate points. Filtering algorithms (such as Gaussian filtering, median filtering) are used to remove the noise points in the point cloud, interpolation algorithms (such as K-nearest neighbor interpolation, Laplace interpolation) are used to fill the holes in the point cloud, and smoothing algorithms (such as moving average method, bilateral filtering) are used to smooth the point cloud data to improve the data quality.

[0052] The normalization module 220 is used to preprocess the three-dimensional point cloud data output by the three-dimensional scanner 210, aggregate the point cloud data from different perspectives and align them to the same coordinate system to ensure the consistency of the data and avoid errors caused by inconsistent coordinate systems.

[0053] The key point recognition module 230 is used to extract the features of the key points in the same coordinate system using a deep learning model, and identify the corner points and edge points of the mold as key points.

[0054] The parameter analysis module 240 is used to calculate vectors based on the coordinates between key points, and represent the inclined surface angle, thickness, and aperture parameters of the mold through the angles between vectors and the distances between points. For example, the inclined surface angle is calculated through the angle between two vectors, the thickness is calculated through the distance between two parallel planes, and the aperture is calculated through the distance between the edge points of the hole, improving the accuracy and efficiency of mold parameter calculation and realizing the intelligence and automation of mold manufacturing.

[0055] Furthermore, the key point recognition module 230 uses a three-dimensional corner detection algorithm to recognize the corner points of the mold, including the following steps:

[0056] The three-dimensional coordinates of each key point are expressed as: , where represents the th point, recognize the three-dimensional intensity gradient in the area around the key point, calculate the three-dimensional covariance matrix in the area around the point, and recognize the corner point according to the product of the eigenvalues, realizing the recognition of the corner point by calculating the covariance matrix of the three-dimensional intensity gradient , eigenvalue , , satisfies , and the criterion for the corner point is: , which can reflect the change situation of the local area in different directions. The larger the product of the eigenvalues, the more significant the intensity change in this area in multiple directions, that is, the corner point;

[0057] The key point recognition module 230 uses a three-dimensional edge detection algorithm to recognize the edge points of the mold, including the following steps:

[0058] By calculating the three-dimensional gradient in the area around the point, judge whether it is an edge point according to the modulus length of the gradient. When the modulus length exceeds the threshold, it is determined as an edge point, realizing that by calculating the modulus length of the three-dimensional gradient, it can reflect the edge strength of the local area. When the gradient modulus length exceeds the preset threshold, it means that there is a significant intensity change in this area in a certain direction, that is, the edge point;

[0059] Therefore, the key point recognition module 230 reduces human error and improves the accuracy of parameter calculation by automatically recognizing the corner points and edge points of the mold. The automatic recognition process greatly shortens the data processing time and improves the efficiency of mold manufacturing. The key point recognition provides a reliable data basis for subsequent virtual simulation, helping to simulate the working state of the mold in a virtual environment and discover potential problems in advance.

[0060] Among them: The deviation evaluation unit 300 is used to extract the standard parameter values of the blockchain storage unit 100 and compare them with the current parameter values of the visual recognition extraction unit 200, and calculate the deviation value. If the deviation value matches the standard deviation threshold (the standard deviation threshold refers to the deviation range that does not affect the normal use of the mold, and the deviation value matching the standard deviation threshold means that although the current mold manufacturing has a deviation from the standard parameter value, this deviation value is within the standard deviation threshold range, which means it does not affect the normal use of the mold and can meet the use requirements), it outputs a normal signal. If the deviation value does not match the standard deviation threshold, it outputs a warning signal, which is conducive to timely detecting deviations in the production process, optimizing the production process, improving product quality and production efficiency, and extracting the current parameter values to the condition simulation unit 400, enabling the condition simulation unit 400 to draw a three-dimensional structure diagram of the mold, and using finite element analysis (FEA) software in a virtual environment to calculate and predict the stress distribution and deformation of the mold under different working conditions (different loading conditions are set according to the actual application scenarios, such as pressure, temperature, speed, etc.), running the simulation program, obtaining key parameters such as stress, strain, and displacement of the mold under each working condition, comparing the simulation results with the theoretical standard, and judging whether the mold can meet the use requirements, which is conducive to simulating the working state of the mold in a virtual environment, discovering potential problems in advance, outputting comprehensive simulation data results, and judging whether the mold can meet the use requirements through the comparison of the simulation results with the theoretical standard, avoiding mold failure caused by design defects. If it is a normal use result, the standard deviation threshold of the blockchain storage unit 100 is updated, which is conducive to reducing the scrap rate and improving production efficiency, and forming a complete data closed-loop, realizing continuous optimization and iteration of data from data collection to simulation result feedback. If it is an abnormal use result, the key points corresponding to the deviation value are located, and an adjustment strategy is generated. Through deviation evaluation, problems can be quickly located, the production process can be optimized, the scrap rate can be reduced, product quality and production efficiency can be improved, and the intelligent and efficient mold manufacturing process can be realized;

[0061] On the above basis, the deviation evaluation unit 300 includes a synchronous comparison module 310 and a preliminary warning module 320;

[0062] The synchronization and comparison module 310 is used to read the point cloud data of the standard parameter values and the point cloud data of the current parameter values in the blockchain, align them in the same coordinate system, identify and match key points using feature matching algorithms (such as SIFT, SURF, or deep learning feature descriptors) to ensure that the coordinates at the same position are correctly corresponding. According to the matching key points of the standard mold and the current mold, the inclination angle, thickness, and aperture parameters output from the parameter analysis module 240 in sequence are used to calculate the deviation values of the same category of data using a numerical comparison algorithm. For example, according to two matching key points, the standard inclination angle and the current inclination angle are respectively output in the standard mold and the current mold, and the difference between the standard inclination angle and the current inclination angle is used as the deviation value. The same applies to the thickness and aperture;

[0063] The preliminary warning module 320 is used to retrieve the standard deviation threshold from the blockchain storage unit 100, and again uses a numerical comparison algorithm to calculate whether the deviation value matches the standard deviation threshold. If they match, it means that the current mold does not affect normal use. If they do not match, it means that based on past experience, this mold cannot guarantee accurate use, may not be usable, or may not have been verified. At this time, an initial warning signal is sent to alert the staff, avoiding large errors caused by manual verification and improving accuracy.

[0064] To improve the accuracy of deviation value calculation, the synchronization and comparison module 310 also includes a multi-data evaluation algorithm. The multi-data evaluation algorithm is used to calculate the Euclidean distance for each pair of matching points and statistically analyze the distance distribution of all matching points;

[0065] Output , where, is the Euclidean distance of the pair of matching points, is the set distance threshold, is the total number of matching points;

[0066] If the matching rate exceeds a preset threshold (such as 95%), it is considered that the deviation is within an acceptable range, and a "normal signal" is output;

[0067] Otherwise, calculate , and output a warning signal to indicate the matching point with the largest specific deviation value.

[0068] Immediately afterwards, the condition simulation and simulation unit 400 includes a parameter mapping module 410, a simulation operation module 420, and a parameter acquisition module 430;

[0069] The parameter mapping module 410 is used to receive the point cloud data output by the 3D scanner 210 and convert it into a format supported by the FEA software (such as STL, STEP, IGES, etc.). Format conversion is performed using Python libraries such as `pythreejs` or `trimesh` to map geometric parameters (such as inclined plane angle, thickness, aperture, etc.) to physical properties (such as material properties, boundary conditions, etc.) in the FEA model. A precise 3D model is generated based on the point cloud data using professional CAD software (such as SolidWorks, CATIA) or open-source tools (such as FreeCAD).

[0070] The simulation running module 420 is used to discretize the 3D model into finite element meshes using an automatic mesh generation tool (such as Gmsh, TetGen) to ensure that the mesh quality meets the simulation accuracy requirements, and perform simulations according to different loading conditions (such as pressure, temperature, speed, etc.) in the actual application scenario. For example, in the mold of automotive parts, considering the pressure change during stamping forming, commercial FEA software (such as ANSYS, ABAQUS) or open-source tools (such as CalculiX, Elmer) are used for simulation. The model, material properties, and loading conditions are input, and the simulation program is run.

[0071] The parameter acquisition module 430 is used to pre-define a set of theoretical standards, including the maximum allowable stress, maximum deformation, etc. Key parameters such as stress, strain, and displacement are extracted from the simulation results. A script is written to automatically compare the simulation results with the theoretical standards to determine whether the mold meets the usage requirements. If the simulation results show that the mold is working properly, it is fed back to the smart contract module 120 to update the standard deviation threshold in the blockchain, realizing the update of the historical record according to the results of each simulation, forming a data closed-loop, and reducing the scrap rate in future production. If the simulation results show that the mold is working abnormally, the key points corresponding to the deviation values are located, and an adjustment strategy is generated to optimize the production process.

[0072] Specifically, the condition simulation and simulation unit 400 further includes a deviation positioning module. The deviation positioning module is used to sort the point cloud data in the multi-data evaluation algorithm from largest to smallest based on the deviation range of the average deviation value, output the position information of the key points (such as coordinates, node numbers, etc.), and highlight the deviation area by the depth of color. Text descriptions can also be added to the model, such as "The inclined plane angle deviation here is X%". An interactive interface is developed to allow staff to click on the deviation points to view detailed information (such as deviation values, simulation parameters), and operations such as zooming, rotating, and section cutting are supported to facilitate staff to observe the model from different angles. At the same time, optimization suggestions are generated based on the position of the deviation points and the magnitude of the deviation values. For example, "It is recommended to increase the material thickness in this area to improve the strength", which is beneficial to process adjustment according to the optimization suggestions provided by the system, thereby improving the precision and efficiency of mold manufacturing.

[0073] In summary, considering that the allowable error threshold range is not clear during current mold manufacturing, molds with slight deviations may be misjudged as unqualified, resulting in the molds being directly discarded and wasting resources. Therefore, the blockchain storage unit 100 first obtains known standard parameter values. After the mold is manufactured, the visual recognition extraction unit 200 scans the current mold to obtain the three-dimensional point cloud data on the mold surface, representing the geometric shape of the mold, and calculates the current parameter values of the mold through the coordinate relationship between key points. On the one hand, it can accurately reflect the geometric shape of the mold, provide a reliable data basis for subsequent parameter calculation and deviation evaluation, reduce manual intervention, and improve the recognition accuracy and efficiency. On the other hand, the deviation evaluation unit 300 calculates the deviation value. When the deviation value does not match the standard deviation threshold, the current parameter values are extracted to the condition simulation unit 400, enabling the condition simulation unit 400 to draw a three-dimensional structure diagram of the mold for simulation. This is beneficial for simulating the working state of the mold in a virtual environment, discovering potential problems in advance, and avoiding mold failure caused by design defects. At the same time, if the simulation shows normal use results, the standard deviation threshold of the blockchain storage unit 100 is updated, which is beneficial for reducing the rejection rate and improving production efficiency, forming a complete data closed-loop, from data acquisition to simulation result feedback, and realizing the continuous optimization and iteration of data.

[0074] The above shows and describes the basic principles, main features and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited by the above embodiments. The above embodiments and the descriptions in the specification are only preferred examples of the present invention and are not used to limit the present invention. Without departing from the spirit and scope of the present invention, the present invention will have various changes and improvements, and these changes and improvements all fall within the scope of the present invention claimed. The scope of protection claimed by the present invention is defined by the appended claims and their equivalents.

Claims

1. An information execution management system for mold manufacturing, characterized in that: It includes a blockchain storage unit (100), a visual recognition extraction unit (200), a deviation evaluation unit (300), and a condition simulation unit (400); The blockchain storage unit (100) is used to mine the standard parameter values and standard deviation thresholds of various molds from known data, reflect the mold inclined surface angle, thickness, and aperture through the standard parameter values, and write them into the blockchain for management and storage; The visual recognition extraction unit (200) is used to scan the current mold to obtain the three-dimensional point cloud data on the mold surface, represent the geometric shape of the mold, identify and extract the key points of the mold, and calculate the current parameter values of the mold through the coordinate relationship between the key points, where: The deviation evaluation unit (300) is used to compare the standard parameter values of the blockchain storage unit (100) with the current parameter values of the visual recognition extraction unit (200), and calculate the deviation value. If the deviation value matches the standard deviation threshold, a normal signal is output. If the deviation value does not match the standard deviation threshold, a warning signal is output, and the current parameter values are extracted to the condition simulation unit (400), so that the condition simulation unit (400) draws the three-dimensional structure diagram of the mold, calculates and predicts the stress distribution and deformation conditions of the mold under different working conditions in a virtual environment, runs the simulation program, and outputs the comprehensive simulation data results. If it is a normal use result, the standard deviation threshold of the blockchain storage unit (100) is updated. If it is an abnormal use result, the key points corresponding to the deviation value are located, and an adjustment strategy is generated.

2. The information execution management system for die manufacturing according to claim 1, characterized in that: The blockchain storage unit (100) includes a data mining module (110) and a smart contract module (120); The data mining module (110) is used to collect sufficient historical data, scale the numerical features in the historical data to the same magnitude, and use clustering analysis to find the common features between similar molds to form the standard parameter values of the same type of molds; The smart contract module (120) is used to create and write smart contracts to define the rules and processes for parameter storage, enable authorized and authenticated users to upload new parameter data, and each upload requires multiple confirmations. The standard parameter data is packaged into blocks and added to the blockchain through a consensus mechanism.

3. The information execution management system for die manufacturing according to claim 2, characterized in that: The visual recognition extraction unit (200) includes a three-dimensional scanner (210), a normalization module (220), a key point recognition module (230), and a parameter analysis module (240); The three-dimensional scanner (210) is used to emit laser beams to irradiate the object surface, measure the time difference of the laser reflected back at multiple angles, calculate the three-dimensional coordinates of the points on the object surface, and form three-dimensional point cloud data with multiple three-dimensional coordinates; The normalization module (220) is used to preprocess the three-dimensional point cloud data output by the three-dimensional scanner (210) and align the point cloud data from different perspectives to the same coordinate system; The key point recognition module (230) is used to extract the features of the key points in the same coordinate system using a deep learning model, and identify the corner points and edge points of the mold as key points; The parameter analysis module (240) is used to calculate vectors based on the coordinates between key points, and represent the inclined plane angle, thickness, and aperture parameters of the mold through the included angles between vectors and the distances between points.

4. The information execution management system for die manufacturing according to claim 3, characterized in that: The key point recognition module (230) uses a three-dimensional corner detection algorithm to identify the corner points of the mold, including the following steps: The three-dimensional coordinates of each key point are represented as: , where represents the th point, identify the three-dimensional intensity gradient of the area around the key point, calculate the three-dimensional covariance matrix of the area around the point, and identify the corner points according to the product of the eigenvalues; The key point recognition module (230) uses a three-dimensional edge detection algorithm to identify the edge points of the mold, including the following steps: By calculating the three-dimensional gradient of the area around the point, it is judged whether it is an edge point according to the modulus length of the gradient. When the modulus length exceeds the threshold, it is determined as an edge point.

5. The information execution management system for die manufacturing according to claim 4, characterized in that: The deviation evaluation unit (300) includes a synchronous comparison module (310) and a preliminary warning module (320); The synchronous comparison module (310) is used to read the point cloud data of the standard parameter values and the point cloud data of the current parameter values in the blockchain, align them in the same coordinate system, use the feature matching algorithm to identify and match the key points to ensure that the coordinates of the same position are correctly corresponding, and calculate the deviation values of the same category of data using the numerical comparison algorithm according to the inclined plane angle, thickness, and aperture parameters output from the parameter analysis module (240) in sequence for the standard mold and the matching key points of the current mold; The preliminary warning module (320) is used to retrieve the standard deviation threshold from the blockchain storage unit (100), and again use the numerical comparison algorithm to calculate whether the deviation value matches the standard deviation threshold. If not, an initial warning signal is sent.

6. The information execution management system for die manufacturing according to claim 5, characterized in that: The synchronous comparison module (310) further includes a multi-data evaluation algorithm, and the multi-data evaluation algorithm is used to calculate the Euclidean distance for each pair of matching points and statistically analyze the distance distribution of all matching points; Output , where is the Euclidean distance of the th matching point, is the set distance threshold, is the indicator function, total number of matching points; If the matching rate exceeds the preset threshold, it is considered that the deviation is within the acceptable range, and a "normal signal" is output; Otherwise, calculate , and output a warning signal to indicate the matching point with the largest specific deviation value.

7. The information execution management system for die manufacturing according to claim 6, characterized in that: The condition simulation and simulation unit (400) includes a parameter mapping module (410), a simulation operation module (420), and a parameter acquisition module (430); The parameter mapping module (410) is used to receive the point cloud data output by the three-dimensional scanner (210), convert it into a format supported by the FEA software, map the geometric parameters to the physical properties in the FEA model, and generate an accurate three-dimensional model according to the point cloud data; The simulation operation module (420) is used to discretize the three-dimensional model into finite element meshes using an automatic mesh generation tool, and perform simulations under different loading conditions according to the actual application scenarios, input the model, material properties, and loading conditions, and run the simulation program; The parameter acquisition module (430) is used to pre-define a set of theoretical standards, write scripts to automatically compare the simulation results with the theoretical standards, judge whether the mold meets the usage requirements. If the simulation results show that the mold works normally, it is fed back to the smart contract module (120) to update the standard deviation threshold in the blockchain. If the simulation results show that the mold works abnormally, the key points corresponding to the deviation values are located, and an adjustment strategy is generated.

8. The information execution management system for die manufacturing according to claim 7, characterized in that: The situation simulation unit (400) further includes a deviation positioning module. The deviation positioning module is used to sort the point cloud data in the multi-data evaluation algorithm from largest to smallest based on the deviation range of the average deviation value, output the position information of the key points, and highlight the deviation area by the depth of color, and develop an interactive interface that allows the staff to click on the deviation points to view the detailed information.

Images